Chapter

Orthopaedic Research: What An

Orthopaedic Surgeon Needs to Know
Charles Fisher, MD, MHSc, FRCSC
Marcel Dvorak, MD, FRCSC

Introduction
Although statistics and clinical epidemiology have been
part of the core curriculum for medical schools and
some residency programs, most clinicians have only limited knowledge in these areas. This is not from a lack of
training or interest, but more because statistical and epidemiologic proficiency requires frequent application
and interpretation. Lack of familiarity in these areas
may lead the clinician to feel intimidated or to avoid
statistics and other methodologic aspects of clinical
studies.
Therefore, it is important to be familiar with some of
the essential statistical and epidemiologic concepts and
study design principles necessary to properly conduct
and evaluate clinical research. Although it is impossible
to comprehensively derive or explain these concepts
within the scope of this chapter, hopefully their practicality will stimulate further application within clinical
and academic practice. Surprisingly, many of the errors
or flaws found in clinical research are not complex statistical or methodologic issues but more the neglect of
fundamental principles that are often forgotten.

Its All in the Question

Although biostatistics plays a prominent role, especially
in research, its significance is often overemphasized, particularly relative to issues of study design and methodology. In any research study, the clinician scientist must
first develop a question. On the surface this may seem
simple, but clearly defining the primary study question
is indeed difficult. For example: What are the outcomes
of patients with sciatica treated surgically and nonsurgically? This question may appear adequate, but on further inspection what is meant by outcome? Is the outcome pain, function, or return to work? Studies can
usually only be designed to answer one question precisely and provide probable or possible answers to other
secondary questions. So a better question for this example might be: For patients with sciatica secondary to
lumbar disk herniation treated with either surgery or
nonsurgical treatment, what is the change in neurologic

American Academy of Orthopaedic Surgeons

symptom score from initial assessment to 1 year after

treatment? Once a researcher has a clearly defined
question the research protocol usually is easy to write.
The inclusion criteria, sample size calculation, blinding,
random allocation, follow-up schedule, objective outcome measure, and statistical analysis all become much
easier with a well-defined question.
Once the study question is clearly identified, then
biostatistics becomes fundamental to the understanding
and conduct of medical research. Not only do statistics
enable physicians to compare treatment strategies in
clinical trials, but also in analytical epidemiology the relationship or association between variables can be studied. The classic association of environmental, lifestyle, or
biologic factors to the development of disease has been
analyzed. The relationship of lung cancer and smoking is
probably the most famous example. In clinical research
these same statistical tools can be used to examine the
relationship between various baseline and demographic
variables to an outcome variable of choice. For example,
the association of age, surgery, and disease severity on
the primary outcome of interest (pain) can be determined. Ultimately through this analysis the true effect
or causation is determined; however, before this can be
concluded other possible explanations (bias, confounding, chance, reverse causation) for the associations must
be ruled out, such that causation is a diagnosis of exclusion.
The conclusions that can be reached from a data set
are limited by the question asked and the design of the
study to answer it. Although statistics allow for the summation of data, hypothesis testing, the avoidance or reduction of nonsystematic error, and adjustment for confounding factors, they are not the panacea for a poorly
designed study.

Study Design
In general there are two types of clinical studiesthose
that analyze primary data and those that analyze secondary data. Studies that collect and analyze primary
data include case reports and series, case control, cross

Orthopaedic Knowledge Update 8

Orthopaedic Research: What An Orthopaedic Surgeon Needs to Know

sectional, cohort (both prospective and retrospective)
and randomized controlled trials (RCTs). Analysis of
secondary data occurs in systematic reviews or metaanalysis for the purpose of pooling or synthesizing data
to answer a question that is perhaps not practical or answerable with an individual study. Another way to
broadly characterize studies is as experimental, where
an intervention is introduced to subjects, or observational, in which no active treatment is introduced to
subjects. The methodologic hierarchy or rating of scientific studies is well summarized in the literature.

Case Reports
Case reports are valuable in rare conditions or if they
provide compelling findings that can be hypothesisgenerating for further studies. Case reports are limited
by small sample size, the lack of a control group, and
nonobjective outcome measures. The natural extension
of a case report is a case series, which allows for a more
valid assessment of a clinical course or response to an
intervention. Few conclusions can be made because of
the selection bias, subjective assessment, a small, often
ill-defined number of subjects (n), and lack of a comparison group. Case series can be improved by addressing
some of these limitations such as using objective outcome measures and clearly defining their inclusion criteria, which makes them very similar to cohort studies.

Prevalence Studies
Cross sectional or prevalence studies are common in
public health but are rare in the surgical realm. They
provide a snapshot of the health experience of a population at a specified period of time. These studies can provide a relatively quick assessment of health status or
health needs. They can be hypothesis-generating for illdefined diseases and are a good design for common diseases of long duration such as osteoarthritis. Unfortunately, they cannot establish temporality, so these
studies are prone to reverse causality or protopathic
bias. A cross-sectional analysis can be blended with a
retrospective cohort study and be quite effective. An example would be to retrospectively define a patient cohort such as cervical burst fractures and then do a crosssectional outcome analysis on function or quality of life.
The follow-up will occur at various times relative to
when the patient was injured but can still provide valuable objective long-term outcome information.

Case Control Studies

Case control studies usually involve a cross-sectional
analysis on similar subjects and classically compare certain patient groups (cases) with control patients for the
presence of risk factors. This design is ideal for assessing
etiologic or risk factors for rare diseases and is useful in
studies of prognosis. Although these studies are efficient

from a time and cost perspective, there are limitations.

Finding appropriately matched controls and defining inclusion and exclusion criteria that are similar for both
cases and control patients are steps taken to control for
confounding variables. Because both the exposure and
disease have already occurred and there is different recall bias between cases and control patients, proving
causation is difficult. An example of a surgical case control study would be comparing patients who have had
anterior cruciate ligament reconstruction and developed
the complication of a stiff knee (cases), to patients without postoperative knee stiffness (controls). The results
may allow the identification of risk factors for knee
stiffness. Ideally the control:case ratio should be 1:1 up
to a maximum of 4:1.

Cohort Studies
Cohort studies can be retrospective or prospective, with
prospective studies providing better scientific evidence.
Cohort studies are similar to case series, but more
tightly controlled. They require a time zero, strict
inclusion/exclusion criteria, standardized follow-up at
regular time intervals, and efforts to optimize follow-up
and reduce dropouts. For these reasons prospective cohort studies are expensive and time-consuming. Cohort
designs are ideal for identifying risk factors for disease,
determining the outcome of an intervention, and examining the natural history of a disease. The Framingham
cohort study examining cardiovascular disease is one of
the more famous cohort studies. Prospective cohorts can
be compared with historical controls but problems with
data quality, selection bias, outcome parameters, and
temporal trends make this less desirable than a nonrandomized prospective outcomes study.
Retrospective studies have the advantage of being
less expensive and time consuming. The records (usually
charts) are made out without knowledge of exposure or
disease and therefore recall bias is not an issue. However, because the records used for data are collected for
other reasons and in a nonstandardized manner, critical
information such as confounders is almost always missing. The incorporation of a cross-sectional outcome
analysis to a retrospective cohort study provides a standardized outcome, but many subjects may be deceased
or lost to follow-up, leading to poor response rates.

Randomized Controlled Trials

RCTs are justifiably recognized as the gold standard in
obtaining clinical evidence; however, they have well recognized disadvantages including high costs, administrative complexity, prolonged time to completion, and difficulty ensuring methodologic vision. Furthermore, RCTs
in surgery are complicated by difficulties in blinding,
randomization, technique standardization, and generalizability. Nevertheless, the ability to control for known

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 1 Orthopaedic Research: What An Orthopaedic Surgeon Needs to Know

and unknown bias outweighs these disadvantages. Randomization is unrivaled in ensuring the balancing of the
experimental and control groups for unknown confounders. Known confounders are also well balanced if
the group sizes are large enough; however, if the sample
sizes are small the balancing of known confounders can
easily be performed through stratification. Blinding is
designed to induce comparability in the handling and
evaluation of the participants, it preserves the integrity
of the randomization, and allows for objective collection
and analysis of data. Surgical RCTs are difficult to perform; hence, there is a paucity of information about
them in the orthopaedic literature. Practical and ethical
issues limit their use in surgery, but above all they are
extremely difficult, time consuming and expensive to
perform. These factors should not deter clinical scientists from pursuing this study design so that needed answers to important questions can be obtained.

Basic Terms and Concepts in Biostatistics

Population and Sample
A population is a complete set of homogenous individuals with a specified set of characteristics. A sample is a
subset of a population. The population and sample represent the starting point for all analysis. Total populations are very difficult if not impossible to study. The
majority of studies are based on subpopulations or samples of the population of interest. It is the parameters
(measurements) of the sample that are used not only to
accurately describe the larger population of interest, but
also to help answer scientific questions about whether
interventions affect these measurements. For instance,
the variable might be age and the parameter average
age. The parameters used describe the location and variability among the members of a population and are discussed in the following paragraphs.

Mean and Median

Systematic Reviews
A systematic review provides a rational synopsis of the
available literature. By summarizing all relevant literature on a particular topic, the systematic review tends to
be a tremendous asset to the busy clinician. A systematic review attempts to overcome the bias that is associated with the majority of traditional reviews or more
appropriately termed narrative reviews. Through the
application of rigorous methodology, potential bias is
minimized. A properly conducted systematic review will
ensure all published and unpublished literature is considered, will evaluate each study for its relevance and
quality through independent assessment, and then synthesize the remaining studies in a fair and unbiased
manner. A good systematic review is transparent. Transparency implies openness by the authors so that the
reader can determine the validity of the conclusions for
themselves. A properly conducted systematic review
should allow a second group of authors using the same
methodology to arrive at the same conclusion(s).
Component studies of a systematic review may be
combined qualitatively, or quantitatively with RCTs.
When a quantitative synthesis is performed it is termed
a meta-analysis. Meta-analysis refers to the statistical
technique used for combining independent studies. A
meta-analysis is particularly useful when combining several small studies whose results may be inconclusive because of low power. Meta-analyses have a greater ability
to detect uncommon but clinically relevant end points
such as mortality.
After determining the question being asked in a
study, and having an overview of the types of studies
used in clinical research, the next step is to look at data
analysis, which validates the answer to the question.

American Academy of Orthopaedic Surgeons

A simple figure that provides a measure of central tendency or an average of variability for symmetric or normative data is the mean, which is defined as the sum of
all the observations in a sample divided by the number
of observations. For nonsymmetric data, a better measure of central tendency or average is the median. The
median is the point that divides the distribution of observations in half, if the observations are arranged in increasing or decreasing order. This is relevant because
statistical procedures vary depending on the distribution
of the population. In other words, certain tests assume a
normal distribution, and if the data distribution is not
normal, alternative statistical tests must be used to ensure accurate results.

Standard Deviation
The standard deviation (SD) is one of several indices of
variability used to characterize the distribution of values
in a sample for symmetric data. Numerically, the SD is
the square root of the variance. The SD is conceptually
easier to use than the variance, which is defined as the
average squared deviation from the mean. Degrees of
freedom are used in the calculation of SD and are often
misunderstood or confusing. They are used in the mathematical formulas that construct tables to determine
levels of significance. Specifically, they represent the
number of samples and sample size, which are factors in
determining significance.
If there were a greater range (maximum to minimum) in the measured variable, then the SD would be
larger. Assuming a normal distribution, about 95% of
the population falls within 2 SDs of the mean. Therefore, the mean and SD provide a concise summary of a
particular variable within a symmetrically distributed
population. If a population does not follow a normal

Orthopaedic Research: What An Orthopaedic Surgeon Needs to Know

distribution then it would be more appropriate to report
the median and percentiles.
Another term of variability worth mentioning that is
often reported in the literature is standard error of the
mean (SEM). SEM estimates the accuracy of the mean
computed from the sample compared with mean of the
actual population from which the sample was taken. It
quantifies the uncertainty in the estimate of the mean;
however, it says nothing about the variability of the
population itself. The SD measures the variability of the
population and is always larger then the SEM, and thus
makes the data less statistically appealing.

Randomization
Randomization is a process that arbitrarily assigns subjects to two or more groups by some chance mechanism,
rather than by choice. It ensures that each subject has a
fair and equal opportunity to be assigned to each group.
Randomization is necessary to avoid systematic error
(bias) that may produce unequal groups with respect to
general characteristics, such as gender, age, ethnicity,
and other key factors that may affect the probable
course of the disease or treatment. Depending on the
distribution of the data and the size of the sample there
is a chance that the sample will not be representative
just by chance alone.

Variables and Types of Variables

Any division of measurement or classification on which
individual observations are made is called a variable. In
general, there are two types of variables, qualitative and
quantitative. A qualitative variable is subdivided into
nominal and ordinal variables. With ordinal variables,
the categories have an obvious rank order, such as the
stages of bowel cancer. Nominal variables allow for only
qualitative classification; for example, gender or occupation. Quantitative variables are either discrete (length of
hospital stay) or continuous (age). Continuous variables
allow not only ranking of the order of observations that
are measured, but also quantification and comparison of
the size of differences between observations.

Hypothesis
A hypothesis is a supposition made as a basis for reasoning, without assumption of its truth, or as a starting
point for further investigation. In statistics there are two
kinds of hypotheses. The null hypothesis (Ho) assumes
no effect or differences, whereas the alternate hypothesis (Ha) postulates there is an effect or difference. Statistics are designed to test for the Ho. When the probability of the observed data patterns cannot support Ho, a
researcher would reject Ho in favor of Ha. This does not
mean that Ho is absolutely incorrect, only that the data
at hand cannot support it.

Orthopaedic Knowledge Update 8

Hypothesis and Significance Testing: P Value

These terms are best explained by an example. A researcher wants to test a new drug for postoperative pain
control. The population will be patients undergoing lumbar fusion. Two samples will be randomly chosen from
the population, with one getting the standard drug (control) and the other getting the new drug (experimental).
The two groups will then rate their pain and the mean
and SD pain score for each group will be calculated. Is
the observed difference in mean pain scores caused by
the drug or by random assignment to the two study
samples? This question is answered by quantifying the
difference by means of a test statistic. The actual specific
statistical test the researcher uses will vary depending
on the question and population, but one number will be
calculated. The greater the difference between the two
means the greater the test statistic number; however,
what is the cutoff value for this number between true
and random difference?
To obtain the distribution of the test statistic, the researcher will assume the drug has no effect (null hypothesis) and in theory run the same experiment a
multitude of times using all possible samples from the
population. Most test statistic numbers will be small because there is no difference; by chance alone some of
these experiments will demonstrate a large test statistic.
Say 5% of them will be above this cutoff point even if
the Ho is correct.
Having established the cutoff for the test statistic, the
experiment is redone with a drug that the researcher hypothesizes will improve pain. If the calculated test statistic is larger than the 5% cutoff point that was determined
above, then the chance of the observed difference in pain
scores used to calculate this big test statistic number being due to chance alone is 5%. Therefore the experimenter would reject the Ho and accept the Ha at a significance
level of 5%. It should be noted that there is still a 5%
chance of a false-positive result. The p value is the probability of reporting a difference when one does not exist.
The p value should not be regarded as rigid and can be
adjusted depending on the question being asked and the
clinical implications of the Ho being falsely rejected. The
researcher should specify the p value before the study is
commenced.

Type I () and Type II () Error

A type I error occurs in a comparative study when a statistically significant finding occurs by chance alone. In
other words, a studys Ho is rejected when it is true and
a false-positive result occurs. The probability of a type I
error is represented by , and is widely accepted at 5%.
Not as well recognized is the type II error or falsenegative finding of a study. In this situation no difference is identified between experimental and control
groups, when in fact there is a difference between the

American Academy of Orthopaedic Surgeons

Chapter 1 Orthopaedic Research: What An Orthopaedic Surgeon Needs to Know

Orthopaedic Knowledge Update 8

Table 1 | Influence of the Magnitude of Sample Size Determinants on Sample Size Required
Determinants

Sample Size

Variability
Effect change

Increase
Decrease
Decrease
Decrease

Decrease
Increase
Increase
Increase

groups. In medicine, a type II error occurs when a study

is underpowered; the study has failed to include
enough patients to detect a predetermined magnitude of
difference between treatments. The chance of making a
type II error, represented by , is commonly accepted to
be 20%. The chance of detecting a true positive result is
the power of the study (1-) and would be 80%. Unfortunately, it is common for RCTs published in peerreviewed journals to neglect the power of the study by
not reporting a sample size calculation a priori. When a
study reveals no difference between groups but does not
report the power, the chance of making a type II error is
essentially unknown. This significantly limits the validity
of a studys findings.

Sample Size or Power Calculation

A power calculation determines the number of patients
or experimental units required for the study to detect a
difference, if one is present. Too large a sample results
in a waste of time and money; conversely, too small a
sample leads to a lack of precision in the results (type II
error). Formulas are available to determine sample size
and vary depending on the statistical analysis being
done. The determinants of sample size are , power
(1-), effect change, and variability (Table 1).
One final consideration is a one- or two-tailed significance test. Convention would suggest that all comparative studies be two-tailed such that A analyzed to be
better than B and vice versa. However, if it is important
to know that A is better than B, a one-tailed test can be
done and sample size reduced because is theoretically
larger. An example where a one-tailed test is useful
would be a study of a new intervention that is more expensive and the only way it will replace the control is if
it is superior to it; equivalence or worse is not of
interest.

Statistical Versus Clinical Significance

As discussed, many factors affect statistical significance,
such as the magnitude of the effect (effect change), the
sample size, the reliability of the effect, and the reliability of the measurement instrument. A clinically significant difference varies depending on factors associated
with the intervention and variable of interest. The ge-

American Academy of Orthopaedic Surgeons

neric definition of clinical significance that permits its

application to numerous studies is a difference large
enough for the clinician to want to choose one treatment over the other, after considering all factors (such
as cost or side effects). Determining a clinically significant difference is often difficult, especially if it involves
an outcome such as health-related quality of life. It is an
important decision because it is an integral component
of the sample size calculation.

Confidence Interval
A confidence interval (CI) gives an estimated range of
values that are likely to include the unknown population parameter being sought. The CI is usually calculated at 95%, but 99% CIs are also used. The interval
describes the confidence with which the true difference
in mean values from each group or intervention is
within the CI. The CI then describes both the size of the
treatment effect and the certainty of the estimation of
the treatment effect. The CI can be used for hypothesis
testing for if the interval contains zero, then the Ho cannot be rejected. If the interval is 95% confidence this
would be analogous to a p value of 0.05. The advantage
of using the CI for hypothesis testing is that it provides
information about the size of the effect. For example,
something might be statistically significant because of a
large sample size, but when one sees the quantitative effect it may be clinically insignificant.

Statistical Analysis
The goal of this section is to create familiarity with
some of the common statistical methods used in clinical
research. Most of these methods are used for hypothesis
testing.

t Test (Students t test) and the Signed-Rank Sum Test

The t test procedure is used to test the Ho where the
mean of a single variable of the population from which
the sample was drawn is equal to a specified value. The t
test is applied if the sample values are independent and
normally distributed. If not, a nonparametric method is
used, commonly the signed rank sum test. Researchers
want to test whether the average score of the Short
Form-36 (SF-36) at 1 year postoperative for 10 patients
(mean = 39.8, SD = 10.3) differs from published normative population data (mean = 52.0; SD = 8.0) (Ha). Or, if
it is likely that the mean score at 1-year follow-up is
equal to the normative data (Ho). From a computer statistical package or tables, the exact p value based on a
two-sided test is 0.005. This is a significant difference at
the 5% level and suggests that the decrease of the SF-36
score in patients at 1-year follow-up is unlikely due to
chance and Ho is rejected.

Orthopaedic Research: What An Orthopaedic Surgeon Needs to Know

The Paired t Test and Wilcoxon Two-Sample Test

The paired t test is generally used when measurements
are taken from the same subject or related subject over
time or in different circumstances. This implies that the
data are from within the same sample and that measurements are related to each other. For example, a paired t
test can be used to determine the difference in blood
pressure before and after treatment. The paired t test assesses the average change that an intervention produces.
For the paired t test, the observed data are from the
same subject or from a matched subject and are drawn
from a population with a normal distribution. If these
assumptions are not met, then one of the nonparametric
alternative tests should be applied, for example, the Wilcoxon two-sample test.

Comparisons of Two Means (t Test and the

Mann-Whitney U Test)
The t test and the Mann-Whitney U test are appropriate
where the data have been obtained from two independent samples. The two-sample t test determines whether
two samples are likely to be from the same population.
In addition, the two different samples are assumed to
come from populations with the same variance. The
t test would be commonly used in RCTs comparing the
mean score from each of two groups. If the assumption
of normality or equality of variance is not met, a
distribution-free test that is an alternative to the independent sample t test is used. This is called the MannWhitney U test and uses ranked scores.
One of the common problems that arise with t tests
is that of multiple comparisons. The issue of multiple
comparisons actually goes beyond t tests into issues of
subgroup analysis and hypothesis testing versus hypothesis generating analysis. From the explanation of the
p value earlier in the chapter, the chance of obtaining a
false-positive result is 5%. If a researcher has three
groups to compare and does a t test for A versus B,
B versus C, and A versus C, then the chance for a type I
error is greater. For three tests it is 5% + 5% + 5% so it
becomes 15%; if it were four groups it would be 30%. In
other words the chances of the researcher obtaining a
highly coveted significant difference are increased, and
the problem is that it may not be a true difference. To
avoid this problem there are two approaches. One is to
use multiple comparison procedures such as the Bonferroni or Holm t test that use appropriate correction factors to address the changing p value. The simplest and
initial approach, however, should be to use the analysis
of variance (ANOVA) when the experimental design involves multiple groups.

Analysis of Variance
The ANOVA takes the t test from two groups to three
or more groups. It accomplishes this by replacing multi-

Orthopaedic Knowledge Update 8

ple t tests with the F test. The F test assumes all the underlying group population means are equal. Therefore, if
the groups have a common population mean, then the
group or sample means should all lie near the population mean. If the group means are sufficiently different,
then the F statistic will be large, and it can be concluded
that at least one of the population means for the group
varies from the others. The ANOVA does not differentiate which group differs from the others. To make that
determination, multiple comparison techniques must be
used. Beyond the analysis of studies with more than two
groups and one factor of interest, ANOVA is valuable
for testing whether treatment groups have comparable
population means for variables that may influence treatment, but are not the primary variable of interest. Classic examples are age, gender, or comorbidities. In welldesigned RCTs this type of analysis will be done to
ensure comparable treatment groups.

Rates and Proportions

The tests and examples given above have been concerned with quantitative data made up of continuous or
discrete data. Many scientific studies use data that are
measured on a nominal scale. These variables are easily
described as proportions and are analyzed by a distinct
set of statistical tools. Analogous to the t test for proportions is the z statistic. The z statistic is derived along the
same lines as a t test and is only adequate for hypothesis
testing when there are two outcomes of interest. The
more commonly used test is the chi square test, which is
the proportion equivalent of the ANOVA, but is also
used when there are only two outcomes of interest.
Comparisons of Two Proportions (Chi-Square Test and
the Exact Test)
In a study assessing neurologic outcome in spinal cord injuries, a total of 123 patients were eligible but only 71 consented. Because there were a large number of patients
who refused, the researchers wanted to ensure there was
no bias introduced based on the subjects who did and did
not consent. They wanted to check if there was an association between gender distribution and participation.
The chi square test summarizes the differences between the expected and observed frequencies. The expected frequencies are the number of subjects that would
be expected in each group subdivision (consenting/
nonconsenting males, consenting/nonconsenting females) if the treatment had no effect, or in this case if
there was no relationship between gender and participation. The Ho is that there is no relationship between the
two categorical variables, gender and participation. The
chi square test is used to test this hypothesis.
For the chi square test to be accurate, all the expected frequencies for each subgroup or cell should exceed 5. If the study is small and expected frequencies

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 1 Orthopaedic Research: What An Orthopaedic Surgeon Needs to Know

are less than 5, two methods of improvement are widely

used: the application of continuity correction; and the
calculation of Fishers exact probability. When the expected frequencies are particularly small, say, less than
1, the Fishers exact test should be applied.
Relative Risk and Odds Ratio in a Cohort and
Case-Control Study
A classic cohort study usually makes comparisons between a group of individuals exposed to some factor
and a group not exposed. The relative risk (RR) is the
probability of the disease in the exposed group divided
by the probability of the disease in the unexposed (control) group. RR measures the increased risk (if any) of
incurring a particular disease in the exposed individuals
compared with the unexposed individuals. For example,
the association of lung cancer with smoking is a RR of
10, which would mean that smokers are 10 times more
likely to develop lung cancer. If the RR is 1, the risks
are equal and if less than 1, then the risk in the exposed
group is lower.
In a cohort study, the result of the study can be summarized as a 2 2 contingency table illustrated by the
generic form shown in Table 2.
The RR of a specific disease for individuals exposed
as compared with those not exposed is given by
RR =

a/(a + b)
c/(c + d)

The 95% CI for RR can also be calculated.

Case-control studies are retrospective; therefore, disease rates in the exposed sample and the nonexposed
sample cannot be estimated, thus RR cannot be calculated. The data can again be summarized in a 2 2 table. In such a case, if the disease in both exposed and
nonexposed groups is rare, or a and c are small in comparison with b and d, respectively, the odds ratio (OR)
will be used to estimate the relative risk.
OR =

a/b ad
c/d = bc

The OR is the probability of an event occurring, divided by the probability of an event not occurring. The
OR is the ratio of the odds of disease for the experimental group relative to the odds of disease in the control group. Similar to the RR, an OR above 1 implies
that exposure to the factor under investigation increases
the risk of disease, while a value below 1 means the factor reduces the risk of disease.

Statistical Modeling (Regression Analysis)

A statistical model is a detailed mathematical specification of a hypothesis. Modeling helps in examining the
relationship or association between variables. For example, do age and weight affect function after total hip re-

American Academy of Orthopaedic Surgeons

Table 2 | Presentation of Data From a Cohort Study

Disease Status
Characteristic
Exposed
Nonexposed

Total
Disease
a
c
a+c

No disease
b
d
b+d

a+b
c+d
n

placement? Ultimately, causation is the factor of interest, but before causation can be determined association
must be studied in more detail. Specifically, there are
five possible explanations for an association between
two variables (A and B). (1) Chance: luck has resulted
in the sample that was chosen not accurately reflecting
the situation in the true population. (2) Bias: structured
or systematic error in design analysis or conduct resulting in an alternative explanation for an observed relationship. There are numerous types of bias. Some of the
more common ones include recall, reporting, and selection bias. (3) Confounding: a relationship between A
and B reflects the effect of a third variable that is more
exactly related to both A and B. For example, a medication (A) designed to decrease postoperative pain (B) in
surgical patients, some of whom have narcotic dependence (C). C affects both A and B and therefore would
be a confounder in the evaluation of postsurgical pain
control. Age and gender are classic confounders as they
influence many health variables. (4) Causation: A is a
cause of B. (5) Reverse causation: B is a cause of A. For
example, the statement that drinking tea causes people
to live longer could be attributed to the result of older
people drinking more tea rather than tea affecting life
expectancy.
It is prudent to think of causation as a diagnosis of
exclusion. Once chance has been minimized, confounders have been controlled, and there is appropriate timing with cause preceding effect, then causation is supported. Statistical modeling facilitates this process, but
cannot compensate for bias, which is more a product of
study design.
One final issue that statistical modeling can detect
and deal with is interaction. Interaction is where As influence on B can change in the presence of a second
variable; a hypothetical example is that continuous passive motion and an epidural anesthetic may increase
postsurgical knee range of motion more than either intervention alone.
The variables described earlier in the chapter form a
hierarchy from continuous to binary. Higher, more descriptive variables can be converted to a lower variable;
for example, grouping a continuous variable such as age

Orthopaedic Knowledge Update 8

Orthopaedic Research: What An Orthopaedic Surgeon Needs to Know

power perspective the basic rule is that 5 to 10 subjects

are needed for each independent variable being assessed.

Survival Analysis (Kaplan-Meier Method)

Figure 1 Hypothetical survival probability for medical versus surgical trial. Group A =
medical treatment. Group B = surgical treatment.
into 5-year intervals. This will result in loss of detail but
can facilitate data analysis.
In the analysis of variables it is important to characterize them into one of two groups depending on the
role they play in the evaluation. The dependent or outcome variable is the quantity of interest whose variation
in a population is being explained. The independent or
explanatory variables are the ones being analyzed to determine if they influence the dependent variable. Independent variables are also referred to as covariates. The
researcher determines the status of each variable; it is
not an intrinsic property of the variable itself. A dependent variable in one evaluation could be independent in
another.
Regression modeling is any statistical model that involves independent or explanatory variables. Regression
statistics essentially allow the relationship between the
dependent variable and the independent variables in a
population to be uncovered, despite the presence of
other factors affecting outcome. These principles are
needed primarily in observational studies to control for
confounders, but regression statistics can be equally
valuable in RCTs to adjust for imbalances in treatment
groups and known confounders.
Simple linear regression involves one dependent and
one independent variable, such as trying to predict the
weight of a child by height. Multiple linear regression
implies that there are at least two independent variables. For example, is neurologic recovery in spinal cord
injury (dependent variable) is associated with surgery,
age, gender, and energy of injury (independent variables)? In linear regression the dependent variables are
continuous; however, when a dichotomous dependent
outcome is selected, then a logistic regression analysis
must be done.
Details of regression modeling are beyond the scope
of this chapter, but regression or statistical modeling is a
very powerful tool that can be used both in observational or cohort studies as well as in RCTs. From a

10

Many clinical trials accrue patients over time, and in certain situations patients are followed for varying lengths
of time. Often, event rates such as mortality or frequency of aseptic loosening of a total joint arthroplasty
are selected as primary response variables. Analyzing
two groups for event rates such as these could be done
with chi-square or the equivalent normal statistic for
comparing two proportions, but because the length of
observation for each subject is variable, estimating an
event rate is complicated. Moreover, a basic comparison
of event rates may be misleading. For example, Figure 1
shows the survival pattern for two different groups. Although their mortality rate at 5 years is nearly identical,
the mortality pattern is quite different. This could represent a medical versus surgical trial where surgery might
carry a high initial mortality secondary to perioperative
and intraoperative risks.
Kaplan-Meier or survival analysis is a nonparametric
method to examine and compare the distribution of
times between two events. These techniques can provide
three main functions. The first is to estimate the cumulative risk of an event (usually adverse) over time. This is
plotted as a cumulative proportion of subjects remaining event free and is termed the survival curve. The pattern of this curve becomes smoother and more accurate
with increasing numbers of subjects. Second, it can compare the position of two survival curves using proper
statistical hypothesis testing. For example, a study to
compare the survival patterns between two types of total knee arthroplasties would use the Mantel-Haenszel
(log rank) Statistic. Finally, survival analysis techniques
can study the influence of various baseline variables on
the underlying risk of the event. This can be done in the
setting of a natural history study or to adjust for imbalances in prognostic variables (confounders) between
two treatment groups. The Cox proportional hazards
model allows for adjustment for these imbalances.

Outcomes
Outcomes research and other epidemiologic issues historically have been of little interest to the clinician, but
as pressure builds for therapeutic accountability clinicians find themselves in the unfamiliar territory of
health-related quality of life (HRQOL), cost effectiveness and other patient based outcomes. Use of these
patient-based measures will continue to grow as patients, administrators, policy makers, and professional
organizations demand evidence-based medicine.
HRQOL focuses on many different dimensions of
health including physical, mental, pain, function, and sat-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 1 Orthopaedic Research: What An Orthopaedic Surgeon Needs to Know

Table 3 | Interobserver Variability in Clinical Assessment

Lumbar Spine Interpretation by
Musculoskeletal Radiologists

Self-Administered Patient
Questionnaires

Reading

Kappa*

Questionnaire

Kappa*

Any abnormality
Facet joint sclerosis
Any narrowed disks

0.51
0.33
0.49

Sickness impact
Medical history

0.87
0.79

*Kappa statistic or intraclass correlation coefficient ranges from 0 to 1, with 1 representing perfect agreement and 0 representing random agreement

isfaction. Despite this specificity, clinicians, especially

surgeons, often regarded HRQOL outcomes as soft
measures that dont provide the hard distinct data surgeons covet, but patients or society have little interest
in. They are perceived as subjective and easily influenced by a patient who exaggerates symptoms or disability. These patient influences are not limited to questionnaires, because mood, motivation and other
psychosocial issues can prejudice physiologic measures
such as range of motion, strength, and ability to perform
functional tasks. Outcomes instruments are developed
through a collaborative effort of clinicians and social
scientists using vigorous psychometric principles. Even
the so-called objective tests such as radiographs that
take patient factors out of the equation can be outperformed by the so-called softer questionnaires (Table 3).
In the mid 1980s several articles pointed out the
glaring deficiencies in the orthopaedic literature and the
somewhat shaky foundation on which treatment decisions are based. The majority of orthopaedic studies
were retrospective cohort or case control studies with
flawed outcome measures and uncontrolled bias. A 1985
study brought to light the tremendous variability in outcome that could occur if various traditional surgical
measures were used on the same patient group. Terms
such as excellent, good, or satisfactory were not precise
in their definitions and varied depending on the context
in which they were used (ie, pain, function, return to
work). By combining multiple dimensions of outcome
with subjective and unclear questions it was difficult to
determine whether surgery was successful or not, nor
could results between studies be compared. Furthermore, the importance of a qualified independent patient
evaluator was made clear; patients report better results
when the surgeon or his/her delegate asks the questions.
There are three general psychometric criteria that
should be established in HRQOL measures before they
are endorsed. Reliability is the ability of the tool to be
reproducible and internally consistent over time. Validity ensures that the instrument is accurately measuring
what it is supposed to be measuring. Responsiveness is

American Academy of Orthopaedic Surgeons

the ability of the questionnaire to detect clinically relevant change or differences. Responsiveness will vary depending on the type of HRQOL questionnaire and the
patient population being evaluated. For example, if an
outcome tool is designed for the functional assessment
of patients with severe rheumatoid arthritis, and the
outcome tool is given to patients with mild osteoarthritis, it would probably produce all perfect scores, the socalled ceiling effect. Likewise, a questionnaire developed for the general population would not discriminate
among severely impaired patients in a rehabilitation setting, because all patients would end up with the worst
possible score (floor effect).
HRQOL questionnaires may only involve one or
two questions, but generally consist of several items or
questions, organized into domains or dimensions. A domain is an area or experience the questionnaire is trying
to measure. Examples would include pain, disability,
well-being, and mood. There are two basic types of
questionnaires, generic and disease-specific.
Generic instruments attempt to evaluate overall
health status. The SF-36 is probably the most well
known of the generic tools and is made up of eight domains that can blend to form a physical and mental
component score interpreted as the physical and
mental dimensions of health status. The major advantage of generic instruments is that they deal with a variety of areas in any population regardless of the underlying disease. This allows for broad comparisons across
various disease states, enabling an assessment of the impact of health care programs. They can also be very useful in assessing the overall HRQOL after a very specific
intervention, such as surgery. One final advantage of
some of the more frequently used questionnaires is the
availability of normative data. This can be very helpful
when trying to evaluate patients who do not have a
baseline HRQOL score to compare with the postintervention score. For example, in the trauma patient, it is
the surgeons goal to try to return the patient to normal.
One other generic measure of note is utility measurement. These measurements are somewhat more
complicated as they evolved from decision and economic theory. They reflect the patients preferences for
various treatment options and potential outcomes. The
primary value of utility measures is in economic analysis; therefore their use will grow exponentially in the
coming years.
Disease-specific measures are the other major category of HRQOL outcome tools. These questionnaires
concentrate on a region of primary interest that is generally relevant to the patient and physician. As a result
of this focus on a region or disease state, the likelihood
of increased responsiveness is higher. These instruments
can be specific in various ways. Some examples of the
primary focus of these instruments include populations

11

Orthopaedic Research: What An Orthopaedic Surgeon Needs to Know

(rheumatoid arthritis), symptoms (back pain), and function (activities of daily living). The disadvantage of a
disease-specific outcome is that general information is
lost, and therefore, it is generally recommended that
when evaluating patients, both a disease-specific and generic outcome measure should be used.
How are HRQOL outcome tools used? First, an instrument must be selected. Assuming it has passed psychometric scrutiny, the major determinants are the question or the purpose of the study and the feasibility of
applying the instrument. Is the question evaluative or
discriminative? What is the domain of interest, the population being studied? Is function important? For example, in a study population of spinal cord injured patients
the Functional Independence Measure is a much better
choice than the Roland Morris Disability Questionnaire
because the former is designed for neurologically impaired patients and any floor effects will be avoided. In
most studies after addressing the primary question the
physician should be as comprehensive as possible; therefore, the use of both a generic and disease-specific outcome measure is ideal. From a feasibility perspective,
patient and investigator burden must be considered.
Time to complete, patient comprehension, cost to administer, analyze, store and retrieve data, and interpretability of results are only a few of the issues that must
be considered. A concise guide to the appropriate outcome instrument for the anatomic region or disease process being studied is provided on the American Academy of Orthopaedic Surgeons website.
For all the attributes of HRQOL questionnaires
there are numerous problems, such as compliance, cost,
and collection of too much data. Caution must also be
exercised not to alter questionnaires or develop creative
scoring systems, as these threaten the validity of the instrument, the ability to compare across studies, and ultimately the results of the study. Multiple comparisons
are another danger with HRQOL outcomes. Because of
the wealth of information they provide in numerous domains, it is tempting for the researcher to go off on a
fishing expedition looking for the elusive variable that
will result in statistical significance.
One of the most difficult challenges is to determine
what magnitude of change in score represents a clinically significant change when evaluating a particular intervention. Some outcome measures such as the American Academy of Orthopaedic Surgeons/North American
Spine Society Low Back Instrument suggests that a 20%
difference is clinically significant. Other questionnaires
provide no guidelines and it is left up to the clinical researcher to decide. This issue is not only relevant from
an evaluative perspective but is also germane to the
power calculation of a study. A 10% to 20% effect
change would generally be considered appropriate.

12

Orthopaedic Knowledge Update 8

Evidence-Based Medicine
In the current era of increasing accountability for health
care services, the issues of quality and access to health
care are of overriding importance. A major focus to date
has been on efficiency of care, with a view to maximizing productivity and optimizing resource utilization. Recently, the National Academy of Science Institute of
Medicine has defined quality of health care as the degree to which health services for individuals and populations increase the likelihood of desired health outcomes
and are consistent with current professional knowledge. Thus, improvements in accountability must
clearly include indicators of efficiency. Accountability
requires that increased emphasis be placed on bringing
patterns of clinical practice in line with current scientific
evidence and that the effectiveness of current health
services at producing desirable health outcomes be determined.
As pressure grows for increasing accountability in
the use of medical resources, the clinician must play a
greater role in leading the design of the studies that
evaluate care. It is by ensuring that the care is based on
sound evidence, that clinicians will be able to remain
ethical advocates of effective patient care.

Annotated Bibliography
American Academy of Orthopaedic Surgeons Website.
AAOS Normative Data Study and Outcomes Instruments, Table of Contents. Available at: http://
www3.aaos.org/research/normstdy/main.cfm/. Accessed
January 26, 2004.
Eleven functional outcomes instruments related to the
musculoskeletal system are presented. Seven of the instruments contain the generic outcomes instrument, the SF-36.
Not only are the instruments available to assess baseline levels
and responsiveness to treatment, normative data from the general population are provided to serve as a point of reference.

Atlas SJ, Keller RB, Chang Y, Deyo RA, Singer N: Surgical and non-surgical management of sciatica secondary to lumbar disc herniation: Five year outcomes from
the main lumbar spine study. Spine 2001;26:1179-1187.
A well designed prospective outcome study comparing the
results of the two recognized treatment approaches for lumbar
disk herniation is presented. Validated outcome instruments
are used along with appropriate statistical analysis to show
surgical intervention to be superior. The influence of various
baseline variables on these outcomes is also analyzed.

Bailey CF, Fisher CG, Dvorak MF: Type II error in the

spine surgical literature. Spine 2004;29:1723-1730.
The primary purpose of this study was to determine the
frequency of potential type II errors published in the surgical
spine literature. The article clearly defines type I and type II
errors and emphasizes their importance, along with the signifi-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 1 Orthopaedic Research: What An Orthopaedic Surgeon Needs to Know

cance of identifying one primary research question. Only 17%

of the randomized trials had adequate power to determine an
appropriate difference and merely 27% had identified a primary question. The results support appropriate scrutiny of
RCTs before implementing their results.

Fisher CG, Dvorak MF, Leif J, Wing P: Comparison of

outcomes for unstable cervical flexion teardrop fractures managed with halo thoracic vest vs. anterior corpectomy and plating. Spine 2002;27:160-166.
This study uses a retrospective cohort design with a cross
sectional outcome analysis to assess the radiographic and clinical outcomes of two treatment methods. This design eliminates some of the inherent biases of a retrospective study by
obtaining objective long term validated and reliable outcome
measures. Although the study demonstrates improved radiographic measurements with surgery, this does not necessarily
correlate with clinical outcome.

Glantz SA: Primer of Biostatistics, ed 5. New York, NY,

McGraw-Hill, 2001, pp 6-7.
This book provides a superb overview of biostatistics and
is a great reference for both the novice and the part-time researcher. All the essential aspects of statistics are covered in
an easy to read and stimulating way and reinforced with relevant examples. Summary tables provide quick check references for the appropriateness of commonly used statistical
methods.

Institute for Clinical Evaluative Sciences Website. Practice Atlas Series, 2000, Toronto, Ontario. Available at:
http://www.ices.on.ca. Accessed January 26, 2004.
This is a reference for nonbiased evidence-based research in
the areas of health care delivery, service utilization, health technologies, treatment modalities, and drug therapies.The major objective of the Institute for Clinical Evaluative Sciences is to perform population-based health delivery research that is germane
to the clinician and the makers of health policy.

Journal of Bone and Joint Surgery Website. Levels of

evidence for primary questions. Available at: http://
www.ejbjs.org/misc/public/instrux.shtml. Accessed January 26, 2004.

American Academy of Orthopaedic Surgeons

A concise table outlining the hierarchy and spectrum of

the various study designs for clinical research is presented. The
table serves as a guide as to the strength of evidence of a particular study, with level 1 being the gold standard.

National Heart: Lung and Blood Institute Website.

Framingham Heart Study. Available at: http://
www.nhlbi.nih.gov/about/framingham/. Accessed January 27, 2004.
A comprehensive review and current status of the landmark cohort study assessing variables that influence cardiovascular disease is presented. Design, objectives, results, and ancillary studies are discussed.

Classic Bibliography
Begg C, Cho M, Eastwood S, et al: Improving the quality of reporting of randomized controlled trials: The
CONSORT statement. JAMA 1996;276:637-639.
Bombardier C, Kerr M, Shannon H, Frank J: A guide to
interpreting epidemiologic studies on the etiology of
back pain. Spine 1994;19(suppl 18):2047S-2056S.
Deyo RA, Andersson G, Bombardier C, et al: Outcome
measures for studying patients with low back pain.
Spine 1994;19(suppl 18):2032S-2036S.
Gartland JJ: Orthopaedic clinical research: Deficiencies
in experimental design and determinations of outcome.
J Bone Joint Surg Am 1988;70:1357-1364.
Howe J, Frymoyer J: Effects of questionnaire design on
determination of end results in lumbar spine surgeries.
Spine 1985;10:804-805.
Lieber RL: Statistical significance and statistical power
in hypothesis testing. J Orthop Res 1990;8:304-309.
Markel MD: The power of a statistical test. What does
insignificance mean? Vet Surg 1991;20:209-214.
Sledge CB: Crisis, challenge, and credibility. J Bone Joint
Surg Am 1985;67:658-662.

13

Chapter

Soft-Tissue Physiology and Repair

Robin Vereeke West, MD
Freddie H. Fu, MD

Introduction
Progressive advances including research on stem cells,
growth factors, and tissue engineering continue to be
made in the treatment of soft-tissue diseases and injuries. Although cells have been cultured in the laboratory
for many years, the replication of the function and the
structure of complex human tissues is a relatively recent
development. There are currently many advances being
made regarding tissue engineering. Continuing progress
also is being made in the understanding of the genetic
basis of diseases such as degenerative joint disease, and
the biologic response to soft-tissue injury.

Articular Cartilage
Structure and Function
Articular cartilage is a highly organized viscoelastic material composed of chondrocytes, water, and an extracellular matrix (ECM) and is devoid of blood vessels, lymphatics, and nerves. Complex interactions between the
chondrocytes and the ECM actively maintain tissue balance.
Chondrocytes from different cartilage zones vary in
size, shape, and metabolic activity. All chondrocytes are
active in the homeostasis of their surrounding matrix
and derive nutrition from the synovial fluid. The chondrocytes sense mechanical changes in their surrounding
matrix through intracytoplasmic filaments and short
cilia on the surface of the cells.
The ECM consists primarily of water (65% to 80%
of its total wet weight), proteoglycans, and collagen. The
predominant collagen is type II (95%), but smaller
amounts of other collagens (types IV, VI, IX, X, XI)
have also been identified. The exact function of the
other collagens is unknown, but they may be important
in matrix attachment and stabilization of the type II collagen fibers.
The collagen forms a three-dimensional network
that encases proteoglycan molecules, predominantly
chondroitin and keratan sulfates. This lattice framework
is responsible for the structural properties of articular
cartilage, including tensile strength and resiliency. The

American Academy of Orthopaedic Surgeons

proteoglycans are negatively charged and attract water

along with cations. These sulfated proteoglycans are
linked to hyaluronate and are responsible for the high
water content of the ECM. The rate of deformation of
articular cartilage is directly correlated with how quickly
water is discharged, and therefore is affected by the
structural integrity of the collagen.
The normal thickness of articular cartilage is determined by the contact pressures across the joint. The
higher the peak pressures, the thicker the cartilage. The
patella has the thickest articular cartilage in the body.
Articular cartilage can be divided into four distinct
layers: superficial, middle, deep, and calcified (Figure 1).
These layers differ in cellular morphology, biomechanical composition, and structural properties. In the superficial layer, the collagen orientation is parallel to the surface. It changes to a more random, less densely packed
array in the middle zone. The collagen bundles are perpendicular to the joint surface and subchondral bone in
the deep and calcified layers. The deep zone has the
largest collagen fibers, the highest concentration of proteoglycans, and the lowest concentration of water.
The morphology and arrangement of the chondrocytes differ in each zone. In the calcified zone, the chondrocytes are small and randomly arranged. The chondrocytes transition to a columnar arrangement and
spherical shape in the deep zone, to a more random array of cells in the middle zone, and to a flat, parallel array of cells in the superficial zone (Figure 2). The tidemark, a thin, basophilic line that is seen on light
microscopy sections of articular cartilage, represents the
boundary between the calcified and uncalcified cartilage.

Injury and Repair

Injuries to articular cartilage can be divided into three
distinct types: type 1 injuries involve microscopic damage to the chondrocytes and ECM (cell injury) and
type 2 injuries involve macroscopic disruption of the articular cartilage surface (chondral fractures or fissuring).
These two injury types have an extremely poor healing

15

Soft-Tissue Physiology and Repair

Orthopaedic Knowledge Update 8

Figure 1 Articular cartilage layers.

potential because they do not penetrate the subchondral
bone and therefore do not bring forth an inflammatory
response. Type 3 injuries involve disruption of the articular cartilage with penetration into the subchondral
bone (osteochondral fracture). Type 3 injuries produce a
significant inflammatory process. A fibrin clot is formed,
and mesenchymal undifferentiated cells produce a reparative tissue, which is not normal articular cartilage
but fibrocartilage that consists primarily of type I collagen and has a tendency to undergo early degenerative
changes.
Articular cartilage injuries have frequently been observed in conjunction with anterior cruciate ligament
(ACL) injuries. An occult osteochondral lesion or bone
bruise may be detected with MRI in up to 80% of patients. The most common locations of these lesions are
within the lateral compartment of the knee, on the lateral femoral condyle at the sulcus terminalis, and at the
posterolateral tibial plateau (Figure 3). Although the
area may appear normal during arthroscopic evaluation,
in vivo histologic studies have shown significant disruption of the articular cartilage. Current data strongly indicate that chondrocyte apoptosis can be stimulated by
the application of a single, rapid impact load and that
the extent of apoptosis is related to the amount of load
applied.

Treatment
Surgical treatment options for full-thickness cartilage
defects are simple arthroscopic dbridement, abrasion
arthroplasty, microfracture, autologous chondrocyte cell
implantation, and mosaicplasty with either autologous
tissue or fresh allograft. In a recent study of autologous
osteochondral mosaicplasty in the treatment of fullthickness cartilage defects, good to excellent results
were achieved over a 10-year period.
Actual regeneration of articular cartilage is accomplished when the present cells become mature chondrocytes that are capable of restoring the biomechanical

16

Figure 2 Basic structural anatomy of articular cartilage. (Reproduced from Browne

JE, Branch TP: Surgical alternatives for treatment of articular cartilage lesions. J Am
Acad Orthop Surg 2000;8:180-189.)
and structural integrity of the articular surface. Current
research is focused on inducing the newly attracted or
transplanted cells to become mature chondrocytes using
growth factors (polypeptides that bind to cell surface receptors). These growth factors act in a paracrine manner
and have a variety of regulatory effects on cells. Bone
morphogenetic proteins (BMPs) are members of the
transforming growth factor superfamily (except BMP-1)
and have a regulatory role in the differentiation of
cartilage-forming and bone-forming cells from pluripotent mesenchymal stem cells. More than a dozen BMPs
have been discovered.
Acceleration of cartilage healing has been shown in
vivo with the implantation of genetically modified chondrocytes, expressing BMP-7. Large articular cartilage
defects in the patellofemoral joints of 10 horses were
implanted with either controls or the genetically modified chondrocytes. The lesions that were treated with the
BMP-7 chondrocytes showed an accelerated healing response at the 4-week biopsy. The 8-month biopsy revealed a similar healing response in both the control
and the study groups. A recent study was performed to
evaluate the clinical outcomes and histologic results of
isolated femoral articular cartilage defects treated with
either a microfracture or an autologous chondrocyte implantation. No significant differences were found between the two groups regarding the clinical and biologic
outcomes.

American Academy of Orthopaedic Surgeons

Chapter 2 Soft-Tissue Physiology and Repair

Orthopaedic Knowledge Update 8

Figure 3 MRI of common occult bone bruises associated with ACL injuries in the sulcus terminalis of the lateral femoral condyle (A) and the posterolateral tibial plateau (B).
Alternatives to surgery are also being promoted in
the treatment of arthritis. Viscosupplementation, or
intra-articular injections of hyaluronic acid, has been
used to treat osteoarthritis. The proposed mechanisms
of action result from the physical properties of the hyaluronic acid, as well as the anti-inflammatory, anabolic,
local analgesic, and chondroprotective effects. Hyaluronic acid has both viscous and elastic properties. At
high shear forces, the molecules exhibit increased elastic
properties and reduced viscosity. At low shear forces,
the opposite effects are seen. The anti-inflammatory effects of hyaluronic acid include inhibition of phagocytosis, adherence, and mitogen-induced stimulation. The anabolic effects have been demonstrated in vivo with
studies showing that intra-articular injections of hyaluronic acid may stimulate fibroblasts. The antiinflammatory effects may explain the analgesic effect.
Although hyaluronic acid has been shown to stimulate
cartilage matrix production, the chondroprotective effects have not been confirmed. Several studies have
failed to show a statistically significant benefit for hyaluronic acid injections when compared with a placebo.
Furthermore, viscosupplementation is relatively expensive, with the cost of a series of injections at more than
$500 per knee.
Numerous studies have investigated other potential
chondroprotective agents (substances that are capable
of increasing the anabolic activity of chondrocytes while
suppressing the degradative effects of cytokine mediators) on cartilage. These agents include chondroitin sulfate, glucosamine sulfate, piroxicam, tetracylines, corticosteroids, and heparinoids. Glucosamine serves as a

American Academy of Orthopaedic Surgeons

substrate for the biosynthesis of chondroitin sulfate, hyaluronic acid, and other macromolecules located in the
cartilage matrix. Chondroitin sulfate, which is covalently
bound to the proteins as proteoglycans, is secreted into
the ECM. The load-bearing properties of the cartilage
are attributable to the compressive resilience and affinity for water that the proteoglycans possess. Studies
have supported the effectiveness of glucosamine and
chondroitin sulfate for the relief of symptoms of osteoarthritis based on clinical trials and short-term
follow-up. These studies have shown a progressive and
gradual decline of joint pain and tenderness and improved motion; few side effects have been reported.
However, many questions surround the long-term effects, the most effective dosage and delivery route, and
the purity of glucosamine and chondroitin sulfate products. Prospective studies that use validated outcome
measures for pretreatment and posttreatment and that
stratify major confounding variables are needed.

Meniscus
Structure and Function
The meniscus is a specialized viscoelastic fibrocartilaginous structure capable of load transmission, shock absorption, stability, articular cartilage lubrication, and
proprioception. The meniscus is more elastic and less
permeable than articular cartilage and is composed of a
complex three-dimensional interlacing network of collagen fibers, proteoglycan, glycoproteins, and fibrochondrocytes that are responsible for the synthesis and maintenance of the ECM. The meniscus is composed of 75%

17

Soft-Tissue Physiology and Repair

Figure 4 Collagen fiber ultrastructure with longitudinally and radially oriented fibers.
collagen, 8% to 13% noncollagenous protein, and 1%
hexosamine. Type I collagen is predominant (90%);
small amounts of types II, III, V, and VI are present.
The collagen fiber ultrastructure influences the loadbearing role of the meniscus. Radial collagen bundles
run from the periphery to the center of the meniscus
and large collagen fibers have a predominately circumferential arrangement (Figure 4). Collagen bundles on
the superficial surfaces of the meniscus have no organization. The rate of fluid exudation from the meniscal tissue determines the rate of creep. The collagen proteoglycan matrix and the applied load affect the
deformation of the meniscus.
The peripheral meniscus obtains its blood supply
from a circumferentially arranged perimeniscal capillary
plexus from the superior and inferior geniculate arteries. This capillary plexus penetrates up to 30% of the
medial and 25% of the lateral meniscus. The inner two
thirds of the meniscus are essentially avascular and receive nutrition from the synovial fluid. Free nerve endings and corpuscular mechanoreceptors have been
found in meniscal tissue, concentrated at the root insertion sites and the periphery.
Kinematic analysis has shown that the meniscus is a
dynamic structure, moving anterior with knee extension
and posterior with knee flexion. Because the peripheral
attachments of the lateral meniscus are interrupted by
the popliteus, it has greater mobility than the medial
meniscus.
Through its shape and structure, the meniscus provides
several important functions in the knee joint. The shape
of the meniscus improves the congruency of the articulating surfaces and increases the surface area, thus aiding in
load transmission across the joint.The meniscus is responsible for transmitting 50% of the joint force in knee extension and 90% of the joint force in deeper flexion.

Meniscal Tears and Repairs

The detrimental effects of partial and complete meniscectomy have been shown in numerous studies. Partial
and complete meniscectomies have been shown to in-

18

Orthopaedic Knowledge Update 8

crease contact pressure by as much as 65% and 200%,
respectively. Meniscal loss leads to alterations in load
transmission and accelerated rates of articular cartilage
degeneration. The healing potential of meniscal tears
varies according to patient age and the location/
chronicity of the tear.
The success of meniscal repairs depends on multiple
factors, including vascular supply, tissue stability, tear
pattern, associated injuries, and chronicity of the tear.
The types of tears that can be repaired include acute,
peripheral, and unstable tears in the vascular zone of
the meniscus. Complex, radial, or flap tears are not usually appropriate for repair because excision is often necessary. Partial-thickness or stable tears usually do not
require repair.
The poor healing potential of central meniscal tears
has led to the investigation of methods to provide blood
supply to the injured area. These methods include the
use of fibrin clot, fibrin glue, endothelial cell growth factor, vascular access channels, and synovial pedicle flaps.
Studies support the use of exogenous fibrin clots to promote healing of meniscal tears in the avascular zone.
The clot provides chemotactic and mitogenic factors,
such as platelet-derived growth factor and fibronectin,
which stimulate the cells involved in wound repair. The
clot also provides a scaffold for the support of the reparative process.
A stable knee is important for successful meniscal
repair and healing. Meniscal repairs have a success rate
of as high as 90% to 100% when associated with a concomitant ACL reconstruction. In contrast, the success
rate drops to 30% to 70% when the repair is performed
in an unstable knee. Healing rates of meniscal repairs
are lower in ACL-intact knees when compared with
healing rates in ACL-reconstructed knees, but are
higher than in ACL-deficient knees. The increased rates
of healing in ACL-reconstructed knees is possibly a result of drilling, which provides an exogenous source of
blood to promote healing of the repair.

Meniscal Replacement
Although autografts, biocompatible prostheses, bioabsorbable collagen scaffold, and synthetic materials have
been used as meniscal replacements, the only currently
available method to replace the entire meniscus is meniscal allograft transplantation. The ideal candidate for a
meniscal transplant is a symptomatic patient with a
prior meniscectomy, persistent pain in the involved compartment, intact articular cartilage (less than grade III
changes), normal alignment, and a stable joint. Localized chondral defects should be treated concomitantly.
Osteotomies or ligament reconstruction can be performed as a staged procedure or concurrently. Additional contraindications include inflammatory arthritis,
obesity, and previous infection.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Other factors that should be considered before performing the transplantation include antigenicity, sterilization techniques, and remodeling capabilities of the allograft. Class I and II histocompatibility antigens are
expressed on the cells of the meniscal allograft, indicating the potential for an immune response. However,
there is only one report of frank immunologic rejection
of a cryopreserved, nontissue-matched meniscal allograft. Immunoreactive cells (B lymphocytes and
T-cytotoxic cells) have been identified in recipients of
fresh-frozen allografts. The effect of the immune response is unknown, but the reaction may stimulate healing, incorporation, and revascularization.
The American Association of Tissue Banks has defined the recommended testing protocol for allograft
screening. Serologic screening is performed for human
immunodeficiency virus (HIV) p24 antigen, HIV-1/
HIV-2 antibodies, human T cell lymphoma/leukemia
virus-1 and -2, hepatitis B surface antigen and core antibody, hepatitis C antibody, and syphilis. Many tissue
banks perform polymerase chain reaction testing, which
can detect 1 in 10 HIV-infected cells. Cultures for aerobic and anaerobic bacteria also are performed. Following the harvest, the tissue can be preserved by one of
four methods: fresh, cryopreserved, fresh-frozen, or lyophilized. Only fresh and cryopreserved allografts contain viable cells at the time of transplantation. However,
it is not known if these cells survive after the transplantation. Several studies have shown that the donor cells
are replaced by the host-derived cells. Additional sterilization with ethylene oxide, gamma irradiation, or
chemical methods has been shown to have deleterious
effects on the graft tissue. The amount of gamma radiation required to eliminate viral DNA (at least 3.0 Mrad)
may adversely affect the material properties of the allograft. Ethylene oxide produces by-products that may
cause synovitis.
Meniscal allografts are repopulated by host-derived
cells that appear to originate from the synovial membrane. The repopulation occurs from the peripheral
zone to the central core. Active collagen remodeling by
the host cells has been shown, but the long-term ability
of these cells to synthesize appropriate matrix proteins
and maintain the ECM is unknown. The transplanted
menisci also undergo a gradual, incomplete revascularization, with new capillaries derived from the capsular
and synovial attachments.

Intervertebral Disk
The intervertebral disk forms the primary articulation
between the vertebral bodies and is the major constraint
to motion of the functional spinal unit. The disk is composed of two morphologically separate parts, the outer
and the inner part. The outer part, the anulus fibrosus, is
made up of fibrocartilage and type I collagen. At the pe-

American Academy of Orthopaedic Surgeons

Chapter 2 Soft-Tissue Physiology and Repair

riphery, the collagen fibers are oriented vertically. They
become more oblique with each underlying layer, with
the fibers in each adjacent sheet running at about 30
angles to each other. Lamination of these layers
strengthens the anulus fibrosus. Some peripheral fibers
extend past the cartilage end plate to insert onto the
vertebral bodies as Sharpeys fibers. Neural fibers are
found in the outer rings of the anulus fibrosus. The
nerve fibers are dorsal branches of the sinu vertebral
nerve; the ventral branches arise from the sympathetic
chain that courses anterolaterally over the vertebral
bodies.
The inner part of the intervertebral disk is the nucleus pulposus. In young people, the nucleus pulposus is
composed of 90% water. The other components are
type II collagen and proteoglycans, which bind water.
With the gradual loss of proteoglycans, the water content of the nucleus pulposus declines with advancing
age. After the third decade of life, there is a gradual
fluid loss and concomitant replacement of the nucleus
pulposus with fibrous tissue. By the sixth or seventh decade of life, the entire nucleus pulposus is replaced by
fibrocartilage.
The disk is separated from the vertebral bodies by
hyaline cartilage end plates. Because the intervertebral
disk is avascular, nutrients and fluid enter the disk by
diffusion through the end plates or the anulus fibrosus.
Glucose diffuses through the end plates and sulfate ions
diffuse mainly through the anulus. The diffusion is influenced by mechanical and biologic factors. An outflow of
fluid occurs with increased load, and a fluid influx occurs with decreasing load. End plate permeability is reduced by patient factors such as smoking and exposure
to vibration and is enhanced with dynamic exercise.
The nucleus pulposus functions to resist compressive
loads. In axial compression, the increased intradiskal
pressure is counteracted by annular fiber tension and
disk bulge. Asymmetric and cyclic loading with combined lateral bend, compression, and flexion is a risk
factor for disk herniation. The lumbar motion segment
can resist a combination of a bending moment and a
shear force of 156 Nm and 620 N, respectively, before
complete disruption occurs. These numbers are much
lower than the failure load in compression. About 35%
of the torque resistance is provided by the disk, whereas
the remainder of the resistance is provided by the posterior elements and ligaments. Therefore, any injury to the
posterior elements may increase the risk for disk failure.
Viscoelastic intervertebral disk changes have been
shown both in vivo and in vitro. Disk height increases
with decreasing compressive loads because the decreased intradiskal osmotic pressure allows water to
flow into the disk. Reported diurnal changes in the
overall height of individuals range from 6.3 mm to
19.3 mm, with an average of 15.7 mm. The average person is 1% shorter in the evening than in the morning.

19

Orthopaedic Knowledge Update 8

Soft-Tissue Physiology and Repair

Information on disk degeneration, herniation, traumatic
injury, and treatment can be found in section 6. The effects of aging on intervertebral disks are discussed in
chapter 6.

Ligament
Structure and Function
Ligaments are dense connective tissues that link bone to
bone. The gross structure of the ligaments varies with
their location (intra-articular or extra-articular, capsular) and function. Some ligaments (ACL, posterior cruciate ligament, and inferior glenohumeral ligaments)
have geometric variations between their bundles.
Ligaments are composed primarily of water. Collagen makes up most of the dry weight of ligaments,
with type I collagen the predominant protein at 90%
and type III accounting for the remainder of the collagen. Type III collagen is often found in injured ligaments. Elastin accounts for about 1% of the dry weight
of ligaments, but is even more prevalent in spine ligaments.
Microscopically, the collagen fibers are relatively
parallel and aligned along the axis of tension. Fibroblasts are located between the rows of fibers and are responsible for producing and maintaining the ECM.
Strength is enhanced by the cross-linked structure of the
collagen fibers. Proteoglycans in the ECM store water
and affect the viscoelastic properties; the rate of deformation is directly related to the amount of stored water.
The direct insertion (for example, the medial collateral ligament [MCL] attachment) is the most common
type of insertion site and attaches the ligament to the
bone through four distinct zones. Zone 1 is made up of
collagen with ECM and fibroblasts. Zone 2 is composed
of fibrocartilage with cellular changes, whereas mineralized cartilage is found in zone 3. Zone 4 is characterized
by an abrupt transition to bone. In indirect insertions
(for example, the tibial attachment to the MCL), the superficial layer connects directly to the periosteum,
whereas the deep layer anchors to bone by Sharpeys fibers.

Stress is defined as force per unit area, and strain describes the change in length relative to the original
length. Under tension, a ligament deforms in a nonlinear fashion. In the initial phases of applied tension, the
coiled nature of collagen and the crimping become
more aligned along the axis of tension; the collagen fibers then become taut and stretch with continued tension. The slope of the linear-elongation curve describes
the tissue stiffness, and the slope of the stress-strain
curve denotes the tensile modulus. The point at which
overload occurs and the tissue fails is the yield point.
The ultimate load and elongation are defined as this
overload point for structural properties, and the ultimate tensile stress and strain are defined as this yield
point for mechanical properties. Besides the nonlinear
nature of these curves, ligaments and tendons also show
a time-dependent viscoelastic behavior. More information on the biomechanics of ligaments can be found in
chapter 4.

Other Influences
An increased prevalence of ACL injuries in females and
gender-specific muscle response to sport-specific maneuvers have been discussed in the literature. Anatomic
features (smaller intercondylar notch, higher Q angle,
low hamstring/quadriceps force ratio), intrinsic factors
(estrogen/relaxin receptors within the ACL), and landing techniques (straight-knee landing, one-step stop
landing with the knee hyperextended) have been suggested as contributing factors to the increased incidence.
Estrogen, progesterone, and relaxin receptors have been
identified in the human ACL. In vitro fibroblast proliferation and collagen synthesis have been shown to directly correspond to estrogen levels. Increasing estrogen
levels lead to a decrease in cellular proliferation and
collagen synthesis of fibroblasts.
Skeletal maturity and age also have been shown to
affect the mechanical and structural properties of ligaments. The load at failure from specimens of older human ACL has been found to be 33% to 50% of that in
younger bone-ligament-bone specimens.

Ligament Properties

Effects of Disuse and Immobilization

The structural properties of ligaments, expressed by the

load-elongation curve, reflect the behavior of the entire
bone-ligament-bone complex, including its geometry, insertion sites, and material characteristics. The mechanical properties of ligaments, characterized by the stressstrain curve, depend on the ligament substance,
molecular bonds, and composition. Over an extended
period of time, ligaments respond to loading with an
overall increase in mass, stiffness, and load to failure. In
addition to these structural changes, the material properties show an increase in ultimate stress and strain at
failure.

Immobilization and disuse dramatically compromise the

structural and material properties of ligaments. In a
2003 study, immobilization led to a significant decrease
in the ability of scars to resist strain. According to another study, after 12 weeks of immobilization, the ACLbone unit showed a significant decrease in maximum
load to failure, energy absorbed to failure, and stiffness
(increased compliance). At 5 months after remobilization, a decrease in ligament strength was still apparent
(stiffness and compliance parameters had returned to
baseline levels). At 12 months, ligament strength had returned to a near-normal level.

20

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

The effect of immobilization on the ligament units
depends on the histologic characteristics of the attachment site. The ACL attachment site, through zones of fibrocartilage, was slightly affected after immobilization;
however, the direct tibial insertion of the superficial
MCL was significantly disrupted. This disruption was
caused by cortical and subchondral bone resorption after the period of immobilization. Collagen degradation
increases and collagen synthesis decreases with longer
periods of immobilization.

Injury and Repair

The healing process of extra-articular ligaments can be
divided into four phases: the inflammatory response, cell
proliferation and fibrin clot organization, remodeling,
and scar maturation. The ratio of type I to type III collagen normalizes during these final phases. The tensile
strength gradually increases. As a result of matrix
changes involving collagen reorganization and crosslinking, there is a gradual increase in tensile strength;
however, the injured ligament is never as strong as the
uninjured ligament. Ligament scarring is dominated by
small-diameter collagen fibers and decreased cross-link
density.
An intrinsic healing response has not been observed
in intra-articular ligaments such as the ACL, because
intra-articular ligaments have a limited blood supply and
the synovial fluid may not allow an inflammatory response. Fibroblast adhesion and migration is different between the MCL and ACL in an inflammatory environment. Platelet-derived growth factor and transforming
growth factor-beta are increased during MCL healing,
whereas the opposite effect is found in the healing ACL
animal model. Low-intensity pulsed ultrasound has been
shown to enhance early healing of transected MCLs in
rats. Superior mechanical properties were found 12 days
after the injury when compared with a placebo group. At
21 days after injury, the mechanical properties were similar between the placebo and ultrasound groups.
Ibuprofen has been shown to have no significant effect on early MCL healing. Studies have shown no significant differences between the mechanical properties
of placebo-treated and ibuprofen-treated rats after
MCL transection. However, cyclooxygenase-2 specific
inhibitors have been shown to impair ligament healing
in rats. When compared with a placebo-treated group,
the rats treated with celecoxib for 6 days after an MCL
transection had 32% lower load to failure.

Tendon
Structure and Function
Tendons function to transmit high tensile loads from
muscle to bone. Tendons are made up of densely packed
parallel-oriented bundles of collagen, composed mainly
of type I and III collagen by dry weight (86% and 5%

American Academy of Orthopaedic Surgeons

Chapter 2 Soft-Tissue Physiology and Repair

respectively). Three collagen chains comprise one collagen molecule, which is then organized into microfibrils
and fibrils. Proteoglycans (which influence the viscoelastic properties of tendons), glycoproteins, and water combine with fibrils into a matrix to form fascicles. The
fascicles come together to form bundles, which are surrounded by endotenon. The vascular supply, lymphatics,
and nerves are supported by the endotenon. Surrounding the collagen bundles are the epitenon and
paratenon.
The composition and organization of tendons make
them ideal to resist high tensile forces. Tendons deform
less than ligaments under an applied load and are able
to transmit the load from muscle to bone.
The toe region on the load-elongation curve, which
represents the structural properties of the bone-tendonmuscle unit, shows the initial stretch of the tendon,
which results from straightening of the cramped fibrils
and orientation of the longitudinal collagen fibers. Because tendons have more parallel collagen fibers than
ligaments and less realignment occurs during initial
loading, the toe region of the curve is smaller in tendons
than in ligaments. In addition, as the amount of crimp
decreases with age, the toe region becomes smaller.
The elastic modulus of the tendon is represented by
a linear region on the load-elongation curve that follows
the toe region. Tendon failure occurs in a downward
curve, representing permanent structural changes. Shortening of the muscle occurs when, during isometric contractions, the length of the muscle-tendon unit remains
constant but elongation of the tendon occurs secondary
to creep. Creep improves muscle function during isometric contractions.
Tendons surrounded by a paratenon receive their
blood supply from the periphery and a longitudinal system of capillaries. The perimysium and osseous insertions also provide a vascular supply to these tendons.
The perimysium, the periosteal insertion, and the long
and short vincula from the proximal mesotendon provide the blood supply for tendons within a sheath.

Factors Affecting Tendon Properties

Tendons become weaker, stiffer, and less yielding as a
result of the vascular, cellular, and collagen-related alterations that occur with aging. Aging causes many
changes in the properties of collagen, including a reduction in collagen turnover and a decrease in water and
proteoglycan content.
Studies have shown that exercise has positive effects
on the mechanical and structural properties of tendons,
whereas immobilization adversely affects their biomechanical properties, resulting in decreased tensile
strength, increased stiffness, and a reduction in total
weight. Therefore, early mobilization should be initiated
whenever possible.

21

Orthopaedic Knowledge Update 8

Soft-Tissue Physiology and Repair

Table 1 | Muscle Contractions

Type

Characteristics

Isometric
Concentric
Eccentric

Fixed load with no joint motion

Joint moves with a load and the muscle contracts
Results in muscle lengthening while controlling a
load during joint motion
Variable load with constant velocity

Isokinetic

(Reproduced with permission from Woo SL-Y, Debski RE, Withrow JD, Janqueshek MA: Biomechanics of knee ligaments. Am J Sports Med 1999;27:533-543.)

Growth factors are cell-secreted proteins that regulate cellular functions and are involved in cell differentiation and growth, including the normal processes of development and tissue repair. Several growth factors,
recently identified as playing a role in tendon healing,
include vascular endothelial growth factor, insulin-like
growth factor, platelet-derived growth factor, basic fibroblast growth factor, and transforming growth factor
beta. In addition, the transcription factor NF-kappaB
has been implicated in the signaling pathways of these
growth factors. The type and timing of cytokine delivery
to facilitate the most rapid and quality repair has yet to
be determined.

Injury and Repair

There are three types of tendon injuries: direct trauma
with transection of the tendon; indirect injury with avulsion of the tendon from the bone, and indirect intrasubstance injury stemming from intrinsic or extrinsic factors. Transections or partial lacerations are associated
with trauma and are most common in the flexor tendons
of the hand. Bone avulsions can occur after overwhelming tensile loads. Degenerative changes within a tendon
can occur from submaximal overload (Achilles tendinitis) or from repetitive pressure against a bony surface.
Tendon healing after an acute injury follows a similar pattern as other soft-tissue healing. The inflammatory response provides an extrinsic source of cells to begin the reparative process. Proliferation of the cells and
increased vascularity follows, whereas collagen synthesis
increases and the tissue matures.
Successful tendon-to-bone healing is necessary for
good outcomes in many reconstructive procedures. As
the tendon heals within a bony tunnel, fibrovascular interface forms and bony ingrowth is started. An indirect
insertion of the tendon collagen fibers then occurs.
Based on the results of animal models, it has been
shown that there is no advantage in repairing a tendon
to a cancellous bone trough compared with direct repair
to cortical bone. In MCL reconstruction in rabbits, the
initial maximum failure load was higher in the tendons
repaired to cortical bone when compared with those repaired to cancellous bone. By 8 weeks after the reconstruction, the tendon attachments in both groups had
matured to almost baseline.
Suture material, type of repair, suture knots, quality
of the tissue, continuous passive motion, gap formation,
and load experienced by the tendon are all factors that
can affect tendon repair. Aggressive early active motion
along with weight bearing has been shown to increase
the rate of tendon re-rupture and gap formation. However, controlled passive motion improves the repair
strength in the early period after repair and decreases
adhesions. An optimum level of stress and motion exists
that promotes healing without resulting in damage to
the repaired tendon.

22

Muscle
Structure and Function
Skeletal muscle constitutes the largest tissue mass in the
body, making up 40% to 45% of total body weight. Skeletal muscle originates from bone and adjacent connective tissue and inserts into bone via tendon.
The microscopic and macroscopic anatomy of muscle includes the motor unit, the muscle fiber bundles, individual fibers, myofibrils within the fibers, and the myofibril contractile unit (sarcomere). Muscle contraction
occurs in response to input via nerve fibers through the
neuromuscular junction. The force of the muscle contraction depends on the number of motor units firing.
The size of the motor unit depends on the number of
muscle fibers that are innervated by the nerve fiber.
Muscle contractions can either be isometric, concentric,
eccentric, or isokinetic (Table 1).
The three basic muscle types are I, IIA, and IIB.
Types I and II are determined by the speed of the contraction. Type I, or slow-twitch oxidative fibers, predominate in postural muscles and are well suited for endurance by aerobic metabolism. They have an ability to
sustain tension, relative fatigue resistance, and have high
amounts of mitochondria and myoglobin. In addition to
a slow contraction rate, type I fibers have a relatively
low strength of contraction. Type II muscle fibers are
fast-twitch, have a fast rate of contraction, and a relatively high strength of contraction. Type II fibers are either A or B, depending on their mode of energy utilization. Type IIA fibers have an intermediate aerobic
capacity, whereas type IIB fibers are primarily anaerobic. Type IIB fibers are common in muscles that require
a rapid generation of power, but they are less capable of
sustaining activity for a prolonged period because of
lactic acid buildup (Table 2).

Factors Affecting Muscle Properties

Like other tissues in the body, skeletal muscle undergoes changes with aging. Muscle mass decreases, the
number and diameter of muscle fibers decrease, stiffness

American Academy of Orthopaedic Surgeons

Chapter 2 Soft-Tissue Physiology and Repair

Orthopaedic Knowledge Update 8

Table 2 | Characteristics of Human Skeletal Muscle Fibers

Type I
Other names

Speed of
contraction
Strength of
contraction
Fatigability
Aerobic capacity
Anaerobic
capacity
Motor unit size
Capillary density

Type IIA

Type IIB

Red, slow-twitch White, fastSlow, oxidative

twitch
Fast, oxidative,
glycolytic
Slow
Fast

Fast, glycolytic

Low

High

High

Fast

centration, resulting in an increased VO2max and improved fatigue resistance. Improvements in endurance
result from changes in both central and peripheral circulation and muscle metabolism. The muscle adapts to use
energy more efficiently.
Anaerobic (sprint) training is high-intensity exercise
that lasts for a few seconds up to 2 minutes. These exercises rely primarily on adenosine triphosphate in the
form of phosphagens and on the anaerobic pathways.

Injury and Mechanisms of Repair

Fatigue-resistant Fatigable
High
Medium
Low
Medium

Most fatigable
Low
High

Small
High

Largest
Low

Larger
High

(Reproduced from Garrett WE Fr, Best TM: Anatomy, physiology, and mechanics of skeletal
muscle, in Simon SR (ed): Orthopaedic Basic Science. Rosemont, IL, American Academy of
Orthopaedic Surgeons, 1994, pp 89-125.)

increases, and collagen content increases with aging. The

loss of muscle size and strength can be diminished with
strength training.
Muscle loses contractile strength and mass with disuse or immobilization. As the muscle atrophies, both
macroscopic and microscopic changes occur, with decreasing fiber size and number, as well as changes in the
sarcomere length-tension relationship.
Changes that accompany immobilization of the muscle are related to the length at which muscles are immobilized. Atrophy and loss of strength are much more
prominent in muscles immobilized under no tension
than muscles placed under some stretch during immobilization. This greater tension induces a physiologic response to load, making the immobilization of muscles in
a lengthened position less deleterious. The profound
strength and mass reduction in inactive muscles can be
reversed quite quickly because of the excellent vascular
supply.
Muscle tissue is capable of significant adaptions.
Muscle training can involve exercises aimed at increasing strength, endurance, and anaerobic fitness. Strength
training (high force, low repetition) leads to an increase
in muscle strength, which is proportional to the crosssectional area of the muscle fibers. The increased muscle
fiber size results in an increased amount of contractile
proteins. There is also a strong neurologic component to
strength training. A poorly conditioned muscle has only
about 60% of the fibers firing simultaneously, whereas a
well-conditioned muscle has over 90% active fibers.
Endurance training, with low tension and high repetition, increases capillary density and mitochondria con-

American Academy of Orthopaedic Surgeons

Muscle injury can result from an indirect injury that

overpowers the muscles ability to respond normally, or
from a direct injury, such as a contusion or laceration.
The indirect mechanism of injury includes muscle
strains and delayed-onset muscle soreness (DOMS).
Muscle strains usually result during eccentric contractions, when the muscle is unable to accommodate the
stretch during the contraction. Muscles that cross two
joints, such as hamstrings, are more prone to strain injury. Muscle injuries most commonly occur at the myotendinous junction, and fatigue has also been associated
with increased rates of strain injury.
The severity of muscle injury varies from microscopic damage, to partial tears, to complete tears. The
healing process is initiated with an inflammatory phase.
Fibroblast proliferation and collagen production lead to
scar formation, with muscle regeneration resulting from
myoblasts stemming from satellite cells. Motion during
healing has been shown to reduce scar size, but healing
muscles are at increased risk for reinjury. Therefore, return to strenuous activity should be delayed until satisfactory healing has occurred.
DOMS is defined as muscle pain that generally occurs 24 to 72 hours after intense exercise. Muscle swelling and pain typically reach maximum levels after 1 to
3 days. Several treatment strategies have been introduced to help alleviate the severity of DOMS and to restore the maximal function of muscles as rapidly as possible. Nonsteroidal anti-inflammatory drugs have shown
dose-dependent effects that may also be influenced by
the time of administration. Massage has shown varying
results that may be attributed to the time of application
and the type of technique used. Cryotherapy, stretching,
homeotherapy, ultrasound, and electric current modalities have demonstrated no effect on the alleviation of
muscle soreness or other DOMS symptoms.
Myositis ossificans (MO) is a benign aberrant reparative process that results in heterotopic ossification in
soft tissues. Myositis ossificans is a misnomer because
the condition does not involve muscle inflammation,
and the process is not limited to muscle. MO circumscripta results from soft-tissue trauma, whereas MO
progressiva is an autosomal dominant genetic disorder.
About 80% of MO occurs in the thigh or arm. Ossifica-

23

Orthopaedic Knowledge Update 8

Soft-Tissue Physiology and Repair

tion occurs in three zones: the central zone is undifferentiated matrix, the surrounding area is composed of
immature osteoid, and the most peripheral zone is made
of mature bone. Treatment consists of immobilization
and the administration of anti-inflammatory drugs. Surgical intervention should only be considered when the
MO is completely mature (6 to 24 months) and is recommended when joint motion is compromised or neurologic impingement occurs.

Nerve
Structure and Function
The axon (conducts information by propagating electrical signals), dendrites (branches from the cell body that
receive signals from other nerve cells), presynaptic terminal (transmits information to cell bodies or dendrites
of other neurons), and the cell body (the metabolic center of the neuron containing the nucleus and organelles
for protein and RNA synthesis) are the main components of the neuron, or nerve cell. The action potential
that originates at the axon hillock is an all-or-none phenomenon. Proteins that support the structure and function of the axon are synthesized in the cell body and
travel along the axon via slow and fast antegrade transport systems. Axons with a myelin sheath that is produced by Schwann cells have a higher conduction velocity than noninsulated axons.
The axon and myelin sheath make up the nerve fiber. The fiber is enclosed with a basement membrane
and connective tissue layer, the endoneurium. The fibers
are further grouped into a bundle, called a fascicle,
which is surrounded by the perineurium. Several fascicles make up the peripheral nerve, which is encased by
the epineurium. The vascular supply of the peripheral
nerves has an intrinsic and extrinsic component. A vascular plexus within the endoneurium, perineurium, and
epineurium comprises the intrinsic component. The extrinsic component is made up of regional blood vessels
that enter the axon at various sites and travel along the
epineurium. The blood-nerve barrier, like the bloodbrain barrier, protects and maintains an appropriate endoneurial environment.
Peripheral nerves demonstrate similar viscoelastic
properties typical of other connective tissues. Their
stress-strain curve shows a linear region at higher stress
and follows a compliant, low-strained toe region. During
normal physiologic function, the peripheral nerves work
in the toe region of the curve. However, low levels of
strain can result in alterations in peripheral nerve conduction and blood flow impairment. Ischemic permanent changes are noted at strain rates as low as 15%.
Nerve repairs can place excess tension on the nerve at
the repair site, and joints at or near the repair site are
frequently immobilized to protect the repair.

24

Injury, Degeneration, and Regeneration

Peripheral nerve injuries can be categorized as (1) having no axonal discontinuity with only temporary loss of
nerve conduction, or as (2) having axonal damage with
proximal and distal degeneration of the axon. The reaction after the second category of injury occurs in two
phases. In the first phase, the axon and myelin sheath
disintegrate along the entire distance distal to the injury
and for some distance proximal to the injury. These
changes are termed wallerian degeneration and result in
denervation. Depending on the severity and location of
the injury, the cell body either regenerates the axon or
autodegenerates (resulting in cell death). Degeneration
of the myelin sheath results in loss of nerve conduction.
Neurapraxia is usually a result of compression and
involves loss of conduction across the injured site without wallerian degeneration. Because the axon is not injured, recovery is complete. Axonotmesis involves disruption of the axon with some preservation of the nerve
connective tissue. Wallerian degeneration occurs with
axonotmesis, but the endoneurial sheath remains intact
and serves to guide the regenerating axon during recovery. Neurotmesis is physiologic discontinuity of the
nerve and is divided into varying degrees of severity.
The likelihood of recovery decreases with each degree
of severity.
The Schwann cells of normal intact myelinated fibers do not divide. However, within 24 hours of nerve
transection, the Schwann cells throughout the distal segment undergo a series of mitosis. Schwann cells maintain cytoplasmic tubes, called bands of Bngner, which
guide the process of regeneration. These cells also produce nerve growth factor and nerve growth receptors.
Macrophages and Schwann cells are responsible for the
phagocytosis of debris.
Factors applied locally or systemically that enhance
regeneration include hormones, proteins, and growth
factors. Patient outcome after peripheral nerve injuries
is quite variable. Factors affecting the outcome of nerve
repair include patient age, the type of nerve injured,
length of the injury zone, timing of the repair, status of
the target organ at the time of repair, and technical skill
of the surgeon.

Nerve Repair
Peripheral nerve repair is used to establish the continuity of the nerve. The best results are achieved when the
repair is done soon after transection in a tension-free
environment. A primary repair requires adequate softtissue coverage, skeletal stability with low nerve tension,
and a good blood supply. However, nerve grafting may
be required with an extensive crush injury, or when excessive gap or tension is present. Mobilization of the
stump during repair should be limited because ischemia
and diminished profusion can result. Better results are

American Academy of Orthopaedic Surgeons

Chapter 2 Soft-Tissue Physiology and Repair

Orthopaedic Knowledge Update 8

typically seen in patients younger than age 50 years, and
in patients with more distal injuries.
There are two types of peripheral nerve repair:
grouped fascicular repair and simple epineurial repair.
Fascicular repair theoretically is preferred over epineurial repair because axon realignment can be more accurate. However, fascicular repair may require additional
dissection, which may result in increased scarring and
decreased vascular supply. A prospective study comparing epineurial to fascicular repair showed no difference
between the two groups.

Nerve Grafting
Nerve grafting is considered in injuries that have a large
gap that may result in undue tension after repair. Autogenous tissue (the sural nerve or the medial and lateral
antebrachial cutaneous nerves) is the most common
source used for nerve grafting. When possible, the fascicular groups on the injured nerve are matched to the
autogenous graft. A single segment is typically used for
grafting smaller nerves, whereas a few segments of graft
may be used with larger nerves. Alternatives to the autograft, including biologic or artificial conduits (such as
arteries, veins, muscle, collagen, or silicone tubes) and
allografts, which avoid donor-site morbidity and nerve
loss. The main complication with the use of allografts
has been the immunogenic host response. The rate of
axonal elongation has been shown to increase if the silicone tubes are filled with extracellular matrix proteins
such as collagen, laminin, and fibronectin.

Annotated Bibliography
Articular Cartilage
Borrelli J Jr, Tinsley K, Ricci WM, Burns M, Karl IE,
Hotchkiss R: Induction of chondrocyte apoptosis following impact load. J Orthop Trauma 2003;17:635-641.
The presence and extent of chondrocyte apoptosis following an impact load to articular cartilage was studied in an in
vivo model. The data suggested that there is a relationship between apoptosis and a single, rapid impact load and that the
extent of the apoptosis is related to the amount of load applied.

Hangody L, Fules P: Autologous osteochondral mosaicplasty for the treatment of full thickness defects of
weight-bearing joints: Ten years of experimental and
clinical experience. J Bone Joint Surg Am 2003;85(suppl
2):25-32.
Outcomes of 831 patients who underwent mosaicplasty
were evaluated with clinical scores, imaging techniques, and biopsy samples. Good-to-excellent results were achieved in 92%
of patients treated with femoral defects, 87% of those with tibial defects, 79% of those with trochlear and patellar mosaicplasties, and 94% of those treated for talar defects.

American Academy of Orthopaedic Surgeons

Hidaka C, Goodrich LR, Chen CT, Warren RF, Crystal

RG, Nixon AJ: Acceleration of cartilage repair by genetically modified chondrocytes over expressing bone morphogenetic protein-7. J Orthop Res 2003;21:573-583.
Large articular cartilage defects in the patellofemoral joint
of 10 horses were implanted with either genetically modified
chondrocytes, expressing BMP-7, or a control group of chondrocytes. The study group showed accelerated healing at
4 weeks and similar healing at 8 months when compared with
the control group.

Knutsen G, Engebretsen L, Ludvigsen T, et al: Autologous chondrocyte implanation compared with microfracture in the knee: A randomized trial. J Bone Joint
Surg 2004;86:455-464.
Autologous chondrocyte implantation was compared with
microfracture in a randomized trial. Short-term clinical results
were acceptable for both methods.

Leopold SS, Redd BB, Warme WJ, Wehrle PA, Pettis

PD, Shott S: Corticosteroid compared with hyaluronic
acid injections for the treatment of osteoarthritis of the
knee: A prospective, randomized trial. J Bone Joint Surg
Am 2003;85:1197-1203.
No significant differences in pain relief or function were
found in patients treated with corticosteroid compared with
hyaluronic acid injection.

Steadman JR, Briggs KK, Rodrigo JJ, Kocher MS, Gill

TJ, Rodkey W: Outcomes of microfracture for traumatic
chondral defects of the knee: Average 11-year follow-up.
Arthroscopy 2003;19:477-484.
Functional outcomes of 72 patients who were treated arthroscopically with microfracture for full-thickness traumatic
defects of the knee were evaluated. Patients under the age of
45 years who underwent this procedure, without associated
meniscus or ligament pathology, showed significant improvement in function.

Meniscus
Cole BJ, Carter TR, Rodeo SA: Allograft meniscal
transplantation: Background, techniques, and results.
Instr Course Lect 2003;52:383-396.
Meniscal allograft transplantation can be effective in
symptomatic meniscectomized patients, alleviating pain and
providing improved function.

Vangsness CT Jr, Garcia IA, Mills CR, Kainer MA,

Roberts MR, Moo TM: Allograft transplantation in the
knee: Tissue regulation, procurement, processing, and
sterilization. Am J Sports Med 2003;31:474-481.
This article is a review of the present issues that surround
the allograft industry, including regulation of tissues and tissue
banks, procurement, sterilization, and storage of allografts.

25

Orthopaedic Knowledge Update 8

Soft-Tissue Physiology and Repair

Yoldas EA, Sekiya JK, Irrgang JJ, Fu FH, Harner CD:
Arthroscopically assisted meniscal allograft transplantation with and without combined anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol
Arthros 2003;11:173-182.
Clinical and subjective outcomes of 31 patients (34 meniscal transplants) who either underwent an isolated meniscal
transplantation or a combined transplantation and ACL reconstruction were reviewed. Relief of symptoms was provided
in carefully selected patients with joint line pain and instability.

Ligament
Bogatov VB, Weinhold P, Dahners LE: The influence of
a cyclooxygenase-1 inhibitor on injured and uninjured
ligaments in the rat. Am J Sports Med 2003;31:574-576.
The cyclooxygenase-1 inhibitor was found to improve the
strength of the uninjured ligament, but was not found to improve the strength of ligament healing.

Takakura Y, Matsui N, Yoshiya S, et al: Low-intensity

pulsed ultrasound enhances early healing of medial collateral ligament injuries in rats. J Ultrasound Med 2002;
21:283-288.
The effect of low-intensity pulsed ultrasound on the healing of medial collateral ligament injuries was studied. This
method was shown to be effective in enhancing early healing.

Thornton GM, Shrive NG, Frank CB: Healing ligaments

have decreased cyclic modulus compared to normal ligaments and immobilization further compromises healing
ligament response to cyclic loading. J Orthop Res 2003;
21:716-722.
The cyclic stress-strain response of normal and healing ligaments to repetitive low loads was examined. Immobilization
significantly decreased the ability of scars to resist strain, with
the majority of immobilized scars failing during repetitive
loading.

Tendon
Oguma H, Murakami G, Takahashi-Iwanaga H, Aoki M,
Ishii S: Early anchoring collagen fibers at the bonetendon interface and conducted by woven bone formation: Light microscope and scanning electron microscope observation using a canine model. J Orthop Res
2001;19:873-880.
Light microscopy and scanning electron microscopy was
used to examine the process of anchoring of collagen fibers to
bone in a canine model. The formation of woven bone was important during the early recovery of the tendon-bone interface
before the completion of fibrocartilage-mediated insertion.

Soda Y, Sumen Y, Murakami Y, Ikuta Y, Ochi M: Attachment of autogenous tendon graft to cortical bone is better than to cancellous bone: A mechanical and histological study of MCL reconstruction in rabbits. Acta Orthop
Scand 2003;74:322-326.

26

The MCL of 33 rabbits was reconstructed using autogenous tendon graft to cortical or cancellous bone. At 8-week
follow-up, the cortical bone group showed a tendency to an increase in maximum failure load. With time, the tendons in
both groups matured.

Muscle
Babul S, Rhodes EC, Taunton JE, Lepawsky M: Effects
of intermittent exposure to hyperbaric oxygen for the
treatment of an acute soft tissue injury. Clin J Sport Med
2003;13:138-147.
Sixteen sedentary female university students were subjected to 300 maximal voluntary eccentric contractions. One
group was treated with 100% oxygen at 2 atmospheres and
the other was treated with 21% oxygen at 1.2 atmospheres absolute. No significant difference was found between the two
groups for pain, strength, or quadriceps circumference.

Cheung K, Hume P, Maxwell L: Delayed onset muscle

soreness: Treatment strategies and performance factors.
Sports Med 2003;33:145-164.
This review article on delayed onset muscle soreness examines the mechanisms of injury and treatment strategies.

Deal DN, Tipton J, Rosencrance E, Curl WW, Smith TL:

Ice reduces edema: A study of microvascular permeability in rats. J Bone Joint Surg Am 2002;84:1573-1578.
This study investigated the relationship between cryotherapy and edema by determining the microvascular permeability
before and after contusion in rats. The application of ice significantly decreased the microvascular permeability after striated
muscle contusion.

Nerve
Midha R, Munro CA, Dalton PD, Tator CH, Shoichet
MS: Growth factor enhancement of peripheral nerve regeneration through a novel synthetic hydrogel tube.
J Neurosurg 2003;99:555-565.
A synthetic hydrogel tube was used to repair surgically
created 10-mm gaps in the rat sciatic nerve. The tubes supported nerve regeneration in 90% of cases, and fibroblast
growth factor-1 enhanced tubes showed significantly better regeneration.

Wang S, Cai Q, Hou J, et al: Acceleration effect of basic

fibroblastic growth factor on the regeneration of peripheral nerve through a 15-mm gap. J Biomed Mater Res
2003;66A:522-531.
Nerve guides embedded with basic fibroblast growth factor were used in the repair of transected sciatic nerves (15-mm
gaps). The recovery and function of the regenerated nerves
was significantly accelerated by basic fibroblast growth factor,
as indicated by an electrostimulation test and histologic assays.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Classic Bibliography
Arnoczky SP, Warren RF: Microvasculature of the human meniscus. Am J Sports Med 1982;10:90-95.
Arnoczky SP, Warren RF, Spivak JM: Meniscal repair
using an exogenous fibrin clot: An experimental study in
dogs. J Bone Joint Surg Am 1988;70:1209-1217.
Buckwalter JA, Mankin HJ: Articular cartilage: Degeneration and osteoarthritis, repair, regeneration, and
transplantation. Instr Course Lect 1998;47:487-504.
Fu FH, Harner CD, Johnson DL, Miller MD, Woo SL:
Biomechanics of knee ligaments: Basic concepts and
clinical application. Instr Course Lect 1994;43:137-148.
Gelberman RH, Manske PR, Akeson WH, Woo SL,
Lundborg G, Amiel D: Flexor tendon repair. J Orthop
Res 1986;4:119-128.

Chapter 2 Soft-Tissue Physiology and Repair

Jackson DW, McDevitt CA, Simon TM, Arnoczky SP,
Atwell EA, Silvino NJ: Meniscal transplantation using
fresh and cryopreserved allografts: An experimental
study in goats. Am J Sports Med 1992;20:644-656.
Kline DG, Hudson AR, Bratton BR: Experimental
study of fascicular nerve repair with and without
epineurial closure. J Neurosurg 1981;54:513-520.
Walker PS, Erkman MJ: The role of the menisci in force
transmission across the knee. Clin Orthop 1975;109:184192.
Woo SL, Hollis JM, Adams DJ, Lyon RM, Takai S: Tensile properties of the human femur-anterior cruciate
ligament-tibia complex: The effects of specimen age and
orientation. Am J Sports Med 1991;19:217-225.

Inaba H: Biomechanical study on contact pressure in

the femoral-tibial joint. Nippon Seikeigeka Gakkai
Zasshi 1987;61:1073-1080.

American Academy of Orthopaedic Surgeons

27

Chapter

Bone Healing and Grafting

George F. Muschler, MD
Chizu Nakamoto, MD, PhD
Jonathan N. Grauer, MD

Introduction
An understanding of the principles of bone healing and
grafting is needed for the treatment of acute fractures,
repair of nonunions, filling of bone voids, healing of osteotomies, correction of angulatory or length deformities, and arthrodeses of the spine and extremities. The
effective use of a broad range of methods and strategies
is required to optimize outcome for each patient.
The formation of bone relies on three crucial processes. Osteogenesis is the ability of grafted cells to
form bone via osteoblastic stem cells and/or progenitor
cells. Osteoinductivity is the ability to modulate the differentiation of stem cells and progenitor cells along an
osteoblastic pathway. Osteoconductivity is the ability to
provide the scaffold on which new bone can be formed.
The many molecular, cellular, local, mechanical, and
systemic variables affecting bone healing must be effectively aligned for the bone healing response to be successful. The surgeon must identify settings in which
these conditions are deficient, and use appropriate
methods to optimize conditions. Bone tissue engineering
has evolved rapidly in the past decade, translating fundamental knowledge and developments in physics,
chemistry, and biology to achieve practical clinical benefits in the form of a variety of new materials, devices,
systems, and strategies.

Variables Affecting Bone Healing

The many variables that are crucial to bone formation
must be considered in the formulation of a treatment
plan for a fracture, the decision to proceed with an elective procedure, and the assembly of a construct during
surgery. Some variables are beyond the control of the
surgeon. However, other variables can be manipulated
to better achieve the desired orthopaedic goal.
Patient characteristics directly affect bone formation. The nutritional status of the patient must be maximized to withstand anabolic events and to limit complications such as infection. Patient use of substances such
as nicotine and nonsteroidal anti-inflammatory medications has been linked to decreased bone formation,

American Academy of Orthopaedic Surgeons

whereas other factors, such as parathyroid hormone,

have been suggested as having beneficial effects.
Local factors also alter the bone healing responses,
justifying an increased emphasis on careful handling of
soft tissue; specifically, limiting local physical, thermal,
and chemical trauma and preserving local blood supply.
Fracture treatment with limited exposure using intramedullary rods or sliding of plates across an injury
zone with minimal direction has been developed to better achieve this goal. The value of soft-tissue coverage
(possibly including skin/muscle flaps) has become more
apparent, particularly in patients with open tibia fractures, where the risk of nonunion is increased.
Mechanical factors must also be considered. Appropriate compressive forces across the site of potential
bone formation have been found to maximize the healing response. During fracture healing and with osteotomies, mechanical alignment must be restored. For example, in spinal fractures, the anterior column, which is
primarily under compression, generally heals more efficiently than the posterior column, which is under tension. Instrumentation and immobilization are effective
methods of holding desired alignment and favorably
modifying mechanical forces.

Phases of Bone Healing

The process of bone formation follows an orderly cascade of events. Initially, the tissue volume in which new
bone is to be formed is filled with a matrix, generally including a blood clot or hematoma. At this hematoma
stage, the matrix within the site is bordered by local tissues, which also are often traumatized resulting in focal
necrosis and reduced blood flow. An effective bone
healing response will include an initial inflammatory
phase, characterized by an increase in regional blood
flow, invasion of neutrophils and monocytes, removal of
cell debris, and degradation of the local fibrin clot.
These phases blend into a revascularization phase
that is associated with the invasion of new blood vessels
formed by endothelial progenitor cells, and subsequently a cell proliferation phase during which there is

29

Bone Healing and Grafting

multiplication of a highly proliferative population of
connective tissue progenitor cells (derived from regional
or circulating precursor cells). These progenitor cells are
capable of forming bone, but also may form fibrous tissue, cartilage, and possibly other tissues, depending on
local conditions at the site. A subsequent matrix synthesis phase coincides with the initial differentiation of
these progenitor cells.
Differentiation is strongly influenced by the local oxygen tension and mechanical environment, as well as by
signals from local growth factors. Regions of high strain
tend to promote fibrous tissue formation. Regions of
low strain and high oxygen tension are associated with
direct formation of woven bone (intramembranous
bone formation) whereas regions of intermediate strain
and low oxygen tension are associated with cartilage
formation. In many instances, cartilage formation may
add to the mechanical stability of the site, improving
strain conditions for regional bone formation, and may
subsequently be remodeled into bone tissue (endochondral bone formation).
Remodeling is the final phase of bone formation. It
is characterized by the systematic removal of the initial matrix and tissues that formed in the site, primarily
through osteoclastic and chondroclastic resorption, and
their replacement with more organized lamellar bone
aligned in response to the local loading environment.
The basic biologic processes and mechanisms associated with these phases of bone repair are similar in
closed fractures, bone graft sites, and sites of distraction
osteogenesis; however, the spatial and temporal relationship between these events may vary significantly depending on the setting. For example, in distraction osteogenesis, the initial inflammatory response is limited
to a relatively small tissue volume at the site of the corticotomy. Revascularization is rapid in the osteotomy
site. Distraction is begun after revascularization and
during the early proliferative phase. Through distraction, the proliferative phase of wound healing is prolonged in the central region of the regenerate, providing
a continuous source of progenitor cells that enter into
matrix synthesis and remodeling processes adjacent to
the distraction zone. Similarly, distraction osteogenesis is
associated with little or no cartilage formation. Angiogenesis at the site is generally rapid enough to prevent
regions of low oxygen tension, and strain at the site is
controlled.
The cascade of bone formation is also altered by
varying degrees of instrumentation and immobilization.
Rigid fixation and a low strain environment (500 to
2000 microstrain) tends to promote osteoblastic differentiation without formation of intermediary cartilage. In
contrast, regions of motion that produce hydrostatic
loading within a healing bone site tend to promote cartilaginous differentiation. Regions of tensile loading promote signaling that leads to fibroblastic differentiation.

30

Orthopaedic Knowledge Update 8

Mass Transport and Metabolic Demand in

Bone Healing
Access to substrate molecules (oxygen, glucose, amino
acids) and clearance of metabolic products (CO2, lactate, urea) are of critical importance to cell survival. The
processes by which molecules are moved in and out of a
fracture or graft site are collectively referred to as mass
transport. This process may be mediated by fluid flow
(convection) or by diffusion of molecules along concentration gradients or in response to electromagnetic
forces. Fluid flow in the circulatory system, as well as
flow induced by muscle contraction, gravitational pooling, and pulsation of arterial structures all are examples
of convection, and, most importantly, passive diffusion
along concentration gradients.
In native bone marrow tissues, the distance between
a cell and a capillary lumen is generally less than 40 m;
however in osteocytes this distance can be up to 200 m
and represents the distance over which extravascular
convection and diffusion must occur. At distances
greater than 200 m, the transport of oxygen and other
nutrients rapidly becomes a limiting factor for cell survival.
In nondisplaced fractures or corticotomies, the distance between osteogenic cells and a local blood supply
may be small, providing an effective mass transport system for much of the fracture hematoma. However, this
situation does not exist in displaced fractures or at most
sites of bone grafting. In these settings, the osteogenic
cells of the fracture hematoma or cells that are transplanted into a graft are subject to significant metabolic challenges. These challenges are imposed by the
imbalance between the rate at which oxygen and other
nutrients (such as glucose) diffuse into the site, and the
rate of consumption of oxygen and other nutrients by
other cells that are present in the site (metabolic demand).
The transport of oxygen into the center of most clinical graft sites must occur over relatively vast distances
(a minimum of 5 mm in most bone grafts, which is approximately 50 times the diffusion distance in normal
tissues). Because of this limitation, transplanted cells are
not likely to survive if they are more than 0.5 to 1.0 mm
from the surface of the graft. The depth of cells that survive is expected to be slightly better with less cellular
grafts, such as bone marrow aspirates; however, even in
these grafts the concentration of transplanted cells limits the predicted cell survival to a maximum depth of
approximately 3 mm. As a general rule, there is an inverse square relationship between cell concentration
and the depth of cell survival, such that fourfold reductions in the concentration of implanted cells will double
the depth of their survival within an implant.

American Academy of Orthopaedic Surgeons

Chapter 3 Bone Healing and Grafting

Orthopaedic Knowledge Update 8

Figure 2 Fracture healing in a rat femur immobilized with an intramedullary rod.

Bone formation, mediated by osteogenic cells, progresses at different rates and
through different pathways of differentiation and remodeling depending on local signals, vascularity, and mechanical forces. Enchondral ossification forms in regions of
higher strain and lower oxygen tension (open arrow). Direct intramembranous bone
formation occurs in regions of lower strain and higher oxygen tension (black arrow).
(Reproduced with permission from Einhorn T: The cell and molecular biology of fracture healing. Clin Orthop 1998;355 (suppl):S7-S21.)

Figure 1 This image of a 10-cm diameter tissue culture dish illustrates colony formation by cells derived from human bone marrow harvested by aspiration. These colonies
were allowed to grow for 9 days before being stained to label those colonies expressing alkaline phosphatase, a marker of early osteoblastic differentiation. The colonies
vary in size, illustrating the variation in biologic potential between individual progenitor
cells. The prevalence of osteoblastic colonies shown here is about average, approximately 50 colonies per million bone marrow derived nucleated cells, or 1 in 20,000.
(Reproduced with permission from Muschler GF, Nakamoto C, Griffith LG: Engineering
principles of clinical cell-based tissue engineering. J Bone Joint Surg Am 2004;86:
1541-1548.)

Osteogenesis
Biology
Stem cells are distinguished from progenitor cells by
their capacity to avoid being consumed or used up. This
characteristic is the result of activation through a process of self renewal, which is accomplished by a process
of asymmetric cell division that produces two daughter
cells. One daughter cell is identical to the original stem
cell and remains available to be activated again by an
appropriate signal. A second daughter cell continues to
divide and produces many additional progenitor cells
that ultimately give rise to new tissues. In contrast to the
stem cell, progenitor cells (also called transit cells) eventually develop into terminally differentiated cells.
Many adult musculoskeletal tissues contain stem and
progenitor cells that are capable of differentiation into
one or more mature cell phenotypes, including bone,
cartilage, tendon, ligament, fat, muscle, or nerve. These
cells are particularly concentrated in bone marrow, periosteum, vascular pericytes, and peritrabecular tissues
and are best characterized in bone marrow aspirates,
where they comprise approximately 1 in 20,000 of the
nucleated cells. Cellularity and the prevalence of precursor cells greatly varies between different individuals and

American Academy of Orthopaedic Surgeons

anatomic sites. Aging and gender are factors that have

been shown to account for only a small portion of this
variation (Figure 1).

Repair Mechanisms
The contribution of individual stem and progenitor cells
to new tissue formation is dependent on their activation, proliferation, migration, differentiation, and the
survival of their progeny. These cell functions are modulated in three-dimensional space by chemical and physical signals, including receptor-mediated signaling
through contact with ligands on other cells or through
ligands presented in the extracellular matrix (Figure 2).
However, repair of bone is also dependent on the number of stem and progenitor cells that are available to
participate in new tissue formation at a given site. Local
bone repair can be limited by a deficiency in the number of osteogenic stem or progenitor cells. This deficiency can be improved by transplantation of autogenous osteogenic cells from other sites, via grafts from
autogenous cancellous bone, bone marrow, periosteum,
or vascularized bone (such as fibula grafts).

Treatment Alternatives
Bone from the ilium is the most commonly used source
of progenitor cells. Grafts from the tibia and rib also are
used, although the number of progenitor cells from
these sources is less characterized than bone and marrow from the iliac crest. Autograft bone also provides
osteoinductive and osteoconductive properties; however, these desirable features are offset significantly by
the morbidity that is associated with the harvest of such
grafts.

31

Orthopaedic Knowledge Update 8

Bone Healing and Grafting

Osteoinductivity
Biology

Figure 3 Individual growth factors often promote osteoblastic differentiation at specific stages. The characteristic genes expressed at each stage are listed and the approximate stage of the principal action of the most active osteotropic growth factors
and hormones are shown. H4 = histone; AP = alkaline phosphatase; OP = osteopontin;
ON = osteonectin; BSP = bone sialoprotein; VDR = vitamin D receptor;
PDGF = platelet-derived growth factor; EGF = epithelial growth factor; bFGF = basic
fibroblast growth factor; BMPs = bone morphogenetic protein family members; TGF-
= transforming growth factor-beta; IGFs = insulin-like growth factors I and II; IL-6 =
interleukin-6; E2 = estradiol; Vit D = vitamin D3; PTH = parathyroid hormone; PTHrP =
parathyroid hormone-related peptide; Cbfa1 = transcription core binding factor-alpha
1; Coll X = collagen type X. (Reproduced with permission from Fleming JE, Cornell CN,
Muschler GF: Bone cells and matrices in orthopedic tissue engineering. Clin Orthop
North Am 200;31: 357-374.)

Bone marrow aspirates have recently become much

more widely used as a safer and less costly alternative
source of transplantable osteogenic cells. Anticoagulated bone marrow can be further manipulated to selectively concentrate osteogenic cells and remove cells that
are unlikely to contribute to a bone healing response
(such as red blood cells and most lymphocytes). Simple
centrifugation and buffy coat isolation can increase the
concentration of marrow-derived nucleated cells by as
much as fourfold. Selective adherence and retention on
some implantable matrices (such as Cellect, DePuy, Warsaw, IN) can also improve local bone repair by increasing the concentration of target cells and potentially improving their survival by reducing the number of
nonosteogenic cells, which then reduces competition for
oxygen and other substrates.
Future sources of osteogenic cells for transplantation
may include cells that are harvested from fat, muscle, or
marrow, and then expanded in tissue culture. This technique can be used to increase the number of cells available for transplantation. In vitro expansion also may
provide an opportunity to modify osteogenic potential
by the selection of optimal subpopulations, selective differentiation, or by genetic engineering techniques. However, these techniques and their potential to improve efficacy will need to be balanced against their cost and
their potential to generate undesirable cell behavior.

32

Osteoinductive materials include a variety of soluble

growth factors that can be used to modulate the behavior
of the osteogenic cells to become activated, to proliferate, and to differentiate into cells that make new bone tissue rather than into cells that remain quiescent or differentiate into other tissues. This multistep process involves
biologic signals expressed in a spatially and temporally
regulated manner during the fracture healing and bone
repair processes. Individual signals can have specific effects on the recruitment, proliferation, differentiation,
and survival of osteoblastic progenitors (Figure 3).
The ability of proteins extracted from demineralized
bone matrix to induce bone formation was first described by Urist in the 1960s. These observations led to
the discovery of a family of signaling proteins now
known as bone morphogenetic proteins (BMPs) that are
part of the larger transforming growth factor-beta
(TGF-) superfamily. Fourteen BMPs are now recognized (BMP-2 through BMP-15). These proteins are secreted as homodimers or heterodimers of 110 to 140
amino acid peptides linked by one disulfide bond (molecular weight of approximately 30 kd).
Other growth factors also can contribute to osteoinduction. Both epidermal growth factor and plateletderived growth factor are capable of inducing colony
formation by osteoblastic stem and progenitor cells. Vascular endothelial growth factors also play several important roles in angiogenesis, osteoclast migration, and osteoblastic activity. Although vascular endothelial growth
factors and TGF- do not induce bone formation directly, they have been shown to act synergistically to enhance BMP activity in fracture healing and during distraction osteogenesis.

Repair Mechanisms
In vivo, these growth factors have autocrine and paracrine effects as soluble factors. Many growth factors are
also embedded in bone matrix where they are believed
to play a role in bone remodeling and the coupling of
osteoclastic and osteoblastic activity. TGF- is most
abundant and is found in bone matrix at a concentration
of approximately 1 mg/kg. The effects of these growth
factors can only be mediated on cells expressing specific
cell surface receptors. For example, cells must express
both a type I and a type II BMP receptor (serine/
threonine kinases) to be responsive to BMPs. To date,
three type I and three type II BMP receptors have been
identified. Each BMP shows a unique binding pattern
for the individual receptors, some of which are also responsive to other TGF- family members. Receptor activation by both TGF- and BMPs results in activation
of an intracellular signaling cascade involving SMAD
pathway leading to changes in gene expression.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

In vitro, BMPs influence differentiation and other
cell functions in a dosage range from 1 to 100 ng/mL.
Wheras BMPs often appear to share similar functions
and act through a limited set of receptors, each is independently regulated in both space and time during embryonic development and fracture healing. The activity
of BMPs is also modulated by specific inhibitors (such
as noggin, chordin, gremlin, and follistatin).

Treatment Alternatives
Demineralized bone matrix preparations are widely
available for clinical use. Because of their mild osteoinductive potential and desirable osteoconductive properties, they are frequently used as bone graft supplements
or secondary extenders, or as a means for transplanting
bone marrow derived cells. Nonetheless, different demineralized bone matrix preparations appear to vary in
biologic activity, depending on the quality of the donor
bone and on the bone processing method. Currently,
there is no generally accepted method for proving or
screening for biologic activity, though several in vitro
and in vivo assays are available. In this regard, demineralized bone matrix preparations lag behind the regulatory standards that have been applied to the purified recombinantly manufactured human osteoinductive
growth factors (such as BMPs) that are becoming available for clinical use.
Available data clearly show that BMPs will provide
valuable tools for enhancing bone healing in some settings. Among the individual recombinant human BMPs
(rhBMPs), BMP-2 and BMP-7 (known as osteogenic
protein-1 or OP-1) have both been developed and are
now approved by the Food and Drug Administration for
use in a limited set of clinical applications. BMP-14
(known as MP52 and growth and differentiation factor-5
or GDF-5) is also under development. BMP-2 in a collagen carrier called InFuse(Medtronic, Minneapolis,
MN) has been evaluated and approved for use in the
setting of anterior interbody fusion using specific metallic and allograft devices. The OP-1 Device (Stryker Biotech, Hopkinton, MA) has been approved under a humanitarian device exemption for use in tibial nonunions
where conventional treatment has failed. Clinical trials
of these and other BMP preparations are ongoing.
One of the principal requirements for optimal performance of BMPs or other growth factors is the presence of a local population of target precursor cells that
express appropriate receptors. For a BMP to be optimally effective, these target cells must be activated in
sufficient numbers to produce the desired result. If an
optimal number of responsive cells are not present, the
biologic response to the protein will be reduced and implantation of a BMP may be ineffective. Very high concentrations of BMP are required to induce a clinically
useful volume of bone in higher animals and humans.

American Academy of Orthopaedic Surgeons

Chapter 3 Bone Healing and Grafting

The current formulations of BMPs deliver approximately 50 times the total amount of BMP that is present
in the human skeleton.
These massive doses of BMPs are currently required
for many reasons, including species-specific differences
in dose response to BMPs, the lower concentration of
osteogenic cells in higher order animals, the decrease in
the concentration of osteogenic cells in human tissues
that occurs with advancing age, and other clinical factors. It is also related to the inefficiency of delivery associated with a declining surface to volume ratio as the
size of graft sites increases. As graft sites increase in
size, more of the implanted growth factor is placed farther away from tissue surfaces where target cells are located. As a result, overdose molecules of BMP become
less likely to interact with a responding cell. This situation results in a progressive decrease in surface-tovolume ratio in larger graft sites. Assuming penetration
of the BMP to the same depth within tissue, increases in
graft volume result in activation of proportionately
fewer target cells per unit volume. Increasing the dose
(BMP concentration in the defect) may increase the
depth penetration of the signal, partly offsetting this
limitation. The concentration or biologic potential of
target cell populations vary in individual patients, which
presents another challenge. Also, the concentration and
prevalence of targeted stem and progenitor cells and
their biologic performance varies significantly from tissue to tissue (bone, bone marrow, periosteum, muscle,
fat) and site because of local or systemic disease. Increasing the BMP dosage may compensate for some of
this variation.
Other opportunities for further advancement and refinement of the use of BMPs and other growth factors
include modification of their molecular structure, dosage, carrier, formulation, delivery kinetics, and on-site
retention.

Osteoconductivity
Three-dimensional porous scaffolds are used extensively
in bone grafting and skeletal reconstruction. The principal function of these implants is to provide a surface
and structure that facilitates the attachment, migration,
proliferation, differentiation, and survival of osteogenic
stem and progenitor cells throughout the implant site.
Scaffolds also may serve as a space holder to prevent
other tissues from occupying the space where new tissue
is desired, improve local mechanical stability (as a supportive block, stent, or strut), and facilitate revascularization.
Structural properties refer to the spatial distribution
of bulk material of the scaffold (the scaffold architecture). These features can be at nanoscale, microscale,
and macroscale (the molecular, cellular, and tissue levels, respectively). Marcoporous features of the scaffold

33

Bone Healing and Grafting

Figure 4 This image shows the macroporous architecture of allograft cancellous

bone, which provides a large interconnected void volume for fluid flow and revascularization and progenitor cell attachment migration, proliferation, and differentiation. The
microporous texture on the surface of the allograft matrix is also shown. Nanostructural features and porosity are too small to be seen in this image. (Reproduced with
permission from Bullough PG, Vigorita VJ: Slide Atlas of Orthopaedic Pathology. New
York, NY, Gower Medical Publishing, 1984.)
define the initial void space that is available for cell migration and ingrowth of new blood vessels. In most settings, where bone ingrowth is needed to depths of 3 to 5
mm, a macropore size between 150 and 1,000 m is optimal. Microstructural features define the surface texture
that cells encounter, which can have important biologic
effects on cell functions. Features at all levels may influence fluid flow and diffusion of oxygen and other nutrients through the scaffold (Figure 4).
The mechanical properties (such as strength, modulus, toughness, ductility) of available scaffolds vary
widely, and are determined by the mechanical properties of the bulk material (material properties) and the
structural properties of the implant. Mechanical properties of a scaffold must be matched to the graft environment and demands. For example, the ability to accommodate early loading may be crucial in some situations
(interbody spinal struts) but not necessary in others
(posterolateral lumbar fusions).

Repair Mechanisms
General Concepts
The conditions at the host-implant interface are defined
by the chemical properties of the implant and by the local response of the host cells and tissues. Biomolecules
derived from the host fluids and tissues are rapidly adsorbed and become the principal mediators of the cellular response to the material, modulating cellular attachment, migration, proliferation, and differentiation
events.
The resorbability of plasticizing agents often has important early effects. Low molecular weight soluble
compounds (such as glycerol and calcium sulfate) create
a hyperosmotic environment that will prevent early cell

34

Orthopaedic Knowledge Update 8

survival in or adjacent to the implant. Dissolution or
degradation may also release acidic by-products (for example, sulfate or carbonate), creating a nonphysiologic
pH in the region in and around the implant. These conditions preclude the use of some materials as a vehicle
for transplantation of viable cells.
Scaffolds that do not degrade (for example, metals
and some ceramics) avoid complications associated with
early solubility or later degradation and can have excellent and durable function when integrated with tissues.
However, these same materials also can compromise
later function in some settings by disrupting tissue remodeling in a way that compromises long-term mechanical function. Stress shielding or stress concentration
around these implants can result in local bone loss or in
increased risk of mechanical failure. Durable materials
also may physically obstruct or complicate future intervention (such as revision surgery). These considerations
have led to interest in biologically resorbable scaffolds
whenever feasible. One example is the use of impaction
grafting for reconstruction of contained and protected
periprosthetic defects. Another example is the recent
shift in bone void fillers from very slowly degradable ceramics (such as hydroxyapatite) to more rapidly resorbed materials (such as tricalcium phosphate, calcium
sulfate, or calcium carbonate).
During the process of degradation of a resorbable
implant, the biologic environment adjacent to the implant will be defined by the degradation products that
are produced and the biologic effects these products
have on local cells. The concentration of degradation
products is determined by the balance between their
rate of release into the graft site and their rate of clearance from the site. Polyesters, such as polylactides and
polyglycolides, are the most commonly used synthetic
degradable surgical materials (surgical meshes, suture
anchors, and screws). However, the release profile of
these materials remains imperfect, despite the fact that
degradation can be extended over a range of weeks to
years because these materials are degraded by hydrolysis. Early degradation of these materials is characterized
by random hydrolysis of long polymeric chains resulting
in a progressive reduction of the molecular weight of
the bulk material and reduction of its material properties. However, the total mass of the polymer remains in
the site until the molecular weight of the fragments that
are created is small enough to make them soluble. When
this occurs, soluble material can be generated rapidly,
liberating the bulk polymer into solution but creating a
profound local decrease in pH. This process is the origin
of late local osteolysis and the formation of sterile cysts
in bone and soft tissue that have been observed after
implantation of these materials. These limitations are
beginning to be addressed by the development of new
materials that are degraded by gradual surface erosion

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

(often cell mediated) without the risk of accelerated release of degradation products.

Treatment Alternatives
The most commonly used scaffold materials are derived
from allograft bone and are available in a broad range
of physical forms including structural blocks, wedges,
rings, dowels, and nonstructural chips, fibers, and powders. Nonstructural allograft preparations have been
combined with plasticizing agents, such as carboxymethylcellulose, hyaluronan, and glycerol, to improve handling properties. Nonbone tissue-derived scaffolds have
been introduced, including allograft and xenograft skin
matrix, and xenograft intestinal submucosa. Biologic
polymers (for example, collagen, hyaluronan, chitin, or
fibrin), ceramic or mineral-based matrices (for example,
tricalcium phosphate, hydroxyapatite, or calcium sulfate), metals (for example, titanium, tantalum, or cobaltchromium-molybdenum alloy), and composites of two
or more of these materials are also widely used as scaffold materials for tissue ingrowth and regeneration. Synthetic polymer materials (for example, polylactide,
polyglycolide, polytyrosine carbonates, polycaprolactone) varying copolymers, and synthetic gel-like polymers (polyethylene oxide-based) are also being adapted
from other surgical uses or developed de novo to improve properties of mechanical performance, degradation, and cell interaction. Other scaffold materials include hybrids of biologically-derived polymers such as
alginate or fibrin that are chemically modified with cell
adhesion peptides or growth factors. Some of these new
scaffold materials are in advanced stages of development in animal studies.
Options for the structural distribution of scaffolds
are almost infinite. The macro structures include regular
geometric shapes (blocks, pellets, and dowels), amorphous structures (randomly packed chips, granules, or fibers), randomly integrated structures (foams or freeze
dried materials), and formally designed regular structures (machined, printed, woven, or assembled structures). Gel or putty preparations can also be made from
powders or fibers by mixing them with plasticizing
agents (such as glycerol, methylcellulose, and hyaluronan).
In some instances, a desirable structure or porosity
can be derived from nature. Allograft bone matrix can
vary significantly according to sample and donor; however, through selective processing (machining, size and
density selection, washing, demineralization), a variety
of relatively uniform and optimized allograft materials
can be made for specialized clinical settings (for example, chips, powders, struts, dowels, rings, wedges, screws).
During the past decade, significant advances have been
made in methods for producing more precise hierarchi-

American Academy of Orthopaedic Surgeons

Chapter 3 Bone Healing and Grafting

cal microstructures from a variety of materials (tantalum, titanium, collagen, synthetic polymers, ceramics),
which are now being applied to the creation of devices
with strategically oriented channels and pores.
Many new types of scaffolds also are being developed to optimize bulk characteristics, three-dimensional
architecture and porosity, mechanical properties, surface
chemistry, initial scaffold environment (osmolarity, pH)
and late scaffold environment (degradation characteristics). Each property has important implications in the
biologic response to a scaffold and its usefulness in supporting the transplantation, homing, or biologic activity
of local stem cells and progenitor cells.

The Evolving Science Leading to Optimized

Bone Healing
Bone healing is an intricate process with many contributing variables. Tools for understanding and manipulating the bone healing process and for optimizing successful bone grafting are becoming increasingly available in
the clinical setting. As a result, autogenous bone grafting
is slowly being replaced as the gold standard against
which all other bone graft alternatives are compared.
Although no material is clinically more effective than
autograft, the significant morbidity involved in harvesting these grafts creates the need for alternative materials that can provide comparable efficacy.
The translation of new materials and strategies into
clinical practice requires significant basic benchtop science, preclinical assessment in animal models, and clinical assessment in either targeted cohort studies or prospective randomized trials. This stepwise approach is
necessary to provide the burden of proof that is needed
to support wide adoption of these new methods.
In addition to implantable or injectable bone graft
substitutes, additional methods of facilitating bone healing and fusion remain equally important. Other methods for promoting bone healing include optimizing patient nutrition, patient cessation of tobacco use, and
application of graded mechanical loading (for example,
progressive weight bearing or dynamization of internal
or external fixation). Biophysical stimulation using electromagnetic field stimulation or ultrasound also may be
effective in reducing healing time and increasing the
rate of union in selected settings.
It is important to make clinical decisions based on
sound biologic and mechanical principles and objective
data. To ensure the most predictable outcome and rapid
patient recovery, attention to clinically proven guidelines, careful patient assessment, monitoring, and the reporting of clinical outcomes are required.

35

Bone Healing and Grafting

Annotated Bibliography
Mass Transport and Metabolic Demand in
Bone Healing
Muschler GF, Nakamoto C, Griffith LG: Engineering
principles of clinical cell-based tissue engineering.
J Bone Joint Surg Am 2004;86-A:1541-1548.
This article reviews the current state of cell-based tissue
engineering, the central engineering principles and strategies
involved in the design and use of cell-based tools and strategies, and examines the challenges of mass transport.

Osteogenesis
Muschler GF, Midura RJ, Nakamoto C: Practical modeling concepts for connective tissue stem cell and progenitor compartment kinetics. J Biomed Biotechnol 2003;
2003:170-193.
Current concepts in stem cell biology, progenitor cell
growth, and differentiation kinetics are reviewed in the context of bone formation. A cell-based modeling strategy is developed and offered as a tool for conceptual and quantitative
exploration of the key kinetic variables and organizational hierarchies in bone tissue development and remodeling and in
tissue engineering strategies for bone repair.

Osteoinductivity
Burkus JK, Gornet MF, Dickman CA, et al: Anterior
lumbar interbody fusion using rhBMP-2 with tapered
interbody cages. J Spinal Disord Tech 2002;15:337-349.
The findings of a multicenter, prospective, randomized,
nonblinded study of patients with degenerative lumbar disk
disease who underwent interbody fusion using two tapered
threaded fusion cages is presented. The investigational group
(143 patients) received rhBMP-2 on an absorbable collagen
sponge. A control group (136 patients) received autogenous iliac crest bone graft. At 24-month follow-up, patients in the
BMP-2 group showed a fusion rate of 94.5%. The patients in
the autograft group showed a fusion rate of 88.7%.

Cheng H, Jiang W, Phillips FM, et al: Osteogenic activity

of the fourteen types of human bone morphogenetic
proteins (BMPs). J Bone Joint Surg Am 2003;85-A:15441552.
A comprehensive analysis of the osteogenic activity of 14
types of BMPs in osteoblastic progenitor cells is presented.
BMP-2, BMP-6, and BMP-9 may play an important role in inducing osteoblast differentiation of mesenchymal stem cells.
Most BMPs are able to stimulate osteogenesis in mature osteoblasts.

Einhorn TA: Clinical applications of recombinant human BMPs: Early experience and future development.
J Bone Joint Surg Am 2003;85-A(suppl 3):82-88.
A review of BMPs and their clinical application is presented.

36

Orthopaedic Knowledge Update 8

Friedlaender GE, Perry CR, Cole JD, et al: Osteogenic
protein-1 (bone morphogenetic protein-7) in the treatment of tibial nonunions. J Bone Joint Surg Am 2001;83
(suppl 1):S151-S158.
A multicenter, prospective, randomized study of patients
with established tibial nonunions who underwent treatment
with bone grafting and intramedullary rod fixation is presented. Patients in the investigational group (63 fractures) received rhOP-1 on a type I collagen carrier. Patients in the control group (61 fractures) received an autogenous iliac crest
bone graft. At 9 months, 75% of the fractures treated with
OP-1 had healed compared with 84% of the fractures treated
with autograft. The rhOP-1 group had a significantly lower incidence of postoperative infection and no donor site pain.

Govender S, Csimma C, Genant HK, et al: Recombinant

human bone morphogenetic protein-2 for treatment of
open tibial fractures: A prospective, controlled, randomized study of four hundred and fifty patients. J Bone
Joint Surg Am 2002;84-A:2123-2134.
This article presents the findings of a prospective, randomized, single-blind study of patients treated for open tibial fractures using intramedullary nail fixation. In the study group, patients were treated with an rhBMP implant (doses of 6 mg or
12 mg in an 8 mL collagen carrier) at the time of wound closure. The control group did not receive an implant. Patients receiving the 12 mg dose of rhBMP had significantly fewer subsequent procedures, more rapid union, fewer hardware
failures, fewer infections, and faster wound healing when compared with the control group or the group who received the
6-mg dose of rhBMP.

Muschler GF, Nitto H, Matsukura Y, et al: Spine fusion

using cell matrix composites enriched in bone marrowderived cells. Clin Orthop 2003;407:102-118.
Posterior spinal fusion model and cancellous bone matrix
was used to compare an enriched cellular composite bone
graft alone, bone matrix plus bone marrow clot, and an enriched bone matrix composite graft plus bone marrow clot.
The union score, fusion volume, and fusion area for the enriched bone matrix plus bone marrow clot composite were superior to the enriched bone matrix alone and the bone matrix
plus bone marrow clot. The addition of a bone marrow clot to
an enriched cell-matrix composite graft results in significant
improvement in graft performance.

The Evolving Science Leading to Optimized Bone

Healing
Hodges SD, Eck JC, Humphreys SC: Use of electrical
bone stimulation in spinal fusion. J Am Acad Orthop
Surg 2003;11:81-88.
A review of the effect of three types of electrical stimulation (direct current electrical stimulation, pulsed electromagnetic fields, and capacitively coupled electrical stimulation) on
spinal fusion is presented. Direct current electrical stimulation
showed positive effects although there was difficulty in deter-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

mining the end point of fusion and incorporating the effect of
patient parameters.

Rubin C, Bolander M, Ryaby JP, Hadjiargyrou M: The

use of low-intensity ultrasound to accelerate the healing
of fractures. J Bone Joint Surg Am 2001;83-A:259-270.
A review of the efficacy of low-intensity ultrasound in
fracture management is presented.

Classic Bibliography
Aronson J: Limb-lengthening, skeletal reconstruction,
and bone transport with the Ilizarov method. J Bone
Joint Surg Am 1997;79:1243-1258.
Bauer TW, Muschler GF: Bone graft materials: An overview of the basic science. Clin Orthop 2000;371:10-27.
Boden SD, Zdeblick TA, Sandhu HS, et al: The use of
rhBMP-2 in interbody fusion cages: Definitive evidence
of osteoinduction in humans. A preliminary report.
Spine 2000;25:376-381.

Chapter 3 Bone Healing and Grafting

Green SA: Ilizarov method. Clin Orthop 1992;280:2-6.
Muschler GF, Boehm C, Easley K: Aspiration to obtain
osteoblast progenitor cells from human bone marrow:
The influence of aspiration volume. J Bone Joint Surg
Am 1997;79:1699-1709.
Rhinelander FW: Tibial blood supply in relation to fracture healing. Clin Orthop 1974;105:34-81.
Stevenson S: Biology of bone grafts. Orthop Clin North
Am 1999;30:543-552.
Urist MR: Bone: Formation by autoinduction. Science
1965;150:893-899.
Wiss DA, Stetson WB: Tibial nonunion: Treatment alternatives. J Am Acad Orthop Surg 1996;4:249-257.
Wozney JM, Rosen V, Celeste AJ, et al: Novel regulators
of bone formation: Molecular clones and activities.
Science 1988;242:1528-1534.

Connolly JF: Clinical use of marrow osteoprogenitor

cells to stimulate osteogenesis. Clin Orthop 1998;
355(suppl):S257-S266.

American Academy of Orthopaedic Surgeons

37

Chapter

Musculoskeletal Biomechanics
Vijay K. Goel, PhD
Ashutosh Khandha, MS
Sasidhar Vadapalli, MS

Introduction
Biomechanics involves the use of the tools of mechanics
(the branch of physics that analyzes the actions of forces)
in the study of anatomic and functional aspects of living
organisms. The musculoskeletal interactions are a good
example of a mechanical system. The primary functions
of the musculoskeletal system are to transmit forces from
one part of the body to another and to protect certain organs (such as the brain) from mechanical forces that could
result in damage. Therefore, the principal biologic role of
skeletal tissues is to bear loads with limited deformation.
To appreciate the mechanical functions that these tissues
must perform, it is necessary to understand the forces that
whole bones normally carry. In most cases, these forces result from loads being passed from the part of the body in
contact with a more or less rigid environmental surface
(such as the heel and the ground surface during walking)
through one or more bones to the applied or supported
load (such as the torso). In addition to the forces transmitted through bone-to-bone contact, large and important
forces exerted by the muscles and ligaments act on the
bones. Most muscle, ligament, and bone-to-bone contact
forces act in or near the bodys major diarthrodial joints.
Basic engineering concepts used in the conventional analysis of mechanical systems and their principles can be applied to the musculoskeletal system.

Concepts and Terminology of Mechanics

Scalars, Vectors, and Tensors
In mechanics, two basic quantities are defined. Scalar
quantities (such as mass, temperature, work, and energy)
have magnitude. A vector quantity (such as force, velocity, and acceleration) has both magnitude and direction.
Unlike scalars, vector quantities add in a tip to tail fashion.
Consider two force vectors, A and B. The length of
the vectors represents the magnitude and arrows point
in the direction of the forces. To replace these two vectors with a single vector that has the same effect on the
body, the two vectors are drawn tail to tail. Next, lines
are drawn parallel to each vector to create a parallelo-

American Academy of Orthopaedic Surgeons

gram. The resultant force (or the force that can replace
the two vectors and still have the same effect on the
body as the two original vectors) is the diagonal of the
parallelogram (R).
An example of vector resolution to predict contact
force at the knee joint is shown in Figure 1. If the direction and magnitude of the muscle and tendon forces are
as shown in the figure, the resultant joint contact force
can be found by using the parallelogram law. The resultant joint contact force is at a 45 angle to the line of action of the tendon force and its magnitude is given by
the vector addition (the square root of the squares of
the two forces), which is 71 lb.
Tensors, defined mathematically, are simply arrays of
numbers, or functions, which transform according to certain rules, under a change of coordinates. In physics, tensors characterize the properties of a physical system. A
tensor consisting of a single number is referred to as a
tensor of order zero, or simply, a scalar. A tensor of order one, known as a vector, is a single row or a single
column of numbers. Tensors can be defined to all orders.
The next order above a vector are tensors of order two,
which are often referred to as matrices. Second order
components are not only associated with magnitude and
direction, but they are also dependent on the plane over
which they are determined. The components of a second
order tensor can be written as a two-dimensional array
consisting of rows and columns as shown below:

xx xy xz

yx yy yz

zx zy zz

Just as vectors represent physical properties more complex than scalars, matrices represent physical properties
more complex than can be described by vectors.

Force, Mass, and Weight

Force may be defined as mechanical disturbance or load
and is associated with the result of muscle activity.

39

Musculoskeletal Biomechanics

Orthopaedic Knowledge Update 8

Figure 1 Vector resolution to calculate joint contact force at the knee.

When an object is pulled or pushed, force is applied to
the object. Force is also exerted when a ball is thrown or
kicked. Force acting on an object can deform the object
and/or change its state of motion. Force is also defined
as a vector quantity and is equal to mass multiplied by
the acceleration. The unit of force is the newton.
A force may be broadly classified as external or internal. When a nail is hit with a hammer, an external
force is applied to the nail. Internal forces are those that
hold a body together when the body is under the effect
of externally applied forces. If the human body is considered as a whole, then the forces generated by muscle
contractions are the internal forces.
A normal force acts in a direction perpendicular to a
particular surface. A tangential force acts in a direction
parallel to the surface. In Figure 2 the weight (W) of the
sliding block acts on the floor as shown. The floor applies a counteracting normal force (N) on the block. The
frictional force ( f ), which develops between the block
and the floor, is an example of a tangential force.
A compressive force tends to shrink the body in the
direction of the applied force, whereas a tensile force
tends to stretch or elongate the body in the direction of
the applied force. Muscles contract to produce tensile
forces that pull together the bones to which they are attached. Muscles cannot produce compressive forces.
Mass and weight are terms that are often confused.
Mass is the amount of matter present in a body and is
an intrinsic property of the body; the mass of an object
always remains the same. Weight, however, is the force
that a given mass feels because of the gravity at its
place. Mass is measured in kilograms, whereas weight is
measured in units of force such as the newton.

Figure 2 Normal and tangential forces when a block slides.

steering wheel of a car. A bending moment is said to be
applied to a structure when the force plane is parallel to
the long axis of the structure. An example of a bending
moment is the bending of a diving board caused by the
weight of a diver. However, the mathematical definition
of moment and torque is the same.
The magnitude of the moment of a force about a
point is equal to the magnitude of the force times the
length of the shortest distance between the point and
the line of the action of the force, which is known as the
lever or moment arm. Figure 3 shows how the moment
caused by a force acting on the wrist is calculated about
a point on the elbow. The direction of moment is per-

Moment and Torque Vectors

In general, torque is associated with the rotational and
twisting action of applied forces, whereas moment is related to its bending effect. A torque is said to be applied
to a structure when the plane in which force acts is perpendicular to the long axis of the structure. Examples of
torque are forces applied on a screwdriver and the

40

Figure 3 Moment of force acting on the wrist about a point on the elbow. (Reproduced with permission from Mow VC, Hayes WC: Analysis of muscle and joint loads,
in Basic Orthopaedic Biomechanics, ed 2. Philadelphia, PA, Lippincott-Raven, 1997,
p 3.)

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 4 Musculoskeletal Biomechanics

it is termed as angular velocity. Velocity is similar to
speed, except that speed is a scalar quantity, whereas velocity is a vector because it has direction in addition to
its magnitude.
Acceleration is defined as the rate of change of velocity and can be linear or angular. An example of linear
acceleration is a driver pressing the accelerator of a car.
Angular acceleration is the rate of change of angular velocity.

Statics and Dynamics

Figure 4 Change in moment arm changes the value of moment.
pendicular to the plane on which the point and force lie,
and is found by the right hand rule. The right hand rule
states, When the fingers of the right hand curl in the direction that the applied force tends to rotate the body
about a point, the right thumb points in the direction of
the moment vector.
Figure 4 shows how a change in the moment arm
changes the moment. When the moment is calculated
about x, the muscle force has a moment arm of 2 inches and therefore, the moment is 100 in-lb. However,
when the force has a moment arm of 1 inch, the moment reduces to 50 in-lb. The change in moment arm
can be the result of surgery.
A couple or pure moment is a system of forces
whose resultant force is zero. The system typically consists of two parallel coplanar forces of equal magnitude
and opposite directions. The magnitude of the couple is
given by the equation: C = F r, where F is the force
and where r is the moment arm. The magnitude and direction of the couple does not change as the point about
which the couple changes, because the moment arm
stays the same.

Pressure
Pressure (P) is defined as the amount of force (F) acting
over a given area (A). Pressure is commonly measured
in Pascal (Pa) or pounds per square inch (psi or lb/in2).
It is a scalar quantity. Stress is measured in the same
units as pressure; however, stress is a second order tensor.

Displacement, Velocity, and Acceleration

Displacement is a vector measure of the interval between two locations measured along the path connecting them.
Velocity is defined as the rate of change of position
of a point or rigid body with respect to a coordinate system or another body. If the motion is translational, then
it is termed as linear velocity. If the motion is rotational,

American Academy of Orthopaedic Surgeons

Statics and dynamics are two major subbranches of mechanics. Statics is an area within the field of applied mechanics dealing with the analysis of rigid bodies in equilibrium. The term equilibrium implies that the body of
concern is either at rest or moving with a constant velocity. Dynamics is the study of systems in which acceleration is present.

Newtons Laws of Mechanics

Newtons first law states that a body that is originally at
rest will remain at rest, or a body moving with a constant velocity in a straight line will maintain its motion
unless an external force acts on the body.
Newtons second law states that if the net force acting on a body is not zero, then the body will accelerate
in the direction of the resultant force. Furthermore, the
magnitude of the acceleration of the body will be directly proportional to the magnitude of the net force
acting on the body and inversely proportional to its
mass. Newtons second law can be formulated as:
Equation 1:
F = ma
where F is the net force acting on the body, m is the
mass of the body, and a is its acceleration. This is also
known as the equation of motion.
Newtons third law states to every action, there is
an equal and opposite reaction. This law is particularly
useful in analyzing complex problems in biomechanics
in which there are several interacting bodies.

Conditions for Equilibrium

If the net force and the net moment acting on a body
are zero, then acceleration (linear and angular) of the
body is zero, and consequently, the velocity (linear and
angular) of the body is either constant or zero. Therefore, two conditions need to be satisfied for equilibrium.
The body is said to be in translational equilibrium if the
net force (vector sum of all forces) acting on it is zero.
Equation 2:

F =0
41

Orthopaedic Knowledge Update 8

Musculoskeletal Biomechanics

tion of the muscle force is assumed to be vertical. The

gravitational forces are vertical as well.
The magnitudes of the muscle tension (FB) and joint
reaction force at the elbow (RF) can be determined by
considering the equilibrium conditions at the forearm.
Considering the rotational equilibrium of the forearm
about the elbow joint (point O) will yield:
Equation 4a:

Mo = 0 :
FB a WF b WO c = 0
1
FB = (bW + cW )
F
O
a
Considering the translational equilibrium of the forearm
along the y direction:
Equation 4b:

Fy = 0 :
RF - F B + WF + WO = 0
RF = F B WF WO
Figure 5 Free-body diagram showing the forces acting on the lower arm while holding
a weight with the arm flexed to 90. (Reproduced with permission from Mow VC,
Hayes WC: Analysis of muscle and joint loads, in Basic Orthopaedic Biomechanics, ed
2. Philadelphia, PA, Lippincott-Raven, 1997, p 6.)
The body is in rotational equilibrium if the net moment
(vector sum of all moments) acting on it is zero.
Equation 3:

M =0
Free-Body Diagrams
Free-body diagrams (FBDs) are constructed to help
identify the forces and moments acting on individual
parts of a system and to ensure the correct use of the
equations of equilibrium. For this purpose, the system is
isolated from its surroundings and proper forces and
moments replace the effects of the surroundings. For example, Figure 5 shows the forearm holding a weight.
Point O is designated the axis of rotation of the elbow
joint, which is assumed to be fixed for practical purposes. Point A is the attachment of the biceps muscle
with the radius, B is the center of gravity of the forearm,
and C is a point on the forearm that lies along the vertical line passing through the center of gravity of the
weight in the hand. The distances between O and A, O
and B, and O and C are measured as a, b, c, respectively.
WO is the weight of the object held in the hand and WF
is the total weight of the forearm. FB is the magnitude
of the force exerted by the biceps on the radius, and RF
is the reaction force at the elbow joint. The line of ac-

42

Equations 4a and 4b can be solved for FB and RF for a

given set of geometric parameters of a, b, c and weights
WO and WF. For example, assume a = 4 cm, b = 15 cm,
c = 35 cm, WO = 80 N, and WF = 20 N. Then from equations 4a and 4b:
FB =

1
0.04

[(0.15)(20)+(0.35)(80)] = 775N (+y)

RF = 775 20 80 = 675N (-y)

FBDs can also be used to calculate external intersegmental forces and moments at different joints during
human locomotion, as shown in Figure 6. In Figure 6, A,
the subscripts f, s and t stand for foot, shank, and
thigh, respectively. The letter I represents the moment
of inertia, a represents the angular acceleration, g
represents the gravitational acceleration, m represents
the mass of the segment concentrated at the centroid,
and a represents the linear acceleration. In Figure 6, B,
the subscripts g, a, k, and h stand for ground, ankle, knee, and hip, respectively. F stands for the force
and T is the resultant torque.
After an FBD is drawn, equilibrium equations are
used to calculate the unknown proximal forces. Calculations proceed from the distal to proximal end and start
at the foot. With the help of equilibrium equations, the
force and torque at the ankle are calculated. The force
and torque at the distal end of the shank are equal and
opposite to the force and torque at the ankle. With the
distal force and torque of the shank known, the proximal end force and torque are then calculated. In most
cases, however, the number of equations available for
the solution of such problems are fewer than the number of unknowns. Therefore, additional assumptions

American Academy of Orthopaedic Surgeons

Chapter 4 Musculoskeletal Biomechanics

Orthopaedic Knowledge Update 8

Figure 6 A and B, model of the lower extremity and the corresponding free-body
diagrams. See text for details. (Reproduced with permission from Mow VC, Hayes WC:
Musculoskeletal dynamics, locomotion, and clinical applications, in Basic Orthopaedic
Biomechanics, ed 2. Philadelphia, PA, Lippincott-Raven,1997, p 44.)

need to be made to obtain a solution. A common

method for solving an indeterminate problem is to seek
an optimum solution (a solution that maximizes or minimizes some process or action).

Kinematics and Kinetics

The general field of dynamics consists of two major areas: kinematics and kinetics. The field of kinematics addresses the description of geometric and timedependent aspects of motion without dealing with the
forces causing the motion. Kinematic analyses are based
on the relationships between position, velocity, and acceleration vectors. These relationships appear in the
form of differential and integral equations.
The field of kinetics is based on kinematics and includes the effects of forces and torques that cause the
motion in the analyses. There are several different approaches to the solutions of problems in kinetics. Different methods may be applied to different situations, depending on the parameter that needs to be determined.
For example, the equation of motion is used for problems requiring the analysis of acceleration, whereas the
analysis of forces related to changes in velocity uses a
different method known as the energy method.
It is a common practice in the study of kinematics and
kinetics to categorize motion as translational, rotational,

American Academy of Orthopaedic Surgeons

or general.Translational or linear motion occurs if all parts

of the body move the same distance at the same time and
in the same direction. An example of a translational motion is the vertical motion of an elevator in a shaft. Linear
motion also may be thought of as motion along a line. If
the line is straight, the motion is rectilinear. If the line is
curved, the motion is curvilinear.
Rotational or angular motion occurs when a body
moves in a circular path such that all parts of the body
move through the same angle at the same time. The angular motion occurs about a central line known as the
axis of rotation, which lies perpendicular to the plane of
motion.
General motion occurs when a body undergoes both
translational and rotational motions simultaneously; it is
more complex to analyze motions composed of both
translation and rotation. Most human body segmental
motions are of the general type. For example, while
walking, the lower extremities both translate and rotate.
Degree of freedom is an expression that describes
the ability of an object to move in space. A completely
unrestrained object, such as a ball, has six degrees of
freedom (three related to translational motion along
three mutually perpendicular axes, and three related to
rotational motion about the same axes).

Stress and Strain

In addition to tensile and compressive forces, there is a
third important category of force, termed shear.
Whereas compressive and tensile forces act along the
longitudinal axis of a bone or other body to which they
are applied, shear force acts parallel or tangent to a surface. Shear force tends to cause a portion of the object
to slide, displace, or shear with respect to another portion of the object. For example, a force acting at the
knee joint in a direction parallel to the tibial plateau is a
shearing force at the knee. During the performance of a
squat exercise, joint shear at the knee is greatest at full
squat position, which places a large amount of stress on
the ligaments and muscle tendons that prevent the femur from sliding off the tibial plateau.
The area moment of inertia is a measure of the distribution of the material in the cross section of a structure characterizing its bending stiffness and strength.
The cross section of a bone, shown in Figure 7, is divided into many small elements of area (A).
The area moment of inertia about the X and Y axes
are given by:
Equation 5a:
Ix = y2 A
Equation 5b:
Iy = x2 A

43

Musculoskeletal Biomechanics

Orthopaedic Knowledge Update 8

A = bh
3

bh
I = 12

A = ro2
4
I = 4 r0

J = r04
2
A = (ro2 ri2)
4 4
I = 4 (r0 ri )

J = 2 (r04 ri4)

Figure 8 Area moment of inertia and polar moment for three geometries: rectangle

Figure 7 Area and polar moment of inertia defined for a cross section of bone.

The polar moment of inertia is a measure of the distribution of material in the cross section of a structure
characterizing its torsional stiffness and strength. The
mathematical expression for the polar moment of inertia (J) about the centroid of the cross section is as follows:
Equation 6:
J = r2 A
Material distribution in a structure has an impact on
strength and bending (Figure 8). A solid cylindrical rod
of 5 mm radius has an area moment of inertia of
491 mm4. Redistributing the same material into a hollow tube of 1-mm thickness results in an outer radius of
13 mm and an area of moment of inertia of 6,146 mm4.
The bending strength (the area of moment of inertia divided by the radius) of the two different geometries
now can be calculated. It is found that the 13-mm tube
is 4.8 times as strong as the 5-mm rod. Human bone is
an excellent example of this type of construction. Bones
are hollow and cancellous on the inside and hard and
cortical on the outside. This construction provides maximum strength to the bone.

Stress
An important factor affecting the outcome of the action
of forces on the human body is the manner in which the
force is distributed. Whereas pressure represents the dis-

44

(a), solid cylinder (b), and hollow cylinder (c). a = Area ; i = area moment of inertia; j =
polar moment of inertia; NA = neutral axis. (Reproduced with permission from Ozkaya
N, Nordin M: Fundamentals of Biomechanics, ed 2. New York, NY, Springer-Verlag,
1998.)

tribution of force external to a solid body, stress represents the resulting force distribution inside a solid body
when an external force acts. Stress () is quantified in
the same way as pressure. Stress = force (F) per unit
area (A) over which the force acts.
Equation 7:
F

=A
A given force acting on a small surface produces
greater stress than the same force acting over a larger
surface. When a blow is sustained by the human body,
the likelihood of injury to body tissue is related to the
magnitude and direction of the stress created by the
blow. Compressive stress, tensile stress, and shear stress
are terms that indicate the direction of the acting stress.
From a mechanical perspective, long bones can be
compared to structural beams. A bones ability to resist
a shear force is more important than its ability to resist
an axial force. A bone can be subjected to three-point
bending and four-point bending as shown in Figure 9.
When a bone is bent, it is subjected to stresses occurring
in the longitudinal direction or in a direction normal to
the cross section of the bone. Based on the loading configuration shown in Figure 9 A, the distribution of these
normal stresses over the cross section of the bone is
such that it is zero on the neutral axis (NA), negative
(compression) above the neutral axis, and positive (tensile) below the neutral axis.

American Academy of Orthopaedic Surgeons

Chapter 4 Musculoskeletal Biomechanics

Orthopaedic Knowledge Update 8

Figure 10 Uniaxial strain for a rod in elongation.

Strain
Strain, which is also known as unit deformation, is a
measurement of the degree or intensity of deformation.
Consider a rod with initial length L which is stretched
to a length L' (Figure 10). The strain measure , a dimensionless ratio, is defined as the ratio of elongation
with respect to the original length.
Equation 9:

Figure 9 Types of bending. A, three-point and B, four-point.

For a bone subjected to pure bending, the stress x is
given by the following equation known as the flexural
formula.
Equation 8:
x =

(M) (y)
(I)

M is the bending moment, y is the vertical distance between the neutral axis and the point at which the stress
is sought, and I is the area moment of inertia of the
cross section of the beam about the neutral axis.
For example, if the bending moment has a value of
10,000 N-mm4, the distance from the neutral axis at
which stress is to be calculated is 10 mm, and the area
moment of inertia of the cross section is 200,000 mm,
the stress is calculated as:
x = 10000 10/200,000
= 0.5 N/mm2 = 0.5 MPa

American Academy of Orthopaedic Surgeons

L' L
L

A common term encountered in biomechanics is

strain rate (the rate at which strain occurs). Mathematically, it is obtained by dividing strain (dimensionless) by
the time (seconds) of application of load, thus imparting
the unit of measurement as second1 (s1).

Stress-Strain Diagram
When any stress is plotted on a graph against the resulting strain for a material, the resulting stress-strain diagram has several different shapes, depending on the
kind of material involved. As an example of a stressstrain diagram, Figure 11 illustrates the behavior of a
particular metal when subjected to increasing tensile
(stretching) stress. On the first portion of the curve (up
to a strain of less than 1%, 0a), the stress and strain are
proportional. This condition holds until the point a,
the proportional limit, is reached. It is known that stress
and strain are proportional because this segment of the
line is straight (Hookes Law). Youngs modulus is essentially the slope of the straight line on the stress-strain
diagram. Every material has a unique Youngs modulus
value. The larger the Youngs modulus for a material,
the greater stress needed for a given strain. The greater
the Youngs modulus for a material, the better it can
withstand greater forces.
From points a to b on the diagram, stress and
strain are not proportional; however, if the stress is removed at any point between a and b, the curve will

45

Musculoskeletal Biomechanics

Orthopaedic Knowledge Update 8

Figure 11 Stress-strain diagram. (Adapted from Examining the effects of space flight
on the skeletal system, Houston, Texas, National Space Biomedical Research Institute.)

Figure 13 Finite element model of the L3-L5 lumbar segment.

Figure 12 Stress-strain behavior of different materials.

be retraced in the opposite direction and the material
will return to its original shape and length. In the region
ab, the material is said to be elastic or to exhibit elastic behavior and the point b is called the elastic limit.
If the material is stressed further, the strain increases
rapidly, but when the stress is removed at some point
beyond b, say point c, the material does not revert
to its original shape or length but returns along a different path to a different point. Further increase of stress

46

beyond c produces a large increase in strain until

point d is reached and fracture occurs. From points
b to d, the metal is said to undergo plastic deformation. If the plastic deformation that takes place between
the elastic limit and the fracture point is large, the metal
is said to be ductile. Such materials are capable of being
drawn out like a wire or hammered thin like gold leaf.
If, however, fracture occurs soon after the elastic limit is
passed, the metal is said to be brittle. Figure 12 compares the stress-strain behavior of different types of
materials and includes characteristic mechanical properties.
It is often necessary to compute stresses and displacements in different regions of biologic structures
that have complex geometries. Usually, it is not feasible
to make in vitro or in vivo measurements of these parameters. A technique called finite element modeling is
useful in such cases. As the name suggests, this technique involves constructing a complex geometric model
using a finite number of elements of different shapes,
sizes, and properties, and then analyzing the structure.
Mathematically, it is defined as a procedure to obtain
approximate numerical solutions to equations that express the fundamental laws of physics (such as equilibrium and balance of mass and energy). Figure 13 shows
a validated finite element model of the L3-L5 ligamentous spinal lumbar segment.

American Academy of Orthopaedic Surgeons

Chapter 4 Musculoskeletal Biomechanics

Orthopaedic Knowledge Update 8

The Poissons ratio for most metals falls between 0.25 to

0.35. Rubber has a Poissons ratio close to 0.5 and is
therefore almost incompressible. Theoretical materials
with a Poissons ratio of exactly 0.5 are truly incompressible, because the sum of all their strains leads to a
zero volume change. Cork, on the other hand, has a
Poissons ratio close to zero.

Torsion

Figure 14 Shear force (F) deforms a cube into a parallelepiped.

Youngs Modulus, Shear Modulus, and Poissons Ratio

The modulus of elasticity in tension, also known as
Youngs modulus (E), is the ratio of stress to strain on
the loading plane along the loading direction when
the material is in the elastic limit.
Equation 10:

E=

Torsion occurs when a structure is caused to twist about

its longitudinal axis, typically when one end of the structure is fixed. Torsional fractures of the tibia are common
in football and skiing injuries in which the foot is held in
a fixed position while the rest of the body undergoes a
twisting motion. The magnitude of the shear stress () is
related to the magnitude M of the applied torque, the
cross-sectional area of the shaft, and the radial distance
(r) between the center line and the point at which shear
stress is to be determined.
Equation 13:
= (M) (r)
(J)

xx

xx

Common sense (and the second law of thermodynamics)

indicates that a material under uniaxial tension must
elongate in length. Therefore, the Youngs modulus (E) is
required to be nonnegative for all materials (E > 0).
A similar property known as shear modulus is used to
define material stiffness. Shear modulus is the ratio of
shear stress to shear strain of a material. Shear modulus
(G) is measured in Pa or newton per square meter (N/m2).
In the cube shown in Figure 14, application of a shear
force (F) deforms the cube into a parallelepiped (a solid
with parallel sides that are not perpendicular to each
other).This results in shear stress, (F/A) and shear strain,
(D/L).
Equation 11:
(F / A)

Shear modulus, (G) = (D / L) = /

Also, a rod-like specimen subjected to uniaxial tension
will exhibit some shrinkage in the lateral direction for
most materials. The opposite is true for the compression
force. The ratio of lateral strain and axial strain is defined as Poissons ratio ().
Equation 12:

yy
v=
xx
The Poissons ratio for most materials will fall in the
range,
1

0v2

American Academy of Orthopaedic Surgeons

Mechanical Properties of Tissues

Biomechanical principles guide the manner in which
components of the musculoskeletal system respond to
trauma, stresses, and loads and therefore determine the
success of this system to withstand external forces that
may threaten the more vulnerable organs. These principles also apply to body motions such as bending, lifting,
and fine manipulation. It also is important to realize
that the behavior of the muscles and tissues depends on
intrinsic material properties. In general, materials respond differently to different loading configurations. For
a given material, there may be different mechanical
properties that must be considered in analyzing its response to, for example, tensile loading as compared with
loading that may cause bending or torsion. To apply biomechanical principles to the musculoskeletal system, it
is essential to understand the mechanical properties of
tissues.

Extensibility and Elasticity

The properties of extensibility and elasticity are common to many biologic tissues. Extensibility is the ability
to be stretched and elasticity is the ability to return to
normal length after extension or contraction.
The elastic behavior of muscle consists of two major
conceptual components. As shown in Figure 15, the
muscle membranes represent the parallel elastic component, which provides resistive tension when a muscle is
passively stretched. The tendons represent the series
elastic component, which acts as a spring to store elastic
energy when an active muscle is stretched. These com-

47

Musculoskeletal Biomechanics

Figure 15 Conceptual components to explain muscle elasticity. PEC = parallel elastic

component; SEC = series elastic component. (Reproduced with permission from Hall
SJ: The biomechanics of human skeletal muscle, in Basic Biomechanics, ed 3. New
York, NY, McGrawHill, 1999.)

ponents of muscle elasticity are so named because the

membranes and tendons are respectively parallel to and
in series with the muscle fibers, which provide the contractile component. The elasticity of human skeletal
muscle is believed to result primarily from the series
elastic component.

Irritability and the Ability to Develop Tension

Another characteristic property of muscle is irritability,
which is the ability to respond to a stimulus. Muscles are
affected by either electrochemical stimuli (such as an
action potential from the nerve) or mechanical stimuli
(such as an external blow to a portion of a muscle). If
the stimulus is of sufficient magnitude, the muscle responds by developing tension.

Viscoelasticity
Viscoelasticity is the time-dependent mechanical property of a material. As the name suggests, two basic components of viscoelasticity are viscosity and elasticity. Viscosity is a fluid property and is a measure of resistance
to flow. Elasticity is a solid material property. A viscoelastic material is one that possesses both fluid and
solid properties. It has been experimentally determined
that most biologic materials, such as bone, ligaments,
tendons, and passive muscles exhibit viscoelasticity.
Creep and relaxation are two characteristics of viscoelastic materials that are used to document their behavior quantitatively. Creep is a phenomenon in which a
material or a structure deforms as a function of time under the action of a constant load. A creep test involves
the application of a sudden load, which is then maintained at constant magnitude. Measurements of deformation are recorded as a function of time. When an individuals height is measured in the morning and again
at night after weight bearing throughout the day, the
nighttime measurement is less than the morning measurement. This change in height (deformation) is not
caused by any additional weight the person might have
gained during the day, but rather by creep of the intervertebral disks, and to a lesser extent by the effects of
compression of the cartilage of load-bearing joints.

48

Orthopaedic Knowledge Update 8

Relaxation is a phenomenon in which stress or force
in a deformed structure decreases with time, while the
deformation is held constant. A relaxation test involves
the application of sudden deformation, which is then
maintained during the test. The force- or stress-time
curve obtained is called the relaxation curve and is used
to document the viscoelastic nature of the material or
structure.
Another characteristic that is typical of viscoelastic
materials is sensitivity to the rate of loading. The stressstrain curve of a viscoelastic material is dependent on
the rate of loading. A higher rate of loading generally
results in a steeper curve. For example, a slow pull on a
material such as Silly Putty can produce a deformation
of as much as several thousand percent before failure.
The behavior of this material is classified as ductile. A
fast pull, however, will break the putty at less than 5%
deformation, and will require a relatively large force.
The material property may now be called brittle. The literature of biomechanics has established that bone, ligaments, tendons, and passive muscles are sensitive to the
rate of loading.
The other major characteristic of a viscoelastic material is hysteresis or energy dissipation. This property
means that if a viscoelastic material is loaded and unloaded, the unloading curve will not follow the loading
curve. The difference between the two curves represents
the amount of energy that is dissipated or lost during
loading.

Anisotropy of Materials
A material is called anisotropic if its mechanical properties (such as strength and modulus of elasticity) are different in different directions. Experiments have shown
that bone is highly anisotropic. For example, strength of
bone in the mid-diaphysis of a long bone varies in different directions. Generally, the axial strength for bone
is the greatest, followed by radial strength, and then the
circumferential strength.

Biomechanical Evaluation of Human Tissues

Most biologic tissues are composite materials (consisting of materials with different properties) with nonhomogeneous and anisotropic properties. The mechanical
properties of living tissues may vary from point to point
within the tissue, and their response to forces applied in
different directions may be different. Among the common components of biologic tissues, collagen and elastin
fibers have the most important mechanical properties
affecting the overall mechanical behavior of the tissue
in which they are present.
Collagen is a protein made of crimped fibrils that aggregate into fibers. The mechanical properties of collagen fibrils are such that each fibril can be considered a
mechanical spring and each fiber as an assemblage of

American Academy of Orthopaedic Surgeons

Chapter 4 Musculoskeletal Biomechanics

Orthopaedic Knowledge Update 8

Table 1 | Ultimate Strength for Human Femoral Cortical

Bone

Table 2 | Elastic and Shear Moduli for Human Femoral

Cortical Bone

Loading Mode

Youngs Moduli/ Elastic Moduli, E

Longitudinal
Tension
Compression
Shear
Transverse
Tension
Compression

Ultimate Strength
133 MPa
193 MPa
68 MPa
51 MPa
133 MPa

(1 MPa = 106 Pa; 1 GPa = 109 Pa)

(Reproduced with permission from Mow VC, Hayes WC: Biomechanics of cortical and trabecular bone, in Mow VC, Hayes WC (eds): Basic Orthopaedic Biomechanics, ed 2. New
York, NY, Raven Press, 1997, p 85.)

springs. The primary mechanical function of collagen fibers is to withstand axial tension. Because of their high
length-to-diameter ratios (aspect ratio), collagen fibers
are not effective under compression. Whenever a fiber is
pulled, its crimp straightens and its length increases.
Like a mechanical spring, the energy supplied to stretch
the fiber is stored, and it is the release of this energy
that returns the fiber to its unstretched configuration
when the applied load is removed. Collagen fibers exhibit viscoelastic behavior and possess relatively high
tensile strength.
Among the noncollagenous tissue components, elastin is another fibrous protein with material properties
that resemble the material properties of rubber. Fibers
containing elastin are highly extensible, and their extension is reversible even at high strains. Elastin fibers behave elastically with low stiffness.

Biomechanics of Bone
Bone is the primary structural element of the human
body that protects internal organs, provides kinematic
links, provides muscle attachment sites, and facilitates
muscle actions and body movements. Bone is also
unique in that it is self-repairing. Bone also can alter its
shape, mechanical behavior, and mechanical properties
to adapt to the changes in mechanical demand. The major factors influencing the mechanical behavior of bone
are the mechanical properties of tissues comprising the
bone; the size and the geometry of the bone; and the direction, magnitude, and rate of applied loads.
At the macroscopic level, all bones consist of two
types of tissues. The cortical or compact bone tissue is a
dense material forming the outer shell (cortex) of bones
and the diaphysial region of long bones. The cancellous,
trabecular, or spongy bone tissue consists of thin rods
and plates (trabeculae) in a lattice-like structure that is
enclosed by the cortical bone.

American Academy of Orthopaedic Surgeons

Longitudinal
Transverse
Shear Modulus, G

17.0 GPa
11.5 GPa
3.3 GPa

(1 MPa = 106 Pa, 1 GPa = 109 Pa)

(Reproduced with permission from Mow VC, Hayes WC: Biomechanics of Cortical and
trabecular bone, in Mow VC, Hayes WC (eds): Basic Orthopaedic Biomechanics, ed 2. New
York, NY, Raven Press, 1997, p 85.)

The mechanical response of bone can be observed

by subjecting it to tension, compression, bending, and
torsion. Various tests to implement these conditions are
performed using uniform bone specimens or whole
bones. Results from these tests indicate that the compressive strength of bone is greater than its tensile
strength. Also, bone is much better at resisting rapidly
applied loads than at resisting slowly applied loads.
Bone is stiffer and stronger at higher strain rates. The
energy absorbed by bone tissue, which is proportional to
the area under the stress-strain curve, increases with an
increasing strain rate. It should be noted that during
normal daily activities, bone tissues are subjected to a
strain rate of about 0.01 s-1.
The stress-strain behavior of bone is also dependent
on the orientation of bone with respect to the direction
of loading. Cortical bone has a larger ultimate strength
(indicating that it is stronger) and a larger Youngs modulus or elastic modulus (indicating that it is stiffer) in
the longitudinal direction than the transverse direction.
Bone specimens loaded in the transverse direction fail
in a more brittle manner (without showing considerable
yielding) compared with bone specimens loaded in the
longitudinal direction. The ultimate strength values for
adult femoral cortical bone under various modes of
loading and its elastic and shear moduli are listed in Tables 1 and 2.
Long bones have a tubular structure in the diaphyseal region, which provides considerable mechanical advantage over solid circular structures of the same mass.
Tubular structures are more resistant to torsional and
bending loads when compared with solid cylindrical
structures. Furthermore, a tubular structure can distribute the internal forces more evenly over its cross section
compared with a solid cylindrical structure of the same
cross-sectional area.
Bone will fracture when the stresses generated in
any region of bone are larger than the ultimate strength
of bone. Fractures caused by pure tensile forces are seen
in bones with a large proportion of cancellous bone tissue. Fractures caused by compressive loads are com-

49

Orthopaedic Knowledge Update 8

Musculoskeletal Biomechanics

Table 3 | Stresses in Tendon During Various Activities

Stress (MPa)

Tendon

Activity

23
29
59
110

Goat patellar tendon

Goat patellar tendon
Human Achilles tendon
Human Achilles tendon

Walking
Trotting
Walking
Running

Figure 16 Lines of fracture vary depending on whether the load is a combination of

compressive and bending forces (a), bending force (b), torsional force (c), compressive
force (d), and tensile force (e).

(Reproduced with permission from An Kn: In vivo force and strain of tendon, ligament, and
capsule, in Guilak F, Butler DL, Goldstein SA, Mooney D (eds): Functional Tissue Engineering. New York, NY, Springer, 2003.)

monly encountered in the vertebrae of elderly patients,

whose bones are weakened as a result of aging. Clinically, most bone fractures occur as a result of complex
combined loading situations rather than simple loading
mechanisms. Figure 16 shows the differing lines along
which fractures in bone occur depending on the type of
load encountered.

Compared with tendons, ligaments often contain a

greater proportion of elastic fibers that account for their
high extensibility but lower strength and stiffness. Like
tendons, they are viscoelastic and exhibit hysteresis, but
deform elastically up to strains of about 0.25 (yield
strain), which is about five times as much as the yield
strain of tendons.

Biomechanics of Tendons and Ligaments

Biomechanics of Articular Cartilage

Tendons and ligaments are fibrous connective tissues.

Tendons help execute joint motion by transmitting tension from muscles to bones. Ligaments join bones and
provide stability to the joints. Unlike muscles, which are
active tissues and can produce mechanical forces, tendons and ligaments are passive tissues and cannot actively contract to generate forces. Tendons and ligaments do not rupture easily. Damage to tendons and
ligaments usually occurs at their junctions with bones.
Compared with skeletal muscles, tendons are stiffer,
have higher tensile strengths, and can endure larger
stresses. The relatively low ultimate strength of muscles
would require that they have relatively large crosssectional areas to transmit sufficiently high forces to
bones without tearing. Tendons are better designed to
perform this function. Therefore, around the joints
where the space is limited, muscle attachments to bones
are made by tendons. Tendons are capable of supporting
very large loads with very small deformations.
Tendons are believed to function in the body at
strains of up to about 0.04, which is believed to be their
yield strain. Tendons rupture at strains of about 0.1 (ultimate strain), or stresses of about 60 MPa (ultimate
stress). The magnitude of tendon tension has been estimated by various analytical methods, and also by combined experimental and analytical methods. Direct measurements also have been attempted. The most
commonly studied tendons include the patellar tendon
of the knee, the flexor tendons of the finger (the flexor
digitorum superficialis and flexor digitorum profundus),
and the Achilles tendon. Tendon stresses and forces reported in the literature are listed in Table 3 and Table 4
respectively.

Cartilage covers the articulating surfaces of bones

within the diarthrodial (synovial) joints. The primary
function of cartilage is to facilitate the relative movement of articulating bones. Cartilage reduces stresses

50

Table 4 | Tendon Forces During Various Activities

Tension
(N)
207
801
999
13
50
82
2,600
9,000
4,000
480661
1,8952,233
3,786
1-9
35
120
8-16

Tendon

Activity

Goat patellar tendon

Goat patellar tendon
Goat patellar tendon
Rabbit flexor digitorum profundus
Rabbit flexor digitorum profundus
Rabbit flexor digitorum profundus
Human Achilles tendon
Human Achilles tendon
Human Achilles tendon
Human Achilles tendon

Standing
Walking
Trotting
Standing
Level hopping
Inclinal hopping
Walking
Running
Running
Cycling

Human Achilles tendon

Squat jump

Human Achilles tendon

Human flexor digitorum profundus
Human flexor digitorum profundus
Human flexor digitorum profundus
Human flexor digitorum superficialis

Hopping
Passive motion
Active motion
Pinch
Keystroke

(Reproduced with permission from An Kn: In vivo force and strain of tendon, ligament, and
capsule, in Guilak F, Butler DL, Goldstein SA, Mooney D, (eds): Functional Tissue Engineering. New York, NY, Springer, 2003.)

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

applied to bones by increasing the area of contact between the articulating surfaces and reduces bone wear
by reducing the effects of friction. From an engineering
perspective, the mechanical behavior of cartilage is considered to be remarkable.
During daily activities, the articular cartilage is subjected to tension and shear stresses as well as compressive stresses. Under tension, cartilage responds by realigning the collagen fibers that carry the tensile load
applied to the tissue. The higher the collagen content, the
higher the tensile strength of cartilage. Shear stresses on
the articular cartilage are caused by frictional forces between the relative movements of articulating surfaces.
The importance of cartilage to bear load and to maintain
its mechanical integrity becomes clear when the magnitude of the forces involved at the human hip joint are
considered. The hip bears a load of about five times the
body weight during ordinary walking; the load is much
higher during running and jumping.
An essential step toward a better understanding of
the mechanics of articular cartilage is the development
of appropriate constitutive models for this tissue that relate the state of stress to the state of strain, and fluid
and solute fluxes to electrochemical gradients. These
mathematical models, when validated through experiments, can provide valuable insight into the functioning
of this tissue.

Joint Lubrication
The design and lubrication of human joints are major
factors in the capacity for swift and prolonged mobility.
The loads on many joints are as great as two to four
times body weight during ordinary activities. These
loads increase to even higher multiples in other activities. One of the most critical biomechanical characteristics of human joints, which allows them to function for
70 to 80 years, is their system of lubrication. Several attempts have been made to explain the underlying mechanisms.

Basic Concepts and Definitions

When a joint in the body is subjected to external forces,
in the form of external loads and muscle forces, the internal reaction forces acting at the contact surfaces are
called the joint reaction forces. For example, the contact
surface of the metacarpal portion of a metacarpophalangeal joint has two components of joint reaction
forces acting on it. One is perpendicular to the contact
surface and is called the normal joint reaction force. The
other force is parallel to the surface and is called the
tangential joint reaction force or the frictional force. The
normal component is always compressive and is generally large. The tangential component is generally small.
The ratio of the two components (tangential to normal)
is called the coefficient of friction.

American Academy of Orthopaedic Surgeons

Chapter 4 Musculoskeletal Biomechanics

For example, a skater can glide effortlessly on ice because the ratio of effort (the tangential or frictional
force) to body weight (the normal compressive force) is
very small. This interbody action has a rather low coefficient of friction. Typical values range from 0.01 (such as
a bearing with lubrication) to 0.75 (such as a stone on
the ground).
Another important joint parameter is the contact
area, which is the area of contact between the two joint
surfaces. In general, the greater the contact area the
smaller the contact stress or the stress at the contacting
surface. An injury, joint misalignment, or disease process
may lead to a decrease in the contact area and, therefore, to an increase in contact stresses and wear of the
joint surfaces. The contact area depends on the congruency of the mating surfaces, physical properties of cartilage and supporting bone, and the forces that press the
two surfaces together. The contact area also increases
when either the contacting surfaces decrease in stiffness
or the joint forces increase. In the hip joint, the diameter
of the femoral head is slightly larger than that of the acetabulum resulting in contact that begins around the periphery. As the load is increased, the contact area of the
femoral head extends toward the center. With such a design, the contact area increases with increasing load,
thus keeping the contact stresses relatively constant.
This factor is especially important when the load
reaches its maximum value. At that time, the load is
spread all the way to the periphery, thus reducing the
peak stress in the central regions of the head.
In a lubricant, viscosity is essential for providing lubrication between two loaded joint surfaces and is defined as the property of a material to resist shear load.
The lubricant is sheared by the sliding action of one surface against the other. Experiments showed that the
friction in an animal joint increased dramatically when
the synovial fluid was replaced by water, which is relatively less viscous. The coefficient of viscosity is a measure of the viscosity of a fluid. A more viscous fluid has
a higher coefficient of viscosity. Water, for example, has
a low coefficient of viscosity and motor oil has a high
coefficient of viscosity. The coefficient of viscosity may
be constant or it may vary with the speed of shearing.
Fluids that have constant viscosity coefficients are called
newtonian fluids and include water and glycerin. Fluids
that decrease in viscosity at higher shearing rates are
called thixotropic and include synovial fluid.

Hydrodynamic Lubrication
Hydrodynamic lubrication is a mechanism that decreases
the friction between two sliding surfaces by maintaining
a fluid film caused by the sliding motion between the surfaces. Consider a circular metal shaft placed in a slightly
larger hole in a metal block or bearing. As shown in Figure 17, A, when the shaft is stationary and carries an ex-

51

Musculoskeletal Biomechanics

Orthopaedic Knowledge Update 8

Elastohydrodynamic Lubrication
Elastohydrodynamic lubrication is a mechanism that decreases friction between two sliding surfaces by maintaining a fluid film between the surfaces caused by both
the sliding motion and the elastic deformation of the
surfaces.
In some machine parts, the loads are too high and
the geometry of the bearing surfaces is not ideal for
producing a dynamic wedge of fluid film. An example of
this situation is the gear teeth in an automotive gear box
in which both the surfaces are convex, making it difficult to produce a wedge film. However, the measured
friction in the gear teeth is low, which can be explained
by the effects of elastohydrodynamic lubrication. The
basic idea is that the high loads carried by the two convex surfaces deform the surfaces significantly on a microscopic scale to produce a geometry of the mating surfaces that is suitable for the development of a fluid film.
Human cartilage can deform elastically under load and
thus provides the potential for elastohydrodynamic lubrication.

Weeping Lubrication
Weeping lubrication is a mechanism by which the joint
load is borne by the hydrostatic pressure created by the
water phase of the synovial fluid escaping from the cartilage. Because cartilage is permeable, the water phase
of the synovial fluid can move in and out of it. Under
the application of load, the water phase of the synovial
fluid is released from cartilage. As the fluid is pushed
into the joint cavity, it separates the two cartilage surfaces because of its hydrostatic pressure and, thus, decreases the friction. The reverse happens when the joint
is unloaded and the water phase of the synovial fluid is
sucked into the cartilage, thus completing the cycle. This
squeeze-out/suck-in occurrence provides a self pressurizing and load-bearing mechanism that is not dependent
on the speed of sliding. This theory is based on some experimental studies of animal joint lubrication.
Figure 17 Hydrodynamic lubrication. (Reproduced with permission from Panjabi MM,
White AA III: Joint friction, wear, and lubrication, in Biomechanics of the Musculoskeletal System. New York, NY, Churchill Livingstone, 2001.)

Squeeze Film Lubrication

ternal load, the lubricant is squeezed out and there is direct contact between the shaft and its bearing. However,
when the shaft starts rotating, the fluid is brought in between the shaft and the bearing, creating a fluid wedge
(Figure 17, B). This action has the effect of lifting the
shaft above the bearing and thus decreasing friction and
wear. Although this lubrication mechanism explains the
low friction found in high speed and highly loaded machinery (for example, the bearings of the automotive
crankshaft), it does not explain the lubrication of a body
joint, which has a sliding velocity that is too slow to generate any significant wedge of fluid film.

Squeeze film lubrication is a mechanism by which the

joint load is carried by the fluid pressure developed between the two surfaces as a result of the relative movement of the approaching surfaces. As the two surfaces
approach each other, the fluid trapped between the cavities of the cartilage is squeezed out, thus creating a
cushion and providing a load-bearing capacity for the
joint. When the surfaces touch, there is no longer a
squeeze film. Thus, this mechanism cannot explain the
joint lubrication under normal physiologic conditions.
However, in situations where a joint is suddenly loaded,
such as landing on the ground after a jump, the squeeze
film mechanism may be responsible for reducing shock
and injury to the cartilage.

52

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 4 Musculoskeletal Biomechanics

Simultaneous Fluid Exudation and Imbibition

Although the weeping and squeeze film theories provide explanations of some aspects of animal and human
joint lubrication, they do not give a detailed description
of the actual interaction between the cartilage and synovial fluid during joint motion under load. Backed by
mathematical description and experimental evidence,
some recent work has attempted to describe the actual
fluid flow in the cartilage, an essential step toward understanding joint lubrication.
Consider a sheet of sponge soaked with water. When
a finger is pressed into the sponge, water exudes out
from the region around the finger (Figure 18, A). Accompanying this wet region, surrounding the finger, is a
drier region under the finger. Pressing the finger into
the sponge simply displaces the water from under the
finger to its surroundings. By lifting the finger, the water
is reabsorbed by the sponge, thus restoring the original
condition. If motion is added to the above experiment
(Figure 18, B), it is found that the wet region that is
ahead of the finger motion is wetter, and the wet region
that is behind the finger is now drier. The motion of the
finger produces a flow of the fluid from behind to in
front of the finger. The reverse will occur if the direction
of motion is reversed. Researchers have provided a
mathematical model for this process. A diagrammatic
representation of the concept as it applies to the cartilage is shown in Figure 18, C. As one cartilage-covered
bone moves over another, the water phase of the synovial fluid is squeezed out (exudation) and pushed
ahead of the moving contact area. At the same time, the
water phase of the synovial fluid is sucked in (imbibition) behind the contact area. Because the permeability
of the cartilage is significantly reduced under load in the
contact area, there is enhancement of the fluid film to
carry the load.
The interpretation of experimental results requires
the proper formulation of a lubrication theory for cartilage. Any such formulation must be able to predict a variety of experimental outcomes using physically and
mathematically consistent equations. The formulation of
such a theory is a great challenge in the field of biotribology; research is still needed to fully understand the
physiologic lubrication processes in vivo.

New Approaches and Future Perspectives

Over the past few years, substantial progress has occurred not only in measurement techniques of biomechanical parameters, but also in the techniques used to
aid healing of injured tissue.

Imaging Techniques
To better document the load environment in various
components of the musculoskeletal system, more direct
in vivo measurements are essential. Newer techniques

American Academy of Orthopaedic Surgeons

Figure 18 Diagrammatic representation of a new concept to describe fluid flow in

cartilage. (Reproduced with permission from Panjabi MM, White AA III: Joint friction,
wear, and lubrication, in Biomechanics of the Musculoskeletal System. New York, NY,
Churchill Livingstone, 2001.)
should enable in vivo signals to be measured less invasively than with conventional strain gauges and transducers. Noninvasive techniques based on image analysis
are useful tools for such measurements.
A technique for quantifying two-dimensional strain
using MRI has been developed. Images are collected on
a 1.5 Tesla scanner using a two-dimensional spin-echo
technique, which uses a pattern-matching algorithm to
quantify tissue deformation from the unstrained and
strained images. No markers are placed on or within
the tissue as had been done using traditional methods.
Instead, this approach uses patterns naturally inherent
to the tissue, treating small rectangular regions of the
image as discrete, unique markers. Using this technique,

53

Musculoskeletal Biomechanics
the strain in the human patella has been measured noninvasively. The in vivo three-dimensional velocity profiles for the patellar tendon were measured during a
low-load extensor task using cine phase contrast MRI.
The data were used to calculate patellar tendon elongation and strain.
A newly developed technology, magnetic resonance
elastography, provides great potential for noninvasive in
vivo investigation. Magnetic resonance elastography
provides images of the response of tissue to acoustic
shear waves to determine the shear modulus and tension in the muscle. Other advancements, such as those in
micro-CT technology are making the quantification of
microarchitectural parameters possible.

Microelectromechanical Systems Technology

Microelectromechanical systems technology is also
promising. Scientists and engineers are trying to design
smaller measurement devices that can be placed within
the tissue rather than on its surface where there is a
possibility of interference from adjacent tissues. Transducer materials selected for these devices minimize encapsulation and rejection by the body. These transducers
also need to be durable to permit signals to be recorded
for longer periods of time after surgery and may use telemetry rather than hard wiring to prevent lead wire
and connector damage. By measuring signals up to
weeks rather than days after surgery, researchers can expect to obtain more realistic force and deformation
measurements without the adverse effects of surgery.

Tissue Engineering
The inability of natural healing and surgical repair to
truly regenerate normal soft tissue has been the impetus
for tissue engineering. Combining principles of engineering and biology, tissue engineering endeavors to
fabricate new tissues in the laboratory that are designed
to rapidly restore tissue form (three-dimensional architecture and composition of normal tissue) and function
(normal structural and material properties). The tissue
engineering process typically involves introducing living
cells and a carrier into a wound site or mixing these
cells and a carrier with a natural or man-made scaffold
material. Collagen gels often serve as carriers for these
cells; in the past, scaffolds have included carbon fibers,
collagen fibers, polylactic acid, polyglycolic acid, and
Dacron sutures. Biologic scaffolding materials, such as
small intestine mucosa obtained from porcine small intestines, are gaining wide acceptance. When these constituents are combined, cellular recruitment and tissue
ingrowth are encouraged in the implants. The stiffer
scaffold material protects the cells and newly forming
repair tissue from high forces during the early phases of
repair. To avoid stress shielding, however, the scaffold
should degrade at the same rate as the rate of increase

54

Orthopaedic Knowledge Update 8

in the capability of the newly formed tissue to bear significant load. The degrading scaffold should pose no
threat of rejection.

Functional Tissue Engineering

Functional tissue engineering extends the goals of tissue
engineering by taking into consideration the actual
functional demands placed on the implants that are intended to repair the damaged tissue. Tissue engineers
then incorporate these requirements into meaningful
parameters that are designed into the constructs before
surgery. Presuming that tissue-engineered implants
alone will not possess sufficient strength and stiffness,
biomaterials experts and tissue engineers should work
together to formulate scaffolds with the correct mechanical integrity, degradation characteristics, and biocompatibility to facilitate an aligned extracellular matrix of
the correct composition. Novel biologic and bioengineering techniques, including application of growth factors, gene transfer, and cell therapy offer the potential
to improve the quality of healing tissues.
Growth factors are small proteins that are known to
induce a variety of cellular responses, including cell proliferation and matrix synthesis. The induction of these
biologic effects is mediated by binding of growth factors
to their specific receptors on cell surfaces. In vivo studies have shown that treatment of platelet-derived
growth factor-BB affects the biomechanical properties
of the rabbit femur-medial collateral ligament complex.
Specifically, this growth factor increased the ultimate
load, energy absorbed to failure, and ultimate elongation of the rabbit femur-medial collateral ligament complex.
Recent advances in genetic engineering are providing excellent advantages for recreating and investigating
various human skeletal disease states in transgenic (organisms whose genome has been altered by the transfer
of a gene or genes from another species or breed) mice
animal models. Various studies have shown the profound effect of several noncollagenous proteins through
the alteration of the normal mineralization process or
collagen fibrillogenesis. Transgenic mouse models also
make it possible to investigate the effect of genetic disease, such as osteogenesis imperfecta, on the mechanical
function of bone. Using gene transfer technology, it is
possible to increase cellular production of specific proteins (such as growth factors) that are functionally important for tissue healing. Two approaches, ex vivo and
in vivo, are often used for gene transfer. With the ex
vivo approach, genes are placed into cells in culture and
the cells are later transferred back to the host tissues of
interest. With an in vivo approach, genes are transferred
to cells in host tissues by direct injection of a carrier
into the host tissues.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

It has been established that mesenchymal progenitor
cells (MPCs) from both humans and animals can differentiate into a variety of cell types, including fibroblasts,
osteoblasts, and endothelial cells. These cells are responsible for tissue wound healing. Bone marrow, blood,
muscle, or adipose tissue are potential sources of MPCs
for cell therapy. For cell therapy, MPCs are isolated
from the bone marrow, cultured, and finally transplanted into host tissues. MPCs appear to retain their
developmental potential even after extensive subculturing. Implantation of MPCs into an injured tendon significantly improves its mechanical properties.
The basis for future treatments eventually may use a
combination of approaches to treat injuries that include
seeding cells on a scaffold that is conditioned with the
right combination of mechanical stimuli, growth factors,
and gene therapy. Because these experiments will require a combination of techniques that simultaneously
address the structure, biochemistry, and biology of a tissue to enhance healing, a significant effort must be
made to encourage collaboration among investigators
from different disciplines. By combining appropriate engineering mechanics with other basic sciences, improved
outcomes can be achieved.

Annotated Bibliography
An KN: In vivo force and strain of tendon, ligament and
capsule, in Guilak F, Butler DL, Goldstein SA, Mooney
DJ (eds): Functional Tissue Engineering. New York, NY,
Springer, 2003, pp 96-105.
In vivo force and strain measurements in soft tissue were
presented in six groups: tendon tension, ligament deformation,
capsular pressure, tendon surface friction, soft-tissue stress,
and soft-tissue strain.

Ateshian GA, Hung CT: Functional properties of native

articular cartilage, in Guilak F, Butler DL, Goldstein
SA, Mooney DJ (eds): Functional Tissue Engineering.
New York, NY, Springer, 2003, pp 46-68.
Important mechanical properties of the cartilage including
viscoelasticity, anisotropy, tension-compression nonlinearity,
and inhomogeneity were assessed. Valuable measures of the
functionality of the tissue-engineered constructs were included.

Butler DL, Dressler M, Awad H: Functional tissue engineering: Assessment of function in tendon and ligament,
in Guilak F, Butler DL, Goldstein SA, Mooney DJ
(eds): Functional Tissue Engineering. New York, NY,
Springer, 2003, pp 213-226.
This chapter discusses the issues that confront researchers
in fabricating structures from cells. State-of-the-art applications of biologic-microelectromechanical systems, imaging, and
other technologies are described.

American Academy of Orthopaedic Surgeons

Chapter 4 Musculoskeletal Biomechanics

Efunda: Engineering Fundamental Website. Available
at: http://www.efunda.com. Accessed June, 2004.
This website provides information that is helpful in the understanding of biomechanics. Information includes formulas
for beam deflection, bending, and twisting and on stress and
strain parameters for materials.

Keaveny TM, Yeh OC: Architecture and trabecular

bone: Toward an improved understanding of the biomechanical effects of age, sex and osteoporosis. J Musculoskeletal Neuronal Interactions 2002;2:205-208.
This study reviews what is known about architectural
changes in trabecular bone associated with age, gender, and osteoporosis and the role of these changes in the mechanical
properties of tissue. Recent developments in three-dimensional
high-resolution imaging technologies have provided more accurate measures of quantitative metrics of architecture, thereby
providing new data and raising questions about earlier conclusions.

NSBRI: National Space Biomedical Research Institute

Website. Available at: http://www.nsbri.org. Accessed
June, 2004.
This website provides a description of areas of research
such as bone loss, muscle atrophy, and other current research
projects.

Panjabi MM, White AA III: Biomechanics of the Musculoskeletal System. New York, NY, Churchill Livingstone,
2001.
This book describes the principles of biomechanics in a
simple manner and is suitable for orthopaedic residents and
fellows.

Puska MA, Kokkari AK, Nrhi TO, Vallittu PK: Mechanical properties of oligomer-modified acrylic bone
cement. Biomaterials 2003;24:417-425.
Mechanical properties of oligomer-modified acrylic bone
cement with glass fibers were studied. The three-point bending
test was used to measure the flexural strength and modulus of
the acrylic bone cement composites using analysis of variance
between groups. A scanning electron microscope was used to
examine the surface structure of the acrylic bone cement composites.

Woo SL, Abramowitch SD, Loh JC, Musahl V, Wang JH:

Ligament healing: Present status and the future of functional tissue engineering, in Guilak F, Butler DL, Goldstein SA, Mooney DJ (eds): Functional Tissue Engineering. New York, NY, Springer, 2003, pp 17-34.
Properties of normal ligaments, including their anatomic,
biologic, biochemical, and mechanical properties, and changes
that occur after injury, were described. Functional tissue engineering methods and preliminary findings were presented.

55

Musculoskeletal Biomechanics

Classic Bibliography
Burstein AH, Reilly DT, Martens M: Aging of bone tissue: Mechanical properties. J Bone Joint Surg Am 1976;
58:82-86.
Ding M, Dalstra M: Danielsen CC, Kabel J, Hvid J,
Linde F: Age variations in the properties of human tibial trabecular bone. J Bone Joint Surg Br 1997;79:9951002.

Orthopaedic Knowledge Update 8

Martin BR, Burr DC, Sharkey NA: Skeletal Tissue Mechanics. New York, NY, Springer, 1998.
Mow VC, Ateshian GA: Lubrication and wear of diarthrodial joints, in Mow VC, Hayes WC (eds): Basic Orthopaedic Biomechanics, ed 2. New York, NY, Raven
Press, 1997, pp 275-316.
Ozkaya N, Nordin M: Fundamentals of Biomechanics, ed
2. New York, NY, Springer-Verlag, 1998.

Hall SJ: Basic Biomechanics, ed 3. New York, NY,

McGrawHill, 1999.

56

American Academy of Orthopaedic Surgeons

Chapter

Biomaterials and Bearing Surfaces in

Total Joint Arthroplasty
Timothy M. Wright, PhD

Introduction
Synthetic materials play a prominent role in orthopaedic surgery because of the continuing need to replace, stabilize, or augment damaged musculoskeletal
tissues. Materials used for orthopaedic devices must be
safe and effective; therefore they must be biocompatible, resistant to corrosion and degradation, possess superior mechanical and wear properties, and meet high
quality standardsall at a reasonable cost. The interplay between synthetic materials and the surrounding
environment is an important factor to consider when using such materials to temporarily stabilize a fracture or
to permanently replace a structure such as the hip joint.
Implant wear is the major complication that limits
the longevity of total joint arthroplasties. Substantial
clinical evidence exists that links the release of large
amounts of submicron particulate debris from articular
and modular interfaces in implant components to subsequent osteolysis and implant loosening. Bioengineering
solutions to the wear problem are based on two approaches: (1) replacing the conventional metal-on-ultrahigh molecular weight polyethylene (UHMWPE) articulation with alternative combinations of bearing
materials with improved wear resistance, and (2) using
improved designs aimed at lowering contact stresses and
sliding distances between moving parts. Only long-term
clinical experience will ultimately establish the efficacy
of these approaches; however, laboratory results and
short-term clinical experience suggest that such approaches are beneficial in reducing implant wear.
Appreciation of the basic science behind these and
other uses of biomaterials in orthopaedic surgery requires knowledge of the basic structure and composition
of these materials and an understanding of how a materials structure and composition determine its ability to
meet necessary performance criteria essential to its clinical efficacy. For many applications of biomaterials in orthopaedics, the failure criteria for the material can be
measured and directly compared with the expected mechanical burden the material will be subjected to for use

American Academy of Orthopaedic Surgeons

in vivo. For many other applications, however, the situation is not as straightforward.
The goal of improving implant wear resistance is a
difficult task. The introduction of new forms of existing
materials or alternative bearing materials is hampered
by the lack of strong scientifically based relationships
between specific material properties (measured using
standardized laboratory tests) and in vivo wear behavior. Knowledge has been gained by analyzing the
stresses that occur in bearing materials under realistic
geometries and loading conditions; however, the link between specific wear mechanisms and the controlling material properties remains circumstantial. Thus, material
selection cannot be made on the simple basis of specifications such as elastic modulus, fracture toughness, or
yield stress. Similarly, standard laboratory experiments
on simple geometries, such as pin-on-disk tests, do not
adequately recreate the mechanical stresses that the material will experience in vivo.
Because of these limitations, screening tests using
hip and knee joint simulators have become the accepted
approach for providing preclinical test data on wear performance. Wear is measured gravimetrically on the basis
of periodic measurements of the small amount of weight
that is lost as material is worn from the articular surface
during the simulator test. Joint simulators are validated
in the sense that, under certain kinematics and loading
conditions, they produce worn polyethylene surfaces
and generate wear particle sizes and shapes similar to
those observed on retrieved implants. Hip simulators
also produce wear rates over the course of the test that
generally agree with clinical wear rates determined from
component thickness changes observed on serial radiographs (assuming one million cycles of test equals 1 year
of clinical use).
The reliance on joint simulator tests is especially important because the clinical ramifications of improved
wear are not known until many years after the introduction of a new material. If a bearing material is intended
to produce a meaningful reduction in wear, its effectiveness can only be shown through long-term studies; however, this requirement makes it difficult for both indus-

57

Orthopaedic Knowledge Update 8

Biomaterials and Bearing Surfaces in Total Joint Arthroplasty

Table 1 | Benefits and Disadvantages of Bearing Material Combinations

Bearing Combination

Benefits

Disadvantages

Metal-on-polyethylene

Long-term clinical results

Design effects well known
Long-term clinical results
Low wear
Excellent resistance to degradation

Osteolysis
May be unsuitable for young patients
Limited shelf life before implantation
Wear rates not as low as elevated cross-linked
polyethylene

Ethylene oxide or gas plasma sterilized

polyethylene-on-metal or ceramic

Excellent resistance to degradation

Wear rates not as low as elevated cross-linked

polyethylene

Polyethylene sterilized in low oxygen/inert

environment

Excellent resistance to degradation

Low wear

Wear rates not as low as elevated cross-linked

polyethylene
Possible residual free radicals

Elevated cross-linked polyethylene-on-metal or

ceramic

Excellent wear resistance

Excellent resistance to degradation

Short-term clinical use

Reduced toughness could be problematic
High cost

Metal-on-metal

Usually low wear

Possible lubrication film

Sometimes high wear

Local and systemic accumulation of metallic
debris and ions

Ceramic-on-polyethylene

Low wear
Good resistance to third body wear

Component fracture/difficult revision

High cost

Alumina-on-alumina

Usually low wear

Excellent biocompatibility
Possible lubrication film

Sometimes high wear

Component fracture
High cost

(Adapted from McKellop HA: Bearing surfaces in total hip replacements: State of the art and future developments. Instr Course Lect 2001;50:174.)

try and the Food and Drug Administration (FDA) to

bring improved technologies to the marketplace safely
and expeditiously. Current evaluation of the performance of new bearing materials must be centered on
their intended benefits and disadvantages (Table 1) and
the results of laboratory and short-term clinical tests.

Metallic Materials
Stainless steels, cobalt alloys, and titanium alloys have
been used in orthopaedic devices for decades. They are
generally corrosion resistant and biocompatible and
possess mechanical properties sufficient for use as structural load-bearing implants. These materials are fabricated using a wide variety of techniques (including casting, forging, extrusion, and hot isostatic pressing) which
lend flexibility in terms of both mechanical properties
and shape.
Stainless steels are predominantly iron-carbon alloys. Carbon is added to allow the formation of metallic
carbides within the microstructure. Carbides are much
harder than the surrounding material, and a uniform
distribution of carbides provides strength. Additions of
other alloying elements, such as molybdenum, stabilize
the carbides. Chromium provides the stainless quality to
stainless steel. It forms a strongly adherent surface ox-

58

ide, which serves as a protective passive layer between

the corrosive body environment and the bulk steel.
Cobalt alloys are composed primarily of cobalt,
chromium, and molybdenum. Chromium is added for increased hardness and corrosion resistance. As is the case
with stainless steel, the chromium forms a strongly adherent passive oxide film. Molybdenum is added to provide high strength. Standard cobalt alloy contains significant amounts of carbon that allows for the formation of
hard carbides, which strengthen the alloy and improve
its wear resistance. The addition of nickel forms an alloy
suitable for forging, which further enhances the materials strength.
The most commonly used titanium alloy for orthopaedic applications is titanium-aluminum-vanadium alloy, often referred to as Ti-6Al-4V because the primary
alloying elements, aluminum and vanadium, comprise
about 6% and 4%, respectively, of the alloy. Titanium
has the ability to self passivate, forming its own oxide
that imparts a high degree of corrosion resistance. Oxygen concentration is kept very low to maximize strength
and ductility. Despite long-term clinical evidence of excellent biocompatibility, concern that the release of vanadium (a cytotoxic element) could cause local and systemic complications has led to the introduction of other

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 5 Biomaterials and Bearing Surfaces in Total Joint Arthroplasty

titanium alloys in which vanadium is replaced by more

inert elements such as niobium.
Stainless steel is the material that is most susceptible
to both galvanic and crevice corrosion. The area between the screw heads and countersunk holes in stainless steel bone plates is a common location for crevice
corrosion. Continual removal of the passive layer can be
caused by fretting and can lead to the corrosion of all
three metallic alloys. Modular metallic devices, including
femoral components for total hip replacement and
stemmed components for revision knee replacement,
provide both the micromotion that causes fretting and
the enclosed environment needed for crevice corrosion.
Damage caused by corrosion and fretting often is found
on retrieved modular components, and is associated
with local and systemic accumulations of alloy elements
such as titanium and chromium.
Corrosion and strength concerns have largely relegated the use of stainless steel components to temporary implant devices, such as fracture plates, screws, and
hip nails; however, with appropriate designs, stainless
steel is still used in permanent implants such as Charnley style femoral components for hip replacement. Cobalt alloys, because of their high strength and their ability to be polished to a very smooth surface finish, are
the most common materials used for metallic total joint
arthroplasty components.
Titanium alloys possess roughly one half of the elastic modulus of stainless steel or cobalt alloy materials;
therefore, for the same component design, a titanium
version will have one half the structural stiffness
(whether subjected to axial, bending, or torsion loads).
This property makes titanium alloys attractive in situations in which more load is to be shared with the surrounding bone. The low modulus and excellent strength
of these alloys make them useful in a host of devices for
trauma, spinal fixation, and arthroplasty. The latter
application, however, no longer extends to articular surfaces because significant scratching and wear of titanium alloy femoral heads has been observed, particularly in the presence of third body wear.

Metal-on-Metal Bearings
The problems of wear and osteolysis in total joint arthroplasties have led to a resurgence of interest in
metal-on-metal bearing surfaces. Metal-on-metal articulations were among the first to be used in total hip arthroplasty and had clinical success in the 1960s and
1970s in designs such as the McKee-Farrar (Howmedica,
Limerick, Ireland) hip replacement. Recent studies suggest that many of these early implants have performed
well, even after implantation intervals approaching 30
years.
Metal-on-metal hip joint replacements fell from favor when the clinical results of polyethylene-on-metal

American Academy of Orthopaedic Surgeons

joints proved superior and as carcinogenic concerns

over local and systemic accumulation of particulate and
ionic metallic debris grew. These early failures were
partly the result of poor metallurgy and less than optimum implant design. Casting of metallic alloys often resulted in microstructures with large grains and poorly
distributed carbides, resulting in inhomogeneity in surface hardness. Early metal-on-metal designs often had
small head to neck ratios; therefore, impingement between the neck of the femoral component and the rim
of the acetabular components was a common occurrence.
More recently, improved metallurgy and manufacturing techniques have led to the reintroduction of metalon-metal bearings for hip replacement. Cobaltchromium alloys with well-controlled grain sizes and
finely distributed carbides provide superior hardness
and wear resistance compared with earlier versions of
the alloy and to implants made of stainless steel and titanium alloy. Laboratory evidence from hip joint simulator studies has confirmed that these improved bearing
surfaces can provide low friction and low wear articulations. Clearance and conformity between the mating
surfaces are now recognized as important parameters
that must be controlled as part of the design and manufacturing processes.
The strength of cobalt-chromium alloys (compared
with polyethylene) and their increased toughness over
ceramics provides additional benefits from the standpoint of hip implant design. For example, the wall thickness of solid metallic acetabular components can be
smaller than modular polyethylene and metal or ceramic and metal implants, so larger head sizes can be incorporated, providing an advantage for patients where
joint stability is a concern. Similarly, the ability to manufacture large metallic shells allows for surface replacement of the hip joint, which is a bone-conserving operation with particular indications for young, active patients
with good bone stock in the femoral head and neck.
Laboratory and analytical evidence suggest that
wear rates of metal-on-metal bearings can be markedly
affected by head size. Unlike metal-on-polyethylene
bearings, in which larger head sizes tend to increase
wear because of increased sliding distances between the
articulating surfaces, wear decreases significantly in
metal-on-metal bearings with larger heads. Large femoral head-acetabular component combinations seem to
form a protective lubricating film that separates the surfaces. Smaller head sizes ( 28 mm) have much less or
no separation and therefore increase the propensity for
adhesive and abrasive wear.
Corroborating clinical evidence of the effect of head
size on wear rates in metal-on-metal hip replacements
does not yet exist. With most modern versions of alternative bearings, clinical results are available only for
short- to intermediate-term usage. To date, few serious

59

Biomaterials and Bearing Surfaces in Total Joint Arthroplasty

wear-related problems such as osteolysis have been
found. Nonetheless, serum concentrations of metallic elements including chromium, cobalt, and molybdenum
are found in increased levels in patients receiving metalon-metal hip replacements compared with control
groups and patients receiving metal-on-polyethylene
joint implants. The same cytokines found in abundance
around failed metal-on-polyethylene joints have been
identified around failed metal-on-metal joints, suggesting that the same biologic pathways for osteolysis exist
in these joints. Laboratory evidence also suggests that
increased adhesion can occur between metal-on-metal
bearing surfaces at the startup of motion after a resting
period (common in patients daily activities); the clinical
ramifications remain unknown.
Only longer and more extensive clinical experience
will determine if the reduced volumetric wear with metalon-metal bearings will lower the incidence of osteolysis
and wear-related failures. Data will soon become available
because metal-on-metal hip joints have received approval
by the FDA for commercial distribution in the United
States. Surface replacements using metal-on-metal bearings are currently in the investigational stage.An early report (2- to 6-year follow-up) showed a revision rate of 3%
because of loosening and femoral neck fracture.
Metal-on-metal bearings have not been used for
other joints, such as the knee. Most other joints require
very different design concepts and are subject to additional wear mechanisms for which metal-on-metal surfaces have few advantages.

Ultra-High Molecular Weight Polyethylene

Clinical follow-up for polyethylene-on-metal bearings
now extends beyond 25 years, which is longer than for
some other bearing types. Considerable biologic and
clinical evidence shows that polyethylene wear debris
elicits a deleterious biologic response culminating in osteolysis. Biomaterial approaches to combat polyethylene
wear that do not require the introduction of alternative
bearing materials are attractive because they can capitalize on the relatively low cost of manufacturing polyethylene components, the material biocompatibility and
toughness in bulk form, and an easier path to regulatory
approval. In the past, polyethylene components have
been sterilized by exposure to gamma radiation at a
higher dose than 25 kGy in an ambient environment. It
is now known that this technique causes oxidative degradation. The exposure to radiation energy causes chain
scission, chain cross-linking, and the creation of free
radicals in the material, resulting in a decrease in molecular weight, an increase in density, and detrimental alterations in mechanical properties that add significantly
to the wear problem.
In response to these findings, several alternative sterilization techniques have been introduced with the aim

60

Orthopaedic Knowledge Update 8

of eliminating degradation. These techniques do not
have the same beneficial impact on wear. Sterilization
techniques that eliminate irradiation, such as exposure
to gas plasma and ethylene oxide, also eliminate the potential benefit of the additional cross-linking produced
by irradiation. Hip joint simulator studies have shown a
higher rate of wear for components sterilized with gas
plasma and ethylene oxide compared with gamma irradiation; however, the long-term clinical relevance of
these differences remains unknown. Little is known
about the impact of irradiation-free sterilization techniques on the pitting and delamination surface damage
that occurs in total knee polyethylene components.
Studies of retrieved tibial components originally sterilized in ethylene oxide have shown less evidence of
cracking than conventionally sterilized components.
Other sterilization techniques that use irradiation in
an inert or low-oxygen environment are more beneficial
in improving wear resistance, based on improved wear
rates found in hip simulator studies. The improvement
in wear behavior is probably the result of the crosslinking of polyethylene chains that occurs with irradiation. However, free radical production is still a detrimental result, and postirradiation thermal treatment is
performed to quench free radicals and prevent degradation that could occur with subsequent exposure to an
oxygen environment.
The fabrication technique used in manufacturing
polyethylene total joint components also affects wear
behavior. Direct compression molding in which polyethylene powder is heated and pressed in molds to form a
finished product results in polyethylene components resistant to oxidative degradation. In clinical use for more
than 20 years, directly molded hip and knee replacement components have shown favorable wear resistance, based on clinical evidence, studies of wear damage in retrieved components, and laboratory wear tests
compared with components machined from extruded
polyethylene stock. The lower elastic modulus that results from compression molding may be a key factor.
Under the same load conditions, a lower modulus polyethylene will have larger contact areas and will experience lower stresses than a higher modulus polyethylene
under the same load conditions.
The most significant recent alteration in fabrication
techniques for manufacturing polyethylene joint arthroplasty components is the inclusion of elevated levels of
radiation to induce even higher levels of cross-linking
than occur with the conventional sterilization dose
(25 kGy). Advantages of elevated cross-linking include
significantly reduced wear (as shown in hip and knee
joint simulator test) and, in the case of knee joints, a resistance to pitting and delamination. Wear rates of essentially zero have been shown, for example, in hip simulator tests for elevated cross-linked polyethylenes
(Figure 1). Few well-controlled clinical studies in the

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 5 Biomaterials and Bearing Surfaces in Total Joint Arthroplasty

peer-reviewed literature exist to date on the in vivo

wear behavior of these materials. Short-term results suggest an improvement in wear; however, a longer
follow-up period is required.
The improved abrasive and adhesive wear resistance
that accompanies elevated cross-linking provides other
potential benefits. Backside wear in modular implants,
for example, is an important complication that could be
reduced with the use of elevated cross-linked instead of
conventional polyethylene. Increased wear resistance
also has renewed interest in the use of larger femoral
heads for total hip replacement. The risk of dislocation
is markedly reduced with a larger head size. However,
because sliding distance between the bearing surfaces
and the amount of wear is higher with a larger head,
conventional polyethylene-metal bearings usually have
a small diameter ( 32 mm). Larger head sizes are now
available with matching larger diameter, elevated crosslinked polyethylene acetabular components. The wall
thicknesses of these components are thin (less than
5 mm in some instances), making the strength and
toughness of elevated cross-linked polyethylenes an important consideration.
Changes in mechanical properties that accompany
increased cross linking may pose the biggest threat to
the clinical efficacy of these materials. Reduced toughness and resistance to fatigue crack propagation have
been shown in several laboratory studies using standardized specimens under controlled loading conditions.
These findings suggest greater susceptibility to pitting
and delamination wear damage, to gross failure should a
crack propagate entirely through a component, and to
dissociation that could result from the failure of a locking mechanism that relied on the structural integrity of
the polyethylene. The clinical relevance of standardized
fracture and fatigue test results is difficult to interpret,
because the fracture conditions in implant components
depend on geometry and loading conditions that differ
from those of standard laboratory specimens. Analytical
attempts to interpret performance with consideration of

Figure 1 The wear rates of acetabular cups tested in a hip simulator decrease with
exposure to higher doses of radiation applied to the components prior to testing.
(Adapted with permission from McKellop H, Shen FW, Lu B, Campbell P, Salovey R:
Development of an extremely wear-resistant ultra high molecular weight polyethylene
for total hip replacements. J Orthop Res 1999;17:157-67.)
lowered toughness and crack propagation resistance are
hampered by the lack of good nonlinear material models for polyethylenes stress-strain behavior. Close monitoring of data from the clinical experience will be
needed to establish the extent of this problem.
Wear tests of knee-like geometries (with nonconforming bearing surfaces) and knee joint simulator studies show that elevated cross-linked materials perform
well in comparison with conventional polyethylene.
Given their decreased fracture resistance, the hypothesis
would be the oppositewear rates and pitting and
delamination damage should be worse for elevated
cross-linked materials. Lowering of the elastic modulus
that accompanies the post cross-linking thermal treatment used to quench free radicals may explain these
positive findings. As in the case of compression-molded
polyethylene, the lower modulus creates larger contact
areas, lower stresses, and better resistance to wear damage (Figure 2).

Figure 2 Wear tracks from a conventional polyethylene (1050 resin irradiated at 25 kGy) (A) and an elevated cross-linked polyethylene (1050 resin irradiated at 65 kGy) (B),
which were tested on a knee-like wear apparatus for 2 million cycles under identical conditions show more severe damage in the conventional material. The elastic moduli were
1.0 GPa for the conventional and 800 MPa for the elevated cross-linked material. (Reproduced with permission from Furman BD, Maher SA, Morgan T, Wright TM: Elevated
crosslinking alone does not explain polyethylene wear resistance, in Kurtz SM, Gsell R, Martell J (eds): Crosslinked and Thermally Treated Ultra-High Molecular Weight Polyethylene for Joint Arthroplasties ASTM STP 1445. West Conshohocken, PA, ASTM International, pp 248-261.)

American Academy of Orthopaedic Surgeons

61

Biomaterials and Bearing Surfaces in Total Joint Arthroplasty

Recent findings of increased fatigue crack propagation in elevated cross-linked polyethylenes may prove
problematic for in vivo wear behavior. These results
suggest that if cracks form in vivo after a large number
of cycles (larger than the number typically used in laboratory tests), failure might progress rapidly. Thus, even if
crack initiation is retarded because stresses are lower,
the end results could still be less favorable.

Ceramics
Ceramic materials are solid, inorganic compounds consisting of metallic and nonmetallic elements. Held together by ionic or covalent bonding, ceramics are stiff
(high elastic modulus), hard, brittle, very strong under
compressive loads, and possess excellent biocompatibility and exceptional wear resistance. Ceramic materials
usually have polygranular microstructures similar to metallic alloys. Their properties depend on factors such as
grain size and porosity (for example, strength is inversely proportional to both grain size and porosity).
Fully dense ceramics, such as alumina and zirconia,
are used in total joint arthroplasty components specifically because they provide more wear-resistant bearing
surfaces; they have few other mechanical advantages for
joint arthroplasty. Ceramic-on-polyethylene bearings
have been commercially available for some time as alternatives to conventional metal-on-polyethylene.
Ceramic-ceramic bearings have only recently received
regulatory approval for commercial distribution in the
United States. Because of their hardness and strength,
ceramics can be polished to a very smooth finish and resist roughening while in use as a bearing surface. They
possess good wettability, suggesting that lubricating layers may form between ceramic couplings, thus reducing
adhesive forms of wear.
The most significant disadvantage of ceramics is low
toughness, which resulted in a significant number of ceramic head fractures during early clinical use in total hip
replacement. More recently, however, improvements in
ceramic quality, most notably increased chemical purity
and reduced grain size, has led to a dramatic reduction
in head fractures. Toughness is also a concern in the use
of ceramic-on-ceramic acetabular components for hip
replacements. Retrieval studies of polyethylene acetabular components suggest that impingement between the
neck of the femoral component and the rim of the acetabular component is a common occurrence. With the
use of ceramic components, impingement could cause
significant damage and eventual fracture. Recent laboratory tests suggest, however, that with the improved
quality of ceramic materials the possibility of
impingement-related failure is quite low.
Three types of ceramic bearing materials are commercially available. Bulk implant materials made from
alumina and zirconia have been used for decades,

62

Orthopaedic Knowledge Update 8

mainly for femoral heads in total hip arthroplasties.
Long-term clinical experience with these materials has
been published. More recently, an oxidized zirconium
material has been introduced into both hip and knee replacement components for articulation against polyethylene; however, few results are available and clinical experience with this material is short term.
Long-term experience with alumina-on-polyethylene
bearings for hip replacement shows reduced wear rates
compared with metal-on-polyethylene bearings and an
associated decrease in the presence of osteolysis, suggesting that these types of bearings are beneficial in improving clinical performance. Alumina-on-polyethylene
bearings for knee replacements have a much more limited use. Mid-term results are excellent; however, the
absence of direct comparisons with conventional metalon-polyethylene bearing surfaces of the same design
and the lack of long-term results make it difficult to assess the clinical benefits.
The use of zirconia as a bearing surface against polyethylene has not been proven as clinically successful as
alumina-on-polyethylene bearings. High wear rates,
which resulted in catastrophic failure as early as 5 years
after implantation, have been reported; however, these
failures may have resulted from the use of an enhanced
form of polyethylene (Hylamer, DePuy-Dupont, Newark, DE) in the acetabular component, that has been
prone to rapid wear. More convincing data exist from a
direct comparison between alumina-, zirconia-, and
metal-on-conventional polyethylene bearings in patients
with total hip arthroplasties. Data show the highest wear
rate in the zirconia group, which is consistent with an increased monoclinic content on the surface of retrieved
zirconia heads from the same series. The propensity for
zirconia to transform from a tetragonal crystalline form,
which is stable at elevated temperatures, to the less
tough monoclinic form is a disadvantage of this material
and has led to FDA warnings against autoclave resterilization of zirconia heads. The decrease in toughness
makes the material more susceptible to roughening and
increased wear. Manufacturing problems led to a high
incidence of fracture and prompted a recent voluntary
recall of nine batches of zirconia femoral heads, which
further undermined confidence in this material.
Ceramic-on-ceramic bearings have been used extensively in total hip arthroplasty in Europe, though hip implants using this bearing combination have only recently
received regulatory approval in the United States.
Alumina-on-alumina joints have shown very low clinical
wear rates; however, the results are design dependent
(these bearings can show excessive wear if incorporated
into an inferior design). Recent reports also show excellent wear resistance of alumina-on-alumina joints in
young patients with no measurable wear and no evidence of osteolysis even at more than 10-year follow-up.
Furthermore, head fractures have not been observed in

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 5 Biomaterials and Bearing Surfaces in Total Joint Arthroplasty

this high-demand patient population, which supports the

supposition that alumina-ceramic materials have improved mechanical properties. Zirconia-on-zirconia
bearings have not performed well based on laboratory
findings of high wear.
The use of ceramic bearings in other joint arthroplasties is more limited in scope. For implant designs
dominated by the potential for abrasive and adhesive
wear (bearing surfaces that remain conforming throughout the useful ranges of motion of the implant), ceramic
bearing surfaces may have all of the potential benefits
of ceramic hip replacement components. For joint designs such as conventional knee replacements that require nonconforming articulations to adequately function, the advantages of ceramic materials are less clear.
Although knee joint simulator data show substantial decreases in wear rate (for example, an oxidized zirconia
surface bearing against polyethylene), the clinical benefit of this decreased wear rate remains uncertain. Osteolysis has not been as significant a complication in
knee replacement as in hip replacement surgery; longterm clinical data will be required to substantiate a clear
advantage. Fatigue-related pitting and delamination is
the main wear problem after knee replacement. This
type of wear is caused by the large cyclic stresses generated at and near the polyethylene surface of the tibial
component as the more rigid femoral component moves
across it. Ceramic materials provide no advantage in
such a situation.
Bulk ceramic components for knee replacement have
been used primarily in Japan, where both ceramic-onpolyethylene and ceramic-on-ceramic bearings have been
used clinically. Evidence of the efficacy of either bearing
combination is limited by the length of follow-up (< 9
years), the lack of comparative data for other bearing surfaces with the same design, and patient selection (for example, one study was limited to patients with rheumatoid
arthritis). Whereas satisfactory outcomes have resulted
with the use of ceramic bearings in total knee arthroplasty,
no data yet exist to prove them superior to the more common metal-on-polyethylene articulations.

Cements
Polymethylmethacrylate (PMMA) bone cement has
been the polymer of choice as a grouting agent to secure implant components to bone since its introduction
by Charnley in the 1970s. The basic principles of the in
situ polymerization remain the same. Liquid methylmethacrylate monomer with the addition of hydroquinone (to inhibit premature polymerization) and
N,N,-dimethyl-P-toluidine (to accelerate polymerization
once mixing commences) is mixed with prepolymerized
PMMA, which also contains dibenzoyl peroxide (to initiate the polymerization process) and a radiopaque material (usually barium sulfate or zirconia). Variations on

American Academy of Orthopaedic Surgeons

this basic approach include powders that are a blend of

a copolymer (PMMA and polystyrene or PMMA and
methacrylic acid) to increase toughness.
The heat given off during polymerization is substantial. The rise in temperature that occurs in the adjoining
tissues is dictated by the volume and shape of the cement bulk and the heat transfer properties of the surrounding structures. Bone cement temperatures during
in vivo polymerization are usually below that required
to denature proteins or kill bone cells. The long-term
successful use of such cement in an array of orthopaedic
applications shows that the thermal properties do not
affect clinical efficacy.
Antibiotics can be added to bone cement to provide
prophylaxis or treatment of infection. The elution of antibiotic from the cement is dictated by the preparation
technique, chemistry, and surface area of the cement.
The elution of gentamycin from commercially available
cements varies significantly depending on the brand
used. Bone cement properties can be detrimentally altered by the addition of antibiotics during the mixing
process, an important clinical concern when antibiotics
are added to the cement at surgery. The FDA has recently approved commercially prepared antibiotic cements for use in the United States; this will markedly
reduce potential complications.
The performance of cement as a grout for fixation of
joint arthroplasty components has been enhanced by
improved protocols in cement handling, bone preparation, and cement delivery. For example, vacuum-mixed
or low viscosity formulations of cement that are delivered into the bone under pressure using a cement gun
have a significantly reduced porosity compared with
hand-mixed cement. Reducing the porosity results in
more cement in the mantle and increased structural
strength, even though the material properties of the cement itself remain unaltered. Although difficult to prove
clinically, a reduction in porosity should reduce the
chances of cement mantle fracture and subsequent implant loosening.
PMMA cement has often been used solely as a
structural component to fill bone defects and as a loadcarrying element of the resulting composite. Vertebroplasty and kyphoplasty are gaining acceptance as treatments for painful osteoporotic compression fractures.
Vertebroplasty was first used for osteolytic spinal conditions secondary to cancer; injection of cement into the
vertebral body can prevent further collapse and relieve
pain. Kyphoplasty, an extension of vertebroplasty, uses
an inflatable bone tamp to restore the vertebral height
while creating a cavity to be filled with bone cement.
Short-term clinical results with both procedures are
promising; however, several biomechanical questions remain, including the appropriate properties for the cement, the amount of fill, and the effects of vertebra reinforcement on the adjacent spinal motion segments.

63

Orthopaedic Knowledge Update 8

Biomaterials and Bearing Surfaces in Total Joint Arthroplasty

Serious complications related to cement leakage, including soft-tissue damage and nerve root pain and compression, have prompted the FDA to issue a public
health notification concerning this off-label use of cement.
Interest in using bioactive cements in these types of
applications is growing. Cements based on hydroxyapatite and calcium phosphate have been studied to determine their ability to provide strength and stiffness to osteoporotic vertebral bodies compared to that achieved
with PMMA. Handling and radiopacity difficulties have
hampered the clinical use of these cements in spinal applications. These cements have been found to be effective as augments to standard fixation in several fractures, including those of the distal radius and femoral
neck.

Biodegradable Polymers
Biodegradable polymers degrade chemically in a controlled manner over time. Orthopaedic applications for
these materials include sutures, screws, anchors, and pins
designed to slowly lose their mechanical function as
they resorb and the surrounding tissue heals. The tissue
assumes its normal mechanical role, while the resorption
of the polymer eliminates the need for a second surgical
procedure to remove the device. Resorbable polymers
also are used for drug delivery, and considerable research is underway to develop biodegradable scaffolds
for tissue engineering. This last application is especially
challenging because the scaffold must provide a suitable
biologic environment for the cells to be delivered within
the material and a suitable mechanical environment so
that they manufacture extracellular matrix with the appropriate biomechanical properties. Biocompatibility is
a very important consideration; the polymer must degrade without adversely affecting the cells or the tissues
that replace it.
Common bioresorbable polymers include polylactic
acid, polyglycolic acid, polydioxanone, and polycaprolactone. Their mechanical properties span large ranges, depending on polymer type, the addition of copolymers,
molecular weight, fabrication technique, and the addition of reinforcing materials such as fibers. Orthopaedic
applications are limited by the strengths of these materials, which are insufficient for many load-bearing situations common in the musculoskeletal system. Bioresorbable polymers are most often used in applications
involving trauma; such as the fixation of small cancellous bone fractures. The most common use reported in
the literature is in association with malleolar fractures.
Hydrogels are a soft, porous-permeable group of
polymers that are nontoxic, nonirritating, nonmutagenic,
nonallergenic, and biocompatible. They readily absorb
water (and thus have high water contents), and have
low coefficients of friction and time-dependent mechan-

64

ical properties that can be varied through altering the

materials composition and structure. Hydrogels have
been considered for use in a wide range of biomedical
and pharmaceutical applications; orthopaedic applications include tissue engineering of cartilage and bone
and for drug delivery.

Interfaces
In orthopaedic applications, biomaterials are generally
used to fabricate entire structures for specific purposes
(for example, a fracture plate or a tibial knee replacement insert). Difficulties in achieving long-term permanent fixation of orthopaedic devices to the skeleton has
led to the development of specific materials and technologies intended to enhance biologic fixation. For example, tantalum is a highly biocompatible, corrosionresistant, osteoconductive material. Recently, porous
forms of tantalum deposited on pyrolytic carbon backbones have been suggested as superior structures for
bone ingrowth. Orthopaedic applications include coatings for joint arthroplasty components (acetabular cups
and tibial trays) and as spinal cages. Experimental work
in animal models and randomized trials in humans suggest that this may be a useful material for achieving fixation to bone.
Hydroxyapatite, a ceramic, has been available as a
coating on joint arthroplasty implants for some time.
Animal studies showed that such coatings increase fixation strength by preferential deposition of new bone at
the interface. Despite this in vivo evidence, no clear advantage has been shown in the clinical literature;
whereas some reports suggest enhanced fixation, others
show no benefit. This disagreement may be the result of
several factors, including patient selection, implant design, and the quality and type of hydroxyapatite used.

Summary
A clinical improvement in wear rate is possible with the
introduction of alternative bearing materials. In all
cases, however, compromises exist; although wear behavior can be improved, other properties are altered in
potentially detrimental ways (Table 1). The use of preclinical assessment tools such as joint simulators are vital to establishing efficacy, but the shortcomings of these
tools must be considered when interpreting the clinical
relevance of test results. Long-term follow-up data from
well-controlled studies remain the only real test of efficacy.

Annotated Bibliography
General Reference
McKellop HA: Bearing surfaces in total hip replacements: State of the art and future developments. Instr
Course Lect 2001;50:165-179.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 5 Biomaterials and Bearing Surfaces in Total Joint Arthroplasty

This review article discusses the advantages and disadvantages of bearing surfaces including the new polyethylenes,
modern metal-metal, and ceramic-ceramic bearings. The goal
of this review is to provide surgeons with the information
needed to assess the risk-benefit ratios of each of the new
bearing combinations.

Santavirta S, Bohler M, Harris WH, et al: Alternative

materials to improve total hip replacement tribology.
Acta Orthop Scand 2003;74:380-388.
This review article examines proposed improvements in
tribology of bearing surfaces for total hip replacement. The
three approaches examined were highly cross-linked
UHMWPE, aluminum oxide, and metal-on-metal. The findings
in support of their efficacy were emphasized.

Wright TM, Goodman SB (eds): Implant Wear in Total

Joint Replacement: Clinical and Biologic Issues, Materials and Design Considerations. Rosemont, IL, American
Academy of Orthopaedic Surgeons, 2001.
This monograph provides answers to important clinical, biologic, and bioengineering questions concerning the complications caused by wear in total joint arthroplasties. The answers
were drafted by invited attendees at a workshop sponsored by
the National Institutes of Health and the American Academy
of Orthopaedic Surgeons and are supported by a comprehensive bibliography.

Metal-on-Metal Bearings
Amstutz HC, Beaule PE, Dorey FJ, Le Duff MJ, Campbell PA, Gruen TA: Metal-on-metal hybrid surface arthroplasty: Two to six-year follow-up study. J Bone Joint
Surg Am 2004;86:28-39.
Encouraging results were found in this follow-up study
(average follow-up 3.5 years) of 400 hips treated with metalon-metal surface arthroplasties. Survival rates of the arthroplasties at 4 years were 94.4%; however, a 3% revision rate for
loosening of the femoral component and femoral neck fracture were also found.

Goldberg JR, Gilbert JL, Jacobs JJ, Bauer TW, Paprosky

W, Leurgans S: Links: A multicenter retrieval study of
the taper interfaces of modular hip prostheses. Clin
Orthop 2002;401:149-161.
This study found that corrosion and fretting of modular
taper surfaces was significantly worse for mixed alloy couples
compared with similar alloy couples and were also dependant
on implantation time and flexural rigidity of the neck. The results suggest that this damage was attributable to a mechanically assisted, crevice corrosion process.

Smith SL, Dowson D, Goldsmith AA: The effect of femoral head diameter upon lubrication and wear of metalon-metal total hip replacements. Proc Inst Mech Eng
[H] 2001;215:161-170.
A hip joint simulator study of metal-on-metal joints with
different head diameters showed no surface separation with

American Academy of Orthopaedic Surgeons

smaller heads and significantly higher wear rates compared

with those found with larger diameter heads. In the larger
heads, surface separation occurred, suggesting the formation
of a protective lubricating film.

Ultra-High Molecular Weight Polyethylene

Burroughs BR, Rubash HE, Harris WH: Femoral head
sizes larger than 32 mm against highly cross-linked polyethylene. Clin Orthop 2002;405:150-157.
In vitro wear testing and anatomic studies, together with
prior clinical studies in which large femoral heads were used,
support the hypothesis that highly cross-linked UHMWPE allows the use of femoral heads larger than 32 mm.

Collier JP, Currier BH, Kennedy FE, et al: Comparison

of cross-linked polyethylene materials for orthopaedic
applications. Clin Orthop 2003;414:289-304.
Physical and mechanical properties of six commercially
available cross-linked polyethylene materials were obtained
from tests in a single laboratory and reported along with wear
data and summary descriptions of the materials by the manufacturers.

Digas G, Karrholm J, Thanner J, Malchau H, Herberts P:

Highly cross-linked polyethylene in cemented THA:
Randomized study of 61 hips. Clin Orthop 2003;417:126138.
Highly cross-linked polyethylene components (irradiated
to 95 kGy) were compared with conventional polyethylene
components in a randomized study of cemented hip replacements. At 2-year follow-up, the highly cross-linked cups
showed 50% reduction of proximal wear based on radiostereometric examinations; however, the follow-up period was
short and the results preliminary.

DLima DD, Hermida JC, Chen PC, Colwell CW Jr:

Polyethylene cross-linking by two different methods reduces acetabular liner wear in a hip joint wear simulator. J Orthop Res 2003;21:761-766.
A comparison of wear reduction by two different crosslinking methods, 9.5 Mrad electron beam and 10 Mrad gamma
irradiation, showed no clinically significant differences. Both
methods were superior to conventional nominally cross-linked
(gamma sterilized) polyethylene in wear resistance.

Endo MM, Barbour PS, Barton DC, et al: Comparative

wear and wear debris under three different counterface
conditions of crosslinked and non-crosslinked ultra high
molecular weight polyethylene. Biomed Mater Eng
2001;11:23-35.
The wear of grade GUR 1020 cross-linked, GUR 1120
cross-linked, and noncross-linked GUR 1020 UHMWPE
(Ticona, Summit, NJ) was compared in multidirectional wear
tests using three different counterface conditions. Better wear
resistance was found with cross-linking for smooth counterfaces, but with rough and scratched counterfaces, the GUR

65

Orthopaedic Knowledge Update 8

Biomaterials and Bearing Surfaces in Total Joint Arthroplasty

1020 cross-linked material produced significantly higher wear
rates than the noncross-linked material.

more severe wear. No change in wear rate occurred in the alumina group, which showed considerably less wear.

Hopper RH Jr, Young AM, Orishimo KF, Engh CA Jr:

Effect of terminal sterilization with gas plasma or
gamma radiation on wear of polyethylene liners. J Bone
Joint Surg Am 2003;85:464-468.

Urban JA, Garvin KL, Boese CK, et al: Ceramic-onpolyethylene bearing surfaces in total hip arthroplasty:
Seventeen to twenty-one-year results. J Bone Joint Surg
Am 2001;83:1688-1694.

Patients who received hip replacements with polyethylene

acetabular liners sterilized with gamma radiation (61 hips at
mean follow-up of 5.2 years) experienced an average of 50%
less wear than patients with noncross-linked liners sterilized
with gas plasma (63 hips at mean follow-up of 3.9 years).

Findings showed that long-term use of ceramic-onpolyethylene bearings implanted with cement by one surgeon
resulted in outstanding long-term clinical and radiographic results and wear rates lower than those previously reported for
metal-on-polyethylene bearings. These findings support the
use of such bearings in total hip arthroplasty.

Muratoglu OK, Bragdon CR, OConnor DO, Perinchief

RS, Jasty M, Harris WH: Aggressive wear testing of a
cross-linked polyethylene in total knee arthroplasty.
Clin Orthop 2002;404:89-95.
A knee simulator study comparing elevated cross-linked
and conventional tibial inserts showed extensive delamination
and cracking as early as 50,000 cycles for conventional polyethylene inserts, whereas the cross-linked polyethylene inserts
did not show any subsurface cracking or delamination at
0.5 million cycles.

Cements
Cassidy C, Jupiter JB, Cohen M, et al: Norian SRS cement compared with conventional fixation in distal radial fractures: A randomized study. J Bone Joint Surg
Am 2003;85:2127-2137.
A prospective, randomized multicenter study of closed reduction and immobilization with and without the use of calcium phosphate bone cement for the treatment of distal radial
fractures, found significant clinical improvement during the
first 3 months of treatment with cement augmentation.

Ceramics
Allain J, Roudot-Thoraval F, Delecrin J, Anract P, Migaud H, Goutallier D: Revision total hip arthroplasty
performed after fracture of a ceramic femoral head: A
multicenter survivorship study. J Bone Joint Surg Am
2003;85:825-830.
One hundred five surgical revisions to treat a fracture of a
ceramic femoral head, performed at 35 institutions, were studied. Although the fractures were potentially serious events,
treatment with total synovectomy, cup exchange, and insertion
of a cobalt-chromium or new ceramic femoral ball minimized
the risk of early loosening and the need for one or more repeat revisions.

Hamadouche M, Boutin P, Daussange J, Bolander ME,

Sedel L: Alumina-on-alumina total hip arthroplasty: A
minimum 18.5-year follow-up study. J Bone Joint Surg
Am 2002;84:69-77.
A retrospective long-term study showed a 20-year survival
rate of 85% for alumina-on-alumina hip replacements with
minimal wear rates and limited osteolysis, provided that the
acetabular component remained well-fixed. No alumina head
or socket fractures occurred and osteolysis was moderate in
about 10% of the patients.

Hernigou P, Bahrami T: Zirconia and alumina ceramics

in comparison with stainless-steel heads: Polyethylene
wear after a minimum ten-year follow-up. J Bone Joint
Surg Br 2003;85:504-509.
Comparison of ceramic and metal bearing surfaces showed
increased polyethylene wear at 5- to 12-year follow-up for zirconia and metal heads. The group with the zirconia heads had

66

Garfin SR, Reilley MA: Minimally invasive treatment of

osteoporotic vertebral body compression fractures.
Spine J 2002;2:76-80.
A literature review with presentation of early results of a
national, clinical study showed 95% of patients treated for
painful osteoporotic compression fractures had significant improvement in symptoms and function after kyphoplasty or vertebroplasty. Only kyphoplasty improved vertebral height and
kyphosis.

Lewis G: Fatigue testing and performance of acrylic

bone-cement
materials:
State-of-the-art
review.
J Biomed Mater Res 2003;66B:457-486.
A literature review of the fatigue behavior of acrylic bonecement formulations showed that an increase in molecular
weight leads to an increase in fatigue life; the effect of vacuum
mixing on fatigue life remains controversial.

Biodegradable Polymers
Grande DA, Mason J, Light E, Dines D: Stem cells as
platforms for delivery of genes to enhance cartilage repair. J Bone Joint Surg Am 2003;85(suppl 2):111-116.
Stem cells transduced with either bone morphogenetic
protein-7 or sonic hedgehog gene were delivered to osteochondral defects in rabbits using bioresorbable scaffolds. The
addition of either factor enhanced the quality of the repaired
tissue, showing the utility of tissue-engineering gene therapy
strategies.

Hovis WD, Kaiser BW, Watson JT, Bucholz RW: Treatment of syndesmotic disruptions of the ankle with bio-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 5 Biomaterials and Bearing Surfaces in Total Joint Arthroplasty

absorbable screw fixation. J Bone Joint Surg Am 2002;

84:26-31.
Syndesmotic disruptions associated with malleolar fractures in which the syndesmosis was fixed with poly-L-lactic
acid screws healed without displacement of the syndesmosis or
widening of the medial clear space. No episodes of osteolysis
or late inflammation secondary to the hydrolyzed polylactide
occurred.

Interfaces
Kim YH, Kim JS, Oh SH, Kim JM: Comparison of
porous-coated titanium femoral stems with and without
hydroxyapatite coating. J Bone Joint Surg Am 2003;85:
1682-1688.
At a mean follow-up of 6.6 years postoperatively, the clinical and radiographic results associated with proximally
porous-coated femoral prostheses with identical geometries
that differed only with regard to the presence or absence of
hydroxyapatite coating were found to be similar.

Boehler M, Plenk H Jr, Salzer M: Alumina ceramic

bearings for hip endoprostheses: The Austrian experiences. Clin Orthop 2000;379:85-93.
Chan FW, Bobyn JD, Medley JB, Krygier JJ, Tanzer M:
The Otto Aufranc Award. Wear and lubrication of
metal-on-metal hip implants. Clin Orthop 1999;369:
10-24.
Dorr LD, Wan Z, Longjohn DB, Dubois B, Murken R:
Total hip arthroplasty with use of the Metasul metal-onmetal articulation: Four to seven-year results. J Bone
Joint Surg Am 2000;82:789-798.
Dorr LD, Wan Z, Song M, Ranawat A: Bilateral total
hip arthroplasty comparing hydroxyapatite coating to
porous-coated fixation. J Arthroplasty 1998;13:729-736.
Fenollosa J, Seminario P, Montijano C: Ceramic hip
prostheses in young patients: A retrospective study of 74
patients. Clin Orthop 2000;379:55-67.

Wigfield C, Robertson J, Gill S, Nelson R: Clinical experience with porous tantalum cervical interbody implants
in a prospective randomized controlled trial. Br J
Neurosurg 2003;17:418-425.

Jacobs JJ, Skipor AK, Patterson LM, et al: Metal release

in patients who have had a primary total hip arthroplasty: A prospective, controlled, longitudinal study. J Bone
Joint Surg Am 1998;80:1447-1458.

A prospective randomized study of the ability of porous

tantalum implants to achieve cervical interbody fusion showed
inferior end-plate lucency in four patients raising concerns
about delayed or nonfusion. Fusion subsequently occurred in
all patients at 12 months after surgery.

McKellop H, Shen FW, Lu B, Campbell P, Salovey R:

Development of an extremely wear-resistant ultra high
molecular weight polyethylene for total hip replacements. J Orthop Res 1999;17:157-167.

Classic Bibliography

Oonishi H, Kadoya Y: Wear of high-dose gammairradiated polyethylene in total hip replacements.

J Orthop Sci 2000;5:223-228.

Bartel DL, Rawlinson JJ, Burstein AH, Ranawat CS,

Flynn WF Jr: Stresses in polyethylene components of
contemporary total knee replacements. Clin Orthop
1995;317:76-82.

Ritter MA: Direct compression molded polyethylene

for total hip and knee replacements. Clin Orthop 2001;
393:94-100.

Bobyn JD, Toh KK, Hacking SA, Tanzer M, Krygier JJ:

Tissue response to porous tantalum acetabular cups: A
canine model. J Arthroplasty 1999;14:347-354.

Rokkanen PU, Bostman O, Hirvensalo E, et al: Bioabsorbable fixation in orthopaedic surgery and traumatology. Biomaterials 2000;21:2607-2613.

American Academy of Orthopaedic Surgeons

67

Chapter

Physiology of Aging
Susan V. Bukata, MD
Mathias Bostrom, MD
Joseph A. Buckwalter, MD
Joseph M. Lane, MD

Introduction
Aging is a complex process that involves changes in
many of the physiologic functions of the body; aging
also may create changes in social and economic conditions that influence lifestyles and daily routines. The accumulated effects of past diseases, side effects of medications, and environmental factors can change an
individuals ability to perform daily activities. Older
people often have difficulty sleeping and many experience at least some recent memory loss. The heterogeneity of the aging body also makes patient response to a
particular medication or treatment much more variable
than in a younger population. To care for the elderly patient and improve quality of life, it is important to understand the changes that occur with aging and to address the special needs that these changes produce. The
United States Census Bureau projects that by the year
2010, one quarter of the US population will be age 55
years or older. This projection is based in part on increasing life expectancies for men and women. Health
care providers must understand the special needs associated with aging to provide appropriate treatment for
this expanding population of patients.

Changes in Neuromuscular Control

The maintenance of a moderate level of neuromuscular
control is essential for the continuation of normal daily
routines. Many changes that occur with natural aging affect neuromuscular control. These changes affect an individuals ability to perform certain tasks and result in
changes in proprioception that affect balance and increase the risk for falling. Changes in proprioception
stem from a variety of sources, such as sensory changes
that affect vision, hearing, and vestibular function, as
well as age-related changes in the central and peripheral
nervous systems. All of these changes affect coordination and musculoskeletal function by altering the neural
control and stimuli of muscles, which affect mobility.
Many other changes occur in the central and peripheral nervous systems with aging. Brain atrophy of 30%
to 40% in the frontal lobe, temporal lobe, and basal gan-

American Academy of Orthopaedic Surgeons

glia occurs with a decrease in the total number of neurons. A decrease in brain tissue metabolism accompanies a decrease in cerebral blood flow. Processing in
both nervous systems slows with increasing age, with
nerve conduction velocity decreased 10% to 15%. Autonomic and muscle stretch reflexes become less sensitive
and righting reflexes decrease in acuity, resulting in increased body sway that makes it more difficult for elderly people to respond to sudden positional changes.
Reaction times in older adults are 20% longer than
those in young adults, which partially accounts for the
35% to 40% increase in falls seen in adults older than
60 years.
An age-related decrease in the number of spinal motor neurons probably contributes to these changes. Single peripheral motor neurons innervate groups of skeletal muscle fibers forming a motor unit. Accompanying
the loss of spinal motor neurons is a concomitant loss in
the total number of motor units in old muscles and remodeling of other motor units. This decrease in total
motor units is actually a specific loss of fast motor units
and an increase in slow motor units. During normal aging, a reorganization of the motor unit pool for skeletal
muscle occurs, which supports the belief that some fast
fibers may undergo denervation and others may be reinnervated by sprouting nerves from the slow motor units.
Changes also occur at the neuromuscular junction,
which affect the recruitment of muscle fibers for coordinated activity. With aging, there is a degeneration of the
neuromuscular junction that prevents or slows the transmission of neural stimuli to muscle fibers. All of these
changes contribute to slower reaction times.
Many sensory changes also occur with aging that can
make it difficult for eldery people to respond to changes
in their environment. Increasingly poor vision, macular
degeneration, stereopsis, cataracts, and poor night vision
commonly occur with aging. Hearing loss with a decrease in high-frequency acuity (affecting the ability to
distinguish words from background noise) affects almost
30% of patients older than 70 years and almost 90% of
nursing home residents. Impaired vestibular function
also affects many older patients and may play a signifi-

69

Physiology of Aging

Figure 1 A, Osteoporotic trabecular bone. B, Normal trabecular bone. (Adapted with

permission from Dempster DW, Shane E, Horbert W, Lindsay R: A simple method for
correlative light and scanning electron microscopy of human iliac crest bone biopsies:
Qualitative observations in normal and osteoporotic subjects. J Bone Miner Res 1986;
1:15-21.)
cant role in balance impairment and falls. Neuropathy
secondary to diabetes and nerve injuries can limit sensation in the feet and impair the ability to respond to uneven terrain. Impaired cognition secondary to medications, dementia, depression, or stroke also can impair an
older persons ability to respond appropriately to their
environment and is a risk factor for appendicular fracture.

Changes in Body Composition and

Functional Reserve
Age-related changes occur in body composition that result in decreased bone mass, decreased muscle mass, and
an increase in body fat. Sarcopenia, or muscle mass loss,
reduces basal metabolism and is the predominant cause
of decreased muscle strength seen with aging. This loss
of strength can have a profound effect on an elderly
persons ability to perform simple tasks such as walking
and rising from a chair. Peak muscle mass is attained at
approximately 30 years of age. After age 50 years, approximately 15% of muscle mass is lost per decade, increasing up to 30% per decade after age 70 years. Muscle atrophy occurs from both a reduction in fiber size
and quantity. The equal loss of type I (fast twitch) and
type II (slow twitch) fibers accounts for most of the atrophy associated with aging; however, the type II fibers
also decrease in size more than the type I fibers and
subsequently compose a smaller percentage of the muscle cross-sectional area. A 26% reduction in the size of
type II fibers occurs from age 20 to 80 years and accounts for a large proportion of age-related muscle mass
loss. This decrease in fiber size is not seen with normal
disuse atrophy. Overall muscle cross-sectional area de-

70

Orthopaedic Knowledge Update 8

creases 40% and the total number of fibers declines
39% between the ages of 20 and 80 years. In addition,
greater muscle mass is lost in weight-bearing muscles
than in nonweight-bearing muscles. These combined
changes in muscle mass lead to a prolongation of twitch
contraction and a reduction in voluntary strength. Poor
quadriceps function has been identified as a key risk
factor for falls and hip fracture.
Aging also causes many significant changes in the
cardiovascular system, such as changes in functional reserve capacity that affect an individuals ability to respond to physiologic stresses. Increased vascular stiffness, hypertension, and other factors result in a
decreased ability to increase cardiac output in response
to physiologic stress. Maximal oxygen consumption, a
measurement that serves as a good measure of athletic
fitness, declines 5% to 15% each decade after age 25
years. Endurance training is able to slow this decline up
to 50% until approximately age 70 years. Older patients
are able to respond to exercise and to improve muscle
strength and tone and cardiovascular response; however, even very fit older adults may have a maximal
functional reserve that only is equivalent to that of
younger sedentary individuals.
Bone mass also changes as a part of normal aging.
Peak bone mineral density is reached between age 25
and 30 years and begins to decline in the fourth and
fifth decades of life. Bone mineral density declines 0.3%
to 0.5% per year and the rate of loss may increase for
women to 2% to 3% per year in the 7- to 10-year period
of perimenopause. The rate of bone loss and the associated increase in fracture risk is related to a complex
combination of genetics, body mass characteristics, nutritional factors, and diseases. A complex relationship,
commonly referred to as coupling, exists between osteoblasts and osteoclasts, which balances bone formation
with bone resorption. It is ultimately this relationship
between the osteoblasts and osteoclasts that determines
the rate of bone mineral density loss. After peak bone
mass is attained, the full bone remodeling unit is not
completely replaced during the process of normal bone
turnover, which results, over time, in decreased bone
mass. During the perimenopausal period, the activity of
osteoclasts is enhanced and osteoblasts are unable to replace bone at the same pace. Architectural changes
eventually occur as bone trabeculae decrease in number
and lose continuity. Whereas cortical bone thins, the
losses are greater in the trabecular bone mass (Figure
1). More information about bone mineral density and
bone remodeling can be found in chapter 18. All of
these processes combine to increase the fragility of aging bone. Fracture repair is not delayed or compromised, but fixation failure is more common because of
inadequate bone stock. Almost 98% of elderly patients
who sustain a hip fracture have some level of bone deficiency. Low-energy hip fractures are associated with a

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

20% increase in mortality within the first 6 months after
injury. Although patients with two or more vertebral
fractures have a 25% to 30% increase in mortality compared with peers who do not sustain such fractures, this
difference is reflected over the much longer period of
5 years. Because of age-associated bone loss, elderly
people are at greater risk for low-energy fractures and
related mortality.
Normal aging results in many changes in the endocrine system that affect a variety of body tissues and
functions. Changes in glucose metabolism include an increase in peripheral glucose resistance and an increase
in postprandial blood glucose and insulin levels. These
changes are influenced by diet, exercise, and body fat
content, all of which often change with aging. This altered glucose metabolism results in a higher rate of
insulin-resistant diabetes (type II diabetes) in the older
population. Decreased levels of growth hormone, insulin
growth factor, estrogen, and testosterone also occur with
aging. Patients have been treated with growth hormone
supplementation, but the benefits are not clear. Estrogen supplementation was used for many years as the
mainstay of treatment for patients with perimenopausal
bone loss; however, global risk including increases in the
prevalence of breast cancer, strokes, pulmonary emboli,
and blood clots are believed to outweigh the benefits
obtained in fracture reduction and decreased bone loss.
Estrogen supplementation is no longer recommended as
a treatment if its sole purpose is to prevent bone loss.

Age-Related Changes in Articular Cartilage

Most of the cells and extracellular matrix proteins of articular cartilage undergo very little turnover, which results in an accumulation of age-related changes and
damage to the tissue over many years. In the world population, almost 25% of people older than 60 years have
osteoarthritis, which makes age the greatest risk factor
for developing this disease. However, the relationship
between age and osteoarthritis is not well understood.
The age-related changes in articular cartilage are somewhat different from those seen in osteoarthritis; however, these changes affect the biomechanical, biochemical, and cellular characteristics of the tissue and possibly
increase the risk of damage and progression to osteoarthritis. The extent of degenerative changes varies by
joint; however, not all degenerative changes are associated with joint pain and loss of motion. All degenerative
changes do not progress to osteoarthritis, which further
indicates the complex relationship between the two processes. Age-related changes to the synovium, subchondral bone, menisci, ligaments, and the joint capsule may
all contribute to the progression of joint symptoms and
the establishment of osteoarthritis.
Articular cartilage undergoes age-related changes in
thickness, cellular function, matrix tensile properties, ma-

American Academy of Orthopaedic Surgeons

Chapter 6 Physiology of Aging

trix composition, and molecular organization. Localized,
superficial fibrillation of articular surfaces can be first
seen in adolescents as they approach skeletal maturity.
During skeletal growth, there is a noted decrease in chondrocyte density, an associated decrease in synthetic activity, and a change in the proteoglycan component ratios.
With increasing age, there is a decrease in the ratio of
chondroitin sulfate to total glycosaminoglycans, an increase in the production of chondroitin 6 sulfate, and a decrease in the production of chondroitin 4 sulfate. There
also is a decreased response to growth factors, especially
transforming
growth
factor-,
which
exerts
almost no influence on chondrocytes once skeletal maturity is reached, but is a strong stimulant to glycosaminoglycan production (particularly chondroitin 4 sulfate). There
is an age-associated decrease in the water content of cartilage, in direct contrast to the changes that occur with degenerative arthritis. In osteoarthritis, water content initially increases as a result of a disruption in the
architecture of the matrix molecules. Other matrix
changes associated with aging include an increase in decorin concentration, collagen cross-linking, collagen fibril diameter, and variability in collagen fibril diameter.
Chondrocyte proliferation and cell density begin to
decline from birth, and reach a steady plateau by skeletal maturity. Although cell number may not vary significantly with increasing age, the synthetic capabilities and
cell morphology continue to change. Cells accumulate
intracytoplasmic filaments and may lose some of their
endoplasmic reticulum. Cell culture studies using chondrocyte samples extracted from human articular cartilage show a decrease in absolute chondrocyte yield and
in cell proliferation after the age of 30 years. These cell
cultures are responsive to platelet-derived growth factorBB, with a threefold increase in glycosaminoglycan
production up to age 40 years; there is minimal response
after that age. A recent study showed that human articular cartilage chondrocytes become senescent with increasing patient age, which may account for the decreased capacity of chondrocytes to respond to injury. In
addition, oxidative stress to the cartilage tissue may
cause premature chondrocyte senescence, making the
tissue susceptible to injury for a longer period of time.
Chondrocyte cultures maintained in 5% oxygen doubled 60 times before becoming senescent, whereas increasing the oxygen concentration to 21% cut the number of population doublings to 40. Increased oxidative
stress occurs after trauma and with inflammation and
may explain why injury increases a patients risk for osteoarthritis.
Mechanical studies have shown significant agerelated changes in the tensile properties of articular cartilage including decreases in stiffness, fatigue resistance,
and strength. The exact cause of these changes is not
known but it is likely that changes in the collagen matrix, the proteoglycans, and the decrease in water con-

71

Physiology of Aging
tent all play a role. The increase in advanced glycosylation end products (AGEs) and their effect on cartilage
mechanical properties are of significant interest. AGEs
adversely influence cartilage turnover, causing decreases
in matrix synthesis and degradation. Their presence is
accompanied by increased cartilage stiffness and brittleness, possibly because of increased cross-linking of the
collagen molecules by the AGEs. A canine model of anterior cruciate ligament (ACL) transaction showed increased collagen damage, impaired matrix repair, decreased proteoglycan synthesis, and more severe
osteoarthritis in the animals treated to have a higher
concentration of AGEs in their tissues, similar to that
found in older dogs. If there is confirmation that these
findings are correlated with the onset of osteoarthritis,
reversing or preventing the changes modulated by
AGEs may represent a novel approach to preventing
the onset of osteoarthritis. Collagen fibrils also become
larger in diameter and more variable in size with aging;
this change is attributed to a decrease in the type XI
collagen component. The larger, more cross-linked
fibrils are more rigid and may limit the ability of the articular cartilage surface to deform without cracking.

Age-Related Changes in Intervertebral Disk

Low back pain, which affects a substantial portion of
the adult population, is one of the most common reasons for an adult to consult a primary care physician. Although it is difficult to separate age-related changes
from degenerative changes to the intervertebral disk, it
is clear that substantial changes occur (possibly more
than occur to any other musculoskeletal tissue) as a result of aging. There are also substantial individual differences in the rate that these changes occur, which may
explain the difficulty in separating the effects of aging
from those of degeneration. Several general trends link
the blood supply, cellular and matrix proliferation, and
tissue hydration to the structural changes observed in
the intervertebral disk.
Beginning at birth, the abundant notochordal cells
found in the nucleus pulposus begin to disappear,
whereas chondrocyte density increases. The cellular density at the disk end plates begins to increase and disorganization of the cartilaginous regions is noted. By the
second decade of life, blood vessels within the disk begin to disappear and cleft formation in the nucleus pulposus begins. During adulthood, the number of clefts
continues to increase and radial tears develop that extend into the disk periphery and the anulus fibrosus.
There is a marked decline in the number of cells within
the disk, and the rate of cellular apoptosis increases.
Both the volume and shape of the disk change. Mucoid
degeneration with an increase in noncollagenous protein content occurs and there is a concomitant decrease
in proteoglycan and water content. Collagen content de-

72

Orthopaedic Knowledge Update 8

clines and the remaining collagen is composed of fibrils
with increased diameter and diameter variability, and
decreased pyridinoline crosslinks. In later adulthood,
there is increased scar formation within the disk (including the anulus fibrosus), a disruption in the normal
lamellar pattern of collagen fibrillar layers within the
anulus fibrosus, and a decrease in the functional stretch
of these collagen molecules. In elderly people, it becomes difficult to distinguish the anatomic regions of
the disk.
Significant alterations in the blood supply to the disk
appear to parallel the structural changes. With increasing
age, physiologic vessels within the disk disappear. The ossification of the vertebral end plates that occurs with skeletal maturity may significantly affect the nutritional supply and hydration of the adjacent disk.The cells within the
disk begin to depend on diffusion of nutrients from the
periphery, a process that appears to be hindered as the
number of vessels supplying the disk declines and increased intradiskal matrix slows diffusion through the tissues. Decreased hydration of the central nucleus pulposus
leads to stiffening of the surrounding anulus fibrosus and
changes in spine biomechanics that can affect load sharing between the adjacent facet joints, ligaments, paraspinal muscles, and vertebral bodies. Other diseases and lifestyle habits that affect small vessel proliferation and
function, including diabetes and smoking, may accelerate
the rate of change in the disk tissues.

Age-Related Changes in Tendon, Ligament,

and Joint Capsule
Injuries to ligaments, tendons, and joint capsules are
commonly seen in middle-aged and elderly individuals.
These injuries can result in significant impairment in
strength, range of motion, and stability in the regions
where they occur. Significant pain and disability can result from low-energy ruptures of the rotator cuff, quadriceps, patellar, bicep, or posterior tibial tendons, and
form ruptures to spinal and wrist ligaments following
physical activity or even as a result of the normal activities of daily living.
Although the cellular and mechanical properties that
change with aging have not been fully defined in all of
these tissues, similarities in structure and composition
suggest that the known observations are probably applicable to all of these tissues. Fibroblast cells form a major
proportion of ligaments, tendons, and joint capsules. As
fibroblasts age, the cells flatten, elongate, and lose most
of their protein-producing cell organelles, including the
rough endoplasmic reticulum and Golgi apparatus. This
finding suggests that the tissues are less biosynthetically
active, and may be less able to efficiently respond to injury. Several factors may contribute to the age-related
increase in injuries to these tissues. A decreasing blood
supply may cause changes in cell activity and matrix

American Academy of Orthopaedic Surgeons

Chapter 6 Physiology of Aging

Orthopaedic Knowledge Update 8

composition and organization that lead to tissue degeneration. Although biochemical analysis of these tissues
has not identified dramatic changes in the matrix composition, a decrease in collagen cross-links (common in
many musculoskeletal tissues with aging) may contribute to the observed age-related decrease in water content. Collagen bundles in these tissues are highly oriented, a property that contributes to the mechanical
characteristics of the tissues. Animal studies have suggested that changes in the collagen organization, including fibril alignment and fiber bundle orientation, occur
with aging. This change in the organization of the collagen fibers may account for the altered mechanics and
changes in the ultimate load-to-failure. One study of human ACLs showed decreasing tensile stiffness and ultimate load-to-failure with increasing age. Another study
showed that the ligament-bone complexes in people
older than 60 years tolerated only one third the load before failing compared with those of young adults. A recent study of human posterior cruciate ligaments found
a decrease in collagen fiber diameter and an increase in
collagen fibril concentration with aging.

Nutrition
Many changes occur to the gastrointestinal system as a
part of natural aging. Changes in the autonomic nervous
system affect colonic motility and can result in constipation, whereas other neurologic changes to the anorectal
region can result in fecal incontinence. Past diseases also
can have an effect on the gastrointestinal system. Elderly patients with a history of diabetes may have special
dietary needs to regulate their blood glucose levels and
also may have additional bowel motility needs as a result of neuropathic changes. Patients may have difficulty
swallowing (secondary to stroke) and also may have
special dietary needs after abdominal surgery for ulcers,
diverticulitis, or colon cancer. All of these concerns can
make it difficult for elderly patients to obtain adequate
nutrition from food intake alone.
Protein is essential for the maintenance of muscle
mass and for the formation of some of the extracellular
components of bone, especially collagen. Follow-up on
age-related bone loss for patients in the Framingham
study showed lower levels of bone loss if adequate protein was present in the diet. It is widely recognized that
patients with low serum albumin levels have difficulty
with wound healing. Survival rates in elderly patients
with hip fractures also is well correlated with adequate
serum albumin levels. Many elderly patients do not receive adequate protein in their diets because of dietary
choices or social or economic concerns; such patients
should be encouraged to increase protein intake until
serum albumin levels are within normal limits. Some patients also have inadequate caloric intake. If patients are
unable to consume adequate calories with their normal

American Academy of Orthopaedic Surgeons

meals, supplemental nutrition (such as shakes and puddings) should be added to their daily diet.
Older age is associated with a decrease in gastric
acidity, which affects the absorption of calcium and vitamin B12. When prescribing calcium supplementation for
elderly patients, it is important to remember that a high
percentage of them will be naturally achlorhydric or
taking an H2 blocker and are unable to absorb calcium
carbonate supplements. Such patients must take calcium
citrate supplements, which can be absorbed in the absence of stomach acid. A dietary goal for calcium intake
should be 1,500 mg daily for older individuals, which
should be taken in divided doses of 500 mg or less to
optimize absorption from each dose.
Although changes occur in the liver that affect the efficiency of drug metabolism, these changes do not seem
to be the primary cause of vitamin D deficiency in elderly people. Serum 25-hydroxyvitamin D levels decline
with age primarily because of decreased sun exposure
and a decrease in the efficiency of vitamin D production
in the skin. Changes in the kidney do not seem to affect
vitamin D levels, except in patients with renal failure.
Supplementation with 400 to 800 IU of vitamin D daily
is recommended for all older patients. Those who are
vitamin D-deficient (such as patients taking seizure medications, those with sprue, or with irritable bowel syndrome) may require a higher dosage.

Fracture Healing
Fracture healing occurs either through a cartilage callus
(endochondral bone formation) when fracture fragments are not in close apposition, directly onto the surface of existing bone (appositional bone formation), or
along a collagen matrix that does not contain any cartilage (intramembranous bone formation). These modes
of fracture healing occur in people of all ages, but the
speed and efficacy of bone healing declines with increasing age. When skeletal maturity is reached, the periosteum gradually thins and the chondrogenic and osteogenic potential of its mesenchymal cells declines with
age. A recent study with rabbits showed that the percentage of proliferating mesenchymal chondrocyte precursor cells did not change between young and old animals. However, the same study showed that the absolute
number of proliferating cells decreased in the older rabbits because the size and total cell number in the periosteum was decreased. Conflicting evidence has been
found concerning the change in the number and the
proliferative capacity of osteoprogenitor cells. One
study of human bone marrow that was aspirated from
the iliac crest showed an age-related gradual decline in
precursor colony-forming units. Another study of human iliac crest bone marrow also found no change in
precursor cell number or proliferative capacity with advancing age or with the presence of osteoporosis.

73

Physiology of Aging
Delayed fracture healing has been reported both in
humans and in animals with aging. Because most experiments done on fracture healing have involved young or
young adult animals, few data exist on fracture healing
in older animals. Experiments performed to assess the
effect of aging on fracture healing showed no differences in the biochemical parameters of fracture healing.
One recent study comparing 6-week-old rats with
1-year-old rats showed no differences in the messenger
RNA (mRNA) expression of several cytokines and proteins involved in fracture healing. Despite this similarity
in mRNA expression, femur fracture healing was delayed in the older animals. Expression of Indian hedgehog and bone morphogenetic protein-2 was lower in the
older animals at the time of fracture callous formation
and may have contributed to the delay in healing. This
same study found differences in the baseline expression
mRNA levels in the young rats even over the short period of the study, suggesting that the age and metabolic
status of the animal or patient must be taken into account when interventions are considered to enhance
bone repair. In another rat study, fractures in young animals were healed in 40 days and the bones had normal
mechanical properties by 4 weeks. In the older animals,
fracture healing was delayed to 80 days, and normal mechanical properties were not regained until 12 weeks.
The delayed return of mechanical properties may reflect
the increased incidence of hardware failure in older patients with fractures. The bone holding the hardware in
place actually fails in many older patients, rather that
the hardware itself breaking before fracture healing is
complete.
The age-related decline in the rate of fracture healing may be explained by several mechanisms. Aging is
related to a general functional decline in the homeostatic mechanisms of skeletal tissues. There is a decline
in the expression of osteoinductive cytokines and
growth factors both at baseline and with injury in older
animals, caused in part by a reduction in the inflammatory response to injury. In addition, the bone inductive
potential of demineralized bone matrix decreases with
aging. The inductive potential of bone matrix appears to
be growth hormone-dependent; growth hormone secretion decreases with age.

Exercise and Fall Prevention

More than 300,000 hip fractures occur annually in the
United States, and over 90% of these fractures are associated with a fall. Each year, 30% of community dwelling people older than 65 years fall, and nursing home
dwellers average 1.5 falls per person per year. A variety
of factors increase the risk of a fall, including muscle
weakness, abnormalities in gait or balance, poor vision,
decreased sensation in the lower extremities and associated changes in proprioception, functional limitations,

74

Orthopaedic Knowledge Update 8

cognitive impairments, and the side effects of medications. The risk of falling increases with an increased
number of impairments, ranging from 8% in patients
with no impairments to 78% in patients with four or
more impairments. Abnormal body sway and gait are
correlated with falls; several tests can be used to assess
these risks in patients. The average osteoporotic 80-yearold patient should be able to perform a single leg stance
for at least 12 seconds. An inability to perform this maneuver implies compromised balance. Timed get up and
go tests (which evaluate a patients ability to rise from a
chair without using their arms and to transition to beginning gait) and evaluation of heel-toe straight line
walking can be used to identify patients at risk for falls.
Several strategies can be used to prevent falls and
injuries associated with falls. Identification of a patients
risk factors for falling are an important initial step in
choosing preventive strategies. Patients with gait abnormalities and muscle weakness may require an assistive
device such as a cane or walker to improve mobility. The
removal of throw rugs, the installation of safety rails in
the bathroom and on stairs, the use of nightlights, and
the use of proper shoe wear can help to reduce the risk
of falling posed by home hazards. Other preventive
measures may include the withdrawal of medications
that are associated with falls (especially psychotropic
drugs). Exercise programs such as tai chi and physical
therapy can improve balance, muscle strength, and flexibility, reducing falls by 46%. The use of energyabsorbing flooring and hip protectors are indicated for
patients who are at high risk for injury from falls. Hip
protectors are made of molded polypropylene or polyethylene and worn as a part of an undergarment over
normal underwear. The use of hip protectors has been
shown to reduce the average patients overall risk of hip
fracture by 60%, and by more than 80% if the hip protector was being worn at the time of the fall. Patient
compliance in using this device is often less than 60%
because of the bulkiness of the undergarment and the
difficulties some patients have donning and removing it.
Exercise has been shown to decrease the rate of decline of many physiologic changes associated with aging
by improving cardiac function, strength, balance, and
flexibility. Muscle strength reliably increases in older patients who begin resistance exercises, and improvements
in muscle mass and flexibility also are seen. Weightbearing exercises also decrease bone density losses. An
exercise program for older patients must take into account their functional limitations and goals, modifying
the choice of exercises as well as the frequency and duration of the training to increase compliance with the
program. Elderly people can improve their functional
independence, increase mobility, and reduce the risk of
falling by participating in an exercise program. Elderly
patients who maintain a reasonable level of exercise tolerance or who can be rehabilitated to this level of activ-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 6 Physiology of Aging

ity with a proper physical conditioning program can reduce the decline in overall function that occurs with
advancing age.

Martin JA, Lingelhutz AJ, Moussavi-Harami F, Buckwalter JA: Effects of oxidative damage and telomerase
activity on human articular cartilage chondrocyte senescence. J Gerontol A Biol Sci Med Sci 2004;59:324-337.

Annotated Bibliography

Chondrocyte cell lines were grown at 5% oxygen and at

21% oxygen and analyzed for population doublings. Those
grown at 21% oxygen reached senescence at 40 population
doublings, whereas those exposed to 5% oxygen concentration
reached senescence at 60 population doublings. These findings
show that oxidative stress, like that seen in injury and inflammation, can lead to premature chondrocyte senescence. This
finding is also an important consideration in establishing the
conditions necessary during chondrocyte transplantation.

Changes in Body Composition and Functional Reserve

Buckwalter JA, Heckman JD, Petrie DP: An AOA critical issue: Aging of the North American population. New
challenges for orthopaedics. J Bone Joint Surg Am 2003;
85:748-758.
A comprehensive review of the physiologic changes that
occur with aging and the demands these changes create for the
health care system is presented. The authors also address the
response the health care system may need to make to serve
the aging population of North America.

Cummings SR, Melton LJ: Epidemiology and outcomes

of osteoporotic fractures. Lancet 2002;359:1761-1767.
The rate of osteoporotic fractures is increasing with the
aging of the world population. The authors note the need for
epidemiologic research to identify those individuals most at
risk for fracture (especially hip fracture) and for treatment
with medication. Financial resources for treatments could then
be allocated to those most at risk for a disabling fracture.

Stenderup K, Justesen J, Eriksen EF, et al: Number and

proliferative capacity of osteogenic stem cells are maintained during aging and in patients with osteoporosis.
J Bone Miner Res 2001;16:1120-1129.
The results of a study that examined the effect of aging
and osteoporosis on the mesenchymal stem cell population are
presented. Iliac crest bone marrow from 23 younger patients
(age range, 22 to 44 years), 15 older patients (age range, 66 to
74 years), and 13 osteoporotic patients (age range, 58 to 83)
were cultured in vitro. Total colony-forming units, colony sizes,
and cell density per colony were assessed. Results showed no
differences for any group in the number and proliferative capacity of osteoprogenitor cells. The results of this study were
in contrast with some previous studies and suggest that the defective osteoblast functions seen with aging and osteoporosis
are caused by some other factor.

Age-Related Changes in Articular Cartilage

Degroot J, Verzijl N, Wenting-van Wijk MJ, et al: Accumulation of advanced glycation end products as a molecular mechanism for aging as a risk factor in osteoarthritis. Arthritis Rheum 2004;50:1207-1215.
This study examined the role of AGEs in the development
of osteoarthritis. Using a canine ACL transection model, the
affected joints were injected with ribose to mimic the presence
of AGE found in older dogs. The development of osteoarthritis and the degree of collagen damage and proteoglycan release was then analyzed. AGEs were associated with more severe osteoarthritis.

American Academy of Orthopaedic Surgeons

Ulrich-Vinther M, Maloney M, Schwarz EM, et al: Articular cartilage biology. J Am Acad Orthop Surg 2003;11:
421-430.
A comprehensive review is presented of the current information available about the biology of articular cartilage and
the effects on this tissue on various injuries and diseases including osteoarthritis, osteochondral fracture, and microscopic
damage. This article also summarizes the cytokines and growth
factors that are involved in both normal tissue metabolism and
response to injury. A discussion of the rationale and evidence
of effectiveness of currently available treatments for osteoarthritis precedes a discussion of future strategies for treatment
based on new information about articular cartilage biology.

Age-Related Changes in Intervertebral Disk

Boos N, Weissbach S, Rohrbach H, Weiler C, Spratt KF,
Nerlich AG: Classification of age-related changes in
lumbar intervertebral discs: 2002 Volvo Award in basic
science. Spine 2002;27:2631-2644.
A thorough analysis of the histologic age-related changes
that occur to the human intervertebral disk is presented. One
hundred eighty intervertebral disk specimens from a broad
age range of individuals (range, fetal to 88 years of age) were
analyzed for histologic changes in the cells, matrix, blood supply, and overall structure. The study is the first to provide histologic evidence of the changes in blood supply that occur to
the disk and precede the structural and cellular changes that
are associated with aging and degeneration of the intervertebral disk.

Age-Related Changes in Tendon, Ligament, and Joint

Capsule
Sargon MF, Doral MN, Atay OA: Age-related changes
in human PCLs: A light and electron microscopic study.
Knee Surg Sports Traumatol Arthrosc 2004;12:280-284.
This study of 36 specimens from patients of various ages
used light and transmission electron microscopy to analyze ultrastructural differences. Collagen fibers were most variable in
diameter in patients age 10 to 19 years. Aging was associated
with a decrease in collagen fiber diameter and an increase in
collagen fibril concentration with a maximum fibril concentration occurring in patients age 60 to 69 years.

75

Physiology of Aging

Nutrition
Meydani M: The Boyd Orr lecture: Nutrition interventions in aging and age-associated disease. Proc Nutr Soc
2002;61:165-171.
This article presents a review of the physiologic changes
that occur with aging and the concomitant socioeconomic factors that influence the dietary regimen of elderly people. Dietary modifications to maximize the nutritional intake and accommodate the normal changes of aging are discussed.

Fracture Healing
Koval KJ, Meek R, Schmitsch E, Liporace F, Strauss E,
Zuckerman JD: Geriatric trauma: Young ideas. J Bone
Joint Surg Am 2003;85:1380-1388.
A review of the special considerations needed when treating trauma injuries in elderly patients is presented. Discussion
includes the physiologic changes to bone and soft tissues that
occur with aging, as well as patient factors including medical
comorbidities and cognitive status issues that are common in
this population. The article reviews the current information on
the timing of surgery, anesthesia considerations, implant
choices, fixation enhancement, and postoperative care for the
elderly population.

Meyer RA Jr, Meyer MH, Tenholder M, Wondracek S,

Wasserman R, Garges P: Gene expression in older rats
with delayed union of femoral fractures. J Bone Joint
Surg Am 2003;85:1243-1254.
The levels of mRNA gene expression during fracture healing are the same in young rats and older rats, even though a
delay in fracture healing is observed in the older rats. Lower
levels of Indian hedgehog and bone morphogenetic protein-2
were detected in the fracture callous of the older rats, which
may contribute to slower fracture repair.

ODriscoll SW, Saris DB, Ito Y, Fitzimmons JS: The

chondrogenic potential of periosteum decreases with
age. J Orthop Res 2001;19:95-103.

Orthopaedic Knowledge Update 8

ciated with falling included female gender, Caucasian race, the
presence of an increased number of chronic diseases, use of
medications, decreased leg strength, poor balance, slower
walking rates, and lower muscle mass. The authors recommend
that all elderly patients should be assessed for fall risk.

Lin JT, van der Meulen MCH, Myers ER, Lane JM:
Fractures: Evaluation and clinical implications, in Favus
MJ (ed): Primer on the Metabolic Bone Diseases and
Disorders of Mineral Metabolism. Washington, DC,
American Society for Bone and Mineral Research, 2003,
pp 147-151.
This chapter describes the compositional changes that occur to bone with aging and how this affects the ability of bone
to absorb loads. Fall prevention strategies and environmental
interventions to decrease applied loads at the time of a fall are
discussed.

Marks R, Allegrante JP, MacKenzie CR, Lane JM: Hip

fractures among the elderly: Causes, consequences, and
control. Ageing Res Rev 2003;21:57-93.
A comprehensive review of the physiologic changes of aging and the risk factors and treatment options for hip fractures
is presented. Comprehensive references for all of the pertinent
articles published in the past decade, which define risk factors
for hip fracture and physiologic factors related to fracture, are
provided.

Tinetti ME: Clinical practice: Preventing falls in elderly

persons. N Engl J Med 2003;348:42-49.
Case examples are presented to address the potential
causes (use of medication, sensory changes, visual disturbances, and cognitive problems) of falls in the elderly population. The author presents recommendations for assessing a patient during an office visit for fall risk and offers interventions
that can be used based on the deficits noted.

The chondrogenic potential of the periosteum of rabbits

ranging in age from 2 weeks to 2 years decreased with aging.
The most notable change occurred in the total number of cells
and the thickness of the periosteum. Although the percentage
of proliferating cells did not dramatically change with age, the
decrease in the total number of proliferating cells correlated
with the decline in the chondrogenic potential of the periosteum with aging of the rabbits.

Classic Bibliography

Exercise and Fall Prevention

Galea V: Changes in motor unit estimates with aging.

J Clin Neurophysiol 1996;13:253-260.

DeRedeneire N, Visser M, Peila R, et al: Is a fall just a

fall? Correlates of falling in healthy older persons: The
Health, Aging, and Body Composition Study. J Am
Geriatr Soc 2003;51:841-846.
This study evaluated 3,075 high-functioning adults (70 to
79 years of age) who were living independently in the community. Frequency of falls over a 1-year period were studied. Results indicated that 24% of women and 18% of men in this
study fell at least once during that 1-year period. Factors asso-

76

Bemben MG: Age-related alterations in muscular endurance. Sports Med 1998;25:259-269.

Buckwalter JA: Aging and degeneration of the human
intervertebral disc. Spine 1995;20:1307-1314.

Galloway MT, Jokl P: Aging successfully: The importance of physical activity in maintaining health and
function. J Am Acad Orthop Surg 2000;8:37-44.
Hannan MT, Tucker KL, Sawson-Hughes B, Cupples
LA, Felson DT, Kiel DP: Effect of dietary protein on

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

bone loss in elderly men and women: The Framingham
Osteoporosis Study. J Bone Miner Res 2000;15:25042512.

Chapter 6 Physiology of Aging

related and species-related changes. J Bone Joint Surg
Am 1976;58:1074-1082.

Martin PE, Grabiner MD: Aging, exercise, and the predisposition to falling. J Appl Biomech 1999;15:52-55.

Quarto R, Thomas D, Liang CT: Bone progenitor cell

deficits and the age-associated decline in bone repair capacity. Calcif Tissue Int 1995;56:123-129.

Melton LJ: Epidemiology of hip fractures: Implications

of the exponential increase with age. Bone 1996;18:121S125S.

Sherman S: Human aging at the millennium, in Rosen

CJ, Glowacki J, Bilizikian JP (eds): The Aging Skeleton.
New York, NY, Academic Press, 1999, pp 11-18.

Nishida S, Endo N, Yamagiwa H, et al: Number of osteoprogenitor cells in human bone marrow markedly
decreases after skeletal maturation. J Bone Miner Metab
1999;17:171-177.

Taafe DR, Marcus R: Musculoskeletal health and the

older adult. J Rehabil Res Dev 2000;37:245-254.

Noyes FR, Grood ES: The strength of the anterior cruciate ligament in humans and rhesus monkeys: Age-

American Academy of Orthopaedic Surgeons

Woo SL-Y, Hollis JM, Adams DJ, et al: Tensile properties of the human femur-anterior cruciate ligament-tibia
complex: The effects of specimen age and orientation.
Am J Sports Med 1991;19:217-225.

77

Chapter

Managing Patient Complaints,

Patient Rights and Safety
Sally C. Booker
Nena Clark, RN

Introduction
Equal in importance to the skills and knowledge of the
medical practice of orthopaedics are the skills that an
orthopaedic surgeon must have or develop to deal effectively with patients from a social and psychological perspective. Experts confirm what may seem obvious, that
patients are unlikely to become involved in litigation
with physicians whom they like. Also, the overall success
of a physicians practice is highly linked to their ability
to interact with patients on a social level.
Dealing with the disgruntled patient or family can be
difficult. Even the most compassionate and skilled orthopaedic surgeon will encounter unhappy patients. It is
important to examine the factors that lead to the problem and develop effective strategies for dealing with the
disgruntled patient. New guidelines have been developed by patient advocacy groups, hospitals, and the federal government regarding patient rights and responsibilities, patient safety, and patient privacy and
confidentiality.

The Social Side of Patient Care

A successful patient interaction often begins with appearance. A professional, well-groomed physician will
foster patient confidence, whereas a disheveled or distracted physician may lead to a sense of distrust. Good
manners and time-honored etiquette go a long way toward making a patient feel comfortable. Most importantly, the physician should always treat a patient with
respect, dignity, courtesy, graciousness, and compassion.
The patients interaction with office staff is equally
important to the overall success of the medical experience. Patients look for continuity in communication and
view office staff as an extension of the physician. Therefore, ensuring that office staff is fostering an environment that is patient-friendly is of paramount importance.

The Disgruntled Patient or Family

Physicians with excellent bedside manners will encounter unhappy patients, although less often than those

American Academy of Orthopaedic Surgeons

physicians who are perceived as arrogant or uncaring.

An understanding of the many factors that can lead to
patient dissatisfaction is important because many situations can be prevented or addressed early, thus minimizing the degree of frustration. Repeated complaints
should lead the physician to carefully analyze their approach to patients and make improvements in their
practice.
A patient sometimes may be dealing with an array
of emotions (sensitivities, embarrassments, fears, shame,
anger, and antagonism) when interacting with the physician and staff. In addition, health problems produce psychological stress that can bring out traits the patient
would not display in normal situations. Prolonged or
painful conditions can lead to frustration and despair
that heighten the patients emotional response.
Remember that a patients complaint is their perception of events and attitudes or, in other words, is the patients version of the truth. When dealing with an angry
patient, the physician should avoid becoming angry or
defensive and remember that the patients anger may be
based in fear, loss of control, the feeling of being disrespected or misunderstood, personal issues unrelated to
the medical situation, or an organic or psychiatric disorder.
Several identifiable factors contribute to patient
complaints, such as inconsistency in communication,
promises that are not kept, lack of sufficient details regarding the diagnosis and treatment plan, perceived
rudeness, lack of understanding regarding known
procedural/surgical complications, the perception that
the physician and staff are too busy to be concerned
with the patients problem, long wait times, and frustration with the inability to fix a painful condition.

Complaint Avoidance
Complaint avoidance often begins with a well-run office. Avoiding long waits and working with courteous
staff go a long way toward avoiding unhappy patients.
The physicians behavior is also paramount in setting
the stage for a positive patient interaction. The physi-

81

Managing Patient Complaints, Patient Rights and Safety

Table 1 | Template for Letter of Response to a Patient

Complaint
Paragraph One
Acknowledge the complaint. Thank the patient for taking the time to
express thoughts or concerns. Assure the patient that a complaint represents an opportunity for the physician to grow. Let the patient know that
his/her concerns have been reviewed and discussed.
Paragraph Two
Address the issue. Explain clinical facts in a language the patient can
understand. If appropriate, tell the patient what steps have been taken
to avoid such an occurrence in the future. If the issue involves potential
liability, a Risk Manager should be consulted before responding to the
patient. Do not respond to the patient in anger.
Paragraph Three
Thank the patient again. Assure the patient that his or her feedback will
or has been used to discuss the concern among staff, that the concern
has given you the opportunity to examine your practices, service delivery,
and others. Tell the patient that it is through patient feedback that your
practice or institution can grow.

cian should be warm and cordial, balancing concern and

interest. Generally, it is best for the physician to sit
down and speak to the patient at their level.
The physician needs to be an excellent listener and if
he/she senses a problem, it is generally best to address it
immediately. A prompt response and action may solve
the problem immediately and prevent the issue from deteriorating to a frustrating situation.

How to Manage a Patient Complaint

When a complaint cannot be avoided or diffused, it is
important to know how to manage a formal complaint.
The first thing to remember is that all complaints must
be taken seriously. The physician is required by law to
respond to all formal complaints in writing. Often additional information about the complaint is needed and
may be obtained from the patient or an institutional patient representative. It is advisable at this time to provide the patient the phone numbers to the patient representative department and allow the patient to take the
initiative to contact the department. The patient should
be asked if he/she would like to receive a telephone call
or a visit from the patient representative. It is always
best to allow the patient to make the decision.
When speaking with dissatisfied patients, always listen attentively to their concern. Avoid the pitfalls of interrupting or becoming defensive. Acknowledge the patients concerns, even if they are unreasonable. Provide
accurate information to help clarify the situation. Be
frank and honest. Avoid negative comments about other
health care providers.
As with all medical care, documentation is vital
when dealing with an unhappy patient. First, the medical assessment should be accurately performed and doc-

82

Orthopaedic Knowledge Update 8

umented. Next, the nature of any discussions with the
patient and the outcome of the discussion should be recorded. Finally, all serious complaints should be formalized by sending a written response. The Center for
Medicare and Medicaid Services specifies as a condition
of participation that all complaints be addressed with a
written response to the patient in a timely manner. A
formal letter provides acknowledgment of the complaint
and furthers documents complaint management. When
writing a letter, it is important to address the complaint
frankly and courteously. In addition to providing the basis of good medical practice, accurate and detailed documentation is vital for cases that result in litigation.
To create a good system for complaint management,
all complaints and records regarding their management
should be kept in a secure location, and labeled quality
improvement or quality assurance. Surveys may be
useful tools and provide feedback about issues of concern to patients that may never reach the formal complaint process. Complaints should be viewed as a process of learning or quality improvement for the
physician and his/her practice.

Which Complaints Require a Written

Response?
A formal written complaint requires a written response
to the patient. A formal complaint is one where the patient explicitly asks that a written report be generated.
A formal complaint also exists when there is a dispute
about charges based on the patients perception of quality of care. If the patient chooses to file a formal complaint, the patient representative in most health care organizations will record the patients feelings and
perceptions and submit that report to the attending physician in charge of the patients care, or to the manager
of the area where the problem occurred. The physician
or his/her designee is then obligated to respond to the
patient in writing (Table 1). In a teaching hospital, it is
always preferable for the attending physician to respond
to the patient rather than the medical or surgical resident.

Health Insurance Portability and

Accountability Act of 1996
The Health Insurance Portability and Accountability
Act of 1996 (HIPAA), Public Law 104-191 required the
Secretary for the US Department of Health and Human
Services to implement Sections 261 through 264 of
HIPAA. The purpose of these sections, known as the
Administrative Simplification provisions, is to improve
the efficiency and effectiveness of the health care system by mandating national privacy standards to enable
electronic exchange of consumers health information.
The Standards for Privacy of Individually Identifiable
Health Information or the Privacy Rule requires cov-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 7 Managing Patient Complaints, Patient Rights and Safety

Table 2 | What Privacy Rule Says About Disclosure of Protected Health Information

Enforcement of Compliance/Penalties for

Noncompliance

Uses and Disclosures of Patient

Health Information Situations

Compliance with the Privacy Rule standards is voluntary. The OCR of the Department of Health and Human Services will perform the enforcement of compliance. The OCR will investigate reports of violation and
covered entities will be subjected to progressive disciplinary actions to demonstrate compliance. The OCR
may impose civil penalties of $100 for each act of noncompliance of the standards and up to $25,000 per year
for multiple identical violations. However, for criminal
penalties, the Department of Justice will perform the investigation, and may impose a fine depending on the severity of violation from $50,000 to $250,000 and imprisonment of 1 to 5 years (OCR Privacy Brief, 2003, pp 1718).

Required disclosures

To the individual (or individuals representative)

who is the subject of information
To Department of Health and Human Services
during a compliance review, investigation or enforcement action
Permitted uses and To the individual who is the subject of information
disclosures (with- For treatment, payment, and health care operations
out an individuals By simply asking for permission from the individual
authorization)
which then give the individual the opportunity to
agree or object
Use or disclosure of information incident to an
already permitted use or disclosure of information
as long as the covered entity has adopted reasonable safeguards and the release of information is
limited to the minimum amount of protected
health information needed to accomplish the
pur pose as required by the Privacy Rule
Public interest and benefit activities (eg, court
order, serious threat to health or safety, essential
government functions)
Limited dataset for the purposes of research, public health or health care operation
No restriction
De-Identified Health Information (removal of individually identifiable health information)
Not covered under Employment records held or maintained by a covprivacy rule
ered entity as an employer
Education and other records defined in the Family
Educational Rights and Privacy Act

ered entities (such as health plans, health care clearinghouses, and health care providers) and its business associates to implement the national standards to protect
the security and privacy of all individually identifiable
health information. The Privacy Rule also requires covered entities to provide individuals an adequate notice
of its privacy practices and a description of their individual rights. Furthermore, covered entities are to make a
good faith effort to obtain a written acknowledgment of
notice of receipt from the individual (Federal Register,
2002, p. 53182.).
Protected health information includes demographic
information such as name, address, birth date, Social Security number, medical record number, and account
numbers that relate to: (1) the individuals past, present,
or future physical or mental health, or (2) the provisions
of health care to the individual, or (3) the past, present,
or future payment for the provision of health care to the
individual. (Office of Civil Rights [OCR] Privacy Brief,
2003, p. 4). Table 2 presents information about uses and
disclosures of Protected health information.

American Academy of Orthopaedic Surgeons

Patient Safety Issues

The Joint Commission on Accreditation of Healthcare
Organizations (JCAHO) is the major accrediting agency
for hospitals. In 1996, JCAHO developed a sentinelevent reporting policy that encouraged an accredited
health care organization to voluntarily report sentinel
events within 5 days and to submit a root cause analysis
within 45 days following discovery. JCAHO expects
health care organizations to perform root cause analyses
of systems and processes of the organization rather than
the performance of individuals. JCAHO defines a sentinel event as an unexpected occurrence involving death
or serious physical or psychological injury or the risk for
which a recurrence would carry a significant chance of a
serious adverse outcome.
Based on the root cause analyses of sentinel events
reported by accredited organizations, JCAHO has identified the need to establish national patient safety goals
(Table 3). JCAHO intends to reevaluate the national
patient safety goals every year. The goals are to be announced in July of each year and to become effective in
January of the following year. In July 2002, JCAHO announced the first National Patient Safety Goals, and in
January 1, 2003, all accredited health care organizations
were required to comply with the established patient
safety goals and its corresponding recommendations.
JCAHO will allow accredited organizations to implement alternatives to specific recommendations as long
as the alternative is as effective as the original recommendation in achieving a specific patient safety goal.
Accredited organizations may submit alternative approaches by completing a Request for Review of an
Alternative Approach to a National Patient Safety Goal
Requirement form available on the JCAHO Website:
http://www.jcaho.org/accredited+organizations/
patient+safety/04+npsg/04_npsg_altform.htm.

83

Orthopaedic Knowledge Update 8

Managing Patient Complaints, Patient Rights and Safety

Table 3 | 2005 National Patient Safety Goals

National Patient Safety Goal

Associated Recommendation

Suggestions for Compliance

Improve the accuracy of patient

identification

Use at least two patient identifiers (neither to be the

patients room number) when taking blood samples or
administering medications or blood products.

Possible identifiers include:

Patients name
Patients birth date
Patients medical record number
Patients Social Security number
Patients address
Example: Ask patient to state his or her name and verify
identification with the patients wrist band. Name and
identification number on the patients wrist band should
be verified and compared with the name and
identification number on the ordered service.
All activities in the operating room should cease in order
for all members of the surgical team to participate. The
surgeon or the circulating nurse must state aloud the
patients name, type of surgery, and location of surgery,
as stated in the patients informed consent form. All
members of the surgical team should respond orally to
affirm that the patients name, procedure, and location
of surgery are correct.
Apply to all verbal or telephone orders (not just for
medication orders) including all critical test results that
are reported verbally or by telephone. Read-back
requirements apply to all including physicians.
The National Coordinating Council for Medication Error
Reporting and Prevention has made recommendations
for improving the use of verbal orders, some of which
are:
Verbal orders should be limited to urgent situations
Entire verbal orders should be repeated back to the
prescriber.
Verbal orders should be documented in the patients
medical record, reviewed, and countersigned by the
prescriber as soon as possible
JCAHO has released a do not use list of abbreviations
such as:
U should be written as unit
IU should be written as international unit
Q.D., QOD should be written as daily and every other
day
Never write a zero by itself after a decimal point (for
example, 1.0 mg). Trailing zero after decimal point can
be mistaken as 10 mg if the decimal point is not seen.
Always use a zero before a decimal point (for example,
0.5 mg). Lack of leading zero before a decimal dose
(for example, 0.5 mg) can be mistaken as 5 mg if the
decimal point is not seen.
MS, MSO4, MgSO4 should be written as morphine
sulfate or magnesium sulfate
The Institute for Safe Medication Practices has also
published a list of dangerous abbreviations and is
available on the Website at:
http://www.ismp.org/MSAarticles/improve.htm

Prior to start of any surgical or invasive procedure, conduct

a final verification process, such as time-out, to confirm
the correct patient, procedure, and site, using activenot
passivecommunication techniques.

Improve the effectiveness of

Implement a process for taking verbal or telephone orders
communication among caregivers that require a verification read-back of the complete
order by the person receiving the order.

Standardize the abbreviations, acronyms, symbols not

to use.

84

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 7 Managing Patient Complaints, Patient Rights and Safety

Table 3 CONTINUED | 2005 National Patient Safety Goals

National Patient Safety Goal

Associated Recommendation

Improve the safety of using

high-alert medications

Remove concentrated electrolytes (including, but not

Applies to all concentrated electrolytes
limited to, potassium chloride, potassium phosphate,
sodium chloride > 0.9%) from patient care units.
Standardize and limit the number of drug concentrations
available in the organization.
Create and use a preoperative verification process, such
Applies to all invasive procedures performed in the
as a checklist, to confirm that appropriate documents are operating room or special procedures unit (for example,
available such as medical records and imaging studies.
endoscopy unit, interventional radiology) that exposes
patients to more than minimal risk (exception:
venipuncture, peripheral intravenous line placement,
placement of nasogastric tube or Foley catheter).
Implement a process to mark the surgical site and involve Marking the site is required for procedures involving right
the patient in the marking process
or left distinction and multiple structures (fingers, toes)
or levels (spinal procedures). The American Academy of
Orthopaedic Surgeons has developed a checklist for
safety called the Sign Your Site initiative that involves
patients to watch and confirm as the surgeon marks the
surgical site. The checklist is available on the Website:
http://www.aaos.org/wordhtml/papers/advistmt/
1015.htm
Ensure free-flow protection on all general-use and patient
controlled analgesia intravenous infusion pumps used in
the organization.
Implement regular preventive maintenance and testing of
alarm systems.
Assure that alarms are activated with appropriate settings
and are sufficiently audible with respect to distances and
competing noise within the unit.
Comply with current Centers for Disease Control and
Each of the CDC guidelines is categorized on the basis of
Prevention (CDC) hand hygiene guidelines
the strength of evidence supporting the
recommendation. All category I must be implemented for
accreditation purposes. Examples include:
When hands are visibly dirty or contaminated, wash
hands with a nonantimicrobial soap and water or an
antimicrobial soap and water.
If hands are not visibly soiled, use an alcohol-based
hand rub for routinely decontaminating hands in all
other clinical situations.
Decontaminate hands before having direct contact with
patients.
Decontaminate hands after contact with a patients
intact skin, nonintact skin, body fluids or excretions,
mucous membranes, and wound dressings.
Decontaminate hands after removing gloves.
Before eating and after using a restroom, wash hands
with a non-antimicrobial soap and water or with an
antimicrobial soap and water.
The full report on the CDC guidelines is available on the
Website: http://www.cdc.gov/mmwr/preview/mmwrhtml/
rr5116a1.htm
Manage as sentinel events all identified cases of
This is already required for any outcomes that resulted in
unanticipated death or major permanent loss of function unanticipated death or major permanent loss of
associated with a health care-acquired infection.
function.

Eliminate wrong site, wrong

patient, wrong procedure surgery

Improve the safety of using

infusion pumps
Improve the effectiveness of
clinical alarm systems

Reduce the risk of health

care-acquired infections

American Academy of Orthopaedic Surgeons

Suggestions for Compliance

85

Managing Patient Complaints, Patient Rights and Safety

Orthopaedic Knowledge Update 8

Table 4 | Consumer Bill of Rights

General Principles

Consumer Rights

Information disclosure

Consumers have the right to receive accurate and easily understood information about health plan, health care
professionals, and health care facilities.
Suggestions for health care organizations to ensure this right:
Provide reasonable accommodation to meet the needs of patients with language barrier, physical or mental
disability.
Health care providers educational preparation (eg, education, board certification, recertification) and
appropriate experience in performing procedures and services.
Performance measures such as consumer satisfaction.
Provide complaints and appeals processes.

Choice of providers and plans

Consumers have the right to a choice of health care providers that is sufficient to ensure access to appropriate
high-quality health care.
Suggestion for health plans to ensure this right:
Provide sufficient numbers and types of providers to encompass all covered services.

Access to emergency services

Consumers have the right to access emergency health care services when and where the need arises. Health
plans should provide payment when a consumer presents to an emergency department with acute symptoms
of sufficient severity including severe pain such than a prudent layperson could reasonably expect the
absence of medical attention to result in placing that consumers health in serious jeopardy, serious
impairment to bodily functions, or serious dysfunction of any bodily organ or part.
Suggestions to ensure this right:
Health plans to educate their members about availability, location, and appropriate use of emergency and
other medical services.
Emergency department personnel to contact the patients primary care provider or health plan as quickly as
possible to discuss continuity of care.

Participation in treatment decisions

Consumers have the right and responsibility to fully participate in all decisions related to their health care.
Consumers who are unable to fully participate in treatment decisions have the right to be represented by
parents, guardians, family members, or other conservators.
Suggestions for health care organizations/health care providers to ensure this right:
Provide patients with easily understood information and opportunity to decide among treatment options
consistent with informed consent.
Provide effective communication with health care providers for patients with disabilities.
Respect the decisions made by patients and/or representatives consistent with the informed consent process.

Respect and nondiscrimination

Consumers have the right to considerate, respectful care from all members of the health care system at all
times and under all circumstances. An environment of mutual respect is essential to maintain a quality health
care system.
Suggestion for health care organizations to ensure this right:
Provide health care services to patients consistent with the benefits covered in their policy or as required by
law based on race, ethnicity, national origin, religion, sex, age, mental or physical disability, sexual orientation,
genetic information, or source of payment.

Confidentiality of health information

Consumers have the right to communicate with health care providers in confidence and to have the
confidentiality of their individually identifiable health care information protected. Consumers also have the
right to review and copy their own medical records and request amendments to their records.
Suggestion to ensure this right:
Compliance with the Privacy Rule standards.

Complaints and appeals

Consumers have the right to a fair and efficient process for resolving differences with their health plans, health
care providers, and the institutions that serve them, including a rigorous system of internal review and an
independent system of external review.
Suggestion to ensure this right:
Internal and external appeals systems and procedures should be made available to patients and resolution
should be performed in a timely manner and/or consistent as required by Medicare (eg, 72 hours).

86

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 7 Managing Patient Complaints, Patient Rights and Safety

Table 4 CONTINUED | Consumer Bill of Rights

General Principles

Consumer Rights

Consumer responsibility

Some of the consumers responsibilities include:

Maximizing health habits, such as exercising, not smoking, and eating a healthy diet
Become involved in specific health care decisions
Work collaboratively with health care providers in developing and carrying out agreed upon treatment plans
Disclose relevant information and clearly communicate wants and needs
Use the health plans internal complaint and appeal processes to address concerns that may arise
Avoid knowingly spreading disease
Become knowledgeable about his or her health plan coverage and health plan options (when available)
including all covered benefits, limitations, and exclusions, rules regarding the use of network providers,
coverage and referral rules, appropriate processes to secure additional information, and the process to
appeal coverage decisions
Show respect for other patients and health workers
Make a good-faith effort to meet financial obligations

Patient Rights and Safety

The Advisory Commission on Consumer Protection and
Quality in Health Care Industry, appointed by President
Clinton in 1997, drafted a consumer bill of rights.
Health care organizations, both public and private, have
adopted the general principles as cited in the consumer
bill of rights (Table 4).

Annotated Bibliography
Bartlett EE: Physician stress management: A new approach to reducing medical errors and liability risk.
J Health Care Risk Manag 2002;22:3-6.
This article focuses on the scope and effects of medical
stress, conceptual approaches to physician stress control, and
stress reduction programs, resources, and research. The author
concludes that stress reduction programs can result in better
patient relations, improved clinical performance, fewer medical errors, and reduced malpractice risk.

Danner C: Working with angry patients. Behavioral

Medicine Brief, Department of Family Practice and
Community Health. Minneapolis, MN, Family Medicine
and Community Health, University of Minnesota, Issue
19, 2001.
This article discusses how anger can influence the
physician-patient relationship. A three-step model for physicians for effective physical and mental management of critical
and angry patients is outlined. The need for the physician to
successfully manage anger in order to avoid a malpractice suit
and compromised patient care are also discussed.

Hickson GB, Federspiel CF, Pichert JW, Miller CS,

Gauld-Jaeger J, Bost P: Patient complaints and malpractice risk. JAMA 2002;287:2951-2957.
The topic of this article is physicians who receive a large
number of malpractice claims, resulting in high costs. The

American Academy of Orthopaedic Surgeons

premise that a physicians bedside manner and interpersonal

skills can help influence patient satisfaction is discussed.

Joint Commission on Accreditation of Health Care Organizations web site: 2004 National Patient Safety
Goals. Available at: http://www.jcaho.org/. Accessed December 18, 2003.
The JCAHOs National Patient Safety Goals are listed,
which were developed as a result of lessons learned from sentinel events reported by health care organizations.

National Coordinating Council for Medication Error

Reporting and Prevention web site: February 20, 2001.
Available at: http://www.nccmerp.org/council/council
2001-02-20.html. Accessed December 18, 2003.
Valuable information and recommendations on how to reduce medication errors related to labeling and packaging of
drugs and other related products, verbal orders and prescriptions, dispensing of drugs, and drug administration are presented.

Sage WM: Putting the patient in safety: Linking patient

complaints and malpractice risk. JAMA 2002;287:30033005.
Customer satisfaction is connected to clinical safety.
Health care organizations should document information from
patients relevant to patient safety and provide that information to health care professionals who manage and provide
health care.

US Department of Health and Human Services web

site: Office for Civil Rights. OCR Privacy Brief: Summary of the HIPAA Privacy Rule. Washington, DC,
GPO, May 2003. Available at: http://www.hhs.gov/ocr/
privacysummary.pdf. Accessed August 8, 2003.
A summary of the HIPAA Privacy Rule such as statutory
and regulatory background, who is covered by the Privacy
Rule, what information is protected, general principles for us-

87

Managing Patient Complaints, Patient Rights and Safety

age and disclosures, and enforcement and penalties for noncompliance is presented.

US Department of Health and Human Services web

site: Office of the Secretary. 45 CFR Parts 160 and 164
Standards for Privacy of Individually Identifiable
Health Information; Final Rule. Federal Register. Washington, DC, GPO, August 14, 2002. Available at: http://
www.hhs.gov/ocr/hipaa/privrulepd.pdf. Accessed August
8, 2003.
The entire document of the rules and regulations for the
Privacy Rule released by the Department of Health and Human Services is presented.

Presidents Advisory Commission on Consumer Protection and Quality in the Health Care Industry web site:
Consumer Bill of Rights and Responsibilities. Executive
Summary. Washington, DC, GPO, November 1997.
Available at: http://www.hcqualitycommission.gov/cborr/
exsumm.html. Accessed August 3, 2003.

Orthopaedic Knowledge Update 8

Wang EC: Dealing with the angry patient. Permanante J
2003;7:77-78.
Communication skills for physicians faced with an angry
patient are discussed, along with insight on what steps might
be taken for management of this situation.

Classic Bibliography
Levinson W, Roter DL, Mullooly JP, Dull VT, Frankel
RM: Physician-patient communication: The relationship
with malpractice claims among primary care physicians
and surgeons. JAMA 1997;277:553-559.
Rogers C: Communicate, in The American Academy of
Orthopaedic Surgeons, Bulletin. Rosemont, IL, 2003, vol
48, No 6.

A summary of the Consumer Bill of Rights and Responsibilities is presented.

88

American Academy of Orthopaedic Surgeons

Chapter

Selected Ethical Issues in

Orthopaedic Surgery
Angelique M. Reitsma, MD, MA
Kornelis A. Poelstra, MD, PhD

Surgical Ethics in the 21st Century

Ethical values that guide surgeons toward concrete actions in the care of patients are provided by ethicists
and by the professional surgical societies. The ethical
principles of beneficence, nonmalfeasance, respect for
autonomy, and justice were first presented in 1979 as
part of the framework known as principlism. Attention
is given in the literature to virtue as an additional moral
value to be considered. This renewed interest in virtue
as an ethical requirement is one that is in concert with
surgeons perceptions of what it means to be a moral
physician. Professional societies such as the American
College of Surgeons (ACS) and the American Academy
of Orthopaedic Surgeons (AAOS) have adopted statements referring to an honorable surgeons character.
The AAOS adopted ten Principles of Medical Ethics
and Professionalism in Orthopaedic Surgery, first in
1991 and most recently revised in 2002. The principles
are considered standards of conduct, not laws, defining
the essentials of appropriate behavior.
Perceptions of a moral physicians conduct have
shifted somewhat during the past century. For example,
paternalism was perfectly acceptable in medical and surgical practice for many years, but since the mid 1960s
has been considered a breach of patient autonomy. Major improvements in respecting a patients autonomy
were also made in the area of human subject research.
Currently, it would be unthinkable to enroll a patient or
a healthy volunteer into a clinical trial without their informed consent, a requirement that was not obvious to
early investigators. As will be discussed later, informed
consent for some types of surgical research is still not
uniform and is an issue for orthopaedic surgeons to consider carefully.
In some areas of medicine the patient is the sole
medical decision maker, with the physician as mere
counselor, offering a virtual and perhaps overwhelming
assortment of available therapies. In surgical situations,
this approach is not always feasible, because some of the
decision making occurs while the patient is unconscious
and incapable of making decisions. As procedures be-

American Academy of Orthopaedic Surgeons

come increasingly complex and consent forms get

longer, truly informed consent may seem a distant goal.
Perceptions of what constitutes physical harm have
also changed, albeit in many separate and perhaps less
obvious ways. Therapies that were once state-of-the-art
may now be considered obsolete and perhaps even
harmful. As new evidence about outcomes became
available, procedures have been reshaped or abandoned. Evidence-based surgery continues to expand. It
is no longer sufficient to offer a procedure to a patient
merely based on good intentions without some evidence
of benefit. Although modern surgery demands scientific
evidence, this requirement is not always so easy to comply with. Although drugs can be tested in a randomized
controlled trial using a placebo for the control, surgeons
must invent more intricate and sometimes more invasive placebos. Although challenging this goal is not impossible, the use of sham surgery is not without controversy. Despite strong criticism of such research designs,
it must be conceded that surgical controls are a difficult
issue and surgery itself can possibly have quite a strong
placebo effect.
Some predicted developments in the future of surgery are dazzling and will pose new moral challenges.
One example is the use of computers and robotics to
enhance surgical precision and safety. These technologies raise ethical questions about training (Who is qualified to operate such devices, and how much expertise is
required?), responsibility (Who is responsible for operating the device, who is legally responsible for breakdowns?), and perhaps even fears of too much dependence on machinery (What if it breaks down? Should
the surgeon be skilled to perform the procedure without
it? Or will the procedure be abandoned?).
Other advances will challenge conventional thinking.
As discussed in a recent article, some of these developments include human cloning, genetic engineering, tissue engineering, limb morphogenesis, intelligent robotics, nanotechnology, suspended animation, regeneration,
and species prolongation. The potential of these new
technologies to disrupt conventional surgical thinking is
huge. Of direct impact on orthopaedics in the 21st cen-

89

Orthopaedic Knowledge Update 8

Selected Ethical Issues in Orthopaedic Surgery

tury are the ethical issues concerning tissue engineering,
regeneration, and intelligent prostheses. Bioartificial
skin and blood vessel segments already can be found in
laboratories. The anticipated long-term goal is the regeneration of body parts, such as parts of limbs and
eventually, complete prostheses. Will the orthopaedic
surgeons of the future simply replace a diseased hip
joint, intervertebral disk, or leg with a new one? To approach these new technologies, an ethical framework
will be required.

Ethical Approaches to a Patient Dilemma

Every day, physicians are faced with different dilemmas
in patient care. These vary from simple issues such as ordering diagnostic tests or surgical scheduling to more
complex issues involving decisions about terminal illnesses and withdrawal of treatment. Deciding on the
morally best course can be challenging. Although it is
impossible to provide formulaic answers to every ethical
dilemma encountered in clinical practice, ethical theories can guide the orthopaedic surgeon in moral decision making. An actual case is presented to illustrate the
prominent ethical theories and to serve as an example
of how to work through the nuances of ethical thinking.

The Case
Mr. B is an 87-year-old man who has severe injuries, but
no head, severe chest, or abdominal injuries, after a
head-on motor vehicle collision. He is transferred to a
Level 1 trauma center. However, he has several severe
orthopaedic injuries, including bilateral grade IIIB open
pilon and segmental tibial shaft fractures, a right closed
tibial plateau fracture, a displaced femoral neck fracture, and a left closed supracondylar femur fracture. He
also has bilateral comminuted foot fractures and dominant right arm fractures. Prior to the accident he had
lived independently, close to his only daughter.
Mr. B undergoes emergent irrigation and dbridement and A-frame external fixator placement of both
legs. He is then transferred, while intubated and in stable condition, to the surgical intensive care unit for further management. A durable power of attorney had
been granted to his daughter. She is contacted to obtain
consent for urgent right below-knee amputation, repeat
irrigation and dbridement of the left ankle with possible amputation, and hemiarthroplasty replacement of
the right hip. Despite repeated requests, she strongly refuses to give approval for any further interventions that
could prolong her fathers life, despite the understanding of the severity of the open fractures and the risk for
sepsis or pulmonary complications. She states that she
and her father had discussed his wishes many years before, and he had expressed his desire to die at home
without any nursing home care or dependency on a ventilator for a prolonged period of time.

90

The patient is extubated 2 days after the index procedures. Severe tissue necrosis becomes apparent at the
open wounds on both legs. He is unable to tolerate elevation of the head of the bed more than 30 because of
hip pain. Although he recognized his daughter, the patient remains poorly oriented to place and time and coherent conversations with the patient remain very difficult. The Mini-Mental Status Examination is consistent
with early dementia.
Controversy about the management of this case
causes moral distress among staff and leads to heated
discussions. Therefore, a meeting with the ethics committee is requested. Two designees of the committee, a
lawyer from the Department of Risk Management and
a psychiatrist not involved in the case, are present. During this meeting, the daughter continues to strongly
refuse consent for further treatment, stating that those
were his wishes. Although sympathetic to the surgeons
plight, the attorney states that it would be legally inadvisable to continue with the surgical intervention as the
patients earlier wishes were conveyed by the power of
attorney, and it is impossible to clearly establish whether
he had any decisional capacity to counter this. A surgeon on the committee thought the attorneys advice to
be outrageous, stating that not treating this patient
amounted to torturing the patient to death.

Discussion
The approaches that dominate the ethical literature are
deontology and utilitarianism. In addition to these two
major moral philosophies, clinical ethics cannot be understood without taking into account the principles of
biomedical ethics: self-determination, beneficence, nonmaleficence, and justice. Important ethical principles are
listed and defined in Table 1.
The deontologic approach is also called duty-based
ethics, considering the duties that people have toward
one another. In this case, it is reminiscent of the familiar
notion that the surgeon has certain special duties toward the care of the patient with whom a therapeutic
relationship has been established. From a strictly deontologic point of view, the right approach would be for
the orthopaedists to perform the surgeries they believe
are indicated, as it is their duty as Mr. Bs treating physicians to give him the necessary treatments. This is what
the orthopaedic surgeons in this case suggested be done
and requested consent for, to no avail, leading to the
ethical controversy in the first place.
The utilitarianism approach is the view that actions
are to be morally evaluated according to the amount of
well being they promote. This approach is consistent
with preoccupation with the outcomes or consequences
of an intervention, and is why one treatment is recommended over another if it is more likely to provide a desired result. Although this approach has been criticized

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

for its the ends justify the means image, utilitarianism
must be taken into account in public policy as it strives
to emphasize the social consequences of an action. In
the case of Mr. B, the utilitarian approach may suggest
that the intervention is not warranted if the results cannot be expected to be favorable. Of course, exactly what
the outcome of an intervention will be is not always
clear, leading to controversy such as in the present case.
Self-determination or respect for autonomy is expressed through an informed consent process. By giving
an informed consent, the patient chooses the treatment
that they wish according to his or her own values. If a
patient is not capable of making or of communicating
decisions and if an advance directive is not available, a
surrogate must make treatment decisions on behalf of
the patient. In this case, Mr. B was considered decisionally incapacitated. His daughter was the indicated surrogate decision maker as having been appointed by the
durable power of attorney. Even if she had not been appointed by the patient, the daughter would have probably been the obvious surrogate by legal standards as the
closest next of kin. Although in the absence of written
advance directives it is not possible to know without a
doubt what a patient would have wanted, the durable
power of attorney is designed to convey previously uttered (verbal) wishes. In this case, respecting the patients autonomous decision implies that the request for
withholding further surgery should be honored.
If Mr. Bs prior wishes were unknown, his daughter
could have used the best-interest standard to make decisions on his behalf. The best interest for a patient is calculated by weighing the benefits against the burdens of
treatment. From a legal perspective, it is unlikely that
Mr. B (or his daughter) would be forced to continue
with the surgical treatment against his wishes. If the
daughter had wanted to continue aggressive limb salvage treatment despite a limited chance of success it is
likely that that decision, too, would have been respected. Increasingly, however, physicians are questioning their responsibilities to patients in situations where
they consider the intervention to have no medical benefit.
Beneficence, or the obligation to do good for the patient, is closely related to the traditional Hippocratic obligation to avoid harm, or the principle of nonmaleficence. The discussion of beneficence here will include
nonmaleficence. In this case, a beneficent act might constitute the amputation of both lower extremities to prevent pain and possibly reduce the risk for septicemia.
Justice is the point of reference for discussion of access to health care and of distribution of (scarce) resources as a matter of social policy, and will be left out
of the discussion in this case.
In the case of Mr. B, it was decided to honor the verbal advance directive given to the daughter. Plans were
developed to provide sufficient medication and pallia-

American Academy of Orthopaedic Surgeons

Chapter 8 Selected Ethical Issues in Orthopaedic Surgery

Table 1 | Principles of Biomedical Ethics

The four clusters of moral basic principles that guide ethics are:
Respect for
A norm of respecting the decision-making capacity of
autonomy
autonomous persons. This principle is expressed by
the doctrine of informed consent.
Nonmaleficence A norm of avoiding the causation of harm. This
principle requires intentionally refraining from
actions that cause harm, and is expressed by: One
ought not to inflict evil or harm.
Beneficence
A group of norms for providing benefits and balancing
benefits against risks and costs. This principle is
expressed by: One ought to prevent evil or harm;
one ought to remove evil or harm; and one ought to
do or promote good.
Justice
A group of norms for distributing benefits, risks, and
costs fairly. Applicable to both medical care and
biomedical research with human subjects.
Two relevant moral philosophical theories that further guide decision
making in biomedical ethics are:
Utilitarianism
A consequence-based moral theory holding that
actions are right or wrong according to the balance
of their good and bad consequences. The right act in
any circumstance is the one that produces the best
overall result, as determined from an impersonal
perspective.
Deontology
A duty-based moral theory holding that some features
of actions other than or in addition to consequences
make actions right or wrong.
(Adapted with permission from Selbst SM: The difficult duty of disclosing medical errors.
Contemp Pediatr 2003;20:51-53.)

tion to ensure the patients comfort during the following

days in the hospital.

The Role of the Institutional Review Board in

Modern Orthopaedics
According to the Department of Health and Human
Services, research is defined as any systematic investigation designed to develop and contribute to general
knowledge. Human subject is defined as a living individual about whom an investigator obtains either
(1) data through interaction or intervention (such as
surgery), or (2) identifiable private information. Research involving human subjects is socially important
but morally perilous because it can expose subjects to
risks for the advancement of science. Ethically justifiable research must satisfy several conditions, including
(1) a reasonable prospect that the research will generate
the knowledge that is sought, (2) the necessity of using
human subjects, and (3) a favorable balance of potential
benefits over risks to the subjects, and (4) fair selection
of subjects. Only after these conditions have been met is
it appropriate to ask potential subjects to participate.

91

Selected Ethical Issues in Orthopaedic Surgery

Assessing if a proposed study met all the criteria
above was initially left to the discretion of the
physician-investigator, but the dual roles of research scientist and clinical practitioner pull in different directions and present conflicting obligations and interest,
endangering the objectivity of the assessment. Therefore, the responsibility to assess the morality of clinical
research was given to outside reviewers. In its 1989 revision of the Declaration of Helsinki, the World Medical
Association sets as the international standard for biomedical research involving human subjects this requirement: Each experimental procedure involving human
subjects should be clearly formulated in an experimental protocol which should be transmitted for consideration, comment and guidance to a specially appointed
committee independent of the investigator and sponsor.
The purpose of the Institutional Review Board
(IRB) is to ensure that research with human volunteers
is designed to conform to the relevant ethical standards.
These standards concern protecting the rights and welfare of individual research subjects, ensuring voluntary
informed consent is obtained before participation in a
study, and an evaluation of risks and benefits. Since
1978, an additional moral standard has been added by
the National Commission for the Protection of Human
Subjects of Biomedical and Behavioral Research (The
National Commission). The Commissions Belmont
Report required IRBs to guarantee equity in the selection and recruitment of human subjects.
Across the United States, IRBs may have local
names, such as Human Investigations Committee or
Committee for Research Subject Protection, and have
a decidedly local character, although they are required
to comply with federal regulations when reviewing activities involving Food and Drug Administrationregulated investigational drugs or devices, and when
reviewing research supported by federal funds. Furthermore, all institutions that receive federal research grants
and contracts are required to file a statement of assurance of compliance with federal regulations. In these
assurances virtually all institutions voluntarily promise
to apply the principles of federal regulations to all research they conduct, regardless of the source of funding.
The IRB system has long been subject to criticism,
accused of stifling creativity and impeding progress. In
many institutions, there is a paucity of surgeons that sit
on IRBs, thus risking that decisions about surgical studies will be made with insufficient surgical knowledge.
IRBs have also been accused by bioethicists and patient
advocates of providing inadequate protection to prospective research subjects. Even the US federal government criticized the work of several IRBs through its Office of Human Research Protections (OHRP). The
OHRP was reconstituted by the Department of Health
and Human Services in 2002 and is responsible for im-

92

Orthopaedic Knowledge Update 8

proving the system of oversight and enforcing the departments regulations.
Regardless of the criticism they endure, IRBs are
currently the only formal and federal mechanism to
safeguard the interests of human research subjects. As
such, they are an important line of protection to safeguard human participants of clinical research. Most
IRBs have a home page on their institutions Website,
on which they provide information and submission
forms to download and complete.
Although the regulations for submitting research
proposals to an IRB are fairly straightforward, uncertainties remain as to what protocols should and should
not be reviewed. This is especially the case among clinicians who do not regularly perform research with human participants. One common false assumption is that
retrospective studies do not require IRB review. Retrospective studies whose results are to be published or
otherwise professionally shared do need to be reviewed
by an IRB. Another common misperception is that a
study that is likely to be exempt does not have to be reviewed. Officially, only an IRB can make the determination that a study is exempt from full review, although
many IRBs will have an expedited review process for
studies that are exempt. The important principle to remember is that all human research should be submitted
for IRB review, whether it entails a simple chart review
or an extensive intervention.
Another area of debate is the use of innovative or
experimental surgery. The introduction of surgical innovations frequently occurs under the header of therapeutic intervention, without protocols or consents. This scenario has the potential to produce an informal study
without specific study consent, which is currently being
examined by the surgical and ethical communities alike.
In the future, federal regulations or at least professional
standards will likely be developed to address this area
of surgery to ensure that new innovations are conducted
with appropriate protection of human subjects while
promoting research and technical advances within the
field. Presently, the issue of what constitutes a significant enough innovation as to require an ethical review
or specific consent is left to the discretion of the surgeon.

New Technology in Orthopaedic Surgery

Developing new technologies has been the driving force
for improvement in surgery. Innovation is necessary if
future gains in patient care are desired. Orthopaedists
are among the most innovative surgeons in the United
States; therefore, attention needs to be paid to the ethics
of innovation in orthopaedic surgery. This seems especially important given the promise of emerging new
technologies such as cell therapy or the use of growth
factors, for which there currently are no clear federal

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

regulations, nor are there any clear guidelines pertaining
to innovative surgical procedures.
On a professional level, some attempts have been
made to address the issue. The ACS has adopted selfimposed guidelines for emerging surgical technologies
and their application to the care of patients, as formulated by the Committee on Emerging Surgical Technologies and Education. The introduction of new technology
to surgeons and the public must be done ethically in accordance with the ACS Statement on Principles. These
Principles require prior and continued IRB (or equivalent) review of the protocol, full description of the procedure, and informed consent of the patient.
As major innovators, orthopaedic surgeons should
assume an active role in thinking through the issue of
new surgical technologies. By doing so, orthopaedists
can provide practical guidance for their colleagues as
well as take on a leadership role within the entire surgical community. Differing views often exist on how to
best introduce surgical innovations in an ethically responsible way. Although there is no uniform policy to
date, some truths appear valid and of practical use to
the innovative orthopaedic surgeon. First, the innovator
should be familiar with existing rules and regulations
that govern human subject research, and know if and
when an innovation is in fact a research activity. This determination includes the question of whether IRB submission and the patients informed consent is necessary.
When a proposed surgical procedure falls short of being
definite research but yet is experimental in a narrower
sense, or when expected risks are deemed significant or
risks and benefits are largely unknown because of the
very novelty of the procedure, additional precautions
appear prudent and moral. Reasonable precautions are
prior consultation (for example, with peers, surgeon-inchief, and department chair or division head) and full
disclosure of the experimental nature of the procedure
to the patient with perhaps a separate consent form. It
would also be necessary to carefully monitor the outcome of the new technology, if only to assess if further
formal evaluation in a trial would be warranted.

Other Selected Ethical Issues

in Orthopaedics
The HIV-Positive or Hepatitis B- or C-Positive Health
Care Worker
There still is considerable debate concerning the management of health care workers infected with hepatitis
B virus (HBV), hepatitis C virus (HCV), or human immunodeficiency virus (HIV). In 1991, the Centers for
Disease Control and Prevention (CDC) issued guidelines for health care workers infected with HBV or HIV.
These guidelines set restrictions that still apply today.
The CDCs guidelines were established with the
premise that the risk for transmission of HBV and HIV

American Academy of Orthopaedic Surgeons

Chapter 8 Selected Ethical Issues in Orthopaedic Surgery

to patients is greatest during certain definable
exposure-prone procedures. Exposure-prone procedures include insertion of a needle tip into a body cavity, or the simultaneous presence of a health care workers fingers and a needle or other sharp object in a
highly confined anatomic site. Health care workers who
are positive for HIV or HBV should not perform
exposure-prone procedures unless they have obtained
expert counsel regarding the circumstances under which
they may perform such procedures. Furthermore, health
care workers should inform patients of their infection
status before conducting exposure-prone procedures.
Arguably, there are some limitations to the CDC recommendations. First, they do not specify exposure-prone
procedures, leaving such determinations to the expert
counsel, allowing for a diversity of opinions and decisions impacting individual infected surgeons or health
care workers. In addition, there are no recommendations that restrict professional activities of health care
workers infected with HCV, although five cases of viral
transmission from health care workers to patient have
been documented to date. Of these cases of viral transmission, three involved surgeons; one gynecologist and
two cardiac surgeons. Perhaps the most important shortcoming is that serostatus disclosure as proposed by the
CDC does not improve patient safety, while it violates
physician privacy.
In addition to the CDC recommendations, other
professional societies such as the American Cancer Society, American Medical Association (AMA), and
AAOS have issued guidelines. Unfortunately, most
guidelines offer no, minimal, or fairly nonspecific guidance as to what constitutes exposure-prone procedures.
Also, the issue of disclosure to patients remains controversial. Most organizations do not recognize a need for
disclosure; some guidelines still favor either postexposure or preprocedure disclosure and/or identification of
the source of infection. It has been argued that the law
should not require health care workers to disclose their
HIV status (or HBV or HCV status) because this is
seen as an invasion of privacy.
Regardless of the ongoing controversy, some relevant experiential information is available to the HIV- or
HCV-infected orthopaedic surgeon. The AAOS provides practical guidance for HIV-, HBV-, and HCVinfected orthopaedic surgeons to ensure the maximum
safety of the patient. Orthopaedic surgeons infected
with HCV should always follow strict aseptic technique
and vigorously adhere to universal precautions. In addition, they should seek medical evaluation and treatment
to prevent chronic liver disease. Specific recommendations for the prevention of HCV transmission from infected health care workers may be developed as more is
learned about the virus and its associated risks. HIV carries a lower risk of transmission than either HBV or
HCV. There are only two known cases of HIV transmis-

93

Selected Ethical Issues in Orthopaedic Surgery

sion occurring from an infected health care worker to a
patient, and one instance where transmission is suspected. In 1990, a cluster of six patients was infected by
a dentist in Florida. In 1997, an orthopaedic surgeon in
France transmitted HIV to one of his patients during a
total hip joint arthroplasty. A third case where transmission is suspected concerns an instance of HIV transmission from an infected nurse to a surgical patient in
France.
Although current data indicate that the risk of transmitting a blood-borne pathogen in a health care setting
is exceedingly low, some risk is still present. The orthopaedic surgeon should therefore be familiar with the established guidelines of the CDC and the AAOS. It is important that all recommendations to prevent the
transmission of blood-borne pathogens are consistently
followed. In general, the guidelines for preventing transmission from patients to health care workers also apply
to preventing transmission from health care workers to
patients. Additionally, health care workers are encouraged to know their own HBV, HCV, and HIV infection
statuses. Voluntary and confidential testing of health
care workers for blood-borne pathogens is recommended. Preventing injuries to health care workers and
subsequent blood exposure to patients offers the greatest level of protection. Health care workers who have
preexisting conditions, such as exudative lesions or
weeping dermatitis, should refrain from direct patient
care until the condition is resolved. The affected health
care worker should also refrain from handling patientcare equipment and devices used to perform invasive
procedures. If a member of a surgical team is injured
during a procedure, the instrument responsible should
be removed from the surgical field without being reused
on the patient until appropriately resterilized. Additionally, any disposable items that come into contact with a
health care workers blood should be removed from the
surgical field and discarded into an appropriate biohazard bag or container.
Finally, it is advisable for any orthopaedic surgeon
infected with HIV or HCV to seek expert legal counsel
in addition to medical treatment. An infected surgeon
may encounter problems with malpractice insurability,
even while in remission and expected to practice. Also,
legal charges have been leveled against seropositive surgeons by former patients, even in the absence of infection, claiming psychological stress.

Care of the Uninsured Patient

The AAOS Opinion on Ethics and Professionalism:
Care and Treatment of the Medically Underserved, originally formulated in 1998 and recently revised, discusses
the dismal health care insurance situation in the United
States, noting that increasing percentages of Americans
are uninsured, underinsured, and have inadequate ac-

94

Orthopaedic Knowledge Update 8

cess to medical care. The obligations of society and the
medical profession to treat the medically underserved
are also discussed. These recommendations are based on
the 1993 AMA Council on Ethical and Judicial Affairs
guideline.
Also noted in the document, it is stated that physicians should be encouraged to devote some time to the
provision of care for individuals who have no means of
paying.
Caring for the uninsured, however, can have a considerable impact on the orthopaedic surgeons practice.
One challenge surgeons encounter when attending to
the uninsured or underinsured is how to discuss treatment plans. According to the literature, the economic
constraints on the available care are beyond the physicians control, yet they raise specific ethical issues for
the physician. The question is whether the physician is
obligated to disclose all potentially beneficial medical
treatment, even the ones that exceed ordinary standards
of care, and which probably will be unaffordable or unavailable.
Another more pressing problem is the question of
how much care an orthopaedic surgeon is required to
offer the uninsured. This has become more of an issue
since the implementation of the 1986 Emergency Medical Treatment and Active Labor Act (EMTALA). This
act mandates that emergency departments provide medical screening examinations to every person seeking
treatment regardless of their ability to pay or whether it
is the appropriate point of service. As a result of this
law, many patients have also been seeking and receiving
specialty care for nonemergent problems, although this
was arguably not the original intent of this law. This
consequence of EMTALA has left hospitals and physicians facing a crisis of overcrowded emergency departments and uncompensated care, which in turn threatens
patient access to quality care.
Other current economic and political changes also
have major impact on hospitals and health care providers, such as declines in Medicare payments for graduate
medical education, lower disproportionate share payments, growth of Medicaid managed care, increased
costs resulting from new government regulations such as
the Healthcare Insurance Portability and Accountability
Act (HIPAA) and Center for Medicare and Medicaid
Services rules, increased costs of malpractice insurance,
increased costs resulting from limited house staff duty
hours and decreased payments for services provided
from Medicare and other third party payers. These factors combined place a strain on orthopaedic surgeons,
especially those in public and teaching hospitals that
shoulder a disproportionate share of the care for the uninsured.
It appears appropriate to adopt a prudent stance toward providing services at no cost. Caring for the uninsured is an intrinsic part of orthopaedic surgery in a na-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 8 Selected Ethical Issues in Orthopaedic Surgery

Table 2 | Why Physicians May Conceal a Medical Error

Table 3 | How To Approach Patients After a Medical Error

The medical profession values perfection

Feelings of shame or guilt
The admission may damage the physicians professional reputation
Possible drop in referrals or an impact on income
Desire to maintain the trust of the patients family
Pressure felt from various sources:
Managed care organizations
Hospital administration
Malpractice insurers
Fear of punishment or, in the case of trainees, dismissal
Fear of a malpractice lawsuit

Notify professional insurer and seek assistance from those who might
help with disclosure (for example, attending physician or risk manager)
Disclose promptly what is known about the event; concentrate on what
happened and the possible consequences
Take the lead in disclosure; do not wait for the patient to ask
Outline a plan of care to rectify the harm and prevent recurrence
Offer to get prompt second opinions where appropriate
Offer the option of a family meeting and the option of having other
representatives (for example, lawyers) present
Document important discussions
Offer the option of follow-up meetings and keep appointments
Be prepared for strong emotions
Accept responsibility for outcomes, but avoid attributions of blame
Apologies and expressions of sorrow are appropriate

(Adapted with permission from Selbst SM: The difficult duty of disclosing medical errors.
Contemp Pediatr 2003;20:51-53.)

tion where a great number of people cannot afford even

basic health care.

Disclosure in Medical Mistakes

Each year thousands of injuries and deaths in US hospitals result from medical errors. Errors involving medications have been reported in 4% to 17% of all hospital
admissions. Although most of the information available
about medical errors pertains to hospitalized patients,
errors can occur anywhere in the health care system
(operating room, office, clinic, emergency department,
or elsewhere).
The AMA Code of Ethics provides important guidelines for professional practice including medical mistakes. The Joint Commission on the Accreditation of
Healthcare Organizations requires health care workers
to inform patients when they have been harmed by
medical mistakes. However, this may not always occur
as illustrated by one survey of house officers where they
reported that they told an attending physician about serious mistakes only about half the time; errors were
conveyed to patient or family in only 24% of the cases.
In one study, simulated case scenarios involving
wrong medications were presented to 150 medical students, house officers, and attending physicians. The researchers found that as severity of injury increased, willingness to admit an error declined. About 95% of the
students and physicians said they would admit an error
to a patient when the outcome was minimal. However,
only 79% said they would admit an error that resulted
in the death of a patient. Another 17% said they would
admit the error if they were asked directly about the
event.
Although concealing a medical error violates ethical
codes, fear of professional censure or medical malpractice
lawsuits can pressure health care workers to be less than
forthcoming when a mistake has occurred (Table 2).

American Academy of Orthopaedic Surgeons

(Adapted with permission from Selbst SM: The difficult duty of disclosing medical errors.
Contemp Pediatr 2003;20:51-53.)

In another study, three case scenarios were presented that varied in the degree of outcome severity to
400 patients. The respondents generally indicated that
they would be more likely to file a lawsuit if the doctor
withheld information about a mistake that subsequently
surfaced. About 40% said they would stay with the physician after open disclosure of a mistake was made; however, only 8% said they would continue to see a doctor
who did not disclose a mistake. Only 12% said they
would sue if the physician informed them of a mistake
that did not result in permanent aftereffects. However,
20% said they would sue if they found out about a mistake that the physician tried to cover up.
These results underscore the fact that patients generally appreciate an open, honest relationship with their
physicians. In fact, a good doctor-patient relationship is
one of the greatest factors that reduces the risk of a lawsuit if a poor outcome occurs. Prompt disclosure following a medical error will make the physician appear honest in the event of litigation and trial (Table 3). In
contrast, nondisclosure can have significant negative legal implications as the legal statute of limitations may
be extended if a physician is found to have knowingly
and intentionally concealed information from a patient.

Evaluation of the Risk/Benefits Ratio in Patients With

Difficult-to-Treat Problems
Acute, high-risk surgery involves operations on patients
who often face a significant risk of morbidity and mortality without surgery, but for whom surgery itself involves a significant risk of morbidity and mortality as
well. If nonsurgical management indeed involves a
higher risk than surgery (for example, acute compartment syndrome, unstable open book pelvis injury), the

95

Orthopaedic Knowledge Update 8

Selected Ethical Issues in Orthopaedic Surgery

choice seems simple. However, most situations are not
as straightforward. Informing the patient about surgical
and nonsurgical statistics becomes important so that the
pros and cons of surgery can be carefully weighed. In
general, the reasonable person standard requires physicians to tell patients about a therapys likely complications, especially if surgery is more likely to result in injury, disease, or a patients death than an alternative
treatment. The best interest for patients is then calculated by weighing the benefits against the burdens of
treatment. If the treatment burdens outweigh the benefit, then the treatment is considered not to be in the patients best interests. Benefits of treatment might include
saving or prolonging life, alleviation of pain or restoration of function to an acceptable level or quality of life.

Relationships With Industry

Most orthopaedic surgeons foster active relationships
with industry partners. Representatives of companies
that produce prostheses, instrumentation, and other devices are frequent visitors to orthopaedic departments.
Their companies regularly serve as sponsors of departmental functions and books for residents, and often
fund clinical research when done to study their products. These types of financial relationships occur in every branch of medicine and are no less common in orthopaedic surgery. For orthopaedic surgeons, it is
therefore important to be aware of the potential ethical
problems these relationships can bring about, as such
relationships may cause conflicting interests. These conflicts of interest may seriously compromise the integrity
of clinical medicine, and may hamper the protection and
safety of human research participants. The troublesome
effects of conflicts of interest emerging from the research setting can be far-reaching, as published results
often establish a standard that is followed by physicians
treating patients worldwide.
A conflict of interest is defined here as any financial
arrangement that compromises, has the capacity to compromise, or has the appearance of compromising trust in
clinical care and/or clinical research. There are many
different types of financial relationships that may surface between industry and orthopaedic surgeons and
their institutions. For example, academic orthopaedic
surgeons may be provided grants to study a sponsors
drug or product. Such grants may be a major source of
salary support for investigators and their personnel. It is
also often the case that device manufacturers offer investigators financial incentives for entering patients into
a study. Although these enrollment fees are intended
to cover the cost of the subjects participation, they typically exceed these costs. Academic orthopaedic surgeons
may use the excess funds to support personnel, travel to
meetings, or laboratory supplies and equipment, while in
private practice these bonuses often remain with the re-

96

searcher. Orthopaedic surgeons sometimes serve as consultants to companies whose products they are studying
or join a companys advisory board and speakers bureau, as well as enter into patent and royalty agreements, which may further complicate financial relationships. Arguably, such complex and considerable financial
relationships may create some (perceived) dependency
and/or loss of impartiality for the orthopaedic surgeon.
The appropriate ethical approach to take when
faced with financial conflict of interest has been intensely debated, with editorialized pronouncements appearing in the nations leading medical journals; societies and professional associations have addressed the
issue of financial conflicts of interest in research.
For the moral orthopaedic surgeon, it is appropriate
to be familiar with such professional statements, and to
apply them to everyday practice and research as much
as possible. Although financial conflicts of interest can
probably never be fully eradicated, they can be diminished and brought to a morally more desirable level.
One way this can be achieved is by meticulously disclosing all financial interests to patients and research subjects. Although disclosure is not a curative measure as
advertising financial ties will not break them, it will
lessen the potential for harm. In addition, orthopaedic
surgeons should remain vigilant and cognizant of their
ongoing relationships with industry, and regularly critically evaluate the extent of and the impact such ties
have upon their practice and research.

Annotated Bibliography
General Reference
American Academy of Orthopaedic Surgeons (ed):
Guide to the Ethical Practice of Orthopaedic Surgery, ed
4. Rosemont, IL, American Academy of Orthopaedic
Surgeons, 2003.
The first edition of this booklet was published in 1991. The
Guide provides standards of conduct and the essentials of ethical behavior for orthopaedic surgeons. This book can be considered required reading for the morally conscious orthopaedic surgeon and resident.

American Medical Association Council on Ethical and

Judicial Affairs (ed): Code of Medical Ethics: Current
Opinions With Annotations. Chicago, IL, American
Medical Association, 2000-2001.
This text is presently the most comprehensive and current
guide available to physicians in the United States. It is regularly revised and updated, and includes the seven basic principles of medical ethics and more than 180 opinions of the
AMAs EJA-Council on a wide spectrum of topics.

Beauchamp TL, Childress JF: Principles of Biomedical

Ethics, ed 5. New York, NY, Oxford University Press,
2001, pp 319-320.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

This book is one of the most influential and important basic texts on bioethics, offering the theory of the four guiding
principles: respect for autonomy, beneficence, nonmaleficence,
and justice.

Surgical Ethics in the 21st Century

American Academy of Orthopaedic Surgeons Principles
of Medical Ethics and Professionalism in Orthopaedic
Surgery, 2002. Available at: http//www.aaos.org/
wordhtml/papers/ethics/prin.htm. Accessed August 25,
2004.
These ten principles are standards of conduct that define
the essential aspects of honorable behavior.

Moseley JB, OMalley K, Petersen NJ, et al: A controlled trial of arthroscopic surgery for osteoarthritis of
the knee. N Engl J Med 2002;347:81-88.
In this groundbreaking article the authors use sham surgery as a placebo in the randomized trial, concluding that arthroscopic knee surgery for this particular indication is no better than a placebo.

Satava RM: Disruptive visions. Surg Endosc 2003;17:

104-107.
This visionary article paints a picture of the future of surgery, with technologies that seem like science fiction now but
may very well be available within decades.

The Role of the Institutional Review Board in Modern

Orthopaedics
United States Department of Health and Human Services Website. Protecting Personal Health Information
in Research: Understanding the HIPAA Privacy Rule.
Available at: http://privacyruleandresearch.nih.gov/
pr_02.asp. Accessed February 2004.
The Department of Health and Human Services issued the
Standards for Privacy of Individually Identifiable Health Information (the Privacy Rule) under the HIPAA of 1996 to
provide the first comprehensive federal protection for the privacy of personal health information. Many of those who must
comply with the Privacy Rule complied by April 14, 2003.

Department of Health and Human Services National

Institutes of Health Office for Protection from Research
Risks Website. Available at: http://ohrp.osophs.dhhs.gov/
humansubjects/guidance/45cfr46.htm. Accessed February, 2004.
The Office of Human Research Protections (formerly the
Office for Protection from Research Risks, OPRR), provides
online decision charts, assisting prospective investigators in determining the need for IRB review and/or the subjects informed consent for research.

New Technology in Orthopaedic Surgery

American College of Surgeons Website. Statements on
Principles. Revised March 2004. Available at: http://

American Academy of Orthopaedic Surgeons

Chapter 8 Selected Ethical Issues in Orthopaedic Surgery

www.facs.org/fellows_info/statements/stonprin.html#top.
Accessed August 25, 2004.
This website discusses codes of professional conduct and
relationships between patients and surgeons.

Cronin DC II, Millis JM, Siegler M: Transplantation of

liver grafts from living donors into adults: Too much, too
soon. N Engl J Med 2001;344:1633-1637.
In this article, the rapid implementation of an experimental procedure into transplantation surgery is criticized.

Margo C: When is surgery research? Towards an operational definition of human research. J Med Ethics 2001;
27:40-43.
This article analyzes the vague definition of clinical research in surgery and criticizes the wide implementation of socalled informal research.

Reitsma AM, Moreno JD: Ethical regulations for innovative surgery: The last frontier? J Am Coll Surg 2002;
194:792-801.
This article discusses the regulatory gap between the protection of human subjects involved in research and those undergoing experimental surgery. Results of a survey among surgeons are presented.

Other Selected Ethical Issues in Orthopaedics

ABIM Foundation: American Board of Internal Medicine, ACP-ASIM Foundation, American College of Physicians, American Society of Internal Medicine, European Federation of Internal Medicine: Medical
professionalism in the new millennium: A physician
charter. Ann Intern Med 2002;136:243-246.
This article presents a discussion on professionalism, principles, and responsibilities for physicians.

American Academy of Orthopaedic Surgeons Website.

American Academy of Orthopaedic Surgeons Advisory
Statement: Preventing the Transmission of Bloodborne
Pathogens. Available at: http://www.aaos.org/wordhtml/
papers/advistmt/1018.htm. Accessed February, 2001.
Practical guidance for all clinically active orthopaedic surgeons infected with HIV, HBV, or HCV.

American Academy of Orthopaedic Surgeons Opinions

on Ethics and Professionalism: Care and Treatment of
the Medically Underserved. May 1998, revised May
2002. Available at: http://www.aaos.org/wordhtml/papers/
ethics/1210eth.htm. Accessed August 25, 2004.
The AAOS applies its principles of medical ethics and professionalism in orthopaedic surgery to this pressing issue.

American Academy of Orthopaedic Surgeons Position

Statement. Emergency Department On-Call Coverage.
September 2002. Available at: http://www.aaos.org/

97

Selected Ethical Issues in Orthopaedic Surgery

wordhtml/papers/position/1157.htm. Accessed August
25, 2004.
A detailed statement explaining the various effects of
EMTALA on orthopaedic surgeons and the Academys stance
towards it is presented.

Blendon RJ, DesRoches CM, Brodie M, et al: Patient

safety: Views of practicing physicians and the public on
medical errors. N Engl J Med 2002;347:1933-1940.
A scientific study of the subjective impact of medical errors as viewed by physicians and the public.

Meyer FN: Uninsured healthcare is a growing problem:

Putting economic pressures on physicians. AAOS Bulletin 2003; August 51(4).
This article examines the growing concerns about issues
related to uninsured health care in the United States.

Selbst SM: The difficult duty of disclosing medical errors. Contemp Pediatr 2003;20:51-53.
This article discusses how to manage medical mistakes and
examines the reasons why disclosure is so difficult.

Tereskerz PM: Research accountability and financial

conflicts of interest in industry sponsored clinical research: A review. Account Res 2003;10:137-158.
A discussion of financial conflicts of interest, research accountability, and other aspects of industry relationships are
discussed.

United Kingdom Department of Health Website. HIV

infected health care workers: A consultation paper on
management and patient notification. Available at:
www.doh.gov.uk/aids.htm. A list of exposure-prone procedures available at: www.doh.gov.uk/pub/docs/aids.pdf.
Accessed February, 2004.
Although UK policy obviously does not apply to US surgeons, it is worthwhile to review the detailed list of exposureprone procedures as a reference.

Wears RL, Wu AW: Dealing with failure: The aftermath

of errors and adverse events. Ann Emerg Med 2002;39:
344-346.
This article discusses what happens after a medical mistake has occurred and how patients and health care workers
can be guided through this situation.

98

Orthopaedic Knowledge Update 8

Classic Bibliography
Aronheim JC, Moreno JD, Zuckerman C (eds): Ethics
in Clinical Practice, ed 2. Gaithersburg, MD, Aspen Publishers, 2000, pp 17-50.
McCullough LB, Jones JW, Brody BA (eds): Surgical
Ethics. New York, NY, Oxford University Press, 1998.
Recommendations for preventing transmission of human immunodeficiency virus and hepatitis B virus to
patients during exposure-prone invasive procedures.
MMWR Recomm Rep 1991;40(RR-8):1-9.
Recommendations for prevention and control of hepatitis C virus (HCV) infection and HCV-related chronic
disease: Centers for Disease Control and Prevention.
MMWR Recomm Rep 1998;47(RR19):1-39.
Statements on emerging surgical technologies and the
evaluation of credentials: American College of Surgeons. Bull Am Coll Surg 1994;79:40-41.
Statement on Issues to be Considered Before New Surgical Technology is Applied to the Care of Patients:
Committee on Emerging Surgical Technology and Education, American College of Surgeons. Bull Am Coll
Surg 1995;80:46-47.
Sweet MP, Bernat JL: A study of the ethical duty of physicians to disclose errors. J Clin Ethics 1997;8:341-348.
US National Commission for the Protection of Human
Subjects of Biomedical and Behavioral Research: The
Belmont Report: Ethical Principles and Guidelines for
the Protection of Human Subjects of Research. September 30, 1978. Superintendent of Documents, US Government Printing Office. Washington, DC, DHEW Publication No. 78-0013.
World Medical Association Recommendations Guiding
Physicians in Biomedical Research Involving Human
Subjects (document 17.1). Helsinki, Finland, June 1964.
Wu AW, Folkman S, McPhee SJ, Lo B: Do house officers
learn from their mistakes? JAMA 1991;265:2089-2094.

American Academy of Orthopaedic Surgeons

Chapter

Outcomes Assessment and EvidenceBased Practice Guidelines in

Orthopaedic Surgery
William A. Abdu, MD, MS

Outcomes Instruments in Orthopaedic

Surgery
Traditional approaches to clinical research and reports
of surgical interventions are commonly associated with
the measurement of variables that are either easily
available retrospectively from the medical record, or
variables of primary interest to the researcher. These biologic, physiologic, and anatomic measures include mortality, strength measurement, range of motion, and radiographic findings. These examples of hard outcome
measures or surrogate outcomes have a very weak association with the symptoms and functionality of the end
result of therapeutic interventions and are only of
modest relevance to patients and society. The perceived
attractiveness of hard measures is that they are believed
to be objective (unbiased judgment), have preservability
(radiographs) and have dimension (measurement of radiographic findings). However, if the goal of treatment
is to reduce pain and improve function, these constructs
should be measured directly rather than assuming that
surrogate hard measures will suffice.
More recently, the goals of clinical research have
transitioned to the measurement of variables primarily
critical to the patient, or soft outcome measures.
These patient-reported outcomes of symptoms, physical
function, and health may take several forms and include
the assessment by disease-specific measures, general
health measures, and satisfaction measures. These socalled soft outcomes measures are more reliable and
more consistent than the traditional, hard measures.
Outcomes research often refers to the study of a group
or cohort of patients often with the same diagnosis, and
relates their clinical and health outcomes to the care
they received.
Outcomes research also includes methods of analysis
for small (single site) and/or large (often multiple site)
databases, small-area analysis, structured literature reviews (meta-analysis), prospective clinical trials, decision
analysis, and guideline development. The focus of outcomes research is often on patient-centered outcomes of
care (patient self-report) rather than on the process

American Academy of Orthopaedic Surgeons

measures of care, which are not always under the patients or the doctors control.
An outcomes instrument is the survey tool or instrument used to measure these variables (Table 1). These
measures are not designed or intended to substitute or
replace the traditional measures or clinical end points,
but are to be used in parallel with clinical measures.
An outcomes instrument, in order to be useful,
should have clinical sensibility, meaning that the questionnaire includes relevant content and is appropriate
for both the patient population and the setting in which
it is to be used. The feasibility of the questionnaire is determined in part by its length, degree of respondent burden, ease of scoring and analyzing the results, and the
costs of its use. In constructing a questionnaire, it must
be reliable; the results must be reproducible from one
time to another or between interviewers. The questionnaire must also have internal consistency. The instrument must be validated, meaning that there are correlations in the expected direction and magnitude with a
variety of external measures that are somewhat different but are expected to have predictable associations. In
addition, the instrument must be responsive, or have the
ability to measure and detect small but clinically important differences between groups, or over time.
Outcomes instruments may be used for multiple purposes as outlined in Table 2. The increasing interest in
the use of outcomes measures is relatively new and is a
rapidly evolving methodology. Many issues remain controversial, raising several questions, such as: Which measurements are important? When and specifically how
should they be measured? Are the outcomes instruments valid? Are there controls? What other factors (biologic, physiologic, environmental) may influence the
measurement results? Are generic measures sufficient
or are disease-specific measures also important? Over
half a century ago, Lembcke noted that the best measures of quality is not how well or how frequently a
medical service is given, but how closely the result approaches the fundamental objectives of prolonging life,
relieving stress, restoring function and preventing disability. The goal of outcomes research and the use of

99

Outcomes Assessment and Evidence-Based Practice Guidelines

Table 1 | Common Outcomes Survey Instruments

General health outcomes instruments
Short Form-36 or Short Form-12 general health survey
Health Related Quality of Life (HRQOL)
Sickness impact Profile (SIP Nottingham Health Profile (NHP)
Million Visual Analog Scale (VAS)
Duke Health Profile
Patient satisfaction measures
Patient Satisfaction Questionnaire (PSQ)
Client Satisfaction Questionnaire (CSQ)
Patient Satisfaction Survey (PSS)
Disease or condition specific instruments include
Spine
Oswestry Disability Index (ODI) for low back pain
Roland-Morris Disability Questionnaire (RMDQ)
Waddell Disability Index
Low Back Pain Outcome Score (LBOS)
Clinical Back Pain Questionnaire (CBPQ)
Quebec Back Pain Disability Scale (QBPDS)
Low Back Pain Rating Scale (LBPRS)
North American Spine Society Lumbar Spine Questionnaire (NASS
LSQ)
Resumption of Activities of Daily Living Scale (RADL)
Upper extremity
Constant Shoulder Function Scoring System
Carpal Tunnel Syndrome Evaluation
Upper Extremity Disabilities of the Arm, Shoulder, and Head (DASH)
Shoulder Pain Score
Elbow questionnaire
Lower extremity
Western Ontario McMaster Universities Osteoarthritis Index (WOMAC)
Harris Hip Score
Trauma
Short Musculoskeletal Function Assessment Questionnaire (SMAF)

Table 2 | Purposes for Outcomes Instruments

Study of populations in cross-sectional studies (at one point in time) to
define the characteristics of a particular patient population with a
specific condition, resource utilization, and baseline characteristics.
Determination of longitudinal impact on health change, function, and
satisfaction in an individual patients care for a specific disease process after intervention (such as elective surgery).
As measures for prospective clinical trials to determine the effectiveness
of a particular intervention (for example, surgical versus nonsurgical
treatment of a specific condition).
Development of disease-specific evidence-based medicine clinical pathways and guidelines for unifying the process of delivering effective
medical care.

outcomes survey instruments is to help define the results of interventions and assess these desired end results.
The construct, design, and use of outcomes instruments are rapidly evolving. Key factors about outcomes

100

Orthopaedic Knowledge Update 8

Table 3 | Key Factors in the Construct, Design, and Use of

Outcomes Instruments
Content, Population, Setting, and Purpose
Which outcomes are to be measured (biologic, physiologic, function,
general health, quality of life, satisfaction)?
Which population will be studied and in what setting (age, gender)?
What is the purpose of the study (to describe, predict, measure change,
measure one point in time or change over time, or impact of an intervention)?
Content Validity
What domains (health concepts) and items are included?
Are there important omissions or inappropriate inclusions?
Face Validity
Do these measures make good clinical sense?
Is each question phrased in a suitable way?
Are the response categories appropriate?
Is there an overall score summarizing across questions, and how is
this score calculated?
Is there correlation with other outcomes measures?
Are these measures predictive of future events?
Feasibility
Is the instrument easy to use and understand?
Is the instrument acceptable to the population and the clinician/
investigator?
What is its format (self-administered, telephone, personal interview,
computer, paper/pen)?
How long does it take to administer and report the results?
Responsiveness
Is the instrument able to detect subtle but clinically relevant change
over time or with intervention?

instruments can be divided into categories and are outlined in Table 3.

The major areas of interest in measuring patientcentered self-response measurements using outcomes
instruments include measure of function (the ability to
perform specific tasks, covering the domains of physical,
social, role, and psychological function), general health
perception (integrates the various aspects of health as
reflected by the patients subjective global rating), quality of life (general measure of a patients overall wellbeing), and satisfaction (measure of treatment impact,
process of care, results, and quality of life).
Although these factors are all distinct concepts, they
may or may not tract together throughout the patients
course or treatment. For example, resolution of pain
may not correlate with radiographic findings, ability to
function, return to work, or satisfaction. Thus, outcomes
should be measured in multiple dimensions. Compression of various dimensions of outcome into a single unidimensional scale obscures the detection of these possible variations in responses.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 9 Outcomes Assessment and Evidence-Based Practice Guidelines

Outcomes measures are generally disease-specific or

general health (generic) surveys. Disease-specific instruments such as the Oswestry Disability Questionnaire for
low back pain are designed to measure disease-specific
functional status, and have the advantage of capturing
disease-specific dysfunction in greater detail. General
health or generic surveys such as the Short Form-36 are
important for detecting complications or ill effects of
treatment that extend beyond a specific disease or condition. They also make it possible to compare the impact
of treatments of specific disorders with the impact of
treatment of other diseases. These surveys help determine the cost effectiveness of various treatments. A core
set of survey instruments should include measures of
symptoms, functional status, overall well-being, and
work disability.
Several important factors must be considered for
outcomes measures to be a valid indicator of health care
quality. First, the process of medical care must actually
affect the outcome. Second, the measure of outcome
must be valid, reliable, and responsive to changes in a
patients health status. Third, sufficient information must
be collected about comorbid conditions and patient demographics when studying the outcomes of a particular
condition or treatment. Fourth, patient compliance and
the timing of a survey are critical to obtaining representative measures. Finally, if the variable of interest is a
rare event (such as death), then the population under
study must be large enough to make valid comparisons
between treatments or conditions.

Design of a Prospective Clinical Trial

The major source of information for clinicians is the
published literature, and almost all knowledge in orthopaedics is based on information that has appeared in
texts and journals. However, factors affecting the validity of many currently reported clinical studies include
the lack of randomization of patients, inadequate study
design, and missing sample size calculations, therefore
rendering many studies unable to adequately answer the
research question. Other factors include the lack of
standardized study definitions and measures, poor descriptions of study patients and confounding variables,
inadequate and unclear follow-up, and absence of
patient-centered outcomes measures. Given the absence
of a firm knowledge and research on which to base clinical decisions, it is no wonder that significant practicepattern variation exists.
In order to be most efficient at designing and collecting the appropriate data elements, the clinician must
give careful thought to the exact questions to be addressed and the study design to answer the research
questions. The formulation of an appropriate research
question informs the clinician and researcher about the
most appropriate data elements to be collected, the pri-

American Academy of Orthopaedic Surgeons

Figure 1 Basic design of a randomized clinical trial. IRB = Institutional Review Board;
DSMB = Data and Safety Monitoring Board.
mary outcome of interest. In a prospective clinical trial,
the researcher poses a question, intervenes, and follows
the direction of inquiry forward. The events of interest
occur after the onset of the study (Figure 1).
In clinical trials, the ability to determine the better
of two treatments is the product of the trial hypothesis,
the data elements chosen to evaluate the treatments in
question, the magnitude of change in the scores over
time needed to consider one treatment preference, and
sample size for power of study. All research studies are
subject to invalid conclusions because of bias, confounders, and chance. Bias is the nonrandom systematic error
in study design. It is an unintentional outcome of factors
such as patient selection, performance, and outcome determination. A confounder is a variable having independent associations with both the exposure and the outcome, and thus potentially distorts their relationship.
Common confounders include age, gender, socioeconomic status, and comorbidities. Chance can lead to invalid conclusions based on the probability of type I error (concluding there is a difference when none exists
equal to the p value) and type II error (concluding that
there is no difference when one truly exists). Thus, the
appropriate sample size and power calculation must be

101

Orthopaedic Knowledge Update 8

Outcomes Assessment and Evidence-Based Practice Guidelines

Table 4 | Common Formulas for the Determination of Sample Size

Study Design and Type of Error

Sample Size Formula

Studies using paired t test (before and after studies) with alpha (type I)
error only

N=

(za) (s)
= total number of subjects
(d)2

Studies using t test (randomized controlled trials with one experimental

and one control group, considering alpha error only)

N=

2
2
(za) 2 (s)
= number of subjects/ group
2
(d)

N=

(za+ zb) 2 (s)

= number of subjects/ group
2
(d)

N=

(za+ zb) 2 p(1-p)

= number of subjects/ group
2
(d)

Studies using t test considering alpha and beta errors

Tests of difference in proportions considering alpha and beta errors

N = sample size, za = value for alpha error (equals 1.96 for P = 0.05 in two-tailed test), zb = value for beta error (equals 0.84 for 20% beta error = 80% power in one-tailed test), (s)2 =
variance, p = mean proportion of success, d = smallest clinically important difference to be detected

determined before the initiation of the study to avoid

chance, given the smallest clinically relevant differences
in outcome and the variability in measures. It is necessary to have sufficient statistical power (the ability to
detect a difference of the magnitude required if one
truly exists) to determine whether the efficacy of one
treatment is superior to another.

Sample Size Calculations

One of the more difficult issues determining the success
of a study is sample size calculation, which must be
done prior to the initiation of the clinical study. Sample
size calculations are always an approximation, yet they
are important in providing information about two study
design questions. The first is how many subjects should
participate in the intended study and second, is this
study worth doing if only N number of subjects participate? In some cases the study may not be worth doing if
the likelihood of demonstrating a significant clinical effect or difference between two treatments is very small
or if the population needed to answer the question is
extremely large. Variables critical to determining sample
size include data type (paired data [observations before
and after treatment in the same study group] and unpaired data [observations in an observation and control
group], consideration of error type (beta errors [type II
or false-negative errors] and alpha errors [type I or
false-positive errors]), variance in data set (large and
small), alpha level (usually set at P = 0.05 in the twotailed test and confidence interval of 95%), beta level
(usually set at 0.2), and determining the size of the detectable difference between groups.

Beta Error Importance

Statistical power refers to the sensitivity for detecting a
mean difference in a clinical study, whereas sensitivity
refers to the same difference in a clinical test (statistical
power + beta error = 1.0). This means that the statistical

102

power is equal to 1-beta error. Therefore, in calculating

a sample size, if the investigators accept a 20% possibility of missing a true finding (beta error = 0.2), then the
study should have a statistical power of 0.8 or 80%. That
means that the investigators are 80% confident that
they will be able to detect a true difference of the size
that they specify with the sample size that they determine. By design, the investigator accepts a 20% chance
of false-negative results. The best way to incorporate
beta error into the study is to include it beforehand in
the determination of sample size, which is simple to do,
but it increases the sample size considerably. A large
sample size can be viewed as a means of decreasing the
false-negative results rate when a real treatment effect
exists.
For the comparison of groups in which the data set
variance is high, when the investigators want the answer
to be close to the true value, and when the smallest
clinically significant difference detected is to be extremely small, a large sample size will be required. In
calculating the sample size, investigators must first determine the appropriate formula to be used, based on
the type of study and the type of error to be considered.
Four common formulas for the determination of sample
size are listed in Table 4. As the complexity of the
groups and error types increase, the N correspondingly
increases.
The adverse affects of bias, confounding, and chance
can be minimized by proper study design and statistical
analysis. The purpose of prospective studies is to minimize bias from selection, information, recall, and nonresponder factors. Randomization minimizes selection
bias and equally distributes the confounders. Random
assignment implies that each individual has the same
chance of receiving each of the possible treatments and
that the probability that a given subject will receive a
particular allocation is independent of the probability
that any other subject will receive the same treatment

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 9 Outcomes Assessment and Evidence-Based Practice Guidelines

assignment. The decision regarding study design and

data collection may be the most critical part of planning
a clinical trial. The study design is the map from which
data collection follows, and which enables the physician
to accurately produce data forms and collect data. The
prospective randomized clinical trial with concurrent
controls is the gold standard in clinical research. Traditional randomized trials tell about the efficacy (whether
the treatment can work under ideal circumstances),
whereas outcomes research more typically studies the
effectiveness (how well a treatment works when applied
in routine care). Effectiveness is a function of efficacy,
but also of diagnostic accuracy, physician skill in applying treatment, and patient compliance with prescribed
regimens. Any of these factors may create a gap between efficacy and effectiveness. The prospective randomized clinical trial provides not only qualitative conclusions about whether a treatment is better, but also
quantitative estimates of the extent to which it is better.

Barriers to a Randomized Clinical Trial

There are many barriers that must be overcome to
proceed with a prospective randomized clinical trial involving a surgical procedure (Table 5). History has not
favored the validation of surgical procedures by randomized trials. Many commonly performed surgical procedures were introduced well before randomized trials
became established in medicine. Once a procedure becomes accepted, it is difficult to test it against a control.
For many common procedures performed, orthopaedic
surgeons cannot fall back on history to explain the lack
of rigor in surgical research.
Surgeons can be tempted to ignore evidence that
threatens their personal interests because of competition and prestige. Objectivity about procedures central
to a surgeons reputation is difficult and randomized
clinical trials may seem threatening. The literature must
support and the investigators must feel comfortable
with a state of equipoise, which is a point of maximum
uncertainty, a state of balance or equilibrium between
two alternative therapies such that there is no preference between treatments. Some also refer to this as the
uncertainty principle in which clinicians, researchers,
and patients acknowledge having hunches about a treatments effectiveness, but that the boundaries (or confidence intervals) around their hunches may run from
one extreme (extremely effective) to the other (harmful). If the answer to the clinical research question is
clear, the randomized clinical trial cannot ethically proceed with the comparison of a known successful treatment with a known ineffective treatment.
The infrastructure within which to perform the study
must be well planned and developed to function efficiently. This includes knowledge to obtain funding and
to perform statistical analysis. A lack of education in

American Academy of Orthopaedic Surgeons

Table 5 | Barriers to Overcome Before Proceeding With

Prospective Randomized Clinical Trials Involving a
Surgical Procedure
History
Competition
Equipoise/uncertainty
Infrastructure
Study design and statistics
Surgical trials
Events
Risk assessment
Commitment

clinical epidemiology can affect study design and statistics. The study design and implementation depends
heavily on and will be most successful when a core multidisciplinary group of researchers (including clinicians,
epidemiologists, and statisticians) all contribute their expertise to the project.
The inherent variability of surgery requires precise
definitions of the diagnosis, interventions, and close
monitoring of its quality. Surgical learning curves might
cause difficulty in timing and performing randomized
clinical trials. Blinding is often difficult in surgical trials,
and it is not always possible or necessary. Events are
conditions that may be emergent or life-threatening that
may cause difficulties with recruitment, consent, and
randomization of clinical trials. Multicenter studies are
often required under these conditions. In addition, comparison of surgical and nonsurgical treatments with
greatly different risks causes difficulties with patient
equipoise, and thus recruitment. Finally, all involved
must be absolutely committed to all aspects of the trial
if it is to succeed.
These barriers to randomized clinical trials have
stimulated some to question the need for randomized
clinical trials in surgery, and the debate is substantial.
For many medical questions, a large amount of evidence
has been accumulated through nonrandomized studies.
The risk of nonrandomized studies is that the studies
may spuriously overestimate treatment benefits as they
are more susceptible to unaccounted confounding, yielding misleading conclusions. Observational studies may
provide sufficient evidence of a procedures effectiveness, but the treatment effect of the procedure must be
quite large, and the study well designed to be convincing. However, it is very difficult to use historical controls, obtained under less rigorous scientific standards of
data collection, against which to test a new procedure.
Many previously well established surgical procedures,
once thought to provide significant clinical benefits in
the hands of proponents, have been subsequently
proven ineffective when tested rigorously in well-

103

Outcomes Assessment and Evidence-Based Practice Guidelines

Table 6 | Steps to be Taken Before Patients Are Enrolled in

a Clinical Trial
Establish eligibility criteria that indicate the presence or required eligibility requirements and the absence of exclusion criteria.
Characterize the demographic and general health of the patients eligible
for enrollment into the clinical trial.
Establish a research study database and include baseline data for assessment for changes in the outcome variable to be measured over the
course of the clinical trial.
Recognize and account for stratification required in the randomization
process (such as age, gender, race, diagnosis, smoking, education,
body mass index, socioeconomic factors).
Develop a system for the determination of follow-up outcomes surveys
and tracking patients throughout the trial duration.
Assess performance to the protocol and adherence to the clinical trial
research protocol with quality checks of clinical sites and investigators.
Establish a Data and Safety Monitoring Board to independently monitor
the progress and safety of the clinical trial.

designed studies. When comparisons of well-designed

nonrandomized trials of selected medical topics with
very narrow and specific selection criteria for the quality of the meta-analysis were compared with the results
of randomized trials for the same diagnosis, very good
correlation was observed between the summary odds ratios. However, the nonrandomized studies tended to
show larger treatment effects, and the discrepancies beyond chance were less common when only prospective
studies were considered.
The impact of the randomized clinical trial can be
determined by four factors: the baseline control groups
risk of an outcome event, the responsiveness of patients
to experimental treatment, the potency of the experimental treatment, and the completeness with which the
outcome events are ascertained and included. The goal
of enrollment must be to enroll those most likely to
have the events that are hoped to be prevented with the
experimental treatment. Therefore, to measure the true
success of the experimental treatment, these high-risk
patients become the group most likely to benefit from
participation in the trial and should be the target for enrollment. It is also critical that all attempts be made to
preserve the sample size and prevent erosion of the participants by crossover or dropout in follow up. This allows the detection of events that otherwise might have
been missed, and it increases the chances of being able
to present a worse-case scenario. This involves the
process by which all experimental patients lost to follow
up in a positive trial are assigned bad outcomes, and all
lost controls are assigned a positive outcome. The loss of
minimal patients to follow-up enhances the credibility
of a trials positive conclusions.
To ensure proper data collection to assess the impact
of an intervention, study investigators must agree on the

104

Orthopaedic Knowledge Update 8

details of data collection, and the process by which the
data are collected (paper, telephone, personal interview,
computer). The schedule for data collection must be defined to determine when patients are to be surveyed
throughout the study. To properly select appropriate
outcome measures, the investigators must be thoroughly
familiar with the disease process under investigation.
This allows for the understanding of associated conditions, comorbidities, and other variables that are likely
to influence the clinical results. These other associated
conditions and variables (such as age, gender, body mass
index, smoking, education level, medical comorbidities)
must be measured to provide for statistical control for
their influence (confounding effects) on the primary and
secondary outcomes of interest.

Enrollment in a Clinical Trial

Before patients are enrolled in a clinical trial, several
steps must be taken and these are outlined in Table 6.
The enrollment process is best preceded by pilot testing
of the procedures, survey instruments, and the overall
process on test patients to resolve any potential problems
before the initiation of the clinical trial with actual enrolled and randomized patients. When a patient is considered for the clinical trial, it is necessary to determine the
patients eligibility for participation. It is most efficient if
all the initial required data are collected at
a single visit, the recruitment/preenrollment/randomization visit, during which baseline data are obtained and
treatment assignment occurs. The initial visit must also
include a thorough explanation of the clinical trial, explaining the purpose and goals, and obtaining the informed consent for participation.
Once randomization occurs, the patient is a member
of a treatment group to which they have been assigned.
Postrandomization follow-up visits are then scheduled
at predetermined times according to the data collection
schedule. Completeness of the follow-up visits is critical
to the success of the trial. Incomplete follow-up or withdrawals may bias the results. To ensure maximum
follow-up participation, it is critical to minimize the burden on the patients by requiring only specific instruments to be completed at each visit. Subjects without
telephones or those who plan to move are best excluded, and the addition of contact information including email addresses and names and contact information
of family and friends is critical to maintaining contact.
This information should be updated at each visit. Providing reminders for upcoming visits also ensures compliance with longitudinal follow up. Incentives are also
helpful, whether they are financial or other tokens of
appreciation for continued participation in the clinical
trial.
Data collection and maintenance is critical to the
success of a clinical trial. Data must be secure, safe, and

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 9 Outcomes Assessment and Evidence-Based Practice Guidelines

protected. Computerized data must be backed up regularly. Specific identifying data for each patient is critical
for providing and maintaining confidentiality and consistency. Dedicated personnel whose responsibilities are
specific to data management infrastructure are critical
to this process.
The role of maintaining the safety of the participants
in the clinical trial is performed by the Data and Safety
Monitoring Board (DSMB). This group is commonly
composed of five non participating individuals who are
knowledgeable about the disease process and statistical
and study design. This committee reviews the summaries
of safety, accrual and progress of the trial, the quality of
the data, and blinded interim efficacy and effectiveness
analyses and reports its findings to the principal investigator and executive working group. It is also responsible
for interpreting data on adverse side effects. The DSMB
meets every 6 months and makes recommendations to
the principal investigator and executive working group
regarding actions to ensure patient safety and that participants are not exposed to undue risks. The mandate of
the DSMB should comply with the July 1, 1999 release
of the National Institutes of Health policy for Data and
Safety Monitoring, with the primary function being to
ensure the safety of participants and validity and integrity of the data. The DSMB has direct communications
with the funding agency and can stop a clinical trial
when public health or safety is at risk, or when study
goals are not met.

Evidence-Based Medicine
Physicians seek to base their decisions on the best available evidence. Often this represents experience, teaching, and extrapolations of pathophysiologic principles
and logic rather than established facts based on data derived from patients. The advent of randomized controlled clinical trials has led to an increase in the quality
of evidence concerning clinical treatment interventions,
making clinical reasoning more comprehensible. The
ability to track down and critically appraise and incorporate evidence into clinical practice has been termed
evidence-based medicine. It is the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients using
patient-centered clinical and basic science research data
along with the patients values and expectations.
Evidence-based medicine is not cookbook medicine.
It is the integration of the best external evidence with
individual clinical expertise and patients choice; neither
alone is enough.
Many of these concepts were proposed by Codman in
the early 1900s, who implemented the end result idea.
Codman attempted to put into practice the notion that
each hospital should follow every patient it treats, long

American Academy of Orthopaedic Surgeons

enough to determine whether or not the treatment has

been successful, and then to inquire if not, why not?
Evidence-based medicine is not restricted to randomized clinical trials and meta-analyses. It involves
tracking down the best external evidence with which to
answer the clinical questions. Best available external evidence includes clinically relevant research from both
the basic sciences and patient-centered clinical research.
It is when questions are asked about therapy that the
nonexperimental approaches should be avoided, as
these often lead to falsely positive conclusions about efficacy. The randomized clinical trial is the gold standard
for judging whether a treatment does more harm than
good. A systematic review of several randomized trials
provides even greater degree of confidence in the level
of evidence.
Without current best evidence, a clinical practice
risks the possibility of rapidly becoming out of date, to
the detriment of the patient. Three evidence-based medicine strategies have been developed to help the clinician maintain and expand clinically important knowledge: learning evidence-based medicine, seeking and
applying evidence-based medicine summaries generated
by others, and accepting evidence-based protocols and
practice guidelines developed by others.

Classification of Study Strength

Evidence is a critical component in the decision-making
process for delivering care to patients. For many questions, however, the amount of information is overwhelming and the conclusions often contradictory. Most
surgeons would agree that to make the best decisions,
evidence must be easily available in a comprehensive
and timely fashion.
Levels of evidence, based on the rigor of the study
design, are a way to sort through and evaluate the literature. Levels of evidence may also be used in the development of practice guidelines to weigh the strength of
the recommendations, and should be used by practicing
surgeons to wade through the multiple types of evidence to help determine which information is useful to
them and their patients.
Levels of evidence provide only a rough guide to the
study quality. Level I evidence may not be available for
all clinical questions, and level II and III evidence can
still have great value to the practicing surgeon. No single study will provide all the definitive answers.
The classification of levels of evidence was developed to minimize therapeutic harm to patients by using
evidence that is least likely to be wrong. The Center
for
Evidence-Based
Medicine
(http://cebm.net/
levels_of_evidence.asp) has created the categories for
defining study strength (level of evidence) (Table 7) and
grades (Table 8) for recommendations in common use
today. Similar levels and grades have also been devel-

105

Orthopaedic Knowledge Update 8

Outcomes Assessment and Evidence-Based Practice Guidelines

Table 7 | Oxford Centre for Evidence-Based Medicine Levels of Evidence and Journal of Bone and Joint Surgery: Instructions to
Authors

Level I

Level II

Level III

Level IV
Level V

Therapeutic Studies:
Investigating the Results of
Treatment

Prognostic Studies:
Investigating the Outcome of
Disease

Randomized controlled trial

Significant difference
No significant difference but
narrow CIs*
Systematic review of Level I
randomized controlled trials
(studies were homogeneous)
Prospective cohort study
Poor-quality randomized
controlled trial (eg, < 80%
follow-up
Systematic review
Level II studies
Nonhomogeneous Level I studies
Case-control study
Retrospective study
Systematic review of Level III
studies

Prospective study
Systematic review of Level I
studies

Case series (no, or historical,

control group)
Expert opinion

Diagnostic Studies:
Investigating a Diagnostic Test

Testing of previously developed

diagnostic criteria in series of
consecutive patients (with
universally applied reference
gold standard)
Systematic review of Level I
studies
Retrospective study
Development of diagnostic
Study of untreated controls from a criteria based on basis of
previous randomized controlled consecutive patients (with
trial
universally applied reference
Systematic review of Level II
gold standard)
studies
Systematic review of Level II
studies
Study of nonconsecutive patients
(no consistently applied
reference gold standard)
Systematic review of Level III
studies
Case series
Case-control study
Poor reference standard
Expert opinion
Expert opinion

Economic and Decision

Analysis: Developing an
Economic or Decision Model
Clinically sensible costs and
alternatives; values obtained
from many studies; multiway
sensitivity analyses
System review of Level I studies

Clinically sensible costs and

alternatives; values obtained
from many studies; multiway
sensitivity analyses
System review of Level II studies

Limited alternatives and costs;

poor estimates
Systematic review of Level III
studies
No sensitivity analyses
Expert opinion

*CI = Confidence interval

Table 8 | Oxford Centre for Evidence-Based Medicine

Grades of Recommendation
A
B
C
D

Consistent Level 1 studies

Consistent Level 2 or 3 studies or extrapolations from Level 1
studies
Level 4 studies or extrapolations from Level 2 or 3 studies
Level 5 evidence or troublingly inconsistent or inconclusive studies of any level

oped by this group for evidence and actions in the domains: prognosis, etiology/harm, and economic analysis.

toms, and health-related quality of life as relevant outcomes of clinical care. The growing efforts toward
understanding the clinical study design and the implementation of multicenter prospective randomized clinical trials will provide the best evidence with which orthopaedic surgeons can develop practice guidelines,
decrease practice variability, provide evidence-based
quality care, and allow patients to make informed
choices. The current mandate for quality, necessitating
the measurements of results, provides an opportunity to
develop a more focused approach to the delivery of care
by orthopaedic surgeons and a means to document the
quality of the end result.

The Future of Evidence-Based Approaches in

Orthopaedic Surgery

Annotated Bibliography

Efforts in improving the quality of care delivered to patients remains a continuous work in progress. The routine use of validated and standardized outcomes instruments, not primarily for research purposes but also for
use in routine clinical practice, will provide data with
which to assess interventions. It seems likely that there
will be a growing and continued emphasis on evidencebased approaches in assessing functional status, symp-

Centre for Evidence-Based Medicine: Oxford-Centre

for Evidence Based Medicine Web site. Levels of evidence and grades of recommendation. Available at:
http://www.cebm.net/levels_of_evidence.asp. Accessed
September 1, 2003.

106

This is a comprehensive Web-based resource on evidencebased medicine, including definitions and strategies for searching for literature on evidence-based medicine.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 9 Outcomes Assessment and Evidence-Based Practice Guidelines

Ioannidis JP, Haidich A, Papa M, et al: Comparison of

evidence of treatment effects in randomized and nonrandomized studies. JAMA 2001;286:821-830.
In this review of 45 topics in selected medical topics, analysis of the results demonstrates larger treatment effects and
discrepancy beyond chance nonrandomized studies compared
with randomized trials.

Kocher MS, Zurakowski D: Clinical epidemiology and

biostatistics: An orthopaedic primer. J Bone Joint Surg
Am 2001; Orthopaedic Journal Club: 1-12.
This is a primer of clinical epidemiology and biostatistics
for the orthopaedic surgeon.

McCulloch P, Taylor I, Sasako M, Lovett B, Griffin D:

Randomised trials in surgery: Problems and possible solutions. BMJ 2002;324:1448-1451.
The quality and quantity of randomized trials in surgery is
acknowledged to be limited. Some aspects of surgical trials
present special difficulties for randomized trials and these difficulties and proposed solutions for improving the standards of
clinical trials in surgery are discussed.

Sackett DL: Why randomized controlled trials fail but

neednt: 2. Failure to employ physiological statistics, or
the only formula a clinician-trialist is ever likely to need
(or understand!). CMAJ 2001;165:1226-1237.
Causes of failure in randomized controlled trials is neither
with the teacher nor the trialist, but lies in the mismatch between what is judged necessary to be learned about biostatistics and who learns it. Pitfalls in conducting clinical trials are
discussed, along with appropriate sample size and eligibility
and erosion of study participants.

Daum WJ, Brinker MR, Nash DB: Quality and outcome

determination in health care and orthopaedics: evolution and current structure. J Am Acad Orthop Surg
2000;8:133-139.
Deyo RA, Weinstein JN: Outcomes research for spinal
disorders, in Herkowitz H, Garfin S, Balderston R, Eismont F, Bell G, Wiesel S (eds): Rothman-Simeone: The
Spine, ed 4. Philadelphia, PA, WB Saunders, 1999.
Gartland JJ: Orthopaedic clinical research: Deficiencies
in experimental design and determinations of outcome.
J Bone Joint Surg Am 1988;70:1357-1364.
Hennekens C: Buring J: Interventional studies, in Mayrent S (ed): Epidemiology in Medicine. Boston, MA, Little, Brown and Company, 1987, pp178-212.
Jekel JJ, Elmore JG, Katz DL (eds): Epidemiology, Biostatistics and Preventive Medicine. Philadelphia, PA, WB
Saunders, 1996, pp 160-171.
Kaska SC, Weinstein JN: Historical perspective: Ernest
Amory Codman, 1869-1940: A pioneer of evidencebased medicine. The end result of an idea. Spine 1998;
23:629-633.
Keller RB: Outcomes research in orthopaedics. J Am
Acad Orthop Surg 1993;1:122-129.
Kopec JA: Measuring functional outcomes in persons
with back pain: A review of back-specific questionnaires. Spine 2000;25:3110-3114.
Matthews DE, Farewell VT: Using and Understanding
Medical Statistics. Basel, Germany, Karger, 1985, pp 184195.

Watters WC III: Evidence-based practice and the use of

practice guidelines. AAOS Bulletin, August 2003, pp 1516.

Rosenberg WM, Sackett DL: On the need for evidencebased medicine. Therapie 1996;51:212-217.

Scientific journals, textbooks, and presentations provide

the cornerstone of good clinical decision making and patient
care. The benefits and use of evidence-based medicine to produce and practice guidelines in the context of clinical quality
improvement cycle is demonstrated.

Sackett DL: Evidence-based medicine. Spine 1998;23:

1085-1086.

Wright J: Summarizing the evidence. AAOSBulletin,

June 2003, pp 51-52.
Evidence-based medicine, levels of evidence, and strategies for summarizing evidence are reviewed for the practicing
orthopaedic surgeon faced with determining treatment recommendations.

Classic Bibliography
Bombardier C: Outcome assessments in the evaluation
of treatment of spinal disorders. Spine 2000;25:30973099.

American Academy of Orthopaedic Surgeons

Sackett DL: Why randomized controlled trials fail but

neednt: 1. Failure to gain coal-face commitment and
to use the uncertainty principle. CMAJ 2000;162:13111314.
Sackett DL, Rosenberg WM, Muir Gray JA, Hayes RB,
Richardson WS: Evidence based medicine: What it is
and what it isnt. BMJ 1996;312:71-72.
Ware JE, Snow KK, Kosinski M, Gandek B: SF-36
Health Survey: Manual and Interpretation Guide. Boston, MA, The Health Institute, New England Medical
Center, 1997.
Weinstein JN, Deyo RA: Clinical research: issues in data
collection. Spine 2000;25:3104-3109.

107

Chapter

10

Clinical Epidemiology:
An Introduction
Karel G.M. Moons, MSc, PhD
Diederick E. Grobbee, MD, PhD

Introduction
The concept of evidence-based medicine implies that
the medical care of individual patients should be based
on results obtained from patient-oriented quantitative
research, rather than on qualitative research or clinical
experience. Patient-oriented quantitative research is
also referred to as clinical epidemiologic research.
Traditionally, epidemiologic research focused on the
occurrence of infectious diseases and tried to unravel
determinants of infectious disease epidemics across populations. Over time, however, it has been shown that
methods used for this type of epidemiologic research
(population epidemiology) can be applied in a similar
manner to investigate clinical questions. Clinical epidemiology is a term commonly used for epidemiology
dealing with questions relevant to medical practice. Accordingly, evidence-based medicine is particularly
served by results from clinical epidemiologic research.
The distinction between clinical epidemiology and population epidemiology may be somewhat artificial because both types of studies use largely the same methods for design and analysis. To properly serve clinical
practice and provide for evidence-based medicine, clinical epidemiologic studies should address relevant clinical questions, be validly executed, and should yield results with sufficient precision.

Clinical Epidemiologic Research Questions

Understanding the challenges related to clinical practice
is essential for understanding the objectives of clinical
epidemiologic research. Beginning with the patientphysician encounter, four principal challenges or questions in clinical practice can be defined. These questions
relate back to Hippocrates and form the basis of the
evidenced-based medicine movement. In hierarchical
order these questions are: (1) What is the illness of the
patient given the clinical and nonclinical profile?
(2) Why did this illness occur in this patient at this time?
(3) What will be the course of the patient with this illness and manifestations? (4) How can I, as a physician,

American Academy of Orthopaedic Surgeons

improve the course of the patient with this illness and

manifestations?
The first question asks for the most probable diagnosis or illness given the clinical (such as symptoms,
signs, test results) and nonclinical (such as age, sex, and
socioeconomic status) profile of the patient. The answer
to the second question, which asks for the cause of the
illness, is often impossible to give. Often, adequate care
can be provided without knowing the cause of an illness.
For example, a torn meniscus can effectively be treated
by the orthopaedic surgeon without knowing how the
injury occurred. The most challenging question concerns
a prediction of the clinical course of the patient given
the clinical and nonclinical profile, the underlying illness, and the possible etiology of that illness (question
3). This prediction of the patients clinical course may be
made without considering treatment (baseline prognosis) as well as with the consideration of the effects of
possible treatments. Differences in the estimated course
with and without treatment may guide the physicians
decision on whether to initiate treatment and in the
choice between various treatment options. For example,
if in a patient with a torn meniscus the probability of an
unfavorable outcome (such as pain with activity) is
much higher without surgery (baseline prognosis) than
with surgery, it is likely that surgery would be chosen as
the proper course of action (question 4).
For a physician to meet the challenges of clinical decision making, quantitative or probabilistic knowledge is
required. Diagnostic research yields quantitative knowledge about the probability of disease presence or absence given the patients clinical and nonclinical profile
(addressing question 1). Results from etiologic research
help the physician to answer questions about the probability that a particular disease will occur when a particular risk factor is present (question 2). Prognostic research provides quantitative knowledge about the
probability that a particular outcome (death, complication, recurrence, or improved quality of life) may occur
over a particular period of time in relation to the patients profile, illness, and etiology (question 3). Treatment decisions require knowledge about the probability

109

Orthopaedic Knowledge Update 8

Clinical Epidemiology: An Introduction

Table 1 | Main Design Characteristics of Diagnostic, Etiologic, Prognostic, and Therapeutic Studies
Determinant(s)

Outcome

Occurrence Relation

Domain

Factor(s) under study that are

related to the outcome
occurrence

Health parameter (eg disease,

survival, complication, quality
of life) under study

Association between the

outcome occurrence and the
determinant(s)

Larger, theoretical population

to whom the study results
can be applied (generalized)

Type of research
Diagnostic
Test results under study
research

Presence/absence of target
disease

Etiologic
research

Causal factor under study

Incidence of disease under

study

Patients with particular

symptom/sign suspected of
particular disease
Those potentially at risk to
develop the outcome at
interest

Prognostic
research

Prognostic predictors under

study

Therapeutic
research

Treatment (relative to a control

group) under study

Incidence of disease (eg,

disease, survival,
complication, quality of life)
under study
Incidence of disease (eg,
disease, survival,
complication, quality of life)
under study

Presence (prevalence) of
disease in relation to
combination of test results
Incidence of the outcome in
relation to the causal factor,
accounted for all possible
confounders
Incidence of the outcome in
relation to combination of
prognostic predictors
Incidence of the outcome in
relation to the treatment,
accounted for all possible
confounders

Patients with a particular

disease for whom the
treatment under study is
indicated

General
definition

or extent to which a treatment improves the prognosis

(intended effect) and at what cost (such as treatment
risks [unintended effects] and monetary costs).
In medical practice, few outcomes are certain or
pathognomonic. Hence, the objective of clinical epidemiologic research is to provide probabilistic knowledge
on diagnosis, etiology, prognosis, and treatment of illnesses as a means to enhance patient care. The design,
conduction, and data analysis of clinical studies on diagnosis, etiology, prognosis, or treatment each have their
own characteristics.

Basic Principles of Clinical Epidemiologic

Studies
Research Question, Study Population,
and Generalization
Any research question related to patient care is either
diagnostic, etiologic, prognostic, or therapeutic in nature. Although each type of study has its own characteristics, every research question and thus each type of
study addresses an association between the occurrence
or probability of an outcome on one side and factors
that are related to this outcome occurrence on the other
side. These factors are commonly called determinants.
Accordingly, central to clinical epidemiologic research is
the occurrence relation or the association between the
outcome occurrence and its determinant(s).
To quantify this relation requires empirical data,
which in turn requires a study population. For each
member or patient of the study population, both the
presence of the determinant(s) and of the outcome

110

Patients potentially at risk to

develop the outcome at
interest

needs to be documented. From these empirical data, the

true nature and strength of the association between determinant and outcome is quantified.
To apply the findings with confidence in patient care,
the study results or estimated association need to be
generalized to a larger population (domain) from which
the patients of the study population were selected. The
extent to which the results found in a particular study
population can be applied to a larger (theoretical) patient population is called the generalizability of the
study results.
Therefore, regardless of the question studied, each
type of clinical epidemiologic study (diagnostic, etiologic, prognostic, or therapeutic) has certain basic elements: the occurrence relation, outcome, determinant(s), study population, and domain. These elements
are described later in this chapter according to each
type of study and also are outlined in Table 1.

Causal Versus Descriptive Research

A major distinction first should be made between causal
and descriptive epidemiologic research, because of the
major consequences for study design, conduction, and
data analysis. The difference is directed by the determinant(s) under study. Although determinant is a term
that seems to refer to factors that cause the occurrence
of an outcome, it should be stressed that a determinant,
and therefore clinical epidemiologic research, may be
causal or noncausal (descriptive).
Causal studies quantify whether a specific determinant is causally related to an outcome; for example,

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

trauma as a causal factor for vertebral fractures or immobility as a causal factor for pulmonary embolism. The
goal of causal studies is to explain the occurrence of an
outcome in relation to a causal determinant. Treatment
studies are also causal studies. A treatment cannot be
effective if it does not interfere in some way in the
causal pathway of the outcome, such as progress of the
disease or complications. A treatment is also a causal
determinant, although it does not induce outcome (such
as trauma and immobility) but rather reduces the likelihood or prevents the occurrence of outcomes. The fact
that therapeutic studies are causal studies is very important, as the design and data analysis of causal studies are
quite different from the design and analysis of descriptive studies.
In descriptive studies, the occurrence of the outcome
is only described by the determinant(s) at interest. The
motive of descriptive studies is to quantify whether a
determinant predicts rather than explains the occurrence of the outcome. Causality is not investigated. Consider, for example, a study to quantify whether the skin
color of a neonate (one of the items of the Apgar score)
is predictive for its survival (a prognostic study question). Skin color obviously is not causally related to
mortality. Accordingly, prognostic research that does not
involve questions about the effects of treatment is descriptive research. Diagnostic studies are descriptive as
well. For example, in a patient who reports symptoms of
a knee disorder, abnormalities on the MRI scan of the
knee may predict the presence of knee arthrosis. Results
of the MRI scan obviously are not the cause of the knee
arthrosis, but may predict the likelihood of the presence
or absence of the condition.
Therefore, etiologic and therapeutic research questions address causality between a determinant and outcome. To answer these questions requires a causal study
design and causal data analysis. Diagnostic and prognostic research questions address the predictive value between determinants and outcome. Causality is not investigated. To answer diagnostic and prognostic research
questions requires a descriptive study design and corresponding data analysis. The distinction between causal
and descriptive research is further explained using clinical examples in the next section.

Characteristics of Diagnostic, Etiologic,

Prognostic, and Therapeutic Studies
Diagnostic Research
Diagnosis in practice starts with a patient presenting
with a particular set of symptoms or signs that lead the
physician to suspect a particular disease, the so-called
target disease. The goal of the physician is to estimate
the probability of the presence of the underlying disease. This probability is (often implicitly) estimated using diagnostic test results. These test results are obtained

American Academy of Orthopaedic Surgeons

Chapter 10 Clinical Epidemiology: An Introduction

from patient history (each question, including simple
questions on age and sex, is a different diagnostic test),
physical examination, and additional tests such as imaging, electrophysiology, and laboratory tests. Currently,
many diagnostic tests are used, sometimes with considerable burden to patients and health care budgets.
Therefore, it is important for physicians, patients, and
health care providers to know which additional tests
truly contribute to the distinction between disease presence and absence beyond the information obtained
from patient history and physical examination. Quantitative diagnostic research is needed to acquire this
knowledge. For example, a patient with low back pain
who is suspected of having a herniated disk (target disease) is referred to an orthopaedic surgeon. A possible
diagnostic question could concern which results from
the patients history (age, trauma, pain when seated)
and from the physical examination (notably the Lasgue
test) are relevant to predict or estimate the probability
of the presence of a herniated disk. Also, does the conventional radiograph of the lower back provide added
predictive value?
Occurrence Relation, Study Population,
and Determinants
The occurrence relation of a diagnostic study to answer
the above question would concern the frequency (or
probability) of a herniated disk in relation to patient
history, physical examination, and radiographic results.
To study this relation, the study population should include a sufficient number of patients with low back pain
who have suspected disk herniation. Each patient
should undergo all diagnostic tests under study (all relevant tests indicated from patient history, physical examination, and lower back radiography). These tests are
called the diagnostic determinants under study.
Outcome and Data Analysis
Subsequently, each study patient should undergo the
reference or gold standard test to assess the patients diagnostic outcome or diagnostic truth. The procedure
that is used in practice (at the time of study initiation)
to ultimately determine the presence or absence of the
target disease, should be used as a reference standard in
a scientific diagnostic study. In the example of the herniated disk, MRI of the low back could be the reference
standard. The results of the reference standard should
always be interpreted without knowledge of the diagnostic tests or determinants under study, to prevent information bias. Some diseases may lack a single reference standard and sometimes it may be unethical or too
costly to perform the reference standard in each patient.
In such situations, a combination of tests or the socalled consensus diagnosis is commonly used to determine the diagnostic truth in each study patient.

111

Clinical Epidemiology: An Introduction

Documented patient data from the study in the example can then be used to quantify the probability of a
herniated disk in patients with particular symptoms and
signs, and to quantify whether radiographs increase or
decrease this probability, and to what extent. The latter
reflect the added value of the radiographs beyond the
information obtained from the history and physical findings. Causality between the determinants (test results)
and outcome (final diagnosis) is not an issue. The only
matter of importance is whether they predict the outcome presence. Diagnostic research is therefore descriptive or predictive research. To properly serve clinical
practice, diagnostic studies should address multiple determinants or tests rather than one single test. In practice, any diagnosis is rarely set by a single test. Results
from patient history and physical examination will always be considered before additional tests are applied.
Domain and Generalization
The domain or theoretical target population should be a
defined group (such as all patients presenting in secondary care with symptoms or signs indicative of herniated
disk) to which the results of the diagnostic study can be
generalized. In general, the domain of a diagnostic study
is first of all determined by the symptoms and signs exhibited by the study patients. The domain definition often includes the setting (primary, secondary, or tertiary
care) as well.
The selection or recruitment of any study population
is often restricted by logistic circumstances, such as necessity of patients to live nearby the research center or
the availability of time to participate in the study. These
characteristics, however, are unlikely to influence the
applicability and generalization of clinical study results.
It is important and challenging for the physician to appreciate which characteristics truly affect the generalizability of results obtained from a particular study population. This appreciation usually requires external
knowledge of characteristics of the study population
and of the research setting that may change the nature
and strength of the estimated associations between determinant(s) and outcome. Therefore, generalizability or
definition of the domain of a clinical study is not an objective process that can be framed in statistical terms.
Generalizability is a matter of reasoning.

Etiologic Research
Clinicians and epidemiologists seem to be most familiar
with etiologic research despite its limited direct relevance to patient care and its methodologic complexities.
The goal of etiologic studies is to quantify whether a
single, specific determinant is indeed causally related to
the outcome under study. Consider, for example, a study
to quantify whether frequent falling may cause hip fractures in the elderly.

112

Orthopaedic Knowledge Update 8

Occurrence Relation, Study Population,
and Determinants
The occurrence relation of this study would be the occurrence (probability) of hip fractures in relation to the
frequency of falling. To study this relation, for example,
a group (cohort) of study subjects (men and women
older than 70 years who have not had a hip fracture)
can be recruited and followed for a 5-year period. Each
year, for example, the study subjects visit the research
center so that the number of falls in the previous year
can be assessed. In contrast to diagnostic and prognostic
studies, etiologic studies focus on the quantification of
the causal relation of a single determinant (falling) to
the outcome (hip fractures).
Outcome and Data Analysis
During the study period, whether and when a hip fracture has occurred is documented for each subject. After
the study period, the number of falls and whether a hip
fracture has occurred is known for each subject. In the
analysis, two determinant groups can be formed. One
group includes subjects with a high frequency of falling
(on average three or more times per year) and the other
with a low frequency (less than three times per year).
Accordingly, whether the occurrence of hip fractures is
significantly higher in those with high versus low number of falls can be quantified. In addition, the data can
be analyzed using a cutoff value other than three, or using more than two determinant groups (nonfallers, those
with less than three falls, and those with more than
three falls per year).
Domain
The domain to which the study result can be generalized
will include all patients older than 70 years with no history of a hip fracture. In general, the domain of an etiologic study includes all persons who are potentially at
risk to develop the outcome at interest. As stated earlier, other characteristics of the study population that
may affect the generalizability of the estimated association also need to be considered in the domain definition. Such generalization remains a matter of subjective
reasoning requiring external knowledge of the studied
determinant-outcome or occurrence relation.
Confounders
In contrast to diagnostic and prognostic research, in etiologic research the role of confounders is an additional
issue that must be considered. Suppose the example
study would yield an incidence of hip fractures (within 5
years) of 6% among the frequent fallers and 3% among
the nonfrequent fallers. The initial conclusion would be
that frequent falling causes a twofold increased risk of
hip fractures in subjects older than 70 years. This observation, however, might not be true. The observed difference in frequency of hip fractures could be caused by

American Academy of Orthopaedic Surgeons

Chapter 10 Clinical Epidemiology: An Introduction

Orthopaedic Knowledge Update 8

Table 2 | Hypothetical Example of the Distribution of

Various Potential Confounders Across Subjects With a High
(More Than Three Times Per Year) and Low (Less Than Three
Times Per Year) Frequency of Falling

Confounder

Low Frequency of
Falling
(n = 1,500)

High Frequency of
Falling
(n = 1,000)

Mean age (years)

Current smoker
BMI (kg/m2)

74
35%
29

83*
36%
25*

*P value < 0.001

other risk factors that happened to be different across

the two determinant groups. For example, it is known
from the literature that increased age, smoking, and low
body mass index (BMI) are also risk factors for hip fractures. If (by accident) in the study population these
other risk factors are unequally distributed across the
two groups of fallers, the observed difference of 3% in
hip fractures could be caused by these other risk factors.
In etiologic research these other risk factors of the outcome under study are called confounders.
Table 2 shows hypothetical results of the distribution
of the other three risk factors of hip fractures across the
two groups of fallers. Age was significantly higher and
BMI significantly lower among the frequent fallers,
whereas the smoking distribution was the same. The
higher frequency of hip fractures among the frequent
fallers could obviously not be caused by smoking (which
was the same for both groups) but could be caused by
their higher age and lower BMI compared with the infrequent fallers.
Note that a risk factor can only be a true confounder
in an etiologic study, if it indeed has a causal association
with the outcome, and is also (but not necessarily causally) associated with the determinant. The latter simply
means that the confounder is unequally distributed
across the two determinant groups. This situation is
schematically shown in Figure 1. If association 1 and 2
exist in the data, C is a true confounder of the association between determinant and outcome. Because the
goal of an etiologic study is to quantify whether there is
a causal association between the determinant (falling)
and outcome (hip fractures), the influence of any other
risk factors that might confound this estimation must be
ruled out or accounted for. If confounders are not accounted for in an etiologic study, then an observed association might not be the true association, and is therefore invalid. The association between the causal
determinant and outcome without adjusting for confounders is called the crude association.

American Academy of Orthopaedic Surgeons

Figure 1 Diagram to determine whether a risk factor is a confounder of the

studied association. D = determinant;
O = outcome; C = potential confounder.
1 = the association between the potential confounder and the outcome; 2 =
the association between the potential
confounder and the determinant.
Accounting for Confounders
The most common and successful methods to account
for confounding are randomization and adjustment in
the analysis. Randomization means that the investigator
randomly assigns the study subjects to the two (or
more) determinant groups. Because of this process, the
two determinant groups are by definition comparable
for all other risk factors except for the determinant at
issue, given that the groups are large enough. Randomization, however, is frequently not feasible or ethical. In
the example study it is impossible to randomize subjects
to frequent or nonfrequent falling groups. The same
concept applies to causal studies investigating physiologic determinants such as blood pressure, cholesterol
level, age, and bone mineral density. For ethical reasons,
randomization is commonly not possible if the determinant increases the risk of the outcome at interest. For
example, when studying the causal relationship between
smoking and lung cancer, or alcohol consumption and
hip fractures, subjects cannot be randomly assigned to
years of smoking or alcohol consumption. Randomization is only possible for causal determinants that reduce
the occurrence of the outcome at interest, and is mainly
reserved for therapeutic determinants. Because therapeutic research is causal research, randomized studies
particularly apply to therapeutic research.
In nonrandomized causal studies, accounting for
confounding must be done in the data analysis. During
the design phase, all potential risk factors of the outcome need to be defined beforehand such that association 1 in Figure 1 is fulfilled. This identification of the
potential risk factors should be based on existing knowledge or educated reasoning. These other risk factors
should then be measured in each study subject in addition to the determinant of interest. In the analysis it
should be assessed whether association 2 in Figure 1 is
indeed present in the data. If so, a so-called multivariable analysis is applied to adjust the crude association
between the determinant and outcome (a difference of
3% hip fractures in the example) for all confounders. A
determinant is called causally associated with an outcome, if the association remains after having accounted
for all confounders in the data.
Note that adjustment for confounding might dilute
but could also increase the crude association between

113

Clinical Epidemiology: An Introduction

the determinant and outcome, depending on the distribution of the confounders across the determinant categories and whether the confounder is a risk factor or
preventive factor of the outcome. In the example study,
adjustment for age and BMI would dilute the crude association of 3% difference in hip fractures because increased age and low BMI both increase the occurrence
of hip fractures, and the frequent fallers were on average older and leaner than the nonfrequent fallers. Studies whose purpose is to quantify whether an unintended
effect of a treatment is truly caused by that treatment
are etiologic studies as well and should follow the above
principles.

Prognostic Research
Prognostic research has received limited attention in applied medical research. In practice, to set a prognosis is
to estimate the probability that a patient with a particular illness and clinical and nonclinical profile will develop a particular outcome (death, a complication, recurrence of disease, or a good quality of life) within a
certain period of time. The prognostic probability could
also be estimated given that the patient has undergone a
particular treatment. Here, treatment is considered as
one of the prognostic factors. Prognosis in practice does
not simply imply the typical course of an illness or diagnosis; it refers to the course of a patient with particular
clinical and nonclinical characteristics. Just as with diagnostic questions, it is relevant for physicians to know
which information is truly needed to estimate a patients
prognosis. As with diagnosis, the probability should preferably be estimated with information or determinants
that are easily obtainable from the patient using noninvasive, low-cost methods. The principles of design and
analysis of prognostic research are similar to those of diagnostic research, because both can be grouped under
the heading of descriptive or prediction research. Their
study goal is to quantify the predictive association between the determinant(s) and the outcome; causal explanation of the outcome is not necessary.
An example of a prognostic question is whether preoperative patient characteristics are predictive for the
occurrence of postoperative nausea and vomiting within
24 hours in patients undergoing orthopaedic surgery. If
this occurrence can be estimated preoperatively, preferably using easily obtainable patient variables, the anesthesiologist or orthopaedic surgeon could perform a
timely intervention in high-risk patients to reduce the
risk of postoperative nausea and vomiting. For example,
instead of using isoflurane, desflurane, or sevoflurane
for general anesthesia, intravenous propofol, which has
been proven to cause less postoperative nausea and
vomiting could be used, or preemptive administration of
antiemetics could be done.

114

Orthopaedic Knowledge Update 8

Occurrence Relation, Study Population,
and Determinants
The occurrence relation of a prognostic study to answer
this example question would be the frequency (or probability) of postoperative nausea and vomiting within
24 hours in relation to multiple preoperative patient
characteristics, such as planned duration of surgery, age,
sex, history of postoperative nausea and vomiting, and
smoking status. The study population could include patients at a particular hospital undergoing any orthopaedic surgery who are at risk of developing postoperative nausea and vomiting. All prognostic predictors of
the outcome should be documented from each study patient. Similarly, as in diagnostic research, prognostic research should address multiple predictors rather than a
single prognostic predictor because in practice, no prognosis is set by a single predictor.
Outcome and Data Analysis
Each patient should be followed up to measure whether
the outcome has occurred. In the example, this measurement is done at 24 hours after surgery, using a simple
questionnaire. As with diagnostic research, the person
measuring the outcome must be blinded to the patients
predictor values to prevent information bias. Particularly in prognostic (and therapeutic) research, outcomes
that are important to patients (such as degree of pain
and quality of life issues) should be investigated. Researchers too often rely on so-called proxy outcomes
such as duration of intensive care stay or intraoperatively measured physiologic parameters such as body
temperature or blood pressure rather than on patient
outcomes.
The preoperative predictors that truly have predictive value (that is, may predict the occurrence of postoperative nausea and vomiting within 24 hours) and which
information actually is redundant to determine the patients prognosis can be easily quantified from the data.
Domain
In general, the domain of a prognostic study is determined by the diagnosis or illness that places the patient
at risk of developing the outcome at interest. As with diagnostic research, the setting (for example, primary or
secondary) is often included if it is not inherent to the
study population. Similarly, other characteristics of the
study population that may affect the generalizability of
the estimated association also need to be included. The
domain of the example study would include all patients
scheduled for orthopaedic surgery. However, there may
be no reason to assume that the predictors estimated
from the orthopaedic study population would not also
apply to other surgical patients. Therefore, the results of
the example study could perhaps be generalized to all
types of surgery. However, as discussed previously, this
generalization requires external knowledge and subjec-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

tive judgment. The question to be asked is whether in
these other types of surgery that were not included in
the study population the same predictors would be
found with the same predictive values.
The example study shows that as with diagnostic research, prognostic studies (apart from studies on treatment effects) aim to describe or predict the occurrence
of a future outcome and possible given combinations of
prognostic determinants. Causality between the determinants and outcome is not at issue. A similar approach
as that described above has been applied to predict the
need for intraoperative blood transfusion in orthopaedic
and other types of surgery.

Therapeutic Research
Studies to quantify the intended effect of a treatment,
including surgical treatment (further referred to as therapeutic or intervention studies) are the most popular
form of clinical epidemiologic research. The methods for
design and analysis of intervention studies, including the
well-known Consolidate Standards of Reporting Trials
(CONSORT), has been extensively documented in the
literature. The principles of therapeutic research can be
described by considering the following example of a
study question: does arthroscopic dbridement in patients with osteoarthritis of the knee reduce the occurrence of pain?
The question of how the clinical course of a patient
with a particular illness and manifestations can be modified indicates that therapeutic research is a form of
prognostic research. As stated earlier, the goal of a therapeutic study is to quantify whether an observed intended effect is truly caused by the treatment under
study, excluding any other causes. Therapeutic studies
are also a form of causal research and the characteristics
described for etiologic research similarly apply to intervention research.
Occurrence Relation and Study Population
The occurrence relation of the example study question
(regarding patients with osteoarthritis of the knee)
would be the average level of pain at 24 months postoperatively in relation to arthroscopic dbridement. Study
subjects could include men and women of all ages, with
osteoarthritis of the knee who have an indication for arthroscopy. The study subjects could be selected from
four different hospitals, for example.
Determinant
Proper definition of the determinant in therapeutic
studies requires some explanation. To quantify the effect
of a treatment always requires a control group. A single
group of patients that has been treated with the intervention under study, a so-called case series, is insufficient to properly quantify the treatment effect. Any ob-

American Academy of Orthopaedic Surgeons

Chapter 10 Clinical Epidemiology: An Introduction

served outcome could be caused by other factors such
as natural course or, because the patient is aware of the
treatment, change in patient perception and lifestyle
changes that might influence the outcome occurrence.
The latter is particularly the case when the study outcome is a subjective outcome requiring patient reporting, such as pain and quality of life. If the outcome, for
example, is death, factors such as patient perception and
lifestyle changes have much less influence on the outcome. In situations in which an extreme change in outcome is observed that is unlikely to be related to other
factors, a controlled study may not be necessary. This situation, however, is rare in medicine. However, it may,
for example, apply to the benefits of total hip replacement, which leads to such a dramatic change in mobility
of a patient that the result can validly be assumed to be
the consequence of the intervention. Generally, however, the determinant in a therapeutic study includes a
group that receives the treatment (index group) and a
control group that does not.
The control group may receive either placebo treatment or an existing treatment. A placebo treatment is
exactly the same as the index treatment except that it
does not include the effective part or substance of the
treatment. Placebo treatment for a control group is
common for drug therapies, but is also possible for nonpharmacologic or surgical interventions. The placebo
drug has the same color, size, taste, or mode of use as
the index drug, except that it does not include the pharmacologic substance. Administering one half of the
study population the index and the other half the placebo treatment, and blinding the patient to the received
treatment, excludes the influence of many factors that
could influence the difference in outcome between the
two groups. These factors include patient perception and
lifestyle changes but also treatment-related factors such
as taste, size, or mode of use of a drug. As both patient
groups are unaware of the received treatment, the extent to which these factors operate will be the same
across both groups and cannot cause any differences in
outcome occurrence.
However, the use of a placebo treatment is not always feasible or ethical, especially for surgical therapies.
For example, when quantifying the effect of coronary artery bypass grafting (CABG) it would be unethical to
perform a placebo-CABG. Placebo-CABG would involve the same procedure as the index-CABG, including
undergoing general anesthesia plus sternotomy, but
without the bypassing which is in fact the effective
(pharmacologic) substance of the CABG treatment.
Treatments or diseases that do not allow for a placebo
comparison require comparison with an existing treatment. Existing treatment also could be no treatment
(wait and see).
Besides ethical constraints, using an existing treatment as a control might also be favored to better reflect

115

Orthopaedic Knowledge Update 8

Clinical Epidemiology: An Introduction

clinical practice. Using an existing treatment as control
implies that all patients are aware of the received treatment; the patients are not blinded. The influence of patient perception, lifestyle changes, and treatment factors
other than the effective part, may thus be different
across the new (index) and existing (control) treatment
groups. However, all of these factors are considered as
part of the total effect of each treatment strategy on the
outcome. It merely reflects the real life situation of both
treatments. Any difference in outcome between the two
treatment groups will result from the overall difference
between the two treatment strategies. It will be impossible to unravel which factors are responsible for the observed difference. Therapeutic studies using another
(nonplacebo) treatment as control are called pragmatic
studies. Placebo-controlled therapeutic studies are
called explanatory studies, as any observed difference
can be attributed to or explained by the difference in
effective substance between the two groups.
In the example, if a pragmatic trial would be desired,
the determinant could include arthroscopy dbridement
versus no treatment (or wait and see). If a placebocontrolled study would be desired, the control group
would include placebo arthroscopy, that is, without
dbridement. Such a study has recently been conducted.
Outcome
In the example study, the outcome is pain measured, for
example, 1 and 2 years after surgery using a visual analog or verbal rating scale. As with prognostic research,
patient-driven rather than intermediate outcomes
should be chosen. Furthermore, as in all other types of
research, the person who measures the outcome should
be blinded to the treatment of the patient to prevent information bias, whether or not a pragmatic or placebocontrolled therapeutic study is performed.
Data Analysis and Confounding
As mentioned earlier, one of the most important aspects
of causal research is to account for other risk factors of
the outcome (such as confounders). Randomization is
the best method to ensure that the compared groups are
comparable on all risk factors except for the determinant under study. In therapeutic research, the benefits
of the determinant (treatment) are supposed to outweigh its risks, making randomization feasible and ethical. Nevertheless, despite adequate randomization, an
imbalance of confounders between the two treatment
groups could still occur by chance, particularly when the
number of patients is small. Therefore, in the analysis,
investigators should first check other risk factors of the
outcome to determine if they are indeed equally distributed. If not, adjustment in the analysis may sometimes
be needed to account for confounding.

116

Domain
The domain to which the results of the example study
can be generalized are patients who have osteoarthritis
and an indication for arthroscopy.
Therapeutic research is causal research on the intended effects of treatments. The treatment under study
may be compared with existing treatment including no
treatment, or to a placebo treatment, depending on the
research question. A pragmatic study approach will
quantify the difference in effect between two treatment
strategies. A placebo-controlled study quantifies the efficacy of the effective (pharmacologic) substance of the
treatment under study. Blinding the patient is only at issue in placebo-controlled studies. In pragmatic and
placebo-controlled studies, the treatments always should
be allocated randomly to prevent confounding bias. If
patient numbers are low, adjustment for confounding in
the analysis may sometimes be necessary. Blinding the
observer to the outcome is always desired except when
the outcome under study is not sensitive to observer interpretation.

Summary
Clinical epidemiology attempts to provide quantitative
answers to relevant questions to improve future medical
care. Such questions arise from the patient-physician encounters and require either diagnostic, etiologic, prognostic, or therapeutic knowledge. Accordingly, the first
step when designing a clinical epidemiologic study is to
determine which of these four types of knowledge is addressed. Irrespective of the type of knowledge addressed, definition of the occurrence relation, outcome,
and determinant applies to each type of study question.
With etiologic or therapeutic research, the potential
confounders of the determinant-outcome relationship
should also be defined beforehand. It is also useful to
initially define the domain of the studied occurrence relation because this process helps to select the suitable
study population from that domain. Next, the determinant(s), confounders, and outcome should be measured
from each study subject. Finally, the data should be analyzed, taking into account whether the goal is to predict
the outcome (diagnostic or prognostic research) or to
explain what actually causes the occurrence of the outcome (etiologic or therapeutic research).

Annotated Bibliography
General Reference
Greenhalgh T: How to Read a Paper: The Basics of Evidence Based Medicine, ed 2. London, England, BMJ
Publishing Group, 2001.
This book provides a general introduction on the essentials of evidence-based medicine and guidelines for reading
clinical epidemiologic studies.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Laupacis A: The future of evidence-based medicine. Can
J Clin Pharmacol 2001;8(suppl A):6A-9A.
This article presents a state of the art review of evidencebased medicine and the challenges for the near future.

Clinical Epidemiologic Research Questions

Sackett DL: Clinical epidemiology: What, who, and
whither. J Clin Epidemiol 2002;55:1161-1166.
A general introduction to the history of clinical epidemiology is presented.

Basic Principles of Clinical Epidemiologic Studies

Rothman KJ: Epidemiology: An Introduction. Oxford,
England, Oxford University Press, 2002.
This book, which focuses on etiologic research, covers the
field of epidemiology including history, study questions, and
data analysis. A background in epidemiology is required.

Characteristics of Diagnostic Studies

Chapter 10 Clinical Epidemiology: An Introduction

Characteristics of Etiologic Studies

Hak E, Verheij TJ, Grobbee DE, Nichol KL, Hoes AW:
Confounding by indication in non-experimental evaluation of vaccine effectiveness: The example of prevention
of influenza complications. J Epidemiol Community
Health 2002;56:951-955.
This article illustrates the problems of using data from a
nonrandomized (observational) study to quantify the effectiveness of a therapy.

Kennedy KA, Frankowski RF: Evaluating the evidence

about therapies: What the clinician needs to know about
statistics. Clin Perinatol 2003;30:205-215.
The basics of data analysis of therapeutic studies is addressed in this article.

Lorenz JM, Paneth N: When are observational studies

adequate to assess the efficacy of therapeutic interventions? Clin Perinatol 2003;30:269-283.

Bossuyt PM, Reitsma JB, Bruns DE, et al: Towards complete and accurate reporting of studies of diagnostic accuracy: The STARD initiative: Standards for Reporting
of Diagnostic Accuracy. Clin Chem 2003;49:1-6.

This article presents information on when and how nonrandomized (observational) study data can be useful to quantify the effectiveness of therapies.

This article addresses a list of criteria that will enhance the

reporting and conduction of diagnostic accuracy studies.

Characteristics of Prognostic Studies

Knottnerus JA: The Evidence Base of Clinical Diagnosis. London, England, BMJ Publishing Group, 2002.

This article enhances the understanding of studies that attempt to quantify the prognostic or predictive value of factors
such as patient characteristics, etiologic factors, and diagnostic
tests and results.

This book contains a series of British Medical Journal articles on the essentials of diagnostic studies, from study questions to analysis.

Moons KG, Grobbee DE: Diagnostic studies as multivariable, prediction research. J Epidemiol Com Health
2002;56:337-338.
A brief report discussing the need for studies that attempt
to quantify the added value of new tests beyond existing tests
(rather than estimating a tests sensitivity and specificity) is
presented.

Moons KG, Harrell FE: Sensitivity and specificity

should be deemphasized in diagnostic accuracy studies.
Acad Radiol 2003;10:670-672.
This editorial on the Standards for Reporting of Diagnostic Accuracy initiative discusses why estimating a tests sensitivity has limited value for clinical practice.

Oostenbrink R, Moons KG, Bleeker SE, Moll HA,

Grobbee DE: Diagnostic research on routine care data:
Prospects and problems. J Clin Epidemiol 2003;56:501506.
This article addresses the pros and cons of using routine
care data for the evaluation of diagnostic tests.

American Academy of Orthopaedic Surgeons

Concato J: Challenges in prognostic analysis. Cancer

2001;91(suppl 8):1607-1614.

van Klei WA: Moons KG, Leyssius AT, Knape JT, Rutten CL, Grobbee DE: Reduction in type and screen:
Preoperative prediction of RBC transfusions in surgery
procedures with intermediate transfusion risks. Br J
Anaesth 2001;87:250-257.
This article presents an example study that developed an
easy applicable prognostic rule for surgeons or anesthesiologists to preoperatively predict which surgical patients would
undergo perioperative blood transfusions.

van Klei WA, Moons KG, Rheineck-Leyssius AT, et al:

Validation of a clinical prediction rule to reduce preoperative type and screen procedures. Br J Anaesth 2002;
89:221-225.
This article presents an example study that validated a developed prognostic rule to preoperatively predict which new
surgical patients would undergo perioperative blood transfusions.

Vergouwe Y, Steyerberg EW, Eijkemans MJ, Habbema

JD: Validity of prognostic models: When is a model clinically useful? Semin Urol Oncol 2002;20:96-107.
This article discusses an appealing method to determine
whether a prognostic prediction rule is clinically useful.

117

Clinical Epidemiology: An Introduction

Orthopaedic Knowledge Update 8

Altman DG, Schulz KF, Moher D, et al: The revised

CONSORT statement for reporting randomized trials:
Explanation and elaboration. Ann Intern Med 2001;134:
663-694.

Guyatt GH, Sackett DL, Cook DJ: Users guides to the

medical literature: II. How to use an article about therapy or prevention: B. What were the results and will
they help me in caring for my patients?: EvidenceBased Medicine Working Group. JAMA 1994;271:59-63.

This article addresses a list of criteria that will enhance the

reporting and conduction of randomized therapeutic studies.

Hennekens CE, Buring JE: Epidemiology in Medicine.

Boston, Massachusetts, Little, Brown & Co, 1987.

Helms PJ: Real world pragmatic clinical trials: What are

they and what do they tell us? Pediatr Allergy Immunol
2002;13:4-9.

Jaeschke R, Guyatt GH, Sackett DL: Users guides to

the medical literature: III. How to use an article about a
diagnostic test: A. Are the results of the study valid?:
Evidence-Based MedicineWorking Group. JAMA 1994;
271:389-391.

Characteristics of Therapeutic Studies

This article presents a clear explanation of the differences

between explanatory (such as placebo-controlled) randomized
therapeutic studies and pragmatic randomized therapeutic
studies.

Moher D, Schulz KF, Altman D: The CONSORT statement: Revised recommendations for improving the
quality of reports of parallel-group randomized trials.
JAMA 2001;285:1987-1991.
This article addresses a list of criteria to enhance the reporting and conduction of randomized therapeutic studies.

Moseley JB, OMalley K, Petersen NJ, et al: A controlled trial of arthroscopic surgery for osteoarthritis of
the knee. N Engl J Med 2002;347:81-88.
A good example of an explanatory (such as placebocontrolled) therapeutic study of a surgical intervention is presented.

Classic Bibliography
Apfel CC, Laara E, Koivuranta M, Greim CA, Roewer
N: A simplified risk score for predicting postoperative
nausea and vomiting: Conclusions from cross-validation
between two centers. Anesthesiology 1999;91:693-700.
Charlton BG: Understanding randomized controlled trials: Explanatory or pragmatic? Fam Pract 1994;11:243244.
Evidence-Based Medicine Working Group: Evidencebased medicine: A new approach to teaching the practice of medicine. JAMA 1992;268:2420-2425.
Grobbee DE, Miettinen OS: Clinical Epidemiology: Introduction to the discipline. Neth J Med 1995;47:2-5.
Guyatt GH, Sackett DL, Cook DJ: Users guides to the
medical literature: II. How to use an article about therapy or prevention: A. Are the results of the study valid?:
Evidence-Based Medicine Working Group. JAMA 1993;
270:2598-2601.

118

Laupacis A, Sekar N, Stiell IG: Clinical prediction rules:

A review and suggested modifications of methodological standards. JAMA 1997;277:488-494.
Laupacis A, Wells G, Richardson WS, Tugwell P: Users
guides to the medical literature: V. How to use an article
about prognosis: Evidence-Based Medicine Working
Group. JAMA 1994;272:234-237.
Levine M, Walter S, Lee H, Haines T, Holbrook A,
Moyer V: Users guides to the medical literature: IV.
How to use an article about harm: Evidence-Based
Medicine Working Group. JAMA 1994;271:1615-1619.
McAlister FA, Graham I, Karr GW, Laupacis A:
Evidence-based medicine and the practicing clinician.
J Gen Intern Med 1999;14:236-242.
Moons KG, van Es GA, Michel BC, Buller HR,
Habbema JD, Grobbee DE: Redundancy of single diagnostic test evaluation. Epidemiology 1999;10:276-281.
Randolph AG, Guyatt GH, Calvin JE, Doig G, Richardson WS: Understanding articles describing clinical prediction tools. Crit Care Med 1998;26:1603-1612.
Reichenbach H: The Rise of Scientific Philosophy. New
York, NY, Harper Row, 1965.
Roland M, Torgerson DJ: What are pragmatic trials?
BMJ 1998;316:285.
Sackett DL, Haynes RB, Tugwell P: Clinical Epidemiology: A Basic Science for Clinical Medicine. Boston, Massachusetts, Little, Brown & Co, 1985.
Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS: Evidence based medicine: What it is and
what it isnt. BMJ 1996;312:71-72.
Tinetti ME, Liu WL, Claus EB: Predictors and prognosis
of inability to get up after falls among elderly persons.
JAMA 1993;269:65-70.

American Academy of Orthopaedic Surgeons

Chapter

11

Musculoskeletal Imaging
John A. Carrino, MD, MPH
William B. Morrison, MD

Available Imaging Modalities

All medical imaging modalities that are available clinically have musculoskeletal applications. There have
been significant technologic advancements in the areas
of radiography, CT, ultrasound, nuclear medicine, and
MRI.

Radiography
Digital radiography exists in the form of computed radiography or direct radiography. Image processing and distribution is achieved through a picture archiving and communication system. The images can be placed on a
compact disk with an imbedded viewer. The widespread
availability of computers with compact disk readers allows
this method of image processing to be a more portable
mechanism of transporting and managing images. Viewing the images in the soft copy environment allows for
panning, zooming, windowing, and leveling so that the
viewing experience and diagnostic yield are optimal.
There are certain tradeoffs; the spatial resolution of digital radiography systems is not as great as that with film
screen radiography. However, as additional experience
and data are accumulated, it has been found that the improved contrast resolution is more important than spatial
resolution for diagnostic efficacy, making less defined spatial resolution a reasonable trade-off (Figure 1).

Computed Tomography
The latest generation of CT scanners uses multiple detector row arrays. Multislice CT represents a major improvement in helical CT technology, wherein simultaneous activation of multiple detector rows positioned
along the longitudinal or z-axis (direction of table or
gantry) allows acquisition of interweaving helical sections. The principal difference between multislice CT
and predecessor generations of CT is the improved resolution in the z-axis. More of the photons generated by
the x-ray tube are ultimately used to produce imaging
data. With this design, section thickness is determined
by detector size and not by the collimator itself. Rapid
data acquisition times are possible because of short gan-

American Academy of Orthopaedic Surgeons

try rotation intervals combined with multiple detectors

providing increased coverage along the z-axis. Currently
there are configurations with up to 64 channel detectors
available. The data from a multislice CT scanner can be
used to generate images of different thickness from the
same acquisition. The minimum section thickness is reduced to approximately 0.5 mm and images can be reconstructed at this 0.5-mm interval. Isotropic (equal dimension) voxels measuring 0.5 mm in x, y, and z
directions greatly improve the spatial resolution and the
quality of reconstructing algorithms, allowing generation
of exquisite multiplanar reformats (Figure 1) and threedimensional images. These three-dimensional image
processing and display techniques are particularly useful
for regions of complex anatomy such as the pelvis, or
for bones for which it is difficult to obtain isolated projections without overlap, such as the scapula or the cervical spine (Figure 2). Multiplanar reformats can be created in any plane or using a curved surface to reduce
distortion. Other advantages include increased speed
and increased total volume coverage. Therefore, a single
pass whole body protocol may be feasible, particularly
with 16-detector scanners that can image from the head
vertex to below the hips in less than 1 minute. The ability to acquire high-quality images in the presence of
hardware/joint implants has improved using multislice
CT. Metal artifacts are caused by photopenic defects in
the back projection and are displayed on CT images as
streak artifact. With multislice CT, the holes in the filtered back projection are not as pronounced, resulting
in a less severe streak artifact. This improvement is at
the expense of excess tissue radiation along the penumbra of the beam, which is then picked up by adjacent
detector channels filling in these photopenic defects in
the projection. This technology has forced radiologists
into redefining the image viewing process to a volumetric paradigm rather than a simple tile mode or sectionby-section viewing. In addition, CT protocols have to be
reformulated. There is also an expanded range of CT
applications and indications. The challenges associated
with multislice CT include selection of optimal imaging
sequences, controlling radiation exposure to the patient,

119

Musculoskeletal Imaging

Orthopaedic Knowledge Update 8

Figure 1 Digital radiography and multislice CT multiplanar reformat image quality. A, Anteroposterior projection of the
knee performed with a direct radiography
unit shows high-quality contrast. B, Multidetector CT coronal multiplanar reformat
image shows exquisite trabecular detail
similar to conventional radiography.
Source images were acquired using isotropic 0.75-mm voxels.

Figure 2 Position compensated multiplanar reformat.

Cervical spine multislice CT myelography in a patient
who underwent anterior cervical fusion: Sagittal and
coronal multiplanar reformat images are created from
an axial image (A) while adjusting the planes to correct for any rotation (double oblique prescriptions).
Similar maneuvers are done for the sagittal (B) and
coronal (C) planes to generate more useful sections
based on true anatomic landmarks. Oblique axial images (D) are also used to generate sagittal oblique images (E) through the neural foramen at each level on
each side to compensate for changes in orientation of
the neural canal. Coronal and sagittal oblique multiplanar reformat images show an osseous ridge compressing the nerve roots (arrow in C and E).

and efficiently managing the large amount of data generated. Some disadvantages of multislice CT are high
radiation dose to the tissue and potentially noisy images. Noise is inversely related to the number of pho-

120

tons per voxels; because smaller voxels tend to have

fewer photons, increased image noise is the result. To
keep the noise level reasonable, the exposure (and thus
radiation dose) must be increased.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Ultrasound
Ultrasound is the medical imaging modality used to acquire and display the acoustic properties of tissues. A
transducer array (transmitter and receiver of ultrasound
pulses) sends sound waves into the patient and receives
returning echoes that are converted into an image. Sound
is mechanical energy that propagates through continuous,
elastic medium by the compression and rarefaction of the
particles that compose it (such as air). The resolution of
the ultrasound image and attenuation of the ultrasound
beam depend on the wavelength and frequency.A low frequency ultrasound beam has a longer wavelength and less
resolution but greater depth of penetration. For musculoskeletal imaging of more superficial structures such as
tendons and ligaments, a high frequency beam having a
smaller wavelength provides superior spatial resolution
and image detail. Thus, the creation of appropriate transducers is of critical importance in performing musculoskeletal imaging. Higher frequency transducers ranging
from 7.5 MHz up to even 15 MHz are now available. Modern ultrasound scanners use phased array transducers with
multiple piezoelectric elements to electronically sweep an
ultrasound beam across the volume of interest, thus being
able to create a three-dimensional image.
Applications of ultrasound include evaluation of tendon and muscle abnormalities such as rotator cuff tears.
In addition, ultrasound has also been applied to evaluate
the glenoid labrum and knee menisci. High-resolution ultrasound has the potential to be used for visualization of
articular cartilage. Synovial effusions and proliferation
can be evaluated using color Doppler imaging to determine the degree of hypertrophy and inflammation. Ultrasound has been used more often for diagnostic and therapeutic procedures because of improved transducer
technology being better able to detect infrastructural detail. In addition, ultrasound is a more economical modality for assessing a specific clinical concern. Ultrasound is
best used when the clinical question is specific and well
formulated; the condition is dichotomous (is there a full
thickness tear or not?); for characterizing a soft-tissue lesion (cystic or vascular); or for guiding particular interventions. Performing percutaneous interventions with ultrasound ensures accurate needle tip placement and helps
direct the needle away from other regional soft-tissue
structures and neurovascular bundles.The visualization of
the needle tip in real time allows for reliable placement in
the tendon sheath, bursa, or joint of interest. Intratendinous calcifications, the plantar fascia, and interdigital
(Mortons) neuromas can also be visualized and injected
directly under real-time guidance.

Nuclear Medicine and Positron Emission Tomography

The most significant recent advancement in nuclear
medicine is positron emission tomography (PET) and
the combination of PET CT scanners with important

American Academy of Orthopaedic Surgeons

Chapter 11 Musculoskeletal Imaging

implications for oncology, especially for soft-tissue neoplasms or postoperative cancer patients. PET has only
recently been widely used in clinical oncology. Shortlived positron emitting radioisotopes annihilate to form
two photons with trajectories approximately 180 apart
at a particular energy level (511 kev). The coincidental
detection of these photons by a ring detector is reconstructed via a filtered backprojection (similar to CT) to
form images of tracer distribution. [18F] 2-deoxy-2fluoro-D-glucose (FDG) is a metabolic tracer most
widely used in clinical PET oncology. FDG accumulation reflects the rate of glucose utilization in a tissue because FDG is transported into a tissue by the same
mechanisms of glucose transport and trapped in a tissue
as FDG-6-phosphate, which is a poor substrate for the
further enzyme systems of glycolysis or glycogen storage. The use of FDG in evaluation of the musculoskeletal system is based on increased glycolytic rate in pathologic tissues. Thus, PET has proved to be the gold
standard in metabolic imaging. FDG provides a means
of quantitating the glucose metabolism, with the amount
of tracer accumulation reflecting the glucose metabolism. High-grade malignancies tend to have higher rates
of glycolysis than do low-grade malignancies and benign
lesions; therefore, high-grade malignancies have greater
uptake of FDG than that of low-grade or benign lesions.
PET, as a metabolic imaging technique, has advantages over and complements structural imaging methods
and also shows differences from conventional nuclear
medicine. All of these factors have led to its growth in
clinical applications in recent years. PET applications
are evolving, but the technique has been approved for
the diagnosis, staging, and restaging of many common
malignancies and has shown efficacy for the detection of
osseous metastasis from several malignancies, including
lung and breast carcinomas and lymphoma. However,
the significance of FDG PET in the evaluation of primary bone tumors and tumor-like lesions has not been
extensively elucidated. Several investigators have reported the usefulness of FDG-PET in oncologic applications for primary musculoskeletal tumors. Preliminary
reports suggest a good correlation between glucose consumption measured by FDG PET and the aggressiveness of musculoskeletal tumors. However, a significant
overlap exists between benign and malignant groups;
therefore, PET is not a solo method for differential diagnosis of benign and malignant bone lesions. Both neoplastic and inflammatory processes can cause an increase in FDG activity; therefore, tissue sampling should
be directed to the most metabolically active regions
identified by PET (Figure 3). Therefore, PET has a role
as a useful adjunct to anatomic imaging techniques because it can provide an in vivo method for quantifying
functional metabolism in normal and diseased tissues.

121

Musculoskeletal Imaging

Orthopaedic Knowledge Update 8

Figure 3 PET CT coregistration image fusion.

The imaging studies are from a patient who
was referred for evaluation for metastatic
bone disease. Composite image: Coronal and
sagittal maximum intensity projection FDG-PET
images (A and B) show an area with abnormal
FDG uptake in the pelvis that is not localized
with certainty. Coronal, sagittal, and transaxial
PET-CT fused images (C, D, and E) reveal a hypermetabolic focus corresponding to an osteolytic lesion, which was confirmed histopathologically by percutaneous biopsy to be
adenocarcinoma.

Magnetic Resonance Imaging

Recent MRI technology developments include high
field strength clinical systems for the entire body (3.0
Tesla) and dedicated extremity scanners operating at 1.0
Tesla, with improved gradients and radiofrequency (RF)
coils. Parallel imaging and faster novel pulse sequences
are addressing the time factor involved in obtaining an
MRI scan. High spatial resolution imaging with improved signal-to-noise ratio has generated a new area of
imaging known as magnetic resonance microscopy, used
for the smaller and more complex articulations such as
the hand, wrist, foot, ankle, and elbow. Expanding the
types of three-dimensional pulse sequences available using fast spin-echo techniques afford the opportunity for
isotropic voxel acquisition and better volumetric data
rendering. Examination time can be reduced by using a
single acquisition and reconstructing the plane of interest (similar to multislice CT). Currently the most useful
sequences are two-dimensional. Because musculoskeletal imaging often relies on the depiction of particular
anatomy, several planes are acquired using the same
pulse sequence depending on the articulation, indication, and specific structure of interest.
Imaging of cartilage has been difficult; however,
some new advances in MRI techniques, such as the use
of biologically relevant elements such as hydrogen,

122

oxygen-16, oxygen-17, fluorine-19, sodium-23, and

phosphorus-31, will help in providing and interpreting
these images. Most clinical MRI is hydrogen imaging;
more recently, sodium and phosphorus images have
been obtained in vivo. Sodium imaging can be used for
estimating the glycosaminoglycan content of articular
cartilage. Because of the low natural abundance of sodium in articular tissues, in vivo imaging is limited to
high-field magnets (3.0 Tesla or greater). There are several pulse sequences that have been applied to cartilage
imaging. Gradient-recall echo imaging has been used
because of its three-dimensional capability and ability to
provide high-resolution images. Fat-suppressed techniques are combined with gradient-recall echo imaging
to increase the dynamic range and improve the detection of signal abnormalities in cartilage. However, scan
acquisition time can be substantial (8 to 10 minutes),
thus making the image prone to motion artifact. Magnetization transfer contrast creates tissue contrast by the
exchange of magnetization between protons associated
with macromolecules and bulk water protons by means
of cross relaxation or chemical exchange. Magnetization
transfer contrast acts as an off-resonant saturation pulse
and may make cartilage lesions easier to see. Although
this type of contrast is usually used in conjunction with
the gradient echo techniques, some magnetization trans-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

fer occurs inherently with fast spin-echo or turbo spinecho techniques. The driven equilibrium Fourier transform contrast method is another promising approach for
imaging of cartilage disorders. This method produces
image contrast that is a function of proton density,
T1/T2 and echo time/repetition time (TE/TR). Many tissues have competing T1 and T2 contrast, so the T1 to T2
ratio for driven equilibrium Fourier transform contrast
tends to be synergistic. Driven equilibrium Fourier
transform contrast is well suited to imaging articular
cartilage because synovial fluid is high and articular cartilage is intermediate in signal intensity; the osseous
structures are dark and lipids can be suppressed using
spectral fat saturation techniques. This creates an image
in which cartilage is easily distinguishable from all adjacent tissues based on signal intensity alone, aiding segmentation and volume calculation.
Magnetic resonance arthrography has also been applied to the evaluation of cartilage and may be performed directly (a dilute gadolinium-containing solution
is percutaneously placed into the joint via a needle) or
indirectly (intravenous gadolinium is administered and
allowed to diffuse into the joint). Gadolinium-enhanced
imaging has the potential to monitor glycosaminoglycan
(GAG) content within the cartilage. GAGs are fundamentally important for several reasons: they play a major mechanical support role, they are lost early in the
course of cartilage degeneration, and may need to be replenished during the course of any effective cartilage
treatment regimen. GAGs contribute strong negative
charge to the cartilage matrix and mobile ions will distribute to reflect local GAG concentration. When
GAGs are lost as part of the degenerative process, the
negative charge conferred to the cartilage also is lost.
Consequently, when a negatively charged contrast agent
is administered (either intravenously or intraarticularly) it preferentially distributes into the degraded cartilage and the T1 effect of this contrast material can be qualitatively visualized and quantitatively
measured. With direct magnetic resonance arthrography,
it takes 3 hours for the contrast to sufficiently diffuse
into the cartilage, whereas with indirect magnetic resonance arthrography it only takes approximately 1.5
hours.
Clinical higher field MRI systems, typically 3.0 Tesla
(or 3T) are becoming widely available. The higher intrinsic signal-to-noise ratio of high-field strength MRI
can be used to improve imaging speed or resolution but
changes in relaxation time at 3T as well as increased artifacts must be considered. Nevertheless, 3T MRI offers
the opportunity to explore physiologic imaging of joints
as well as morphologic features. Tissues in a magnetic
field are categorized by two types of relaxation: T1
(spin-lattice relaxation time) and T2 (spin-spin relaxation time). Manipulation of imaging parameters of the
TE and the TR produce the various types of contrast

American Academy of Orthopaedic Surgeons

Chapter 11 Musculoskeletal Imaging

seen in MRI. T2 is fairly constant at different field
strengths, but T1 increases with increasing field strength.
Fat suppression is improved at 3T because the peaks between water and fat are further separated (greater
chemical shift effect; the nature of the chemical shift
phenomenon is that protons associated with long chain
aliphatic fat molecules have precessional frequencies
slightly lower than protons associated with mobile water
molecules, which is field strength-dependent), providing
more robust spectral fat suppression. The US Food and
Drug Administration and manufacturers have mandated
the use of power monitoring systems for 3T because of
increased danger of RF burns. The more problematic sequences are fast spin-echo/turbo spin-echo (unless the
refocusing pulse is reduced to less than 180) and short
TR spin-echo sequences (T1-weighted). Motion artifacts
appear worse at 3T and may be because of the overall
higher signal to noise ratio. Susceptibility artifacts will
be increased, and postoperative imaging may be more
problematic at higher field strengths. Increasing the receiver bandwidth may also help alleviate these artifacts.
There are improvements in speed because the signal to
noise ratio is proportional to the square of the imaging
time, and therefore it is possible to go up to four times
faster at 3T than 1.5T while maintaining an equal signal
to noise ratio.
Another trend is the use of dedicated extremity
scanners that typically operate at low field to midfield
ranges. Low field open scanners or extremity scanners
have been available for several years. However, recently
a higher field extremity scanner at 1.0 T has become
available (Orthone, ONI, Inc, Wilmington, MA). This
unit can image the elbow, hand, wrist, knee, foot, and ankle but cannot acquire images of the shoulder, hip, or
spine. Therefore, it is not considered a stand-alone unit
but may be useful for a high volume site with a backlog
of musculoskeletal patients to supplement a whole body
unit. The site requirements are less than those for a conventional high field strength system, making it a more
economical option. The image quality is much improved
over the typical low field open MRI or very low field
extremity scanners, and this type of device will likely
have a niche role providing high-quality clinical images
for the amenable body parts (Figure 4). It is also a useful option for claustrophobic patients. There are Larger
and smaller diameter RF coils available. Both coils are
quadrature types of volume transmit-receive coils, which
contribute to the improved image quality. There are certain advantages when this method is compared with traditional high field closed MRI systems. The images are
of high quality despite the fact that the field strength is
lower. Loss of signal to noise ratio and artifacts from off
isocenter imaging are minimized. The majority of pulse
sequences are available, including spectral fat suppression.

123

Musculoskeletal Imaging

Orthopaedic Knowledge Update 8

Figure 4 High-field open dedicated extremity MRI system. Elbow common extensor tendinopathy and partial tear: axial T1-weighted image (A) shows intermediate signal at the
common extensor tendon ulnar attachment (arrowhead), coronal oblique T2-weighted image (B) with spectral fat suppression shows a small fluid gap (arrow) but not complete
discontinuity of the common extensor tendon ulnar attachment reflecting a partial tear. Achilles tendinopathy (insertional tendinitis): axial T1-weighted image (C) through distal
Achilles tendon (subjacent to marker) shows tendinosis manifested by enlargement (loss of the normal comma shape) and intermediate signal, sagittal T2-weighted image with
spectral fat suppression. (Images courtesy of Barbara N. Weissman, MD and Rosemary J. Klecker, MD, Harvard Medical School, Brigham and Womens Hospital.)
Parallel imaging is a relatively new class of techniques capable of significantly increasing the speed of
MRI acquisitions. Although a variety of different techniques have emerged, the common principle is to use
the spatial information inherent in the elements of an
RF coil array to allow a reduction in the number of
time-consuming phase-encode steps required during the
scan. Recent technical advances and increased availability to imaging centers place parallel imaging on the
verge of widespread clinical use.

Imaging of Specific Orthopaedic Conditions

Spine
Disk Disease Nomenclature
Spine imaging can exquisitely provide information regarding pathoanatomy of degenerative disk disease but
does not define a specific painful clinical syndrome because of the nonspecific appearance of painful versus
painless degenerative conditions. However, abnormal
imaging findings of the lumbar disks may be degenerative, adaptive, genetic, or a combination of environmental and determined factors. Many findings may simply
represent senescent changes that are the natural consequences of stress applied during the course of a lifetime.
The imaging appearance of lumbar spine degenerative
disk disease has a similar incidence in symptomatic and
asymptomatic patients. Therefore, the appropriate use of
imaging modalities within a defined clinical context is
paramount. For some patients with complicated or recalcitrant symptoms, the most useful aspect for advanced imaging techniques may be in the exclusion of
more serious causes of axial low back pain such as infection, neoplasm, or fracture rather than the inclusion of
any specific degenerative findings.

124

In terms of improving communications between providers there is a multispecialty, multisociety-endorsed nomenclature for the lumbar spine disk disease (some advocate that this system may be used for cervical and
thoracic spine descriptors also). It is important to recognize that the definitions of diagnoses should not define or
imply external etiologic events such as trauma, should not
imply relationship to symptoms and do imply need for
specific treatment. Hence, the following are pathoanatomic descriptors that do not imply a specific pathoetiology or syndrome. Osteoarthritis or osteoarthrosis is a process of synovial joints. Therefore, in the spine it is
appropriately applied to the zygoapophyseal (Z-joint,
facet), atlantoaxial, costovertebral, and sacroiliac joints.
Degenerative disk disease is a term applied specifically to
intervertebral disk degeneration. The term spondylosis is
often used in general as synonymous with degeneration,
which would include both nucleus pulposus and anulus fibrosus processes, but such usage is confusing, so it is best
that degeneration be the general term and spondylosis
deformans a specifically defined subclassification of degeneration characterized by marginal osteophytosis without substantial disk height loss (reflecting predominantly
anulus fibrosis disease). Intervertebral osteochondrosis is
the term applied to the condition of mainly nucleus pulposus and the vertebral body end-plates disease including
annular fissuring (tearing).
Normally, the posterior disk margin tends to be concave in the upper lumbosacral spine (Figure 5, A), and is
straight or slightly convex at L4-5 and L5-S1. The posterior margin typically projects no more than 1 mm beyond the end plate. An annular bulge is described as a
generalized displacement (greater than 180) of disk
margin beyond the normal margin of the intervertebral

American Academy of Orthopaedic Surgeons

Chapter 11 Musculoskeletal Imaging

Orthopaedic Knowledge Update 8

Figure 5 Lumbar disk contour abnormalities; all are axial T2-weighted images at the level of the intervertebral
disk. A, Normal: the posterior disk margin (arrowhead) should have a slight concavity, with the exception of the
lumbosacral junction, which may have a slight convexity. B, Annular bulge: There is generalized displacement
(arrowheads) of greater than 180 of the disk margin beyond the normal margin of the intervertebral disk space
and is the result of disk degeneration with an intact anulus fibrosus. C, Disk protrusion: The base against the
parent disk margin is broader than any other diameter of the herniation. Extension of nucleus pulposus through a
partial defect in the anulus fibrosus is identified (arrow) but the herniated disk is contained by some intact
annular fibers. D, Disk extrusion: The base against the parent disk margin is narrower than any other diameter of
the herniation (arrowhead). There may be extension of the nucleus pulposus through a complete focal defect in
the anulus fibrosus. Substantial mass effect is present, causing moderate central canal and severe left lateral
recess stenosis.
disk (Figure 5, B). The normal margin is defined by the
vertebral body ring apophysis exclusive of osteophytes.
The annular bulge can be the result of disk degeneration with a grossly intact anulus. Disk margins tend to
be smooth, symmetric, or eccentric and nonfocal, and
may have a level-specific appearance in the lumbar
spine. Disk herniation is a localized displacement (less
than 180 of the circumference) of disk material beyond
the normal margin of the intervertebral disk space (Figure 5, C). This material may consist of nucleus pulposus,
cartilage, fragmented apophyseal bone, or fragmented
annular tissue. It is often the result of disk degeneration
with some degree of focal annular disruption. The types
of disk herniation are designated as protrusion, extrusion, and free fragment (sequestration). Protrusion refers to a herniated disk in which the greatest distance in
any plane between the edges of the disk material beyond the disk space is less than the distance between
the edges of the base in the same plane. It is characterized by the following: the base against the parent disk
margin is broader than any other diameter of the herniation; extension of nucleus pulposus may occur through
a partial defect in the anulus fibrosus but is contained
by some intact outer annular fibers and the posterior
longitudinal ligament. The types of protrusions may be
broad based (90 to 180 circumference) or focal (less

American Academy of Orthopaedic Surgeons

than 90 circumference). Extrusion refers to a herniated

disk in which, in at least one plane, any one distance between the edges of the disk material beyond the disk
space is greater than the distance between the edges of
the base in the same plane (Figure 5, D), or when no
continuity exists between the disk material beyond the
disk space and that within the disk space. An extrusion
is characterized by the following: the base against the
parent disk margin tends to be narrower than any other
diameter of the herniation; extension of the nucleus pulposus through a complete focal defect in the anulus fibrosus. Extruded disks in which all continuity with the
disk of origin is lost may be further characterized as sequestrated. Disk material displaced away from the site
of extrusion may be characterized as migrated; it may
stay subligamentous, contained by the posterior longitudinal ligament or may migrate widely. Schmorls nodes
are intervertebral disk herniations (transosseous disk
extrusion). Herniation of the nucleus pulposus occurs
through the cartilaginous end plate into the vertebral
body. These herniations often have a characteristic
round or lobulated appearance. They may enhance after
contrast administration with ring-like enhancement being most common. They are often incidental and likely
to be developmental or posttraumatic rather than
purely degenerative or adaptive. There is now imaging

125

Musculoskeletal Imaging
evidence of a significant genetic association between the
COL9A3 tryptophan allele (Trp3 allele), Scheuermanns
disease, and intervertebral disk degeneration among
symptomatic patients.
There are no formal staging systems for lumbar degenerative disk disease and most physicians will commonly
report findings using the designations of mild, moderate,
and severe disease. However, these designations will hold
different meaning among physicians, especially with respect to degree of disk degeneration. The following
scheme is used to define the degree of canal compromise
produced by disk displacement based on the goals of being practical, objective, reasonably precise, and clinically
relevant. Measurements are typically taken from an axial
section at the site of the most severe compromise. Canal
compromise of less than one third of the canal at that section is mild, between one and two thirds is moderate, and
over two thirds represents severe disease. This scheme
may also be applied to foraminal (neural canal) narrowing with the sagittal images also playing a useful role for
defining the degree of narrowing. Observer interpretations are also made with various degrees of confidence.
Statement of the degree of confidence is an important
component of communication. The interpretation should
be characterized as definite if there is no doubt, probable if there is some doubt but the likelihood is greater
than 50%, and possible if there is reason to consider but
the likelihood is less than 50%.
Positional, Load-Bearing, and Dynamic
(Functional) Imaging
Because imaging in the supine position may not fully reveal the anatomic lesions, there has been an interest in
performing functioning, positional, or load-bearing imaging of the spine. Spine imaging position options available are supine, supine with axial loading (simulated
weight bearing), seated, or standing upright position.
Noncompressive lesions on conventional MRI may
show encroachment and neural element impingement
on dynamic load-bearing (seated) scans. Fluctuating positional foraminal and central spinal canal stenosis has
also been shown in the cervical spine between recumbent and upright neutral position. This situation has led
to a concept of fluctuating kinetic central spinal canal
stenosis (fluctuating fluid disk herniation) that can only
be shown with these different positions. Cervical spine
imaging in the recumbent position showing posterior osteophytes may only reveal cord compression with
upright-extension positioning. Because of the prevalence of back pain that occurs in a nonsupine position
and the inability of routine supine MRI to satisfactorily
reveal clinical syndromes, it is likely that positional imaging will have a role in the future but how it exactly
will be implemented is as yet undetermined; the role of
imaging the hip, knee, and ankle under axial load also
warrants further investigation. If the supine simulated

126

Orthopaedic Knowledge Update 8

weight-bearing paradigm becomes validated, then the
currently installed base of magnets can be used without
having to deploy new, costly space-occupying devices.
Further studies comparing simulated weight-bearing
versus upright imaging will be needed to show whether
new magnets are required for this purpose.

Sports Medicine: Magnetic Resonance Arthrography

For any joint the placement of intra-articular contrast
whether by direct or indirect means can be used to assist the evaluation of ligaments, cartilage, synovial proliferation, or intra-articular loose bodies. MRI provides
cross-sectional and multiplanar imaging for precise spatial delineation and an additional capability to supply
soft-tissue contrast outside of the joint cavity (tendons,
muscles, and bone marrow) unavailable by any other
modalities. Pertinent issues for optimizing diagnostic
yield include technical considerations in performing
magnetic resonance examinations, properly identifying
structures of interest, and the clinical significance of the
incidental findings. Magnetic resonance arthrography
has been most extensively studied in the shoulder and
to a lesser degree in the hip and the postoperative knee.
Other joints in which it has been applied include the elbow and wrist and to a lesser degree the ankle.
Direct magnetic resonance arthrography, most often
done with injection of diluted gadolinium or less often
with saline solution, can be useful for evaluating certain
pathologic conditions in the joints. Gadolinium-based
contrast agents have not been approved by the US Food
and Drug Administration for intra-articular injection
but may be used clinically under the doctrine of the
practice of medicine. These agents are most helpful for
outlining labral-ligamentous abnormalities in the shoulder (Figure 6) and distinguishing partial-thickness from
full-thickness tears in the rotator cuff, demonstrating labral tears in the hip, showing partial- and full-thickness
tears of the collateral ligament of the elbow and delineating bands in the elbow, identifying residual or recurrent tears in the knee after meniscectomy (Figure 7), increasing the certainty of perforations of the ligaments
and triangular fibrocartilage in the wrist, correctly identifying ligament tears in the ankle and increasing the
sensitivity for ankle impingement syndromes, assessing
the stability of osteochondral lesions in the articular surface of joints, and delineating loose bodies in joints. Direct magnetic resonance arthrography has become a
well-established method of delineating various joint
structures that otherwise show poor contrast on conventional MRI. However, direct magnetic resonance arthrography is minimally invasive and usually necessitates fluoroscopic guidance for joint injection (some
authors have described doing blind injections or using
other modalities such as ultrasound or MRI).

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 11 Musculoskeletal Imaging

Figure 6 Direct magnetic resonance arthrography of the shoulder. T1-weighted fat-suppressed images obtained after the intra-articular injection of a dilute gadolinium solution.
A, Buford complex: axial image at the level of the coracoid process shows a deficient anterosuperior labrum (white arrow) with a thick cord-like middle glenohumeral ligament
(black arrowhead) reflecting a normal developmental variant. B, Bankart lesion: axial image caudal to the level of the coracoid process shows contrast intravasation into an
irregular deformed anteroinferior glenoid labrum (arrow) distinct from the middle glenohumeral ligament (arrowhead). C, Superior labral anterior and posterior lesion: coronal
oblique image posterior to the biceps attachment to the glenoid shows an irregular collection of contrast material (arrow) extending into the superior labrum with partial
detachment.

Intravenous gadolinium-based contrast leaks from

the capillary bed into the interstitial space of the synovium and then diffuses through the synovial lining into
the joint cavity. Because the synovial membrane is
highly vascular, intravenous gadolinium-based contrast
material enhances the joint cavity, causing the arthrographic effect. This technique has developed as an alternative to direct magnetic resonance arthrography for
imaging joints. The arthrographic effect is dependent on
the degree of synovial enhancement and volume of synovial fluid. Synovial enhancement is poorer in noninflamed joints. Well-vascularized and inflamed tissues
will show enhancement with this method. One limitation of indirect magnetic resonance arthrography is a
lack of controlled joint distension compared with that of
direct arthrography. Joint distension facilitates recognition of certain conditions such as capsular trauma or
soft-tissue injury concealed by a collapsed capsule.
Joints with a large capacity (knee, hip, shoulder) are less
suited to indirect magnetic resonance arthrography because the larger volume necessitates a longer delay and
leads to less predictable heterogeneous enhancement.
Nonetheless, indirect magnetic resonance arthrography
can provide diagnostic arthrogram-like images of the
shoulder, elbow, wrist, hip, knee, and ankle joint if the
technique is optimized. Fat saturation and joint exercise
before imaging is critical to provide high signal intensity
joint fluid. Diffusion is increased in joints, which are inflamed, or after gentle exercise, because of physiologic
hyperemia. Factors affecting passage of contrast material between the blood vessels and synovial fluid include
pressure differences between these spaces and the viscosity of the intra-articular fluid. The normal joint space
is a low-pressure space, containing joint fluid of relatively low viscosity. Exercise not only increases vascular

American Academy of Orthopaedic Surgeons

Figure 7 Direct magnetic resonance

arthrography of the knee: Sagittal T1weighted fat-suppressed spin echo
images show a horizontal increased
intrameniscal signal intensity similar
to the contrast material extending to
the free margin (arrow) of the posterior horn of the medial meniscus, reflecting a recurrent meniscal tear. The
posterior horn of the medial meniscus
is truncated from a prior partial meniscectomy. Extensive cartilage loss of
the posterior aspect of the lateral
femoral condyle is seen.

perfusion but also increases vascular pressure. Exercise

also leads to motion within the joint fluid, thereby reducing the concentration of contrast adjacent to the synovial membrane and thus facilitating the diffusional effect. Increase in signal intensity of joint fluid has been
reported to be four times greater in joints that were exercised 10 up to 15 minutes before imaging compared
with joints that were not exercised. Without exercise the
signal intensity increases slowly and reaches a maximum
after approximately 60 minutes. In patients with osteoarthritis, maximum enhancement may occur as late
as 90 minutes in the absence of joint motion. As contrast diffusion is based on the concentration difference
between plasma and joint fluid, increasing the dose of
contrast material administered intravenously may increase the diffusion gradient. However, double- and
triple-dose intravenous injections have been shown to
have a limited positive effect on indirect arthrography.

127

Musculoskeletal Imaging

Figure 8 Indirect magnetic resonance arthrography of the wrist: lunotriquetral ligament tear. Coronal T1-weighted fat suppressed spoiled gradient recalled image obtained after the administration of a standard dose of intravenous gadolinium based
contrast material and 10 minutes of exercise shows high signal equal to contrast material in the lunotriquetral interval (arrow) reflecting ligament disruption (normally
there should be a low signal band at the base of the proximal carpal bones). Note the
multicompartment enhancement.
Thus, a standard dose of gadolinium-based contrast injected intravenously is usually sufficient to attain a good
signal to noise ratio and good contrast to noise ratio. It
allows simultaneous assessment of both intra-articular
and extra-articular soft tissues but the physician must be
cognizant of the determinants of contrast enhancement
not to be confounded by normally enhancing structures
(Figure 8). Indirect magnetic resonance arthrography is
a useful adjunct to conventional musculoskeletal MRI,
may be preferable to the more invasive direct magnetic
resonance arthrogram in certain applications, and often
can be performed when direct arthrography is inconvenient or not logistically feasible (outpatient magnet). Although indirect magnetic resonance arthrography has
some disadvantages, it does not require fluoroscopic
guidance or joint injection and it is often superior to
conventional MRI in delineating structures.

Hand/Upper Extremity: Internal Derangements

of the Wrist
Challenges associated with MRI of the hand and upper
extremity are related to the small compact complex
anatomy and the high special resolution requirement.
High signal-to-noise ratios are desired; however, thin
section, small field of view, and high-resolution images
all detract from signal-to-noise ratios. Off isocenter imaging, as is commonly done for the elbow and the wrist,

128

Orthopaedic Knowledge Update 8

can detract from quality because of artifacts related to
reduced field homogeneity. Using a high-field dedicated
extremity magnet can improve overall quality by having
the structure of interest within the isocenter. Highquality surface RF coils should be used for wrist or elbow imaging and can improve the signal-to-noise ratios
threefold to fivefold. Higher field strength with stronger
gradients (increased slew rates) allows for thinner section acquisition, smaller field of views, and reduced
scanner time. Conventional MRI systems have a 3-mm
section limitation for two-dimensional pulse sequences
because the amplifier can only deliver a fixed amount of
power to the gradient coil. The addition of volumetric
three-dimensional imaging can also be useful. The elbow
and wrist are joints that contain compact anatomy and
perform complex motions. Frequently, the structures of
interest are not well shown in the standard planes of imaging and appropriate positioning at optimal planes
needs to be acquired for each area. A three-dimensional
pulse sequence that can be dynamically manipulated
may be advantageous in this context. Contrast resolution issues are also extremely important and are related
to determining the optimal pulse sequence. For certain
structures in the wrist and elbow such as the fibrocartilage, ligaments, and capsule, there is a limited inherent
contrast that normally exists. In this respect intraarticular contrast medium (magnetic resonance arthrography) greatly assists the conspicuity of findings. The
contrast-to-noise ratio tends to improve by fat suppression because this alters the dynamic range by reducing
the contribution of signal from fat with the smaller contrast differences rescaled across a broader range of signal intensity pixel values. The resulting effect is that the
rescaling provides amplified contrast even with nonfatcontaining structures (such as cartilage). Currently, gradient echo sequences can provide high-resolution imaging but there are significant contrast limitations.
Spurious signals are often present in tendons and ligaments, particularly when they approach a 55 angle to
the main magnetic field (causing magic angle artifact, a
T2 shortening phenomenon manifested in short TE images). This is often the case in at least some portion(s)
of some of the tendons given the complex anatomy in
this region.
MRI of the hand, wrist, and elbow is useful for
extra-articular pathology and in this sense is complementary to diagnostic arthroscopy. The reports for diagnostic performance of MRI for ligamentous wrist lesions in the literature are variable. Thus, MRI has not
achieved a respected role in defining the biomechanically clinically significant lesions in the wrist. One mechanism to improve spatial resolution is to use dedicated
high-resolution coils. Thin and contiguous slices are
needed for adequate MRI of the wrist because many of
the larger ligaments around the wrist are no greater
than 1 to 2 mm thick. Recent magnetic resonance tech-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

nology allows the use of a microscopy coil, which provides high-resolution MRI of the hand and wrist. Highresolution MRI with a microscopy coil is a promising
method to diagnose triangular fibrocartilage complex
and other ligament lesions. The limitation of microscopy
coils is that the depiction of deep structures is inadequate. However, this may be resolved by combined positioning with a larger surface coil or a flexible coil. In addition, the limited sensitivity of microscopy coils may
sometimes make accurate coil setting difficult over targeted structure or suspected lesions. The availability of
superconducting coils has also been applied to small
joint imaging. Overall, it is likely that advanced coil development will lead to improved diagnostic performance of MRI because high spatial resolution imaging
is paramount to detect infrastructural features of the
wrist and elbow when evaluating for internal arrangements.

Foot and Ankle: Bone Marrow Edema-Like Lesions

Bone marrow edema (BME)-like lesions are often observed on MRI. Although BME-like signal is not specific on MRI, additional morphologic findings are often
useful to reveal the etiology of many BME patterns.
Normal marrow constituents have three components:
osseous, myeloid elements, and adipose cells. Hematopoietic (red) marrow has approximately 40% fat content and fatty (yellow) marrow has 80% fat content. The
appendicular skeleton tends to have more fatty marrow
than hematopoietic marrow and this serves as a natural
contrast agent, showing bright T1 signal and suppression
on fat saturation images. MRI to detect BME relies on
fluid-sensitive sequences (short-tau inversion recovery
[STIR] and fat suppressed T2-weighted images). T1weighted images can supplement T2-weighted images
and are very specific for the infiltrative process, but are
not as sensitive if there is nonfat marrow or no substantial degree of edema. Gradient echo images are poor for
assessing marrow because of increased susceptibility related to the interfaces between the trabecula and hematopoietic marrow. Gradient echo images can be useful to reveal other diagnoses such as lesions that contain
iron, calcium, or hemosiderin (pigmented villonodular
synovitis). Other novel MRI techniques that have been
variably applied include chemical shift imaging, diffusion weighted imaging, and magnetic resonance spectroscopy.
BME lesions can reflect nonspecific response to injury or excess stress. The pathophysiology is related to
the extracellular fluid, which can be affected by hypervascularity and hyperperfusion (hyperemia), an inflammatory infiltrate causing resorption, granulation (fibrovascular) tissue or an adaptive/reactive phenomena
related to biomechanical alterations (MRI manifestation of Wolffs Law). The pathoetiologies are legion and

American Academy of Orthopaedic Surgeons

Chapter 11 Musculoskeletal Imaging

the signal intensity is directly proportional to the
amount of extracellular water. Contrast enhancement
occurs in areas of BME irrespective of etiology (benign
or malignant, infectious or noninflammatory). The potential etiologies of BME include diseases in the category of trauma, biomechanical, developmental, vascular,
neoplastic, inflammatory, neuropathic, metabolic, degenerative, iatrogenic, and potentially idiopathic conditions
(transient BME syndromes).
Occult injuries result from an acute overt episode of
trauma. The physician should suspect a fracture in these
patients. The traditional modality applied to fracture detection has been radiography, which may be negative or
indeterminate for nondisplaced fractures or a fracture
plane that is not tangential to the x-ray beam. In this
context MRI serves as a more sensitive technique for
fracture detection and characterization. Contusions, also
known as bone bruises, are considered microtrabecular
fractures. On MRI there is no fracture line and the pattern may be a clue or secondary sign of ligament or tendon injury. These fractures often occur in a subarticular
location from osteochondral impaction injuries.
In terms of developmental conditions the normal
conversion of red marrow to yellow marrow sometimes
has areas of slight T2 or STIR hyperintensity but these
usually are not as bright as pathologic lesions. Areas of
a developmental synchondrosis with failure of segmentation of the primitive mesenchyme may cause symptoms via abnormal biomechanics. Symptomatic fibrous
or cartilaginous tarsal coalitions often show reciprocal
areas of BME. Anatomic variants that can present as
painful lesions may be considered a separate but related
category. Those of chronic chondro-osseous disruption
include bipartite patella, dorsal defect of the patella, and
os subfibulare of the ankle. Those in the congenital synchondrosis category include accessory navicular bone
and os trigonum. There are lesions that may predispose
to premature degenerative joint disease; one example is
the os intermetatarsarum, which can contribute to hallux valgus by causing excessive metatarsus varus. When
these variants are symptomatic they often demonstrate
BME-like signal about the abnormality, reflecting altered biomechanics, chronic stress, or sometimes areas
of osteonecrosis. It is thought that MRI can form an objective basis for management of the lesions, particularly
when surgery is considered.
The well-established vascular causes of BME-like
signal may be related to either hyperemic or ischemic
conditions. Of the hyperemic etiologies, inflammatory
disorders that increase vascularity or disuse may cause
subarticular BME patterns. The disuse pattern is also
partly related to increased blood flow, can be characteristic and parallels the radiographic appearance of aggressive osteoporosis with diffuse or multiple rounded
areas of fluid-like hyperintensity in a subarticular and
metaphyseal distribution predominantly in the hindfoot

129

Musculoskeletal Imaging

Orthopaedic Knowledge Update 8

Figure 9 Osteomyelitis superimposed on neuropathic arthropathy. A, Axial T1-weighted spin-echo image of the midfoot reveals disorganization and dislocation of the Chopart
joint, showing replacement of the normal marrow with diffuse infiltration of hypointense signal (arrowheads) in the tarsal bones. B, Sagittal T2-weighted fast spin-echo image
reveals marrow edema in the midfoot and hindfoot bones, tarsus effusions, a rocker bottom deformity and fluid-like signal in the overlying subcutaneous tissues (arrow). C,
Sagittal T1-weighted spin-echo contrast-enhanced image shows rim enhancement around plantar sinus tracts (small arrows) from the ulcer base and extending into midfoot
reflecting a plantar space abscess (large arrow). The marrow edema is enhancing, which is nonspecific, but there is cortical irregularity of the anterior aspect of the cuboid
adjacent to the soft-issue enhancement (arrowhead). The secondary signs of cutaneous ulcer, sinus tract, and cortical interruption have the highest positive predictive value for
osteomyelitis.
and midfoot. In terms of ischemic lesions, the broad category of osteonecrosis (infarct, osteonecrosis) can have
BME early in the course of the disorder associated with
acute, painful symptomatology. Pain improvement usually parallels the resolution of the BME-like signal. The
MRI pattern shows early BME with loss of subchondral
fat signal intensity. The double line sign is specific and
most often identified as a ring of T1 hypointensity and
T2 hyperintensity. This likely reflects a reactive interface
rather than chemical shift artifact. MRI signal of the necrotic segment may be reconstituted and appear fatty
because of the lipid content (the signal is not significantly altered because of the reduced metabolic state).
MRI findings may be seen as early as 10 to 15 days and
for most patients within 30 days after the vascular insult. Transient osteoporosis or the MRI correlate, transient bone marrow edema syndrome, may occur in numerous other low extremity locations including the hip,
knee, talus, cuboid, navicular, and metatarsals. In addition, it may be migratory and occur in a ray pattern.
Some believe that these lesions may reflect salvaged osteonecrosis but it is likely that many of these lesions
may simply be biomechanical in nature.
In the inflammatory category it is well established
that infectious etiologies cause BME. One difficult differential diagnosis in the setting of diabetic neuropathy
is distinguishing osteomyelitis from a Charcot joint.
There are several MRI findings that may help in the differentiation. Osteomyelitis is more common in the phalanges, distal metatarsals, and calcaneus (secondary to
overlying ulceration) whereas neuropathic disease is
more common in the Lisfranc, Chopart, and ankle joints.

130

One notable exception is an ulceration that develops

because of poorly fitting footwear, or foot deformity
and altered weight bearing that can cause atypical location of osteomyelitis. However, there should be a softtissue defect identified over these areas to diagnose osteomyelitis. The epicenters of signal abnormalities can
be useful. Neuropathic disease has an articular epicenter
and usually multiple joints are involved with a regional
instability pattern. Osteomyelitis has a marrow epicenter with focal centripetal spread throughout the bone. It
is important to emphasize that transcutaneous spread is
the route of inoculation in more than 90% of cases of
osteomyelitis of the foot in patients with diabetes.
Therefore, secondary soft-tissue signs are paramount; a
subcutaneous ulcer with interruption of cutaneous signal, cellulitis, soft-tissue mass effect from a phlegmon,
soft-tissue abscess (well-defined rim enhancing fluid collection) and particularly a sinus tract strongly support
infection (Figure 9).
It has been recently recognized that degenerative
conditions are associated with areas of BME. These degenerative conditions may occur with either primary or
secondary osteoarthritis. Geodes (subchondral cysts)
are one of the imaging hallmarks of osteoarthritis and
can be identified on MRI. Early during the course of
disease, ill-defined areas of BME appear and later form
discrete cystic structures. Some of these areas with hyperintensity have been shown by pathologic studies not
to reflect the fluid. It has been hypothesized that some
of these findings are likely mechanical or adaptive responses related to the altered mechanics from the joint
failure and may be considered the MRI manifestation

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 11 Musculoskeletal Imaging

Figure 10 Tendinopathy-associated bone

marrow edema. A and B are T2-weighted fatsuppressed fast spin-echo MRI. Posterior tibialis tendon dysfunction: axial image (A) shows
medial malleolus marrow edema (arrowhead)
immediately subjacent to the posterior tibialis
tendon (arrow). Peroneal tendinopathy: coronal image (B) shows peroneus longus tendon
tenosynovitis (arrow) with underlying subtendinous bone marrow edema in peroneal tubercle of the calcaneus (arrowhead).

of Wolffs Law. In the knee (and possibly in the ankle

and foot) bone marrow findings are strongly associated
with the presence of pain, and moderate or larger fusions in synovial thickening are more frequent among
those with pain than those without pain adjusted for degree of radiographic osteoarthritis. In addition, focal
subchondral BME may be a clue to focal cartilage defects (potential treatable cartilage defects), which presumably are posttraumatic events. The cartilage abnormality itself may be relatively inconspicuous on MRI
pulse sequence selection or spatial resolution and therefore an area of subarticular flame-shaped BME in a
nonarthritic joint can be a helpful secondary sign. Another recently described pattern of BME is a subtendinous location. This has been identified as a response to
tendon abnormality and hypothesized to be from mechanical friction, hyperemia, or because of biomechanical reasons. A subtendinous location is most common in
the lower extremity, particularly in the foot and ankle,
and is most often seen with posterior tibialis tendon
(PTT) dysfunction (Figure 10, A). The areas of edema
related to PTT dysfunction are the medial tibia (malleolus), navicular tubercle, calcaneous, and talus. This finding is not seen in most people with PTT dysfunction but
may be a sign for a more advanced stage and poorer
tendon quality. Less frequently, BME may also be related to peroneal tendinopathy (Figure 10, B): typically
in the lateral fibula (lateral malleolus) or lateral calca-

American Academy of Orthopaedic Surgeons

neus (along the peroneal tubercle). Noninfectious inflammatory enthesopathies such as psoriatic or reactive
arthritis cause prominent flame-shaped BME patterns
at the tendon-bone junction (enthesis) often with an associated erosion that may be better appreciated with radiography.
There are several miscellaneous but important causes
of lower extremity BME patterns. Hematopoietic (red)
marrow can sometimes be confused for an abnormal
BME pattern. Hematopoietic marrow is most prominent
in the pediatric population and there is a conversion pattern progression from distal to proximal. One important
realization is that once an epiphysis is ossified, it should
contain fatty signal with a couple of important exceptions
(reconversion not infrequently occurs in proximal femoral epiphysis). In general, reconversion related to anemia
or other conditions is in the opposite direction. In terms
of marrow replacement disorders (leukemia and lymphoma) the pattern may be a diffuse or focal area of marrow signal abnormality.With infiltrative diseases, the marrow pattern tends to have some asymmetry and
pathologic processes tend to have more T2 prolongation
and higher signal intensity (brighter BME). Neoplasms often show lesional or perilesional BME. The signal intensities are unreliable for histology and there is substantial
overlap between benign and malignant conditions. Metastatic deposits are hematogenous in origin and are predominantly located in red marrow areas (axial skeleton)

131

Musculoskeletal Imaging
but can also be present in the appendicular skeleton, especially for deposits from bronchogenic or breast carcinoma. For primary neoplasms, the degree of BME does
not correspond to malignancy potential.There are several
well-known benign lesions that are characterized by very
prominent BME: chondroblastoma, osteoid osteoma, and
Langerhans cell histiocytosis. Patients who have undergone radiotherapy or chemotherapy, those who are taking bone marrow recovery agents, and patients who have
recently undergone dbridement may show BME depending on the time course of the treatment.

Pediatrics: Physeal Lesions and Growth Arrest

The growing skeleton is susceptible to injury. Advances
in pediatric musculoskeletal radiology have been made
in imaging the cartilage, epiphysis, and physis. Closure
disturbance of the long bones in children is frequently
posttraumatic but also occurs because of physeal, epiphyseal, or metaphyseal ischemia. The growth mechanism represented by the cartilage structures at the ends
of growing bones is not directly visible on radiography
but is well visualized by MRI. Improved definition of
cartilaginous abnormality by MRI may permit earlier
detection and treatment of disorders and thus prevent
bone deformity. The formation of physeal bars (bony
bridges across the growth plate) is one active area of pediatric musculoskeletal radiology research. Premature
bony fusion in children is most often posttraumatic and
disproportionately involves the tibia and femur with
bridges tending to develop as the site of earliest physiologic closures (anteromedially and centrally). The distal
tibia, proximal femur, and proximal tibia physes are disproportionally at risk because of the complex geometry.
The central undulations in the distal femur and Kumps
bump in the distal tibia are sites of initial physiologic
closure and the most frequently involved in the premature fusions. Animal studies with physeal and metaphyseal injuries have shown MRI can identify persistence of
abnormality in the growth cartilage after physeal injuries and evolution of abnormalities after metaphyseal
injury best seen on T2-weighted images. It has been
shown on MRI in animal models that abnormalities in
the physeal cartilage result in development of a transphyseal vascularity that precedes the formation of bony
bridge after trauma. MRI can detect this transphyseal
vascular lesion within the first 2 weeks of injury. In vivo
studies in humans confirm that MRI defect abnormalities in the cartilage are associated with subsequent
growth disturbances and provide accurate mapping of
physeal bridging and associated growth abnormality in
the posttraumatic population (Salter-Harris injuries).
Contrast enhancement can be useful in showing reconstitution of metaphyseal vascularity after injury but does
not reliably enable the detection of transphyseal vascularity after physeal injury until a distinct bony bridge is

132

Orthopaedic Knowledge Update 8

formed. Early MRI evaluation in children with lower
extremity fractures can be prognostic. Physeal narrowing or tethering in the absence of growth arrest lines
was found in those patients who subsequently required
late surgical intervention. The MRI in acute phase provided accurate evaluation of physeal fracture anatomy
and could often augment the staging of the Salter-Harris
classification. The course and level of injury within the
cartilage physeal fracture-separation can be defined
with MRI. Extension into the juxtaepiphyseal region is
another potential risk factor for growth arrest and is detectable by MRI. Early MRI can demonstrate transphyseal bridging or altered arrest lines in physeal fracture
before they become manifest on radiographs. Physeal
enhancement decreases with physeal closure as expected. In the marrow and the extremities, contrast enhancement is greater in the metaphyseal metathesis portion than the fatty epiphyseal portion. In both areas
enhancement decreases as the marrow becomes more
fatty. Local physeal widening in a growing bone may
represent the imprint of the previous or ongoing interference with endochondral ossification. Widening can be
seen on a fluid-sensitive pulse sequences in physeal dysfunction without bridge formation. Physeal widening
with focal palisading morphology, central distribution in
the metaphysis and concomitant epiphyseal signal abnormalities are significant predictors of subsequent
growth disturbance. Therefore, MRI should be considered as part of the evaluation for patients at high risk
for growth disturbance, especially young children with
extensive residual growth potential and those that involve particularly vulnerable growth plates (such as
about the knee) and pediatric patients with severe complex fractures. MRI is now a standard of care and helps
surgical management for these patient populations with
cartilage-sensitive sequences, exquisitely showing the
disturbance and associated abnormality that may follow
physeal injuries.

Orthopaedic Oncology: Metabolic Imaging

Distinguishing benign from malignant soft-tissue lesions
in the extremities is a difficult if not impossible task by
imaging using either signal, morphologic, or enhancement criteria. Follow-up imaging for sarcoma patients is
also complicated by complex prostheses, which can produce artifacts and limit visualization at the surgical site.
PET scanning, discussed earlier in the chapter, holds
promise in showing metabolically active areas and may
be particularly suitable for monitoring patients after
therapy (resection, chemotherapy, or radiotherapy).

Orthopaedic Traumatology: Multislice

Computed Tomography
One of the most recently evident benefits of multislice CT
is in the setting of appendicular and axial trauma. CT

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

greatly improves the anatomic depiction of spinal injury
when compared with projectional radiography. Compared
with single detector helical CT scanners, multislice CT
scanners have increased tube heating capacity and run at
a higher table speed, allowing an increased volume of coverage with the same amount of scanning time. This makes
screening examination of part of the spine or the entire
spine feasible, which may eliminate screening radiographs
in certain settings. Examinations of the thorax and the
lumbar spine can be extracted from a CT examination of
the chest abdomen and pelvis.
There are several pitfalls to be aware of but the most
important image artifacts are not unique to multislice CT.
These pitfalls include metal-induced streak artifact and
patient motion. Because of the higher spatial resolution,
vascular channels of the vertebral bodies are better appreciated and may be mistaken for normal structures. Multislice CT has some risk predominantly related to the radiation dose to the individual patient and to the
population. The radiation dose of the patient increases as
the volume of coverage increases. Multislice CT allows
imaging of very thin sections quickly, much faster than
previously possible, allowing for effective screening of spinal injuries and evaluating extremity injuries. Screening
CT of the entire cervical spine is cost effective if high-risk
criteria, such as focal neurologic deficit referable to the
cervical spine, head injury (skull fracture, intracranial
hemorrhage) or loss of conscious at the time of examination, and high-energy mechanism (motor vehicle accident
at a speed of greater than 35 mm, pedestrian struck by a
car, or a fall greater than 10 feet).

Diagnostic and Therapeutic Procedures

Spinal Injections
Epidural steroid injections, sacroiliac joint injections,
zygapophyseal (facet) joint injections, diskography, and
vertebral augmentation are image-guided procedures
that are important components of a comprehensive
management approach to spine pain syndromes for establishing a diagnosis, directing or administering therapy, and facilitating rehabilitation and functional restoration.
Demand for epidural injections is rapidly expanding.
Patient satisfaction elicited from these procedures is often a direct result of imaging guidance, which can
shorten and simplify procedures, minimize potential for
complications, and verify accurate localization of the
needle to selectively provide the lowest necessary dose
of analgesic to the optimal area. These procedures can
greatly contribute to patient management and surgical
planning by determining sources of pain and treating
pain generators. Participation of an experienced radiologist in performing these procedures optimizes patient
care, because radiologists are trained in anatomy and
image-guided needle localization procedures. However,

American Academy of Orthopaedic Surgeons

Chapter 11 Musculoskeletal Imaging

use of spinal injections should be considered a team effort in conjunction with an experienced clinical diagnostician who can accurately diagnose the patients problem and recommend the appropriate procedure.
Performance of these procedures requires an intimate
knowledge of relevant anatomy, appropriate equipment
and facilities, and apprenticeship with an experienced
practitioner. Selective epidural injections can offer significant diagnostic and therapeutic benefit for patients
with radicular pain. Attention to proper technique will
minimize risk of complications from these procedures
while maximizing their benefit.
Controversy and differences in opinion related to
epidural steroid injections often revolve around choice
of trajectory (transforaminal versus transflaval [translaminar, interlaminar]) and whether to use image guidance. Transflaval injections may be done with or without
image guidance (using loss of resistance techniques),
whereas transforaminal injections are done with image
guidance. Studies reviewing the efficacy of epidural steroid injections favor it as a useful treatment overall.
However, controlled clinical trials performed without
fluoroscopic guidance are not unanimous in demonstrating the benefits of lumbar epidural steroid injections
with a broad range of successful results, ranging from
18% to 90%. This broad range may be related to the actual location of medication deposition. The incidence of
failure to reach the epidural space using a nonimageguided transflaval approach ranges from 13% to 30%
and may target the wrong interlaminar space by one or
more levels. The transforaminal approach demonstrates
superior ventral opacification, whereas the transflaval
method shows predominantly dorsal opacification (using CT as the reference standard to confirm contrast location after fluoroscopically-guided epidurography).
Dorsal deposition may be less effective because the steroid is remote from the source of irritation (for example, disk herniation). There is an increased risk of a
dural puncture (intrathecal administration), spinal headache (may require blood patch treatment), intrathecal
administration of steroid or residual preservatives of local anesthesia leading to nerve root injury, hypotension,
or dyspnea. For several reasons image-guided epidural
steroid injection is favored; it adds only minimal risk
(radiation) and may not add to the overall costs of spine
injections but may even reduce costs by eliminating repeat injections (a nonimage-guided paradigm is to perform two to three successive transflaval epidural steroid
injections because the miss rate of the epidural space is
about 33%).
Intra-articular injections have an established role for
identifying zygapophyseal (facet) and sacroiliac joint articulations as nociceptors. However, intra-articular injections with anesthetic or corticosteroid are often not sufficient for a long lasting therapeutic effect. Once a
zygapophyseal joint has been implicated as a substantial

133

Musculoskeletal Imaging

Orthopaedic Knowledge Update 8

Figure 11 CT characterization after intradiskal contrast injection. Postdiskography transaxial CT images. A, Normal nucleogram characterized by central globule of contrast
material that remains within the expected confines of the nucleus pulposus. B, Annular fissure. Contrast material is noted within the nucleus pulposus, but also extends in a radial
fashion posterolaterally beyond the expected confines of the nucleus pulposus into the region of the anulus fibrosus (arrow). There is also a circumferential component noted in
the anulus fibrosus (arrowhead).
or significant nociceptor then targeted therapy options
exist that may include a neuroablative procedure (medial branch neurotomy) in conjunction with functional
restoration via physical therapy. Sacroiliac joint treatment can be more problematic given the diffuse innervation of the articulation; however, sacroiliac joint fusion is a technique practiced by some orthopaedic
surgeons. For true inflammatory sacroiliitis related to a
spondyloarthropathy, there is good evidence from several clinical trials that intra-articular corticosteroid is
proven to be an effective component of treatment.
However, the data on the efficacy of steroid intraarticular injections for mechanical somatic dysfunction
are conflicting. For an intracanalicular synovial cyst emanating from an adjacent zygapophyseal joint that is
causing lateral recess stenosis with radicular symptoms,
intra-articular injection with corticosteroid assists in decreasing the perineural inflammation, reducing the size
of the cyst, and alleviating the radicular symptoms.

Diskography
The primary purpose for diskography is for documentation of the disk as a significant nociceptor. For patients
who have chronic pain that is predominantly axial, nonmyelopathic, and nonradicular, imaging may be insufficient or equivocal for determining the nature, location,
and extent of symptomatic pathology. A position statement regarding lumbar diskography from the North
American Spine Society was published in 1995. Specific

134

indications include patients with persistent pain in

whom noninvasive imaging and other tests have not
provided sufficient diagnostic information. In patients
who are to undergo fusion, diskography can be used to
determine if disks within the proposed fusion segment
are symptomatic and if the adjacent disks are normal.
Surgeons concerned with limiting the extent of fusion
are interested in obtaining more evidence beyond MRI
abnormalities to document what intervertebral disk levels are contributing to the painful syndrome. In postoperative patients who continue to experience significant
pain, diskography can be used to assist in differentiating
between postoperative scar and recurrent disk herniation (when MRI or CT is equivocal); or to evaluate segments adjacent to the arthrodesis. Postdiskography CT
can also be used to confirm a contained disk herniation
as a prelude to minimally invasive intradiskal therapy
(Figure 11). Diskography is also being used as part of
the selection criteria for many clinical trials assessing
lumbar interbody fusion devices or percutaneous intradiskal treatments.

Annotated Bibliography
Available Imaging Modalities
Carrino JA: Digital imaging overview. Semin Roentgenol
2003;38:200-215.
This article provides an overview of the electronic imaging
environment, including a review of the technologies behind

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

picture archiving and communications systems and radiology
information systems as well as practical information on implementation.

Horton KM, Sheth S, Corl F, Fishman EK: Multidetector

row CT: Principles and clinical applications. Crit Rev
Comput Tomogr 2002;43:143-181.
This article reviews the basic principles of multislice computed tomography. Scanner/detector design, beam collimation/
slice thickness, radiation dose, data manipulation, and display
are discussed.

Jadvar H, Gamie S, Ramanna L, Conti PS: Musculoskeletal system. Semin Nucl Med 2004;34:254-261.
In this article, the diagnostic utility of dedicated PET and
PET combined with CT in the evaluation of patients with
bone and soft-tissue malignancies is reviewed.

Recht M, Bobic V, Burstein D, et al: Magnetic resonance

imaging of articular cartilage. Clin Orthop 2001;
391(suppl):S379-S396.
This article presents a review of procedures regarding
MRI of articular cartilage and cartilage repair. Future directions in imaging strategies and ways to measure cartilage
thickness and volume are discussed.

Sofka CM: Ultrasound in sports medicine. Semin

Musculoskelet Radiol 2004;8:17-27.

Chapter 11 Musculoskeletal Imaging

Recent reports and experience suggest that FDG-PET
cannot be a screening method for differential diagnosis between benign and malignant musculoskeletal lesions, including
many neoplasms originating from different tissues. FDG-PET
might not accurately reflect the malignant potential of musculoskeletal tumors, but rather might implicate cellular components included in the lesions. A high accumulation of FDG can
be observed in histiocytic, fibroblastic, and some neurogenic
lesions, regardless of whether they are benign or malignant.
More specific uses of FDG-PET, such as grading, staging, and
monitoring of musculoskeletal sarcomas, should be considered
for each tumor of a different histologic subtype.

Ecklund K, Jaramillo D: Patterns of premature physeal arrest: MR imaging of 111 children. AJR Am J
Roentgenol 2002;178:967-972.
The purpose of this study was to use MRI, especially fatsuppressed three-dimensional spoiled gradient-recalled echo
sequences, to identify patterns of growth arrest after physeal
insult in children. This method exquisitely shows the growth
disturbance and associated abnormalities that may follow physeal injury, and guides surgical management.

Fardon DF, Milette PC: Nomenclature and classification

of lumbar disc pathology: Recommendations of the
combined task forces of the North American Spine Society, American Society of Spine Radiology, and American Society of Neuroradiology. Spine 2001;26:E93-E113.

Ultrasound plays an important role in the evaluation of injuries and painful conditions of the athlete. With portable ultrasound units, examinations can be performed on the playing field,
immediately at the time of the acute injury, for rapid diagnosis.
Ultrasound can be used to guide therapeutic procedures.

This document provides a universally acceptable nomenclature that is workable for all forms of observation, that addresses contour, content, integrity, organization, and spatial relationships of the lumbar disk; and that serves as a system of
classification and reporting built upon that nomenclature.

Yoshioka H, Stevens K, Hargreaves BA, et al: Magnetic

resonance imaging of articular cartilage of the knee: Comparison between fat-suppressed three-dimensional SPGR
imaging, fat-suppressed FSE imaging, and fat-suppressed
three-dimensional DEFT imaging, and correlation with
arthroscopy. J Magn Reson Imaging 2004;20:857-864.

Karppinen J, Paakko E, Paassilta P, et al: Radiologic

phenotypes in lumbar MR imaging for a gene defect in
the COL9A3 gene of type IX collagen. Radiology 2003;
227:143-148.

In this study, signal to noise ratios and contrast to noise ratios were compared in various magnetic resonance sequences,
including fat-suppressed three-dimensional spoiled gradient
echo imaging, fat-suppressed fast spin echo imaging, and fatsuppressed three-dimentional driven equilibrium Fourier transform imaging. The diagnostic accuracy of these imaging sequences was compared with that of arthroscopy for detecting
cartilage lesions in osteoarthritic knees. Fat-suppressed threedimensional spoiled gradient echo imaging and fat-suppressed
fast spin echo imaging showed high sensitivity and high negative
predictive values, but relatively low specificity.

Imaging of Specific Orthopaedic Conditions

Aoki J, Endo K, Watanabe H, et al: FDG-PET for evaluating musculoskeletal tumors: A review. J Orthop Sci
2003;8:435-441.

American Academy of Orthopaedic Surgeons

The results of this study indicate that the presence of Trp3

allele is associated with Scheuermanns disease and intervertebral disk degeneration. No associations were found for other
radiologic phenotypes.

Ledermann HP, Morrison WB, Schweitzer ME: MR image analysis of pedal osteomyelitis: Distribution, patterns of spread, and frequency of associated ulceration
and septic arthritis. Radiology 2002;223:747-755.
The purpose of this study was to evaluate the anatomic
distribution of pedal osteomyelitis and septic arthritis in a
large patient group with advanced pedal infection and to compare ulcer location with the distribution of osteomyelitis and
septic arthritis. Pedal osteomyelitis results almost exclusively
from contiguous infections and occurs most frequently around
the fifth and first metatarsophalangeal joints. One third of patients with advanced pedal infection show evidence of septic
arthritis on MRI.

135

Musculoskeletal Imaging

Orthopaedic Knowledge Update 8

Philipp MO, Kubin K, Mang T, Hormann M, Metz VM:

Three-dimensional volume rendering of multidetectorrow CT data: Applicable for emergency radiology. Eur J
Radiol 2003;48:33-38.

out injury to the cervical spine in patients with blunt

trauma: National Emergency X-Radiography Utilization
Study Group. N Engl J Med 2000;343:94-99.

This article presents a review of recent literature on

volume-rendering technique and applications for the emergency department.

Jensen MC, Brant-Zawadzki MN, Obuchowski N, et al:

Magnetic resonance imaging of the lumbar spine in people without backpain. N Engl J Med 1994;331:69-73.

Steinbach LS, Palmer WE, Schweitzer ME: Special focus

session: MR arthrography. Radiographics 2002;22:12231246.

Karasick D, Wapner KL: Hallux valgus deformity: Preoperative radiologic assessment. AJR Am J Roentgenol
1990;155:119-123.

Direct magnetic resonance arthrography with injection of

saline solution or diluted gadolinium can be useful for evaluating certain pathologic conditions in the joints. Indirect magnetic resonance arthrography with intravenous administration
of diluted gadolinium may be performed when direct magnetic
resonance arthrography is inconvenient or not logistically feasible. Although indirect magnetic resonance arthrography has
some disadvantages, it does not require fluoroscopic guidance
or joint infection and is superior to conventional MRI in delineating structures when there is minimal joint fluid; vascularized or inflamed tissue will be enhanced with this method.

Zlatkin MB, Rosner J: MR imagingof ligaments an triangular fibrocartilage complex of the wrist. Magn Reson
Imaging Clin North Am 2004;12:301-331.
This article summarizes the current diagnostic criteria that
can be useful in interpreting abnormalities of the wrist ligaments and triangular fibrocartilage complex of the wrist.

Diagnostic and Therapeutic Procedures

Carrino JA, Chan R, Vaccaro AR: Vertebral augmentation: Vertebroplasty and kyphoplasty. Semin Roentgenol
2004;39:68-84.
Vertebroplasty and kyphoplasty (balloon-assisted vertebroplasty) have both led to good pain relief and improved
function with minimal complication rates in appropriately selected patients when high-quality imaging and meticulous
technique are used.

Classic Bibliography
Hoffman JR, Mower WR, Wolfson AB, Todd KH,
Zucker MI: Validity of a set of clinical criteria to rule

136

Lawson JP: International Skeletal Society Lecture in

honor of Howard D. Dorfman: Clinically significant radiologic anatomic variants of the skeleton. AJR Am J
Roentgenol 1994;163:249-255.
Martel W, Hayes JT, Duff IF: The pattern of bone erosion in the hand and wrist in rheumatoid arthritis.
Radiology 1965;84:204-214.
Morrison WB, Schweitzer ME, Batte WG, Radack DP,
Russel KM: Osteomyelitis of the foot: Relative importance of primary and secondary MR imaging signs.
Radiology 1998;207:625-632.
Park YH, Lee JY, Moon SH, et al: MR arthrography
of the labral capsular ligamentous complex in the shoulder: Imaging variations and pitfalls. AJR Am J
Roentgenol 2000;175:667-672.
Stoller DW, Martin C, Crues JV III, Kaplan L, Mink JH:
Meniscal tears: Pathologic correlation with MR imaging.
Radiology 1987;163:731-735.
Weissman BN, Aliabadi P, Weinfeld MS, Thomas WH,
Sosman JL: Prognostic features of atlantoaxial subluxation in rheumatoid arthritis patients. Radiology 1982;
144:745-751.
Winalski CS, Aliabadi P, Wright RJ, Shortkroff S, Sledge
CB, Weissman BN: Enhancement of joint fluid with intravenously administered gadopentetate dimeglumine:
technique, rationale, and implications. Radiology
1993;187:179-185.

American Academy of Orthopaedic Surgeons

Chapter

12

Perioperative Medical Management

Gregory Argyros, MD

Introduction
Appropriate perioperative management of surgical patients is critical in ensuring the greatest likelihood of a
successful patient outcome. It is important to understand the physiologic response to surgery and anesthesia, disease-related and procedure-related risk, prophylactic therapy to prevent perioperative problems, and
postoperative medical complications. The medical consultant must determine exactly what is being requestedwhether for surgical risk assessment, diagnostic or
management advice, or documentation for legal reasons.
The surgeon and the medical consultant must communicate directly to minimize the potential for misunderstanding.
Surgical risk is the probability of an adverse outcome or death associated with surgery and anesthesia.
Surgical risk is categorized into four components: patient-related; procedure-related; provider-related; and
anesthetic-related. Information from the medical history, physical examination, review of available data, and
selectively ordered laboratory tests can be used to make
an estimation of perioperative risk. The following factors must be considered: the patients current health status; if there is evidence of medical illness, how severe it
is and whether it will affect surgical risk; how urgent is
the surgery; if surgery is delayed, will treatment of the
medical illness lessen its severity; and, if there is no reason to delay the surgery, what changes need to be made
perioperatively to maximize the patients overall condition. Assessment of these factors will allow for a decision on whether the patient is in optimal medical condition to undergo the planned surgical procedure.

The Preoperative Evaluation

The practice of extensive testing of all patients before
surgery can be expensive, both in terms of direct costs
of the tests, but also for the follow-up of unanticipated
minor abnormalities, many of which have no clinical relevance. Many studies have shown that preoperative routine testing of all patients is of limited value. Tests
should be ordered based on history and physical exami-

American Academy of Orthopaedic Surgeons

nation findings that identify subgroups of patients who

are more likely to have abnormal results. Tests should
only be ordered if the result will influence management.
If a test result is abnormal, documentation of the
thought process regarding the planned response to the
abnormality is critical. A summary of recommendations
regarding testing before elective surgery is presented in
Table 1.

Cardiac Risk Assessment and Optimization

The American College of Cardiology and the American
Heart Association issued guidelines for perioperative
cardiovascular evaluation of patients undergoing noncardiac surgery in 2002. They propose a stepwise strategy that relies on assessment of clinical markers, prior
coronary evaluation and treatment, functional capacity,
and surgery-specific risk. A framework for determining
which patients are candidates for cardiac testing is presented in algorithmic form in Figure 1.
Clinical markers are divided into major, intermediate, and minor predictors of increased perioperative cardiovascular risk. Major clinical markers include an unstable coronary syndrome, such as a myocardial
infarction, within 1 month of surgery, unstable or severe
angina, evidence of a large ischemic burden by clinical
symptoms or noninvasive testing, decompensated heart
failure, significant arrhythmias, and severe valvular
heart disease. Intermediate predictors are mild angina, a
myocardial infarction occurring more than 1 month before surgery, compensated heart failure, a preoperative
creatinine level greater than 2.0 mg/dL, and diabetes.
Minor predictors are advanced age, abnormal electrocardiogram, heart rhythm other than sinus, low functional capacity, history of stroke, and uncontrolled hypertension.
Functional capacity can be expressed as metabolic
equivalent (MET) levels. Increasing MET levels reflect
a greater aerobic burden. Perioperative cardiac risk is
increased in patients unable to meet a 4-MET demand
during most normal daily activities. Activities such as
light housework, climbing a flight of stairs, or walking

137

Perioperative Medical Management

Table 1 | Recommendations for Preoperative Testing

Test

Indications

Hemoglobin

Anticipated major blood loss or symptoms of

anemia
Symptoms of infection, myeloproliferative disorder
or myelotoxic medications
History of bleeding diathesis, myeloproliferative
disorder, or myelotoxic medications
History of bleeding diathesis, chronic liver disease,
malnutrition, recent or long-term antibiotic use
History of bleeding diathesis

White blood cell

count
Platelet count
Prothrombin time
Partial thromboplastin time
Electrolytes

Know renal insufficiency, congestive heart failure,

medications that affect electrolytes
Renal function
Hypertension, age > 50 years, cardiac disease,
major surgery, anticipated use of medications
that may affect renal function
Glucose
Obesity or known diabetes
Liver function tests No indication
Urinalysis
No indication
Electrocardiogram Men > age 40 years, women > age 50 years, known
coronary artery disease, diabetes, or hypertension
Chest radiograph Age > 50 years, known cardiac or pulmonary disease, symptoms or examination suggestive of
cardiac or pulmonary disease

on level ground at 4 miles per hour involve a 4-MET

energy expenditure.
Surgery-specific risk is related to the type of surgery
and the degree of hemodynamic compromise associated
with the procedure. Most orthopaedic procedures are
intermediate risk, with a cardiac risk generally less than
5%. Prolonged procedures associated with large fluid
shifts and/or blood loss and emergency surgery are high
risk, with a reported cardiac risk of greater than 5%.
Several trials have examined the impact of medical
therapy begun before surgery on reducing cardiac
events. Two randomized, placebo-controlled trials of
-blocker administration have been performed, with
one showing reduced perioperative cardiac events and
the other improved 6-month survival. When possible,
-blockers should be started far enough before elective
surgery so that the dose can be titrated to achieve a
resting heart rate between 50 and 60 beats per minute.

Pulmonary Risk Assessment/Optimization

The type of surgery performed and the anatomic location of the surgery are major determinants of risk in the
development of postoperative pulmonary complications.
Most orthopaedic procedures are associated with a very
low incidence of pulmonary complications. Most of the
complications that do occur involve either oversedation
from pain medications that can result in atelectasis, or

138

Orthopaedic Knowledge Update 8

postoperative deep venous thrombosis and pulmonary
embolism.
Routine preoperative chest radiographs are not recommended for all patients. They should be obtained for
all patients older than 50 years, those with known preexisting cardiopulmonary disease, and those with symptoms or findings on physical examination that suggest a
likelihood of cardiopulmonary disease. Using these criteria, previously unrecognized abnormalities that may
influence perioperative management will be detected in
a small but clinically significant subset of patients.
The purpose of pulmonary function assessment in
patients undergoing orthopaedic surgical procedures is
not to preclude surgery but to help in anticipating and
preventing complications. Preoperative pulmonary function studies are typically not indicated unless the patient
is older than 60 years, has a history of pulmonary disease or smoking, or there is an anticipated anesthesia
time of 2 hours or longer. Routine arterial blood gas assessments are not recommended.
If possible, elective surgery should be delayed if the
patient exhibits evidence of active infection (such as a
change in the character or amount of sputum). Also, surgery should be delayed if the patient has obstructive
lung disease and is in the midst of an acute exacerbation. Although there is conflicting evidence regarding
the benefits and ideal timing of smoking cessation before surgery, abstinence from smoking for at least
8 weeks before surgery is probably ideal for decreasing
postoperative pulmonary complications, but an 8-week
delay in surgery may not be feasible. Patients with obstructive lung disease who are receiving bronchodilator
or inhaled anti-inflammatory therapy should continue
this regimen throughout the perioperative period.
Although several abnormalities in pulmonary function have been noted in obese patients, studies have
shown that obesity does not consistently contribute to
significant pulmonary complications. In patients with
known sleep apnea, perioperative airway management
is paramount because perioperative continuous positive
airway pressure may be required as these patients recover from anesthesia. It has not been determined
whether screening patients with multiple risk factors for
sleep apnea (obesity, increased neck circumference,
craniofacial abnormalities, hypothyroidism) impacts
postoperative pulmonary complications, but concerns
about perioperative airway management difficulties
should be raised.

Renal Risk Assessment/Optimization

In patients with chronic kidney disease, cardiac disease
is the leading cause of death; therefore, a full cardiac
risk assessment as previously described is needed. In addition, close attention must be paid to other manifestations of chronic kidney disease to include volume, elec-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 12 Perioperative Medical Management

Figure 1 Stepwise approach to preoperative cardiac assessment. Subsequent care may include cancellation or delay in surgery, coronary revascularization followed by noncardiac surgery, or intensified care. (Reproduced with permission from Eagle KA, Berger PB, Calkins H, et al: ACC/AHA guideline update for perioperative cardiovascular evaluation
for noncardiac surgery: Executive summary. Circulation 2002;105:1257-1267.)

American Academy of Orthopaedic Surgeons

139

Perioperative Medical Management

Orthopaedic Knowledge Update 8

on microscopic evaluation of the urine. These patients
require volume repletion with a bicarbonate-containing
resuscitation fluid to maintain an alkalotic urine.

Diabetes Mellitus Risk Assessment and Optimization

Figure 2 Summary of preoperative management recommendation based on therapeutic regimen and the complexity and scheduling of the surgical procedure. MDI indicates multiple doses of intravenous insulin; IV, intravenous. (Reproduced with permission from Jacober SJ Sowers JR: An update on perioperative management of diabetes.
Arch Intern Med 1999;159:2405-2411.)

trolyte, and acid/base status, anemia, bleeding diatheses,

the propensity for dramatic swings in blood pressure,
and the need for dosage adjustments for many medications.
The high mortality rate of patients with postoperative acute renal failure makes prevention a key objective in overall management. Before surgery, particularly
for those procedures capable of inducing renal ischemia,
potential risk factors such as volume depletion, hypotension, nephrotoxin exposure, and preexisting chronic kidney disease must be identified. Elective surgery should
be delayed until any abnormalities are improved. Common nephrotoxins include nonsteroidal anti-inflammatory drugs, including selective cyclooxygenase inhibitors, certain antibiotics, angiotensin-converting enzyme
inhibitors, and radiocontrast. The use of these compounds must be monitored very closely and prophylactic regimens such as N-acetylcysteine before radiocontrast administration should be used.
Trauma patients with significant soft-tissue injury are
at risk for developing acute renal failure from the myoglobin released from rhabdomyolysis. Acute renal failure should be suspected when dipstick urinalysis reveals
the presence of heme when no red blood cells are seen

140

Whenever possible, endocrine disorders should be identified and evaluated before surgery. The most common
endocrine disorder by far is diabetes mellitus. Patients
with diabetes have an increased risk of perioperative
complications including infection, metabolic and electrolyte abnormalities, and renal and cardiac complications. The stresses of surgery and anesthesia cause several hormonal changes that contribute to hyperglycemia
during and after surgery. The extent of the metabolic derangements is related to the type and length of surgery.
Several of the most significant consequences of perioperative hyperglycemia include impaired wound healing
and ability to fight infection. Patients with diabetes are
also at risk for hypoglycemia in the perioperative period. This condition may go unrecognized in the patient
under anesthesia if appropriate glucose monitoring is
not performed. Factors that contribute to the risk of hypoglycemia perioperatively include prolonged fasting,
use of oral hypoglycemic medications, inadequate nutritional state preoperatively, and postoperative gastrointestinal problems.
How a patients diabetes is managed during surgery
is dependent on several patient-specific and surgeryspecific factors. Patient-related issues include whether
treatment is with diet alone, with oral hypoglycemic
agents, or with insulin, as well as the degree of glycemic
control. Surgery-specific factors to consider are the type
of anesthesia, whether major or minor surgery is scheduled, and how long the patient is expected to take nothing by mouth. Management algorithms for perioperative
care of patients with diabetes are shown in Figures 2
and 3.

Rheumatologic Disease Risk Assessment and

Optimization
Several unique perioperative issues must be addressed
in the patient with rheumatologic disease. These issues
are related to the disease itself, as well as the drug management of many of these disease processes. Patients
with rheumatoid arthritis are unique in their high rate
of cervical disk disease. When the ligaments, bones, and
joints that maintain C1-C2 stability are eroded by synovitis, C2 can subluxate on C1 and cause spinal cord compression. Manipulating the neck for intubation during
general anesthesia stresses this joint and can result in
spinal cord injury. It is estimated that up to 25% of
rheumatoid arthritis patients have cervical disk disease,
and these patients should be evaluated preoperatively.
Rheumatoid arthritis patients are also at risk for cricoarytenoid arthritis (seen in up to 50% to 80% in some

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

series). This condition can also make intubation very
difficult and, if suspected, indirect laryngoscopy should
be performed preoperatively.
Anti-inflammatory drug and steroid use in patients
with rheumatoid arthritis must be identified preoperatively. Salicylates or nonsteroidal anti-inflammatory
drugs prolong bleeding time by inhibiting platelet aggregation. Relative risks of these medications vary with different surgical procedures, particularly if deep venous
thrombosis prophylaxis is also going to be used. If possible, salicylates and long-acting nonsteroidal medications
should be discontinued 7 to 10 days preoperatively. In
this instance, low-dose prednisone can often control inflammatory symptoms.
Patients who have taken systemic steroid medications for more than 1 week in the 6 months before surgery should be considered for stress dosing of steroids
during the perioperative period secondary to the risk of
hypothalamic-pituitary-adrenal axis suppression. Based
on the complexity/stress associated with the surgery,
dosing recommendations can range from 100 mg of hydrocortisone given preoperatively, then every 8 hours
for the first day, decreased by 50% per day, and then
discontinued by the fourth day, to one dose of 50 to
100 mg of hydrocortisone immediately preoperatively as
a single dose.

Chapter 12 Perioperative Medical Management

Perioperative Care in the Elderly

Figure 3 Management algorithm for oral hypoglycemic agents. SC indicates subcutaneously; IV, intravenous. (Reproduced with permission from Jacober SJ Sowers JR: An
update on perioperative management of diabetes. Arch Intern Med 1999;159:24052411.)

The perioperative management of the elderly has undergone major changes over the past 50 years because
there has been a dramatic population shift. The age
group 65 years and older is the fastest growing segment
of the population in the United States, expected to comprise 20% of the population by 2025.
The contribution of individual patient conditions to
surgical risk is related to a combination of physiologic
changes associated with underlying diseases, combined
to a lesser degree with age-related physiologic changes.
Research has clarified that age alone is at most a minor
risk factor for perioperative complications. Age-related
cardiovascular, pulmonary, and renal changes have to be
considered as well as recognition of altered pharmacokinetics in the elderly that can lead to an increase in
complications and toxicity.
Several postoperative complications are more common in the elderly. Delirium is a clinical syndrome in
which there is an acute disruption of attention and cognition. Orthopaedic patients, especially those with hip
fracture, may have a 28% to 60% incidence of delirium.
The development of postoperative delirium has been associated with increased morbidity and mortality, so it is
critical to identify patients who may be at risk and focus
interventions on this group. Risk factors that have been
identified include a history of drug or alcohol abuse,
preexisting cognitive dysfunction or physical impair-

ment, depression, polypharmacy, and the presence of

metabolic abnormalities. Evaluation of the elderly patient who develops delirium requires consideration of
preoperative, surgical, and postoperative factors. Preoperative risk factors identified above should be corrected
as much as possible, medication lists should be reviewed, unnecessary medications stopped, and careful
questioning about the use of over-the-counter medications and supplements should occur. Intraoperative factors associated with postoperative delirium include the
type of surgery and anesthetic used. Hip fracture surgery (especially for femoral neck fractures) is a procedure with a high risk for delirium. Anticholinergic
agents, barbiturates, and benzodiazepines may also play
a role. Intraoperative hypotension or hypoxemia also
are risk factors. Postoperative risk factors are similar to
preoperative and intraoperative risk factors with reactions to pain medications, particularly meperidine, also
playing a role. Sepsis, myocardial infarction, metabolic
abnormalities, and withdrawal from drugs or alcohol
must be considered. For a patient with preexisting impaired cognitive function, altered sensory input such as
not having eyeglasses or hearing aids available, or environmental changes can contribute to the development
of delirium.

American Academy of Orthopaedic Surgeons

141

Perioperative Medical Management

Immobility, another postoperative complication that
can be devastating in the elderly, can lead to pressure
ulcers, increased risk for osteoporosis, pulmonary risks
such as atelectasis, increased risk of aspiration and
pneumonia, increased risk of venous thromboembolic
disease, and gastrointestinal, genitourinary, and cardiovascular effects. Elderly patients require an aggressive,
multidisciplinary mobilization strategy that starts early
and continues as long as is necessary for maximal functional recovery.
In industrialized nations, the elderly are at perhaps
the greatest risk of being malnourished, and it is important to identify these patients preoperatively. Elderly
patients with preoperative malnutrition may develop
protein-calorie malnutrition from the stress of surgery.
Negative nitrogen balance depletes visceral protein
stores, leading to a loss of muscle mass and less effective
postoperative rehabilitation and ambulation. Nutritional
status should be monitored and addressed from the first
postoperative day and patients must not be allowed to
fall behind nutritionally. Voluntary food intake should
be monitored and nutritional supplements introduced
promptly if necessary. Oral supplementation is usually
adequate and parenteral feeding should be used only as
a last resort for patients with altered gastrointestinal
tract function.

Annotated Bibliography
The Preoperative Evaluation
Arozullah AM, Conde MV, Lawrence VA: Preoperative
evaluation for postoperative pulmonary complications.
Med Clin North Am 2003;87:153-174.
A review of the morbidity, mortality, and risk factors associated with postoperative complications is presented. Indications for preoperative tests and risk reduction strategies are
discussed.

Eagle KA, Berger PB, Calkins H, et al: ACC/AHA

guideline update for perioperative cardiovascular evaluation for noncardiac surgery: Executive summary.
Circulation 2002;105:1257-1267.
This article presents an overview of cardiovascular risk
stratification for patients undergoing noncardiac surgery.

Joseph AJ, Cohn SL: Perioperative care of the patient

with renal failure. Med Clin North Am 2003;87:193-210.
A discussion of recognition and elimination of risk factors
for development of acute renal failure is presented. Perioperative management of patient with end stage renal disease is reviewed.

142

Orthopaedic Knowledge Update 8

Schiff RL, Welsh GA: Perioperative evaluation and
management of the patient with endocrine dysfunction.
Med Clin North Am 2003;87:175-192.
In this article, a discussion of current strategies for the perioperative evaluations and management of patients with diabetes mellitus is presented.

Perioperative Care in the Elderly

Beliveau MM, Multach M: Perioperative care for the
elderly patient. Med Clin North Am 2003;87:273-290.
This article reviews perioperative strategies to decrease
the risk of complications in this patient population.

Classic Bibliography
Cygan R, Waitzkin H: Stopping and restarting medications in the perioperative period. J Gen Intern Med
1987;2:270-283.
Goldman L: Cardiac risks and complications of noncardiac surgery. Ann Intern Med 1983;98:504-513.
Goldman L, Lee T, Rudd P: Ten commandments for effective consultations. Arch Intern Med 1983;143:17531755.
Kellerman PS: Perioperative care of the renal patient.
Arch Intern Med 1994;154:1674-1688.
Kroenke K, Gooby-Toedt D, Jackson JL: Chronic medications in the perioperative period. South Med J 1998;
91:358-364.
Mangano DT, Layug EL, Wallace A, et al: Effect of
atenolol on mortality and cardiovascular morbidity after
noncardiac surgery. N Engl J Med 1996;335:1713-1720.
Morrison RS, Chassin MR, Siu AL: The medical consultants role in caring for patients with hip fracture. Ann
Intern Med 1998;128:1010-1020.
Schiff RL, Emanuele MA: The surgical patient with diabetes mellitus: Guidelines for management. J Gen Intern
Med 1995;10:154-161.
Smetana GW: Preoperative pulmonary evaluation.
N Engl J Med 1999;340:937-944.
Sorokin R: Management of the patient with rheumatic
diseases going to surgery. Med Clin North Am
1993;77:453-463.

American Academy of Orthopaedic Surgeons

Chapter

13

Work-Related Illnesses, Cumulative

Trauma, and Compensation
Owen J. Moy, MD
Robert H. Ablove, MD

Introduction
It is difficult to estimate the overall cost of work-related
illness. There is little doubt that the involved costs are
extraordinary. Most economic studies of work-related
illness focus on direct economic costs and disability duration. These studies place less emphasis on the significant social implications of these injuries, such as the impact on the surrounding family and community. These
costs are also often high, on both a social and economic
level.

The Workers Compensation System

The origin of todays workers compensation system in
the United States dates back to the turn of the 20th century when laws were enacted at both state and federal
levels to serve several purposes: paying workers medical expenses, compensating workers for missed pay, and
providing financial settlements for irrecoverable loss.
Although there is considerable variation from state to
state, the laws generally shield employers from charges
of negligence, thus mitigating the possibility of civil litigation.
Workers compensation tends to function as an advocacy system involving numerous parties, including the
worker and possible legal representation, the treating
physician, the employer, the employers insurance carrier, case managers, and physicians performing independent medical examinations (IMEs). These examinations
are usually performed at the request of either the insurance carrier or the patients legal representation. Unfortunately, the parties involved often have conflicting
goals.
Most patients wish to receive care and get better, although there are often some noninjury-related issues or
barriers. On occasion patients may not want to return to
work, because of issues as far ranging as a hostile work
environment to possible secondary monetary gain. Most
treating physicians desire to administer care as expeditiously as possible. Employers generally want employees
to have as little sick or compensable time away as possible, while their insurance carriers endeavor to resolve

American Academy of Orthopaedic Surgeons

cases with as little expense as possible. The IME physician is usually paid by the insurance carrier, which can
build further bias into the system. The fact that there
are so many parties with different interests can ultimately delay treatment and in some cases affect the ultimate outcome. Several basic terms necessary for a discussion of this system are presented in Table 1.

Assessment of the Injured Worker

Several parties usually perform medical assessment of
the injured worker. The treating physician evaluates the
injured worker as he or she would any other patient,
taking a thorough history of the injury, the patients occupation, and any underlying conditions.
Social and psychological factors can play a large role
in disability. Conflicts with supervisors or associates,
workplace stress, emotional problems, and substance or
alcohol abuse need to be explored and noted.
The physical examination is obviously an important
part of the evaluation. Nonphysiologic findings need to
be explored and noted as potential evidence of malingering. Waddell described a simple list of five physical
signs to help establish a distinction between organic and
nonorganic findings in patients with low back pain (Table 2). Findings of three or more positive signs are considered clinically significant.
Following a thorough physical examination, the physician formulates a treatment plan. In many cases, treatment cannot commence without permission of the insurance carrier. This is often the rate-limiting step in
providing definitive care for an injured worker.
Depending on the cost of proposed treatment, the
carrier may request an IME before initiation of nonurgent care. The physician administering that examination
is theoretically rendering an independent opinion, but is
being paid by the insurance carrier. At times, claimant
attorneys may request additional independent examinations. It is important to remember that the insurance
adjusters main job is to limit the expense to the insurance company of the administered care and that the

143

Orthopaedic Knowledge Update 8

Work-Related Illnesses, Cumulative Trauma, and Compensation

Table 1 |Basic Workers Compensation Terms

Impairment
Disability

Apportionment

Loss of use

Maximal medical
improvement

Deviation from normal function, which may be

either permanent or temporary.
A nonmedical term pertaining to the loss or
diminution of the ability to perform a function
or functions. It is possible to be impaired
without necessarily being disabled. A transcriptionist with a below-knee amputation has
an impairment but is still able to perform all
of his or her job functions. The terms are often
incorrectly used interchangeably. Disability
may be partial or total as well as temporary
or permanent. Disability determinations help
establish the amount of ongoing financial
compensation received during periods of
disability. The amount is usually a percentage
of the regular salary.
A term referring to the relative contribution of
different factors to overall disability. This applies in cases of either underlying illnesses or
prior injuries that make the current disability
materially and substantially greater than it
would otherwise be. Both treating and independent medical physicians are often asked
to assign fixed percentages to the relative
contributing factors.
Usually refers to a permanent relatively static
loss of function, usually of an extremity. This is
a distinct category from disability. In certain
states, loss of use determinations are used to
help determine the appropriate lump sum
settlement necessary to settle a case.
The point at which no further recovery from an
injury or illness is expected.

IME physician may be under pressure to render a favorable decision for the carrier.
Functional capacity evaluations are an attempt to
create an objective measurement of physical capabilities. They are a useful adjunct in determining work
readiness and are at the very least more objective than
arbitrary physician estimates. They also can provide evidence of submaximal or inconsistent patient effort,
which may be of use in detecting malingering. Unfortunately, they also provide an additional layer of expenses
to the care of injured workers.

Points of Documentation
Typically, the first step in the filing of a claim is to fill
out an incident report. However, these reports are at
times filed retrospectively after treatment has already
been sought and possibly initiated, either via out-ofpocket expense or private insurance.
A detailed, specific occupational history including a
list of current and previous occupations is paramount.

144

Table 2 | Waddells Nonorganic Physical Signs in Low

Back Pain
1. Tenderness

2. Simulation Tests

3. Distraction Tests

4. Regional
Disturbances

5. Overreaction

Tenderness related to physical disease should

be specific and localized to specific anatomic
structures.
Superficial: tenderness to pinch or light touch
over a wide area of lumbar skin.
Nonanatomic: deep tenderness over a wide
area, not localized to a specific structure.
These tests should not be uncomfortable.
Axial loading reproduction of low back pain
with vertical pressure on the skull.
Rotation reproduction of back pain when
shoulders and pelvis are passively rotated in
the same plane.
Findings that are present during physical examination and disappear at other times, particularly while the patient is distracted.
Findings inconsistent with neuroanatomy.
Motor nonanatomic voluntary release or
unexplained giving way of muscle groups.
Sensory nondermatomal sensory abnormalities.
Disproportionate verbal or physical reactions.

The occupational descriptions should include a list of

the specific activities performed for each job, as well as
their frequency and duration. It is also important to
note any recreational activity involving significant physical exertion. Obviously, any history of other claims as
well as nonwork-related injuries or illnesses should be
documented.

Interacting With the Caseworker

Many insurance companies hire caseworkers to manage
individual claims. Under ideal circumstances, they can
expedite patient care, and manage the transition from
rehabilitation to work hardening (discussed in more detail later in this chapter) to the ultimate return to work.
The caseworkers main role should be facilitating
communication between the involved parties. To be effective, the caseworker needs to be conversant in orthopaedic and ergonomic issues (see next section). Specific
ergonomic training increases caseworker effectiveness.
The caseworker can accompany the patient to office visits and learn their specific work capabilities. A job site
visit, including both an ergonomic assessment of the
workstation and a discussion with the employer regarding availability of specific light duty positions, can help
speed the return to work.
Patients need to be informed of the role of the caseworker as well as their own right to privacy. Given the
organization of most current workers compensation
systems, caseworkers can be very effective. Because

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 13 Work-Related Illnesses, Cumulative Trauma, and Compensation

the employers insurance carrier employs the caseworkers, it also is possible that bias can be introduced into
the system.

Cumulative Trauma Disorders and Ergonomics

What is the Concept?
Cumulative trauma disorders (CTDs) is a term applying
to health disorders arising from repetitive biomechanical stress. A discussion of ergonomics is central to this
topic. Ergonomics is a relatively recent word, dating to
approximately 1945, defined as a branch of ecology pertaining to human factors in machine design and operation as well as the physical environment. The Occupational Safety and Health Administrations (OSHA)
current definitions of ergonomics listed on their website
include both the science of fitting the job to the worker
and the practice of designing equipment and tasks conforming to workers capabilities.
The Occupational Safety and Health Act of 1970
states that it is the employers responsibility to provide
a workplace free from serious hazards. This includes the
prevention of ergonomic hazards. To meet these goals,
OSHA has developed a series of manuals and programs
to help employers meet workers ergonomic needs that
include a generalized framework of health and safety
programs as well as guidelines for specific industries.
These guidelines have been developed with the assistance of the National Institute for Occupational Safety
and Health, current scientific information, the medical
literature, and OSHAs own enforcement experience.
Specifically, the guidelines include a process for protecting workers as well as a means of identifying specific
problems and implementing solutions. There are also
sections on task-specific training. Appropriately, most of
the focus is on prevention.
CTDs refer to disorders from trauma occurring over
time, rather than from one specific incident. In the early
1990s there was a significant increase in the reporting of
CTDs. According to the Bureau of Labor, statistics from
that time show that CTDs represented half of the occupational illnesses reported in their annual survey. The
very term cumulative trauma disorder is fraught with
controversy because it implies a level of a presumed
knowledge regarding the etiology of certain conditions,
which does not necessarily exist. OSHA literature now
uses the less specific and therefore more benign term
musculoskeletal disorders. OSHA manuals mention low
back pain, sciatica, rotator cuff injuries, epicondylitis,
and carpal tunnel syndrome as common musculoskeletal
disorders, conceding that more needs to be learned
about the connection between the workplace and these
disorders.

American Academy of Orthopaedic Surgeons

Does This Exist and Is It an Epidemic?

There is little doubt that CTDs exist, but that does not
mean that they are easy to describe or quantify. If a
body is exposed to repetitive force over time, given
enough repetitions the body will show signs of wear.
Many CTDs have multifactorial causes and represent
changes that may have occurred over a significant time
period. Controversy exists because an immediate cause
and ultimate responsibility for CTDs can therefore be
hard to determine.
With carpal tunnel syndrome, for example, there is
support in the literature for the role of extreme wrist
position and vibration exposure in the development of
the condition, both of which are potential occupational
causes. Many nonoccupational causes exist, including
gender, obesity, tobacco use, and diabetes. Therefore, it
is difficult to ascribe causation to one particular factor.
The situation is similar with other CTDs. This issue is
paramount in workers compensation because causation
plays a key role in who ultimately pays for care.
Further epidemiologic study is necessary to estimate
relative causation in diseases with more than one risk
factor and ergonomic intervention needs to be preferentially directed toward high-risk groups. High-risk groups
include patients with multiple claims, multiple medical
problems, and prolonged absence from work.

Legal Issues in Occupational Medicine

Workers compensation is generally set up as a no-fault
system. As such, employers are legally shielded from being sued for negligence. Many of the legal issues focus
therefore on causal relationship, degree of ongoing disability, and ultimate disability settlement.
Causal relationship is the key factor that determines
who pays for care. The clinician must be careful to document any evidence of a presumed causally-related illness or injury. It is also important to document any history of prior illnesses and injuries, especially any that
resulted in missed work. Patients recreational and other
nonoccupational activities need to be carefully noted.
The duration and relative number of repetitions of both
work and nonwork activities should be dutifully recorded. If there is any possible evidence that an injury is
not causally related, an insurance carrier is likely to attempt to avoid payment for the condition. Direct communication with the insurance adjuster can sometimes
help expedite care decisions.

Assignment of Disability
Disability Rating
Because disability is not a medical term, disability rating
is not based solely on medical factors. Disability can be
partial or total as well as temporary or permanent. Certain states publish guidelines giving criteria for types of
partial disability. Ultimately, the physician makes a

145

Work-Related Illnesses, Cumulative Trauma, and Compensation

somewhat arbitrary determination of whether or not the
patient is capable of performing his or her job and to
what degree. Administrative panels or workers compensation judges often settle disputes and make final
determinations.
Disputes can arise between worker and employer
over the issue of partial disability. Insurance carriers often urge physicians to state that patients are available
for light duty and thus not totally disabled. Unfortunately, they often make these urgings in the absence of
firsthand knowledge regarding the workplace and the
availability of specific types of light duty. Another problem with so-called light duty designations is that they
are often not job-specific and only provide simple lifting
and time restrictions. Caseworker jobsite visits and communication with the physician, patient, and employer
can help resolve some of these issues.

Role and Certification of the Qualified

Medical Examiner
The theoretical role of the qualified medical examiner is
to provide an independent opinion regarding the patients illness or injury as well as the care they have received. These are most commonly requested and paid
for by the employers insurance carrier. The typical assessment usually includes a history of present illness,
past medical and surgical history, occupational and social history, record review, physical examination, and the
formulation of an opinion. Examiners are usually asked
to comment on specific issues requested by the carrier
and not to offer unsolicited opinions. Commonly, opinions are requested regarding diagnosis, causal relationship, treatment to date, further treatment recommendations, and degree of disability. In certain instances where
there have been prior injuries or underlying illnesses,
examiners may comment on apportionment.
The IME is often a rate-limiting step of workers
compensation care. Insurance adjusters commonly
refuse to authorize diagnostic testing as well as both surgical and nonsurgical care without a concurring independent opinion. If the treating and independent opinions differ, a compensation judge or administrative
panel often has to settle the dispute. These steps can delay care for many months.
There continues to be a lack of consensus regarding
how to manage workers compensation claims. Typically
and predictably, the more say patients have in their care
the more satisfied they are with the outcome. However,
this does not consistently result in measurable outcome
differences. The prevailing fear against allowing more
patient and physician autonomy is that costs will rise.
However, experience has thus far not borne out this
fear. In fact, there has been evidence that increased usage of musculoskeletal specialists tends to decrease
overall costs. Additional prospective study is necessary.

146

Orthopaedic Knowledge Update 8

One area of recent controversy has been documented reports of insurance companies altering reports
as well as requesting specific opinions and wording from
the examining physician. Presumably because of this, the
state of New York recently changed regulations regarding reports, specifying that they can only be issued after
being read and signed by the examining physician.
The certification process for qualified medical examiners is an evolving process and varies considerably
from state to state. In the state of New York, laws have
recently changed requiring that examining physicians be
board certified. Ironically, there is no such requirement
of treating physicians, allowing the possibility that a surgeon may legally operate on compensation patients, but
not be able to render independent opinions.
In addition to orthopaedic board certification, there
are other types of certification such as those by the
board of disability examiners, which enable physician
credentialing, and courses on occupational orthopaedics
and workers compensation.

Return to Work Strategies

The Maximal Medical Improvement Concept
As stated earlier, maximal medical improvement is the
point at which no further improvement is generally expected. When this occurs, the patient is evaluated to determine if there is any permanent impairment. In cases
of permanent impairment, a final settlement is made.
Different criteria exist for establishing these settlements, which can be grouped into three main categories:
functional, anatomic, and diagnostic.
The functional impairment system is based on what
effect the impairment has had on the ability to work.
The anatomic system is based on the loss of a body part.
An amputation is a 100% loss; lesser injuries are quantified via mobility, sensibility, and strength measurements
as compared with normal findings. The diagnostic system is based on the diagnosis code. All of these techniques are incorporated into the Fifth Edition of the
American Medical Association Guides to the Evaluation
of Permanent Impairment.
Generally, most states rely on some form of the anatomic system to determine final settlement sums. The
state of New York divides final determinations into loss
of use and disability. In static cases of extremity injuries
in which patients are not completely disabled and further dramatic deterioration in function is not expected,
a percentage loss of use is determined based on the involved body part. The award is then based on that percentage as well as the occupation and income level of
the injured worker.
Patients with neck and back problems as well as extremity injuries involving ongoing significant disability,
particularly where further deterioration is expected,

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 13 Work-Related Illnesses, Cumulative Trauma, and Compensation

may be classified as permanently disabled on either a

partial or total basis.
Wisconsin uses a modification of the no-fault system.
Awards can be increased or decreased in cases of unsafe
work places and unsafe work practices, respectively.
Minnesota is one of the only states to rely on diagnosis codes to determine impairment, whereas California is one of the only states to rely on functional determination. Regardless of the system, one must remain
cognizant of the distinction between disability and impairment and make a determination based on the guidelines of the particular system.

Work Hardening: How Does This Work?

Work hardening is an intermediate step between therapy and return to work. It has evolved over the past two
decades as an outgrowth of traditional rehabilitation. A
work-like environment is created for the recovering
worker, who is then able to build both physical stamina
and psychological confidence by replicating work activities in a controlled environment.
In this unique setting, the therapist provides
occupation-specific ergonomic instruction, hopefully
minimizing risk factors for re-injury upon return to
work. Although additional prospective controlled efficacy studies of work hardening would be helpful, there
already exists significant documented evidence of its
benefits.

Summary
The intent of the workers compensation system is admirable, namely treating and compensating workers for
work-related injuries. Ultimately that intent is usually
realized. Most of the patient and physician dissatisfaction with the system stems from the mode and timing of
rendered care. Unfortunately, most state systems are
cumbersome with multiple layers of administration,
leading simultaneously to delayed care delivery and increased overall costs.
Several states have experimented with different
models of administered care. Successful strategies include more integrated care, increased physician autonomy, increased specialist involvement, and use of welltrained case managers. Additional study is necessary.
Hopefully, in the coming years, some of these strategies
can be implemented on a larger scale.

Annotated Bibliography
Atcheson SG, Brunner RL, Greenwald EJ, Rivera VG,
Cox JC, Bigos SJ: Paying doctors more: Use of musculoskeletal specialists and increased physicians pay to decrease workers compensation cost. J Occup Environ
Med 2001;43:672-679.

American Academy of Orthopaedic Surgeons

This article studies the effect of using visiting musculoskeletal specialists to assist primary care physicians. Claim costs
were significantly lower in this system than in a typical workers compensation managed care model.

Bernacki EJ, Tsai SP: Ten years experience using an integrated workers compensation management system to
control workers compensation costs. J Occup Environ
Med 2003;45:508-516.
This article examines the positive effect of an integrative
approach to managing workers compensation claims. It presents 10 years of experience using a small group of health care
providers who address both physical and psychological needs.
It also illustrates the importance of maintaining open lines of
communication between all involved parties.

Durand MJ, Loisel P, Hong QN, Charpentier N: Helping

clinicians in work disability prevention: The work disability diagnosis interview. J Occup Rehabil 2002;12:191204.
This article illuminates the multifactorial nature of musculoskeletal disorders. It describes the development of a work
disability diagnosis interview as a means of helping detect
prognostic factors in musculoskeletal pain patients.

Gross DP, Battie MC: Reliability of safe maximum lifting determinations of a functional capacity evaluation.
Phys Ther 2002;82:264-271.
Functional capacity evaluations are evaluated for both interrater and test-retest reliability. Overall, functional capacity
evaluations demonstrated excellent interrater and good testretest reproducibility. The greatest source of variability was inconsistent patient performance.

Harper JD: Determining foot and ankle impairments by

the AMA Fifth Edition guides. Foot Ankle Clin 2002;7:
291-303.
This article describes how to determine impairments based
on the American Medical Association Fifth Edition guide. It
offers a good discussion of the different techniques of determining impairment and the difficulty of accounting for pain as
part of this determination.

Lincoln AE, Feuerstein M, Shaw WS: MillerVI: Impact

of case manager training on worksite accommodations
in workers compensation claimants with upper extremity disorders. J Occup Environ Med 2002;44:237-245.
This is a case control study that illustrates the effect of a
two-day training course on the ability of nurse case managers
to design and implement workplace changes. The trained
nurses offered more specific accommodations and modifications whereas the untrained nurses more often recommended
light duty and lifting restrictions.

Schonstein E, Kenny DT, Keating J, Koes BW: Work

conditioning, work hardening and functional restoration

147

Work-Related Illnesses, Cumulative Trauma, and Compensation

for workers with back and neck pain, Cochrane Database System Review (on CD-ROM). 2003;(1)
CD001822.
This article compares the effectiveness of management
strategies for return to work in patients with back and neck
pain. Patients with chronic back pain benefited from physical
conditioning, even when non-job specific, having a reduced
number of sick days compared with patients who did not undergo physical conditioning.

US Department of Labor: Guidelines for Nursing

Homes: Ergonomics for the Prevention of Musculoskeletal Disorders. OSHA 2003;3182.
This manual is a one of a series for different industries
commissioned by Elaine Chao, United States Secretary of Labor, as part of an overall strategy to reduce ergonomic injuries. It provides both a process for protecting workers and
identifies solutions for specific work-related problems.

US Department of Labor: All About OSHA Occupational Safety and Health Administration. OSHA 2003;
2056-07R.
This manual describes the origin and function of OSHA. It
also provides a concise summary of state programs, standards
and guidance, as well as available programs and services.

Wickizer TM, Franklin G, Plaeger-Brockway R, Mootz

RD: Improving the quality of workers compensation
health care delivery: The Washington State Occupational Health Services Project. Milbank Q 2001;79:5-33.
This is a summary of research and policy adopted in Washington State over recent years to identify and correct problems resulting in poor care and increased disability for injured
workers. It points to a lack of integrative occupational health
services as the most powerful predictor of adverse results.

148

Orthopaedic Knowledge Update 8

Classic Bibliography
Cocchiarella L, Anderson GBJ (eds): Guides to the
Evaluation of Permanent Impairment, ed 5. Chicago, IL,
American Medical Association, 2000.
Creighton J: Workers compensation and job disability,
in Peimer CA (ed): Surgery of the Hand and Upper Extremity. McGraw Hill, 1996, pp 2345-2351.
Gordon SL, Blair SJ, Fine LJ (eds): Repetitive Motion
Disorders of the Upper Extremity. Rosemont, IL, American Academy of Orthopedic Surgeons, 1995.
Lepping V: Work hardening: A valuable resource for the
occupational health nurse. AAOHN J 1990;38:313-317.
Lieber SJ, Rudy TE, Boston JR: Effects of body mechanics training on performance of repetitive lifting. Am
J Occup Ther 2000;54:166-175.
Luck JV Jr, Florence DW: A brief history and comparative analysis of disability systems and impairment rating
systems. Orthop Clin North Am 1988;19:839-844.
Norris CR: Understanding Workers Compensation
Law. Hand Clin 1993;9:231-239.
US Department of Labor: Ergonomics Program Management Guidelines for Meatpacking Plants. (OSHA)
1993;3123.
Waddell G, McCulloch JA, Kummel E, Venner RM:
Nonorganic physical sign in low back pain. Spine
1980;5:117-125.

American Academy of Orthopaedic Surgeons

Chapter

14

Medical Care of Athletes

Mark Anthony Duca, MD

Introduction

The Preparticipation Evaluation

The physicians role in the medical care of athletes has

become increasingly complex as the understanding of
physiology and its application to athletic performance
expands. With ever-developing technologies and ongoing research to improve the care of the athlete, the responsibilities of the team physician have grown rapidly.
The care of an individual athlete often requires a
multidisciplinary approach. Expert opinions may be required from many medical specialties, and it is the role
of the team physician to serve as the gatekeeper and liaison between the athlete and the health care system.
Medical care of athletes now not only requires an
ability to assess and diagnose but a unique skill in coordinating an overall game plan for their management.
The role of the team physician has evolved from one in
which preparticipation qualification and game-time injury management were sole responsibilities to one in
which coordination of medical expertise with other
health-related professionals such as medical specialists,
athletic trainers, nutritionists, pharmacists, physical therapists, and psychologists is essential. This dialog leads to
the safe participation of the athlete and maximizing of
their potential.
The medical management of the athlete not only requires the ability to coordinate the preparticipation assessment with management of on-the-field injuries, but
also the skill to outline a plan for rehabilitation and return to participation after an injury or illness.
The team physician must also be adept with proper
documentation when it comes to the care of athletes
and equally skilled at communicating treatment plans
with coaches, administrators, media, agents, and family
members. The team physician must also communicate
with others in a way that does not violate patient confidentiality and is in strict compliance with Health Insurance Portability and Accountability Act guidelines.
Athletic care from a medical standpoint, whether it
be for an elite, recreational, or nonprofessional athlete,
requires a detailed understanding of basic principles and
standards of care.

The goal of the preparticipation evaluation, defined as

the assessment of an individual athletes qualification to
compete in a particular sport or activity, is multifaceted.
The preparticipation evaluation must be comprehensive,
yet at the same time sport-specific and focused, in an attempt to identify any illness or conditions that predispose the athlete to injury, assess appropriateness for
participation, and identify athletes at risk for specific,
sport-related injuries. For example, a detailed ocular and
visual acuity examination would be much more important to the competitive archer than the Olympic swimmer; similar sports-specific factors should to be taken
into consideration when preparing for the evaluation.
The timing and frequency of the preparticipation
evaluation should be completed at least 4 to 6 weeks before competition. The optimal frequency of the examination has been debated in the literature. Annual examinations with health maintenance and training
counseling have been advocated. Another approach has
been a baseline examination at any new level of competition followed by interval history review at the beginning of each new season. The examination may need to
be done much more frequently, particularly with athletes who participate in multiple sports throughout the
year.
The examination is divided into the history and
physical. Patient history is essential in identifying athletes at risk for a particular injury. A standardized,
health-history questionnaire, completed by the athlete
and reviewed by a health-related professional, can be
very helpful (Figure 1). This tool can be modified for
specific sports and improve the efficiency of the evaluation.
The physical examination is an assessment of overall
general health, with specific focus on organ systems pertinent to the athletes history or central to safe participation in their particular sport. Vital information including height, weight, blood pressure, heart and respiratory
rates, and more recently, body mass index, should be recorded for every potential participant, and common

American Academy of Orthopaedic Surgeons

149

Medical Care of Athletes

Orthopaedic Knowledge Update 8

Figure 1 Sample preparticipation examination health questionnaire. (Reproduced with permission from Fields KB: Preparticipation evaluation of the school athlete, in Richmond JC, Shahaday EJ, Fields KB (eds): Sports Medicine for Primary Care. Ann Arbor, MI, Blackwell Science, 1996, p 68.)

150

American Academy of Orthopaedic Surgeons

Chapter 14 Medical Care of Athletes

Orthopaedic Knowledge Update 8

Table 1 | Classification of Sports by Contact

Table 2 | Medical Conditions Affecting Sports Participation

Contact
Noncontact
Collision/Contact Limited Contact Strenuous
Nonstrenuous

Conditions Requiring Clearance Before Sports Participation

Basketball
Field hockey
Football
Ice hockey
Lacrosse
Martial arts
Soccer
Water polo
Wrestling

Baseball
Cycling
Diving
Fencing
Field
High jump
Pole vault
Gymnastics
Raquetball
Skating
Skiing
Softball
Volleyball

Dancing
Discus
Javelin
Shot put
Rowing
Running/Cross
country
Strength training
Swimming
Tennis
Track

Archery
Bowling
Golf
Rifle

general health problems such as obesity and hypertension should quickly be identified. Multiple studies have
directly linked increased body mass indices with increased morbidity and mortality rates. Recent, stricter
blood pressure screening guidelines established for the
general public will no doubt affect preparticipation examinations as well.
The role of screening tests in the routine medical
evaluation of athletes remains unproven. Many studies,
looking at long-term benefits of screening modalities
such as electrocardiograms, echocardiograms, or even
urinalysis and chemistry screening profiles, have repeatedly failed to show any clear benefit. Thus, at this time
these screening tests cannot be universally recommended unless specifically indicated by the history or
physical examination.
Clearance for participation and fitness assessment is
dependent on the preparticipation physical examination. The results of the examination then must be evaluated in conjunction with the specific demands of the
sport before a final determination on clearance for participation can be made. Sport-specific requirements such
as degree of exertion, degree of contact, agility, and environmental influences play a role in determining eligibility (Tables 1 and 2).
The team physician must be skilled and knowledgeable about acting on any clearance recommendations. A
willingness to clearly communicate and document the
results of the preparticipation evaluation with the athlete must be established. It is preferable that the examining physician coordinate any follow-up required to
complete or change the clearance determination.

Conditioning and Preparation

Several basic principles of conditioning must be understood to provide medical consultation in the develop-

American Academy of Orthopaedic Surgeons

Atlantoaxial instability
Hypertension
Dysrhythmia
Heart murmur/valvular heart disease
Diabetes mellitus
Heat illness history
Hepatomegaly, splenomegaly
History of repeated concussion
Asthma
Absent/undescended testicle
One-eyed athletes or athletes with vision < 20/40 in one eye
Bleeding diathesis
Congenital heart disease
Seizure disorder
Febrile illness
Eating disorders
One-kidney athletes
Malignancy
Organ transplant recipient
Obesity
Dermatologic disorder

ment of a conditioning program. Whether it be sportspecific or for general overall fitness, concepts such as
specificity, prioritization, periodization, and work overload need to be addressed. Also, it needs to be understood that conditioning and readiness is a combined
product of overall general fitness, sport-specific fitness,
and skills specific to the individual sport. Becoming familiar with calculation of workloads for the purpose of
outlining a conditioning and fitness program using heart
rates and metabolic measurements such as maximum
oxygen consumption (MVO2) can be useful as well.
Conditioning specificity refers to the unique conditioning demands of an individual sport and the necessary adjustments an athlete has to make to accommodate those demands. Each sport has differences in
muscle groups needed, aerobic capacity, flexibility, environmental and even psychological factors. Thus, the specifics of a conditioning program are adjusted to meet
these particular demands.
Prioritization refers to different levels of emphasis
placed on certain components of a conditioning program based on the athletes preference with varying levels of input from coaches, trainers, and ultimately, physicians. Basketball players looking to improve their
jumping ability may prioritize lower extremity strengthening and agility training much in the same way a base-

151

Medical Care of Athletes

ball pitcher may emphasize upper extremity and arm
flexibility workouts.
Periodization refers to a planned variation in intensity and duration of a specific workout over a predefined duration of time. A distance runner training for
an upcoming marathon uses periodization as a conditioning method by changing the length in miles and intensity measured in speed at each training session before competition. The ultimate goal of periodization is
obtaining maximal performance based on the concept of
progressive overload, one at which conditioning is
started at moderate tolerable levels of exertion and progressively pushed to maximum exertion. Multiple studies have demonstrated this method of conditioning is superior to training techniques in which intensity and
duration of training are kept constant over a given period of time. Periodization cycles are defined by length
of time and are referred to as macrocycles, mesocycles,
or microcycles. Macrocycles generally refer to an entire
training year (season to season), mesocycles 3 to
6 months, and microcycles 3 to 6 weeks. Multiple variables can be adjusted during a periodization cycle in addition to intensity and duration of a conditioning workout. They can include variations in types of exercises,
number of repetitions of individual exercises, or length
of rest periods between exercises.
Conditioning can be further defined as a product of
aerobic capacity, strength, and flexibility. Aerobic capacity cannot only be subjectively measured by performance, but also objectively by measurement of the
MVO2 or VO2max, referred to as maximum oxygen uptake. MVO2 represents the maximum oxygen-carrying
capacity of the oxygen transport system, which drives
adenosine triphosphate (ATP) synthesis during aerobic
metabolism. The oxygen transport system allows tissue
to extract oxygen from oxygenated blood and synthesize
ATP for energy for muscle activity. A very efficient system allows for maximum performance. The MVO2 is defined as the product of heart rate (HR), stroke volume
(SV), and the difference between arterial and mixed
venous oxygen concentrations (a VO2), or: MVO2 =
HR SV a VO2.
HR times SV is referred to as the cardiac output,
whereas SV is the difference in left ventricular volume
between end diastole and end systole. Routine measurement of MVO2 is not practical because it requires the
use of a calorimeter, which may not be readily available.
Strength is an integral component to overall conditioning and fitness and is defined as the ability of a particular muscle or muscle group to perform work. Although specific strength training programs are beyond
the scope of this chapter, the concepts mentioned earlier
of specificity, periodization, and progressive work overload all apply here as well.
Flexibility training and development represents the
final component of a total conditioning program. Flexi-

152

Orthopaedic Knowledge Update 8

bility is the ability of joints and particular muscle groups
to maximize their natural range of motion without compromising strength and endurance. Certainly flexibility
can vary from athlete to athlete. Thus, no set standard
can be developed for outlining a flexibility program. A
large body of evidence suggests that improved flexibility
decreases risk for injury although few data exist to imply improved performance.
In contrast to flexibility, specific standards and
guidelines can be developed for both aerobic capacity
and strength training. For aerobic training, three methods are preferred to establish training guidelines. The
first, which is relatively easy to calculate, uses the maximum heart rate as a predictor of intensity. The suggested intensity level is expressed as a percentage of the
maximal predicted HR. The target HR for conditioning
should be 60% to 90% of the maximum HR. A more
highly competitive athlete would aim for a larger percentage of the maximum rate, whereas the daily jogger
trying to stay in shape may use a lower percentage. For
example, a 40-year-old recreational jogger may use 75%
of his maximum HR to calculate his target HR. Calculation of the maximum predicted HR is done by subtracting his age from 220. In this instance it would be 180
beats per minute (bpm); 180 is then multiplied by 0.75
for a product of 135, which now becomes a projected
target HR of 135 bpm. The second method requires a
few simple calculations. First, the maximal predicted HR
is established (example: age 40; 220 40 = 180 bpm HR
maximum). Next, the HR reserve is calculated by subtracting the resting HR from the maximal predicted
HR. The HR reserve is then multiplied by an intensity
factor based on an individual athletes goals and level of
conditioning. A beginner should use an intensity factor
in the range of 50% to 65% of HR reserve. A moderate
conditioned or recreational athlete should use 65% to
75% of HR reserve. Finally, a highly conditioned athlete
trying to maximize aerobic capacity for competition
uses 75% to 85% of HR reserve. The product of the HR
reserve and intensity factor (example: HR reserve 140
0.70 for a recreational athlete) is then added to the resting HR to calculate the target exercise HR.
A third method, which requires direct measurement
of MVO2 via calorimetry during maximal exercise stress
testing, is less practical but ultimately more accurate
than the second method. Similar to the HR method, the
oximetry method establishes the target as a percentage
of the MVO2 obtained during maximum exercise testing.
Similar intensity factors are used as guidelines for these
percentages. For example, an elite athlete should target
75% to 85% of the MVO2 as his conditioning intensity
whereas an intermediate or novice athlete would target
ranges from 65% to 75% and 55% to 65%, respectively.
A relationship can then be established between heart
rate and VO2 during exercise testing to determine a target HR for training.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Much in the way HR or MVO2 are central to developing guidelines for aerobic training, the repetition
maximum (RM) can be used for developing strengthtraining guidelines. The repetition maximum, or 1-RM,
is defined as the maximum amount of weight or resistance that can be lifted for one repetition. An RM also
refers to a specific number of repetitions limited by a
particular weight or resistance. For example, a weight
that can be lifted 10 times until fatigue would be 10RM. A preferable utilization of the RM allows the prescriber to establish an RM training zone based on the
target goal of strength training. A training zone is usually a range of three RMs (eg, 8 to 10 RM), which reduces the need to exercise to fatigue with each set.
The concept of specificity mentioned earlier can be
applied to strength training based on the goals and demands of an individual athletes sport. Four common
variables may be adjusted within the training program
to achieve target goals. The RM, number of repetitions,
number of sets, and rest period between sets can all be
adjusted to optimize power, strength, and endurance.
Exercises focusing on improvements in power use lower
RMs (ie, higher weights or resistance), fewer numbers
of repetitions, higher numbers of sets, and longer recovery times between sets. In contrast, training focusing on
muscular endurance uses higher RMs (ie, lower weights
or resistance), much higher number of repetitions, fewer
number of sets and less recovery time in between sets.
Strength training uses lower RMs than does power and
endurance training with repetition numbers, set numbers, and recovery times being similar to that of power
training.

Sideline Medicine
Sideline medicine is a phrase used to refer to the team
physicians approach to handling game-time injuries and
illnesses and developing a well-documented and organized plan for implementation at the site of competition
and practices. From a consensus statement released in
2001 regarding sideline preparedness for the team physician, the definition of sideline medicine is the identification of and planning for medical services to promote
the safety of the athlete. Goals are to limit injury and
provide medical care at the site of competition. In addition, sideline medicine deals with three integrated aspects of athletic competition: preseason planning, gameday operations, and postseason evaluation.
Preseason planning involves the use of the preparticipation evaluation to identify potential problems during
the season. The team physician must have access to any
prior relevant health information before determining eligibility for participation. Timely completion of the
preparticipation evaluation is advantageous because it
allows treatment or follow-up to be initiated before

American Academy of Orthopaedic Surgeons

Chapter 14 Medical Care of Athletes

competition, and involves review of the athletes proposed competition schedule.
Medical care of the athlete involves coordination of
administrative duties such as establishment and documentation of an emergency response plan for the seriously ill or injured competitor. Additionally, a policy to
assess playing condition and environmental factors at
the time of competition must be established, and a protocol involving administrators, competition officials,
coaches, and staff to make an educated decision regarding safe competition conditions should be clearly delineated. Finally, medical record keeping is essential in the
care of athletes and preseason planning should ensure
that this takes place.
Game-day preparations and operations involve the
assessment and management of game-day injuries and
illnesses and final determinations on clearance for participation. A review of game-time playing conditions is
conducted according to the established preseason policy
and any concerns should be promptly brought to the attention of coaches, officials, and relevant staff. A wellprepared medical staff also needs to be familiar with
medical equipment and examination and treatment facilities available. When traveling, game-time review of
locations of examination areas, x-ray equipment, ambulances, and local hospitals with the hometown medical
staff is essential to ensure a quick and efficient response
to any athletic injuries. The team physician also needs to
have a clear vantage point of the competition and easy
access to the athletes competing. Return to competition
decisions are a part of the game-day responsibilities.
Evaluation of concussions, soft-tissue injuries, and eye
trauma are a few examples where return to play decisions are made by the team physician during competition. Game-day assessment also requires a postcompetition review of all injuries and illnesses that have
developed and an action plan to ensure necessary
follow-up. The postcompetition assessment should involve the team physician, trainers, coaches, and relevant
administrators.
The postseason evaluation involves a comprehensive
summary of all athletic injuries and illnesses acquired
during competition. It provides a good opportunity to
collect data and identify trends during recent competition. It also is the time for the team physician to coordinate appropriate follow-up during the off-season.

Dehydration and Heat Syndromes

Dehydration is a loss of all body water caused by decreased intake or increased losses by evaporation and
excretion. Prolonged dehydration can lead to tissue dysfunction and eventual hemodynamic collapse as the circulating intravascular volume drops to a critical level
sufficient to lower cardiac output. As cardiac output decreases, so does blood supply to the skin, resulting in de-

153

Orthopaedic Knowledge Update 8

Medical Care of Athletes

Table 3 | Participants at Risk for Dehydration/Heatstroke

Table 4 | Total Body Water/Na+ Deficit Calculations

General
Poorly acclimated or inexperienced competitors
Obese or poorly conditioned
Elderly
Prior history of dehydration/heatstroke
Medication Usage
Antihistamines
Anticholinergics
Diuretics
Neuroleptics
Illness
Acute febrile illness
Recent vomiting or diarrhea
Uncontrolled systemic condition (for example, diabetes mellitus or
hypertension)

Calculation of Total Body Water

Deficit
[Na+]
TBW = 0.6 (wt in kg) x ( 140 1)

creased heat dissipation and a subsequent rise in core

body temperature. At extremes of dehydration, sweating
mechanisms cease in an attempt to preserve intravascular volume. This in turn can lead to further increases in
core body temperature, which left unchecked can be fatal. As little as a 1% drop in total body water has been
shown to affect performance.
Athletes are particularly at risk for dehydration and
heat illnesses based on the increased demands of competition. It is not unusual for competitive athletes to
lose as much as 1 to 2 L of sweat per hour with vigorous, intense exercise. Athletic participants at risk for dehydration and heatstroke are outlined in Table 3.
Aggressive volume replacement is essential during
exercise. Rehydration also requires electrolyte replacement in addition to volume. Although electrolyte concentrations are highly variable, sodium is the cation excreted in the highest concentration, with lower
concentrations of potassium usually found. Both electrolytes should be replaced, although the necessity is
lessened in low-intensity or short-duration exercise and
in those who consume a normal diet. At least 16 oz of
fluid supplemented with sodium and potassium should
be ingested for every pound of weight lost with exercise.
Hypotonic salt, or isotonic, carbohydrate supplemented
beverages referred to as sports drinks are effective rehydration solutions for palatability and rapid effective
absorption. The presence of increased ingested sodium
and glucose in the intestinal lumen promotes their
cotransport for absorption in the small intestine. This active transport creates an osmotic gradient that enhances
net water absorption by passive mechanisms. Plain water ingestion can be effective for maintaining hydration,
but rehydration after exercise can be slowed by the
drop in serum osmolality it may create. If allowed to
progress, replacement of sweat losses with hypotonic
fluid (water) leads to hyponatremia. Significant de-

154

Calculation of Na+ Deficit

Meq Na+ deficit = 0.6 (wt in kg)
(140 [Na+]) + (140) x (vol def in
L)

TBW = total body water; [Na+] = measured serum sodium concentration (mEq/L)

Table 5 | Clinical Estimates of Degree of Dehydration

% Dehydration (% wt)

Clinical Signs/Symptoms

3% to 5% (Mild to Moderate)

Orthostasis
Thirst
Decreased voiding
Dry mucous membranes
Reduced skin turgor
Dry axillary folds
All of the above plus:
Supine hypotension
Lethargy
Tachycardia
Tachypnea
Hemodynamic collapse

8% to 10% (Moderate to Severe)

12% to 15% (Severe)

creases in serum sodium levels can produce fatigue, lethargy, weakness, confusion, and even hemodynamic collapse and death. Salt loading is not recommended
because of the plasma hypertonicity it may produce. Tables 4 through 7 serve as a brief guide for fluid and electrolyte replacement.
Thirst is a poor indicator of hydration status in humans and should not be used alone as a guide for volume replacement. Measurement of the weight of the
athlete while unclothed, both before and after competition, is an excellent way to monitor fluid losses. Hydration before, during, and after exercise should be emphasized.

Overuse Injuries
Overuse injuries refer to musculoskeletal maladies that
develop as a result of environmental, biomechanical,
and equipment factors. They involve one of the following anatomic structures: bursae, tendons, bones, joints,
and ligaments. Overuse injuries are often classified into
four stages based on degree of pain. Stage 1 is mild pain
that develops only after activity. Stage 2 is moderate
pain that occurs during activity but does not restrict or
interfere with performance. Stage 3 is moderate to severe pain that interferes with performance. Stage 4 is
the most severe form of an overuse injury with signifi-

American Academy of Orthopaedic Surgeons

Chapter 14 Medical Care of Athletes

Orthopaedic Knowledge Update 8

Table 6 | Approximate Deficits in Mild to Moderate

Dehydration (5%)
Type of Dehydration
Isotonic
[Na+] 134-145 meq/L
Hypertonic
[Na+] >145 meq/L
Hypotonic
[Na+] <134 meq/L

[Na+] meq/kg

[K+] meq/kg

4-5

4-5

0-2

0-2

5-6

5-6

*Deficits are doubled for moderate-severe dehydration (10%)

cant pain even at rest. Table 8 outlines common overuse

injuries.

Medical Care of the Female Athlete

Providing optimal medical care for the female athlete
requires an understanding of unique gender-specific issues. A comprehensive review of these topics is beyond
the scope of this chapter but topics such as exercise in
pregnancy, postmenopausal fitness, the female triad (altered menstrual cycle, eating disorder, abnormal bone
metabolism), and athletic injuries seen disproportionately in female competitors will be briefly reviewed.
Exercise during pregnancy is a vital part of the overall health of the mother and fetus. Recent studies have
shown a decreased morbidity and mortality rate among
expectant mothers who established a regular prepartum
exercise regimen. Normal physiologic changes during
pregnancy such as increased weight, increased intravascular volume, increased total body water content, and
low back pain as a result of anatomic changes may
make exercise more challenging during pregnancy. However, the intensity and frequency of exercise can be relatively maintained at prepregnancy levels until late in the
pregnancy. Several modifications should be made. Pregnant women who exercise should be keenly aware of
the propensity for thermal dysregulation and should
make extra effort to ensure adequate hydration and
provide an exercise environment that is optimal for adequate heat exchange. Pregnant women should be aware
of increased caloric requirements, roughly 300 to 400
kcal/day, during pregnancy and adjust diet accordingly.
Furthermore, because of diminished maternal aerobic
capacity, the intensity of exercise may need to be modified. Exercise to exhaustion, even for elite athletes, is to
be discouraged during pregnancy and maternal perceived exertion should be used as a guide for intensity
levels. Certain specific types of exercises during pregnancy should be avoided. Any type of exercise with a
chance of abdominal contact should be avoided. In general, complex exercises associated with a propensity for

American Academy of Orthopaedic Surgeons

Table 7 | Guide to Fluid/Electrolyte Replacement

Isotonic Dehydration: Serum [Na+]134-145 meq/L
Net loss of solute and electrolytes equal to water loss
Ex: Common; healthy athlete with inadequate hydration
Tx: Replace sum of maintenance losses (30-35 mL/kg body weight),
insensible losses (600-800 mL/day) and estimated water/electrolyte
losses over first 24 h; replace 1/2 of volume within first 8 h, rest over
next 16 h
Hypertonic Dehydration [Na+] >145 meq/L
Net loss of water greater than solute
Ex: Inadequate hydration with strenuous exercise
Tx: Repair TBW deficit over 48 h; rate of reduction of serum Na+ not to
exceed 10 meq per/24 h
Hypotonic Dehydration [Na+] < 134meq/L
Net loss of solute/electrolytes less than water
Ex: Unusual; seen in ultraendurance competitors who overhydrate
Tx: Do not replace more than one half total sodium deficit within first
24 h to avoid central pontine myelinolysis
3% NS used in rare situations with seizure
[Na+] of NS 154 meq/L [Na+] of NS 77 meq/L [Na+] of 3% NS 513
meq/L
Ex = example; Tx = treatment; TBW = total body water; NS = normal saline

falls should be avoided. Because of decreased venous

return and a subsequent drop in cardiac output, exercise
requiring prolonged, stationary standing after the first
trimester should be avoided. It is also important to remember that physiologic changes may not occur for 6 to
10 weeks postpartum and that prepregnancy exercise
practices should not be started immediately postpartum.
The benefits of regular exercise during pregnancy
are numerous and include maintenance of a healthy
weight and prevention of excess weight gain, more rapid
postpartum recovery, improved posture and fewer musculoskeletal problems such as low back pain. Less peripheral edema, improved sleep quality, improved energy levels, and an improved overall sense of well-being
contribute to a positive experience during the pregnancy. However, every physician supervising an exercise
program during pregnancy should be aware of absolute
and relative contraindications to exercise. Absolute contraindications include preexisting uncontrolled hypertension, diabetes mellitus, or valvular heart disease. Obstetrical specific contraindications include preeclampsia
or pregnancy-induced hypertension, preterm labor with
the current or previous pregnancy, premature rupture of
membranes, second or third trimester bleeding, or incompetent cervix. Relative contraindications to exercise
during pregnancy include history of multiple miscarriages, precipitous labor, breech fetal positioning, and
multiple gestations.
The benefits of exercise in the postmenopausal female are just as clear. A regular aerobic exercise program

155

Orthopaedic Knowledge Update 8

Medical Care of Athletes

Table 8 | Common Overuse Syndromes

Structures Involved

Symptoms

Treatment

Bursae
Prepatellar bursitis

Anterior/inferior knee pain and swelling

Inferior knee pain

Protective padding, ice, NSAIDs, aspiration,

corticosteriod injection
Rest, ice, PT, local injection
Protective padding, ice, NSAIDs, aspiration,
corticosteroid injection
Ice, NSAIDs, aspiration, corticosteroid injection

Nonlocalized shoulder pain

Localized anterior shoulder pain
Knee pain
Heel pain
Shin pain
Focal medial/lateral elbow pain

PT, rest, NSAIDs, corticosteroid injection

Corticosteroid injection, rest, NSAIDs
PT, rest, NSAIDs
PT, rest, NSAIDs
PT, rest, NSAIDs
Corticosteroid injection, PT, rest, NSAIDs

Localized pain

Rest, correction of precipitating factors

Heel, instep pain

PT, orthoses, splinting, ECSWT

Greater trochanteric bursitis

Olecranon bursitis
Pes anserine bursitis
Tendon
Rotator cuff tendinitis
Biceps tendinitis
Patellar tendinitis
Achilles tendinitis
Anterior tibialis tendinitis
Medial/lateral epicondylitis
Bones
Stress fractures
Insufficiency fractures
Apophysitis
Ligaments
Plantar fasciitis

Focal, lateral hip pain

Elbow pain, swelling

NSAIDs = nonsteroidal anti-inflammatory drugs; PT = physical therapy; ECSWT = extracorporeal shock wave therapy

of intermediate intensity most days of the week combined with strength and flexibility exercises is essential to
the health of the postmenopausal athlete. Improvements
in bone density measurements, lipid profiles, body mass
indices, and sleep quality, all problematic in postmenopausal females, are well established with regular conditioning.
The female athlete triad consists of abnormal bone
metabolism, disordered eating, and abnormal menstrual
function. The exact incidence of the disorder is unknown
but several recent studies suggest that it may be as high
as 12% of all competitive female athletes. Prevalence
rates differ among individual components of the triad existing independently and also differ among individual
sports. A prevalence rate of nearly 50% for menstrual
dysfunction has been reported in some distance runners
compared with a general prevalence rate of 2% to 5%.
Menstrual dysfunction comprises an entire spectrum
of irregularity of menstruation. From delayed onset of
menarche, defined as onset of menses after age 16 years,
to complete amenorrhea, defined as the absence of 3 to
12 consecutive menstrual periods in the absence of
pregnancy, to oligomenorrhea (irregular, infrequent
menses), menstrual dysfunction is common in the female athlete. Females at risk tend to be younger, nulliparous, have lower body weights, and are involved with
more strenuous, high-intensity sports. A poor dietary

156

history is often central to the development of menstrual

irregularities. Menstrual dysfunction is believed to be
caused by a lack of a pulsatile release of gonadotropinreleasing hormone from the hypothalamus, which in
turn leads to diminished release of luteinizing and
follicle-stimulating hormone from the anterior pituitary
gland. The exact cause of diminished gonadotropinreleasing hormone release is unclear but the role of circulating -endorphins, cortisol, catecholamines, melatonin, and androgens are all currently under investigation.
Treatment options include optimizing diet and nutritional status, exercise intensity modification, and hormone therapy.
Eating disorders encompass everything from restriction before competition to frank starvation or binging
and purging, referred to as anorexia nervosa and bulimia nervosa, respectively. Signs and symptoms include
a characteristically thin body habitus, thin hair and hirsute features, parotid gland hypertrophy, dental caries,
subconjunctival hemorrhages, scarring on the dorsum of
the hands, and rectal tears. Treatment depends on the
severity of the condition and often involves a multispecialty approach including physicians, mental health specialists, and nutritionists. The exact prevalence of eating
disorders among female athletes is unknown.
Regular exercise in females leads to increased bone
density by increasing mechanical stress on bone. A low

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

estrogen state seen with delayed or infrequent menarche, as is the case with the female triad, leads to poor
primary bone mass accumulation and the early development of osteoporosis. If allowed to continue, the hypoestrogenic state promotes premature bone resorption,
further compounding the problem. Combined with poor
nutrition status, in particular a deficiency in calcium and
vitamin D, the rate of bone loss can be startling. Several
studies have recorded bone densities of female athletes
in their 20s that were comparable to those of females in
their 60s and 70s.
Although not exclusive to female athletes, several
common musculoskeletal athletic injuries are often seen
in females. Two ligamentous injuries more commonly
seen in females include ankle sprains (particularly anterior talofibular) and anterior cruciate sprains and complete tears. Women are particularly vulnerable to anterior cruciate injuries but the exact cause is believed to
be multifactorial. Hormonally-mediated ligamentous hyperlaxity, decreased ligament size, and a narrowed intercondylar notch have all been proposed as possible etiologic factors. Ankle sprains seem to be more easily
explained because of the relatively reduced stability of
the talus from an inflexible heel cord and the propensity
of the female foot to favor a plantar-flexed position.
Lateral epicondylitis and rotator cuff tendinitis tend
to be more prevalent in females, who are believed to be
more susceptible to these injuries because of postural
differences produced by strength deficiencies and softtissue laxity.
Overuse injuries of bone, particularly stress fractures, are more commonly encountered in females who
comprise the classic triad. Healthy female athletes also
are at increased risk because of hormonal factors and
lower muscle to body mass ratios. Spondylolysis, a stress
fracture to the pars interarticularis of the lumbar spine,
is also seen more frequently in female athletes.
Mechanical low back pain is ubiquitous among females; the etiology is usually multifactorial. Two unique
sources of back pain among women are the aforementioned spondylolysis and pelvic instability produced by
sacroiliac joint dysfunction. Hormonal and mechanical
(particularly the effects of childbirth) factors lead to increased laxity among the ligamentous, muscular, and
joint structures that comprise the pelvic ring. This increased laxity leads to a propensity for subluxation, particularly anteriorly, that leads to pelvic ring asymmetry
and subsequent back pain.
Patellofemoral syndrome is another athletic injury
common in females. The athlete has anterior deep knee
pain with varying degrees of swelling and pain that often is noted to be worsened by walking down steps. The
increased propensity in females is believed to be the result of increased ligamentous laxity leading to abnormal
lateral patellar tracking. A lack of flexibility and weakness of the hamstrings and vastus medialis obliquus also

American Academy of Orthopaedic Surgeons

Chapter 14 Medical Care of Athletes

contribute to abnormal patellar tracking. A wider pelvis
and increased Q angles have also been proposed as possible etiologic factors. A propensity for plantar flexion
leading to internal torsion of the tibia and femur accentuates lateral patellar tracking and is also believed to be
a contributing factor. Treatment consists of biomechanical modification and physical therapy modalities.

Annotated Bibliography
Preparticipation Examination
Chobanian AV, Bakris GL, Black HR, et al: The Seventh
Report of the Joint National Committee on Prevention,
Detection, Evaluation, and Treatment of High Blood
Pressure: The JNC 7 report. JAMA 2003;289:2560-2572.
This article presents new guidelines for hypertension management and prevention.

Carek PJ, Hunter L: The preparticipation physical examination for athletics: A critical review of current recommendations. J Med Liban 2001;49:292-297.
A critical review of current recommendations for the
preparticipation examination is presented.

Seto CK: Preparticipation cardiovascular screening. Clin

Sports Med 2003;22:23-35.
This review article advocates the standardization of the
preparticipation examination.

Conditioning and Preparation

American Academy of Family Physicians, American
Academy of Orthopaedic Surgeons, American College
of Sports Medicine, American Orthopaedic Society for
Sports Medicine, American Osteopathic Academy for
Sports Medicine, American Medical Society for Sports
Medicine: The team physician and conditioning of athletes for sports: A consensus statement. Med Sci Sports
Exerc 2001;33:1789-1793.
This article presents a governing body consensus statement on conditioning principles for athletes.

Ebben WP, Blackard DO: Strength and conditioning

practices of National Football League strength and conditioning coaches. J Strength Cond Res 2001;15:48-58.
A review of conditioning practices in the National Football League reveals 69% of conditioning coaches follow a periodization model.

Kraemer WJ, Knuttgen HG: Strength training basics:

Designing workouts to meet patients goals. Phys Sports
Med 2003;31:39-45.
Human power production capabilities and differences of
physiologic response to varying intensities of exercise are outlined.

157

Medical Care of Athletes

Orthopaedic Knowledge Update 8

Sideline Medicine

Medical Care of the Female Athlete

Mellion M, Walsh WM, Shelton G: The Team Physicians

Handbook, ed 3. Philadelphia, PA, Hanley & Belfus,
2001, pp 126-135.

Brown W: The benefits of physical activity during pregnancy. J Sci Med Sport 2002;5:37-45.

This book chapter presents an outline for game day management for the team physician.

Stricker PR: The sports medicine kit: Basics of the bag.

Pediatr Ann 2002;31:14-16.
A review of essential on-site materials for the team physician is presented.

Dehydration and Heat Syndromes

Burke LM: Nutritional needs for exercise in the heat.
Comp Biochem Physiol A Mol Integr Physiol 2001;128:
735-748.
This article reviews the necessity for carbohydrate, volume, and electrolyte replacement for vigorous exercise.

Overuse Injuries
Maier M, Steinborn M, Schmitz C, et al: Extracorporeal
shock-wave therapy for chronic lateral tennis elbow:
Prediction of outcome by imaging. Arch Orthop Trauma
Surg 2001;121:379-384.
Forty-two patients were assessed before and after extracorporeal shock wave therapy; 84% of men and 52% and of
women showed a good outcome by MRI at 18 months. Tendons that were thickened and swollen were most likely to respond.

McFarlane D: Current views on the diagnosis and treatment of upper limb overuse syndromes. Ergonomics
2002;45:732-735.
Treatment remains focused on modification of biomechanical factors that precipitate these injuries.

Panni AS, Biedert RM, Maffulli N, Tartarone M, Romanini E: Overuse injuries of the extensor mechanism in
athletes. Clin Sports Med 2002;21:483-498.
This article reviews the functional anatomy, pathophysiology, and overall management of overuse injuries of the extensor mechanism in athletes.

158

Maintenance of a regular exercise program during pregnancy leads to improved maternal-fetal outcomes.

Burrows M, Nevill AM, Bird S, Simpson D: Physiological factors associated with low bone mineral density in
female endurance runners. Br J Sports Med 2003;37:
67-71.
This article provides examples of very low bone mineral
densities in a sample of 52 female endurance runners.

Classic Bibliography
American College of Sports Medicine Position Stand:
The recommended quantity and quality of exercise for
developing and maintaining cardiorespiratory and muscular fitness, and flexibility in healthy adults. Med Sci
Sports Exerc 1998;30:975-991.
Anderson SD, Griesemer BA: Medical concerns in the
female athlete. Pediatrics 2000;106:610-613.
Buettner CM: The team physicians bag. Clin Sports
Med 1998;17:365-373.
Clapp JF III: Exercise during pregnancy. Clin Sports
Med 2000;19:273-286.
Eichner ER: Treatment of suspected heat illness. Int J
Sports Med 1998;19(suppl 2):S150-S153.
Grafe MW, Paul GR, Foster TE: The preparticipation
sports examination for high school and college athletes.
Clin Sports Med 1997;16:569-587.
Kurowski K, Chandran S: The preparticipation athletic
evaluation. Am Fam Physician 2000;61:2683-2690.
Shirreffs SM, Maughan RJ: Rehydration and recovery
of fluid balance after exercise. Exerc Sport Sci Rev 2000;
28:27-32.
Warren MP, Shantha S: The female athlete. Baillieres
Best Prac Res Clin Endocrinol Metab 2000;14:37-53.

American Academy of Orthopaedic Surgeons

Chapter

15

The Polytrauma Patient

Hargovind DeWal, MD
Robert McLain, MD

Introduction
The management of the multiply injured or polytrauma
patient requires a multidisciplinary approach integrating
organ- and injury-specific treatment protocols. Multiple
traumainjury to multiple organ systemscan directly
or indirectly trigger processes that may injure specific
organs, disrupt metabolic processes, interrupt normal
endocrine function, create hemodynamic and physiologic instability, and lead to highly lethal systemic diseases and multiple organ failure. By definition, multiple
trauma is a life-threatening disorder.
Successful management of the polytrauma patient
requires a team approach and a broad focus. Within a
few days of injury, the polytrauma patient will be having
or be at risk for a myriad of potentially serious disorders, in addition to their actual, initial injuries. A list of
some of these disorders is found in Table 1.
The concept of a damage control approach to orthopaedic injuries is discussed in the recent literature
and should be observed to minimize the risk of compounding systemic injury through added surgical injury.
A dedicated intensivist, skilled anesthesia staff, trauma
and orthopaedic trauma surgeons, nutritional support
services, infectious disease specialists, and plastic and reconstructive surgeons may all play a role in the care of a
single patient. It is imperative that all of these individuals buy into the principles of trauma management and
communicate well with the other members of the team.
The environment for patient care must support the
level of care required. Access to diagnostic studies, interventions, line care, and respiratory support must be immediate and available around the clock. Nursing staff
must understand the fragility of the patient and recognize that changes in respiratory or circulatory parameters may require immediate attention and response.
Staff must also be familiar with protocols for mobilization, deep venous thrombosis prophylaxis, and pulmonary and bowel care.
After the patients condition is stabilized, attention
to nutrition, infection control, pulmonary function, and
skin care play an often underappreciated role in healing

American Academy of Orthopaedic Surgeons

and rehabilitation. Psychological support, occupational

and physical therapy programs, and multidisciplinary
follow-up all contribute to full and timely recovery, and
improve the likelihood of a satisfactory return to function and community life.
These patients often face some likelihood of permanent impairment and long-term disability. Advances
have been made in all aspects of polytrauma care, ranging from improved prehospital care to more aggressive
resuscitation and surgical management to aggressive
physical therapy and spinal cord injury rehabilitation.
The treating physician must be conversant in all of these
areas, and keep an eye on all aspects of the patients recovery if the best outcomes are to be obtained and the
worst complications avoided.

Assessment of the Polytrauma Patient

An orderly, structured assessment of the polytrauma patient has been shown to improve care and reduce the
likelihood of missed injuries. Patients with multiple injuries typically arrive in the emergency department under
the care of another health care provider, most often a
trained emergency medical technician, who will have assessed the patient in the field, established intravenous
access, and may have intubated the patient to restore or
maintain the airway. They will provide important information on the mechanism of injury, the patients condition at the time of first contact, and evidence of neurologic function, respiratory status, and responsiveness at
the time of initial resuscitation. Their initial observations may provide important perspective as to the patients improvement or deterioration when compared
with the initial assessment in the emergency department.
Once the patient arrives in the emergency department, resuscitation and a primary assessment begin simultaneously. These two processes are interdependent
in that the purpose of the primary assessment is to find
the causes of hemodynamic instability, respiratory impairment, and circulatory collapse at the same time others on the team are trying to restore those functions

159

Orthopaedic Knowledge Update 8

The Polytrauma Patient

Table 1 | Serious Disorders for Which the Polytrauma

Patient is at Risk
Pulmonary contusions, aspiration,
pneumonia
Thromboembolic disease
Urosepsis
Anemia
Systemic hypotension and hemorrhagic shock
Renal insufficiency, acute tubular
necrosis
Immunosuppression
Malnutrition
Pancreatitis
Myonecrosis, myoglobinemia
Coagulopathy
Electrolyte disturbances
Peptic ulcer disease
Gastrointestinal disease
Decubitus ulcers

Cardiac ischemia, contusion, tamponade

Hypothermia
Open, contaminated wounds, burns
Spinal cord injury
Acute pulmonary embolism
Endocrine dysfunction
Hemothorax, pneumothorax,
pyothorax
Compartment syndrome
Multiple organ failure syndrome
Delirium
Ischemic brain injury
Septic shock
Iatrogenic injury
Neurogenic shock

through volume replacement, ventilation, and pharmacologic support. Once the primary survey is complete
and the patients condition begins to stabilize, a secondary, more complete survey is conducted, and the team
can begin to formulate a plan for definitive care.

Primary Survey
Initial management of the polytrauma patient begins
with an assessment of airway, breathing, and circulation,
along with neurologic status (disability) and environmental exposure. Advanced Trauma Life Support guidelines set forth by the American College of Surgeons advocate use of both the primary and secondary survey to
provide an orderly, consistent approach that will rapidly
reveal life- and limb-threatening injuries. The secondary
survey consists of a head-to-toe evaluation and history.
Both the primary and secondary survey should be repeated as needed to ascertain any change in the patients status. Initial radiographs should include those of
the chest, pelvis, and cervical spine, all obtained immediately after the primary survey is complete.
Airway
Assessment of the airway and breathing begins immediately, in the field. The patient must be making an effort
to breath, be successfully moving air, and be adequately
transferring oxygen to the circulating blood. Evaluation
of effort, chest wall excursion, and breath sounds should
be done immediately on arrival. The physician should
look for cyanosis and obtain an arterial blood gas sample. Mechanical obstruction should be addressed immediately, looking for loose teeth, dentures, blood, food, or

160

vomitus, and intubation performed as necessary. The arterial blood gas will assess degree of oxygenation. If oxygenation is inadequate, pulmonary function, including
tension pneumothorax, hemothorax, and flail chest,
should be reinvestigated.
Breathing: Thoracic Injuries
Signs of major thoracic injury during the primary survey, including tension pneumothorax, open pneumothorax, flail chest, massive hemothorax, and cardiac tamponade (discussed in the following section) should be
noted.
Tension pneumothorax develops as air leaks into the
chest cavity either through the chest wall or from the
lung. The air enters via a one-way valve mechanism
and does not exit the cavity. The affected lung collapses
and as air continues to build up, the mediastinum is displaced to the contralateral side, impeding venous return
and compressing the uninjured lung. The diagnosis is
made on the clinical findings of absent breath sounds
and a hyperresonant percussion note. A chest radiograph is not required before treatment is initiated.
Treatment consists of immediate decompression by insertion of a large bore needle into the second intercostal
space in the midclavicular line of the affected side, followed by chest tube placement.
Open pneumothorax results from large defects in
the chest wall. Air will preferentially enter the chest
cavity through the defect rather than the trachea when
the diaphragm contracts. Initial management includes
placement of an occlusive dressing covering the wound
edges, taped on three sides, allowing the dressing to occlude the wound with each inhalation and allowing for
air to escape during exhalation. A chest tube should be
inserted at a site away from the wound as soon as possible.
Flail chest occurs in the presence of multiple rib
fractures and is usually associated with an underlying
pulmonary contusion. The flail chest segment demonstrates paradoxical chest wall motion with inspiration
and expiration, impairing ventilation. The paradoxical
motion is not solely responsible for the associated hypoxia. Pain results in restricted chest wall motion, and
pulmonary contusion contributes significantly to development of hypoxia. Intubation and ventilation may be
necessary if hypoxia is progressive and unresponsive to
initial measures.
Massive hemothorax, the rapid accumulation of at
least 1,500 mL of blood in the chest, may be the result
of blunt or penetrating trauma. The blood loss may contribute to hypoxia, and initial management includes
both restoration of blood volume and decompression of
the chest cavity by chest tube placement. Massive hemothorax often requires thoracotomy to control the
source of hemorrhage.

American Academy of Orthopaedic Surgeons

Chapter 15 The Polytrauma Patient

Orthopaedic Knowledge Update 8

Circulation
Evaluation of circulation involves physical examination
and an assessment of vital signs including blood pressure and heart rate. Intravenous fluid infusion is recommended in all patients, and is usually started before
reaching the hospital. Resuscitation should be monitored by blood pressure, heart rate, perfusion, and urine
output. If the extremities are cold, clammy, and/or cyanotic, the patient should be treated for hypovolemia irrespective of pulse or pressure. If brisk bleeding from an
extremity or penetrating wound is encountered, direct
pressure should be applied immediately.
Goals for urine output are 0.5 mL/kg/h in adults and
1.0 mL/kg/h in children. Central venous pressure will
provide information regarding atrial-filling pressures.
Elderly patients with severe thoracic trauma require a
pulmonary artery catheter. The arterial-alveolar gradient should be calculated to detect ventilation-perfusion
mismatches. Crystalloid infusion is used in the initial
management of these patients, through large bore intravenous access. If intravenous access is not readily available, cutdown on the saphenous, femoral, or cubital
veins may be necessary.
For patients who are experiencing exsanguination, immediate use of universal donor blood group (group O,
Rh negative) is recommended. Thrombocytopenia is
treated at levels below 50,000/mL.
Cardiac tamponade may result in circulatory failure
in the face of normal blood volume. This condition usually results from penetrating injuries. The diagnosis is often difficult, and it must be distinguished from tension
pneumothorax. The classic diagnostic finding of Becks
triad consists of (1) venous pressure elevation, (2) decline in arterial pressure, and (3) muffled heart tones.
Kussmauls sign, a rise in venous pressure with spontaneous inspiration, may be present in cardiac tamponade.
An echocardiogram may aid in diagnosis, but a falsenegative result may be seen in about 5% of patients. Examination of the pericardial sac may also be performed
during a focused abdominal ultrasound. Prompt evacuation of the pericardial blood (usually by pericardiocentesis) is indicated for patients who do not respond to
usual resuscitative measures. A pericardial window, thoracotomy and pericardiotomy, may be necessary.
Head Injury
All trauma patients should receive a minineurologic examination consisting of a Glasgow Coma Scale (GCS)
score. This scoring system has prognostic value with regard to future neurologic function. A decline in the
GCS score may indicate intracranial pathology. Reflexes
of the triceps, biceps, knee, and ankle should be evaluated.
The minimum GCS score is 3 and is seen in flaccid
patients who are unable to open their eyes spontaneously or speak. Patients who do open their eyes, obey

American Academy of Orthopaedic Surgeons

commands, and are oriented score the maximum of 15

points. A GCS score of 8 or less corresponds to the generally accepted definition of coma.
Estimating Injury Severity
Grading the severity of multiple trauma is difficult. The
Injury Severity Score (ISS) was the first scoring system
to use anatomic criteria to assess the extent of injury.
The ISS measures injury severity based on the abbreviated injury scale (AIS), developed in 1971 and revised
in 1985. Injury severity in the AIS is graded on a scale
of 1 to 5 for each organ system. As currently applied,
the ISS is calculated by taking the AIS scores from the
three most severely injured anatomic areas, squaring
them, and adding the resultant figures. An ISS of 16 or
more has been shown to be associated with a mortality
of 10%, whereas a score greater than 40 predicts a 50%
mortality. The ISS score has not been shown to accurately predict outcome for those individuals with a severe injury to a single body area.

Secondary Survey
During the secondary survey, thoracic trauma can be
further defined. Injuries detected through the secondary
survey include simple pneumothorax or hemothorax,
pulmonary and cardiac contusion, tracheobronchial tree
injuries, and diaphragmatic rupture. In all of these injuries hypoxia must be corrected before resuscitation is
successful.
Abdomen
During the primary survey, assessment of circulation, especially in blunt trauma patients, includes a thorough
abdominal examination to rule out hemorrhage. Peritoneal signs such as rigidity and rebound are useful to diagnose a surgical abdomen, but may not always be apparent in obtunded patients.
The Focused Assessment with Sonography for
Trauma examination is now widely used to further evaluate the abdomen. This examination can be done
quickly and does not require the transport of a critically
injured patient. Ultrasound has a sensitivity, specificity,
and accuracy comparable to diagnostic peritoneal lavage and CT scan, but the examination is operator dependent. Its utility is limited in obese patients, in the
presence of subcutaneous air, and in patients who have
had previous abdominal operations. One recent study
found that the focused assessment with sonography for
trauma examination underdiagnosed significant intraabdominal trauma in one group of 372 patients.
CT scan is used only in patients who are hemodynamically stable and who have no immediate indication
for a laparotomy. CT can evaluate the extent of a specific organ injury and can also help in the diagnosis of
retroperitoneal and pelvic organ injuries not readily ap-

161

The Polytrauma Patient

Orthopaedic Knowledge Update 8

parent on clinical examination. CT may also be performed serially to evaluate spleen, liver, and kidney injuries not requiring immediate surgical intervention.
Spine
Injuries to the spinal column should always be sought in
polytrauma patients. Occult spinal injuries may be overlooked in patients with an altered level of consciousness. Inadequate immobilization and excessive manipulation may cause additional damage in a patient with
spinal injury and may worsen the outcome. In hemodynamically unstable patients or patients with respiratory
difficulty, exclusion of spine injury may be deferred as
long as the patients spine is safely immobilized and
protected during the primary survey and initial care.
Moreover, maintaining tissue perfusion and oxygenation
will help stop progression of any existing cord injury.
The secondary spinal assessment should be performed
once life-threatening issues have been dealt with.The goal
of the secondary assessment is to identify and initially
manage neurologically and mechanically unstable spinal
injuries.
Log rolling the patient is essential for an adequate
spinal examination. The soft tissues should be assessed
for swelling, ecchymosis, wounds, deformity, or bogginess. Spinous processes should be palpated individually
with particular emphasis placed on areas of tenderness.
A complete motor, sensory, and reflex examination
should be performed, including tests for perianal sensation, rectal sphincter tone, and bulbocavernosus reflex.
Serial examinations should be performed to document
any progression of neurologic deficits. A neurologic deficit may be classified as complete, in which there is total
absence of motor or sensory function below the level of
injury, or incomplete. Identifying any distal motor or
sensory sparing (incomplete injury) is essential, as these
patients warrant treatment on a more urgent basis.
Spinal shock refers to the flaccidity and loss of reflexes, specifically sacral reflexes, after spinal cord injury.
The return of these reflexes marks the end of spinal shock.
The diagnosis of a complete neurologic injury cannot be
made during spinal shock.
Neurogenic shock manifests itself through hypotension and bradycardia, and must be distinguished from cardiogenic shock, which is characterized by hypotension and
tachycardia. Neurogenic shock should be treated with judicious use of fluid resuscitation and vasopressors. Atropine may be useful to treat the bradycardia.
It is unlikely that an awake, alert, neurologically normal patient without pain or tenderness along the spine
has any spinal injury. However, patients with an altered
level of consciousness (head injury, intoxication, hypoxia) need to have their normal radiographs corroborated via an adequate physical examination before neck
injury can be formally ruled out (Figure 1).

162

Figure 1 Radiograph of a patient who sustained multiple injuries during a head-on

motor vehicle accident. Cognitively impaired because of head injury and intoxication,
the patient was combative, denied neck or extremity pain, and demanded release from
the cervical collar. Cervical precautions were maintained through resuscitation, emergent laparotomy, and multiple emergent studies. Definitive cervical radiographs demonstrated grossly unstable three-column cervical dissociation. The patient was treated
definitively during the secondary stabilization period, with no neurologic injury or impairment.
A full radiographic spinal survey, including cervical,
thoracic, and lumbosacral radiographs, is necessary in all
patients with a suspected spinal cord injury. Patients
with spinal cord injury at one level may have another
injury at a noncontiguous level 5% to 20% of the time.
Lateral cervical radiographs must show the cervicothoracic junction, or a lateral swimmers view or CT should
be obtained through this area. An AP odontoid view
should also be obtained. Although patients with persistent pain despite normal radiographs may eventually
benefit from flexion/extension views to identify ligamentous injury, there is rarely a role for flexion/
extension radiographs in the initial evaluation of the
trauma patient.
CT is useful in delineating the extent of bony injuries detected on plain radiographs. MRI is useful in patients with abnormal neurologic findings. In patients

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

with specific lesions such as facet dislocations and who
are undergoing closed reduction, MRI should be done
to rule out extruded disks that may cause neurologic
damage during closed reduction. MRI is indicated in
any patient with a progressive neurologic deficit, or a
deficit that does not match the level of the recognized
spinal injury (C7 cord deficit in the face of a T10 burst
fracture).
Pelvis
The primary survey of the pelvis involves mechanical
assessment of stability and continuity: the physician
quickly checks for fractures or disruption by medially
compressing the iliac wings, applying an anteriorposterior stress through the ASIS, and by checking stability during hip range of motion. The secondary survey
involves a more thorough history, physical examination,
and analysis of an AP radiograph of the pelvis.
The history should determine the mechanism of injury. Higher energy injuries are more likely to be associated with an increased severity of fracture. Pelvic fractures occur more frequently with lateral impact than
frontal impact. Patients on the side of impact are more
likely to have a severe injury.
Physical findings of pelvic injury may include scrotal/
labial swelling, open lacerations in the perineum and vagina or rectum, associated urologic or neurologic injuries, or excessive internal/external rotation of the lower
extremity. Provocative maneuvers test the stability of
the pelvis to internal and external rotation of the hip.
The pelvis should move as a single unit. If a hemipelvis
moves separately, the ring is disrupted and the pelvis is
mechanically unstable. Once a pelvic injury is determined to be unstable, further manipulation that might
dislodge clots that have formed within the fracture
should be avoided.
Pelvic injuries can result in massive hemorrhage. In
polytrauma patients, intrathoracic and intra-abdominal
injuries are common, causing or contributing to hemorrhage and hypotension. Open wounds and long bone
fractures such as femur fractures also contribute to
blood loss. Hypotension caused specifically by a pelvic
injury is invariably associated with a mechanically unstable pelvis, and may prove difficult to control until the
pelvis is stabilized.
Early control of hemorrhage is crucial, in addition to
staying ahead of volume requirements. Resuscitation of
a hypotensive patient may result in hypothermia and acidosis. These factors may contribute to coagulopathy,
complicating the existing problem and leading to further
bleeding. Persistent hypotension can aggravate pulmonary and neurologic injury, and compromise renal, cerebral, and cardiac function. In addition, the risk of sepsis,
adult respiratory distress syndrome (ARDS), and multiple organ failure is increased in these patients.

American Academy of Orthopaedic Surgeons

Chapter 15 The Polytrauma Patient

Particular attention should be paid to open pelvic injuries because they are associated with exceptional morbidity and mortality and will require emergent dbridement. Rectal and vaginal examinations must be
performed to rule out lacerations. Associated urologic
injuries should be sought. Clinical findings that may indicate injury include blood at the urethral meatus, highriding prostate, or inability to pass a Foley catheter.
These patients will require retrograde urethrogram or
cystogram, depending on the particular injury.
The AP radiograph of the pelvis is used in correlation with the physical examination to determine the stability of the pelvis. In a recent study, it was determined
that physical examination was accurate in identifying injuries of the posterior pelvic ring. Signs of instability included more than 5 mm of displacement of the posterior
sacroiliac joint, the presence of a posterior fracture gap,
and the presence of an avulsion fracture of the transverse process of the fifth lumbar vertebrae. Pelvic ring
injuries can be classified based on anatomic location,
mechanism, or stability. The mechanistic classification
can help predict blood loss and guide management. This
classification as defined by Young and Burgess divides
pelvic injuries into four mechanisms of injury: lateral
compression, anteroposterior compression, vertical
shear, and combined mechanism. Inlet and outlet radiographs as well as CT scan can help further clarify the
degree of pelvic instability.
Classification of these injuries can help guide definitive management. (1) The estimated blood loss for a severe lateral compression injury is approximately 3.6
units whereas blood loss from an AP compression injury
is 14.8 units. (2) AP compression injuries have a higher
mortality and a higher incidence of shock and ARDS
than lateral compression fractures. (3) Moderate lateral
compression injuries have a higher incidence of brain
injury, and vertical shear injuries also have a high incidence of associated injuries as well as mortality.
Lower Extremity Injuries
Femoral shaft fractures are high-energy injuries, usually
occurring in the young patient population. Patients with
bilateral fractures typically have a high ISS, higher mortality, and higher risk of ARDS. Early treatment of
these injuries is important to survival and morbidity. An
unsplinted closed femur fracture can lose up to four
units of blood into the thigh. Tibial fractures are associated with severe soft-tissue and neurovascular trauma
that can render the extremity dysfunctional or even
nonviable.
Initial evaluation includes palpation of the entire extremity to the foot and a thorough neurovascular examination. Assessment of soft-tissue injury should be done
to rule out an open fracture. AP and lateral radiographs
should include the joints above and below the fracture.
Special attention should be paid to the ipsilateral hip to

163

The Polytrauma Patient

Table 2 | Factors in Severely Injured Patients That May

Warrant a Damage Control Treatment Approach
Multiple injuries with an ISS > 20, and a thoracic trauma AIS >2
Multiple injuries with abdominal/pelvic trauma, and hemorrhagic shock
(systolic BP < 90 mm Hg)
ISS > 40
Chest radiograph or CT evidence of bilateral pulmonary contusion
Initial mean pulmonary arterial pressure > 24 mm Hg
Pulmonary artery pressure increase during intramedullary nailing
> 6 mm Hg

rule out an ipsilateral femoral neck fracture because this

can be missed on initial examination.
Tibial shaft fractures can be caused by direct or indirect trauma. The limb should be inspected for evidence
of open fractures. Soft-tissue injury can be classified according to the Tscherne classification: grade 0 has minimal soft-tissue injury whereas grade III represents a
decompensated compartment syndrome requiring fasciotomy. Once the lower extremities have been surveyed, the evaluation should be repeated for the upper
extremities. The physician should reduce dislocations as
soon as possible, dress open fractures and wounds with
saline-soaked gauze, and splint fractures at the first opportunity.
Open fractures are classified according to Gustilo,
from type I (clean punctures) to type III (major disruption of the soft-tissue envelope). Type III injuries can be
further classified according to the extent of neurovascular injury. Type IIIa injuries can be closed while type
IIIb injuries require flap coverage, and type IIIc injuries
require revascularization for limb salvage.
Vascular injury must be considered in any extremity
injury, especially with knee dislocations. Pulses and perfusion must be checked, and if a pulse deficit is present,
all correctable causes should be evaluated: fracture
alignment should be corrected, traction released and restored, compartments checked, and hypotension corrected through resuscitation. Ankle-brachial index
(ABI) may provide information about the perfusion of
the limb. An ABI of 0.9 or higher will ordinarily rule
out arterial injury. Angiography remains the gold standard, however. In the polytrauma patient, formal angiography may not be possible and a limited study may
be performed in the operating room. Management of
life-threatening injury takes priority over limb salvage.

Trauma Management
Management of the polytrauma patient requires a multidisciplinary approach because of multiple injuries requiring intervention from various disciplines including
general surgery, neurosurgery, and orthopaedic surgery.
The trauma surgery team is generally responsible for

164

Orthopaedic Knowledge Update 8

coordinating these efforts and obtaining the appropriate
consultations. Most patients will benefit from rapid skeletal stabilization and mobilization, even when fixation
procedures have to follow abdominal or thoracic surgery. Every patient must be assessed individually, however, to avoid serious complications.

The Concept of Damage Control

Although early stabilization of long bone fractures has
been shown to reduce morbidity and length of hospital
stay, there is a subset of patients who may deteriorate in
the face of early, prolonged surgical intervention. The
cause of this decompensation is always difficult to prove
in patients with so many confounding issues, but several
investigators have suggested that the trauma of surgery,
with its systemic effects, superimposed on the initial
trauma of injury, leads to an increased incidence of
ARDS, multiple organ failure, and death. Patients at
risk for these complications are more seriously injured
and include patients with severe chest injuries and severe hemodynamic shock (Table 2). The development of
these complications is thought to be linked to the proinflammatory cascades that develop as a result of injury,
resuscitation efforts, and surgical interventions.
Although all polytrauma patients develop a systemic
inflammatory response, the more seriously injured patients suffer from an increased inflammatory response
and higher levels of cytokine release (interleukin [IL]-1,
tumor necrosis factor) for longer periods of time. This
prolonged inflammatory response is referred to as systemic inflammatory response syndrome. The inflammatory mediators such as IL-6 that are liberated during
this cascade event may produce deleterious clinical effects, further impairing pulmonary function and precipitating organ failure in other systems. Surgical treatment
of these severely injured patients may result in the release of additional inflammatory mediators, compounding the injury.
An increased awareness of the role of these proinflammatory cascades during surgery has led to the belief
that the second hit or surgical intervention, taking
place after the initial trauma (first hit), should be kept
to a minimum in these severely compromised patients.
This damage control concept was originally developed
as an approach to managing severe abdominal trauma.
As in the original concept, damage control orthopaedic
care is delivered in three stages. The first stage involves
immediate surgery to control bleeding, visceral injury,
gross instability, and contamination. The second stage
focuses on resuscitation and medical optimization. Definitive surgery to provide rigid fracture fixation, articular continuity, and soft-tissue coverage is reserved for
the third stage.
The benefits of early skeletal stabilization are well
recognized, and the impact of surgical trauma on poly-

American Academy of Orthopaedic Surgeons

Chapter 15 The Polytrauma Patient

Orthopaedic Knowledge Update 8

trauma patients is far from proven. Several studies are
available that suggest that other factors may be to
blame for the increase in complications seen among
these more seriously injured patients. Although many
authors suggest that reamed intramedullary nailing is
the culprit, particularly in the face of pulmonary trauma
and contusion, comparative studies have shown no difference in the incidence of ARDS or mortality relative
to that treatment. Similarly, the adequacy of early fluid
resuscitation has long been believed to influence the recovery of these severely injured patients. Finally, studies
of intramedullary fixation among severely injured patients have shown not only that the patients undergoing
nailing had no increased risk compared with patients
treated otherwise, but that, within these groups, the patients treated with immediate rodding had less pulmonary compromise and a lower incidence of ARDS than
patients treated with intramedullary rodding on a delayed basis.
However, it may be inappropriate to attempt definitive procedures in some patients with severe chest
trauma or hemodynamic instability. Damage control
principles are well applied in these circumstances.
The first stage care of orthopaedic injuries consists
of fracture and joint reduction, rapid skeletal stabilization, and control of hemorrhage. Wounds can generally
be washed and superficially dbrided as the patient is
being resuscitated and ventilated, but procedures that
generate more blood loss or tissue damage are avoided.
The second stage of damage control care focuses on
resuscitation and optimization of the patients medical
status. This may require several days in an intensive care
setting, but may also be accomplished over the course of
a few hours without the patient ever leaving the resuscitation area or operating theater. The patients condition
may be considered stable for definitive care whenever
specific parameters are met (Table 3). The trauma team
may find circumstances under which it is believed
worthwhile to push harder for stabilization of specific
injuries to mobilize the patient and obtain a vertical
chest for improved ventilation and pulmonary toilet.
Spinal and pelvic stabilization are sometimes afforded
priority to allow the patient to be safely moved and positioned. Similarly, rapid revascularization of a compromised extremity warrants additional consideration, and
can be accomplished with a temporary shunt and external fixation.
It is in the third and final stage that delayed definitive care of individual fractures is performed. Two methods by which rapid, temporary fracture stabilization can
be performed on the pelvis or long bones are external
fixation and unreamed intramedullary fixation. External
fixation can be accomplished rapidly, with minimal
blood loss, and can be used to span simple, complex, and
segmental fractures as well as traumatized joints. Conversion to an intramedullary nail or fixation plate is usu-

American Academy of Orthopaedic Surgeons

Table 3 | Criteria for Adequate Resuscitation

Hemodynamically stablewarm and well perfused
Stable oxygen saturation
Lactate level < 2 mmol/L
No coagulopathy, INR < 1.25
Normal body temperature
Urine output > 1 mL/kg/h
Not requiring inotropic support
INR= International Normalized Ratio

ally straightforward and carries a low risk of infection

within the first week of treatment. These methods of definitive fixation may be performed after the patient has
achieved optimal medical status.

Prioritizing Orthopaedic Care

Early and stable fracture fixation is of utmost importance in the orthopaedic management of most polytrauma patients. If initial long bone stabilization is delayed the patient could be at risk for greater morbidity
and mortality. It is useful to prioritize the orthopaedic
problems of polytrauma patients with respect to four
relative periods for intervention, as described by
Tscherne: (1) acute care (first 3 hours after injury);
(2) primary stabilization period (1 to 72 hours); (3) secondary stabilization period (3 to 8 days after injury);
and (4) tertiary stabilization or rehabilitation period (after 6 to 8 days).
Acute Care
In the acute period, the primary survey and secondary
assessment and hemodynamic resuscitation are accomplished. Head, chest, abdomen, and pelvic injury are all
recognized and life-saving/limb-saving interventions are
initiated. Significant epidural and subdural bleeding requires immediate evacuation. Once a hemothorax is diagnosed, a chest tube drainage should be placed. If
more than 1,500 mL of blood is obtained through the
chest tube or if drainage of more than 200 mL/h for 2 to
4 hours occurs, surgery should be considered. Continued
hemorrhage into the peritoneal cavity of a hemodynamically unstable patient requires emergent laparotomy.
Bleeding from the pelvic region must be ruled out before laparotomy is done.
External immobilization must be performed if the
pelvic ring is determined to be unstable. Initial external
immobilization consists of sandbags and straps, beanbags, or military antishock trousers. The use of military
antishock trousers, however, has been associated with
compartment syndrome and decreased respiratory ability. In the emergency department, external immobilization has been shown to decrease blood loss and to lower

165

The Polytrauma Patient

ISS dependent mortality. Moreover, transfusion requirements are decreased in patients treated with external
immobilization. External fixation can be applied in the
emergency department or operating room in concert
with other trauma surgery procedures.
Open fractures and joint injuries will require emergent dbridement and stabilization. These wounds
should not be closed primarily, and may require deep
dbridement and repeated irrigation when the patient is
more stable. Open pelvic injuries require emergent
dbridement, and perineal wounds that communicate
with the rectum or colon require diverting colostomy.
Vaginal injuries associated with pelvic ring disruptions
should be repaired to stop hemorrhage and to prevent
the development of abscesses. Degloving injuries of the
skin should be dbrided.
Established or incipient compartment syndromes
should be treated with adequate fasciotomy at the first
opportunity. Although fully developed compartment
syndrome is rarely seen at initial presentation, patients
with severe crushing injuries or prolonged ischemia of
the limb should undergo prophylactic fasciotomy under
their first anesthetic, if at all possible.
Primary Stabilization Period
Maintaining adequate perfusion of the spinal cord helps
to minimize secondary injury to the neural elements. Urgent care then focuses on methods of preventing further
damage and rapid realignment of the spine to decompress neurologic structures. High-dose methylprednisolone is commonly used to treat patients presenting with
a spinal cord injury with no contraindications to use.
Despite widespread use and a perception that steroid
therapy represents the standard of care for spinal cord
injured patients, in reality steroid therapy simply represents a treatment option, and a controversial one at
that. There is scant evidence that this intervention provides consistent or functionally significant improvement
in neurologic outcomes. H2 blockers or proton pump inhibitors should be considered to prevent formation of
gastric stress ulcers.
Compression of the neural elements from spinal
malalignment such as a cervical spine facet dislocation
should be addressed as soon as hemodynamic and respiratory stability has been achieved. Although a controversial issue, the use of MRI before reduction of dislocations to rule out disk extrusion that could compress
the cord after reduction has been advocated. MRI also
provides details of bone or disk fragments causing spinal cord compression or the presence of hematoma.
The issue of when to surgically stabilize patients
with spinal injury and neurologic deficit remains controversial. Emergency surgery for spinal cord injury has not
been clinically proven to be beneficial. However, some
studies indicate that patients with incomplete lesions

166

Orthopaedic Knowledge Update 8

probably benefit from early intervention to relieve persistent neurologic compression.
On the other hand, urgent surgical treatment (< 24
hours after injury) does not increase the risk to the spinal cord injured patient, compared with early care (1 to
3 days after injury), and can improve the overall outcome of the polytrauma patient. Urgent stabilization of
the fractured spine allows immediate mobilization of
the patient, reducing risks of prolonged recumbency
(thrombophlebitis, pulmonary embolism, pneumonia,
urosepsis, and ARDS). This appears to hold true for
even the most severely injured patients. Urgent surgical
stabilization among patients with a mean ISS of 40 or
greater reduced overall mortality from expected, and reduced or eliminated pulmonary complications such as
pneumonia, pulmonary embolus, and ARDS. Although
this does not suggest that all spine fracture patients
should be rushed to the operating room for urgent surgery, it does suggest that it is safe to proceed on an urgent basis when compelled to do so.
The benefits of early long bone stabilization have
been well established. It has been shown that patients
with femoral shaft fractures with an ISS greater than 18
who had early stabilization experienced a decrease in
the incidence of ARDS, pulmonary complications, and
length of intensive care unit stay.
Controversy exists regarding reaming during intramedullary nailing in patients with severe pulmonary
injury. Marrow contents and bone fragments may embolize during reaming, and it has been proposed that embolization of such contents may lead to an inflammatory
response as well as mechanical blockage, exacerbating
the existing pulmonary injury. However, several recent
studies have suggested that the extent of the primary
pulmonary injury is the major determinant of pulmonary morbidity. As noted earlier, a study examining intramedullary fixation versus plating of femur fractures
demonstrated no difference in pulmonary complications.
Vascular injury must be recognized immediately in
injured extremities. Timely diagnosis and treatment can
minimize ischemic injury. Although arterial reconstruction has high priority, bony stability may need to be
achieved before vascular repair. If immediate repair is
not possible, a temporary shunt may be placed. Compartment syndrome should be anticipated and treated
immediately with fasciotomy. In obtunded polytrauma
patients clinical examination may not prove to be reliable. Compartment pressures within 30 mm Hg of diastolic pressure are consistent with compartment syndrome and thus a fasciotomy should be performed.
Treatment of open fractures involves administration
of antibiotics and extensive surgical dbridement. Stable
fixation of the fracture is advocated. Currently, extensive soft-tissue injuries and associated tibia and femur
fractures (grade III) are safely treated with intramedul-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

lary nailing whereas in the past these injuries were
treated with external fixation.
Soft-tissue injuries may require extensive dbridement and reconstruction to ensure adequate coverage of
bone, tendons, vessels, nerves, and implants. If a large
soft-tissue defect is present, the decision on the type of
reconstruction should be made at the second dbridement or second look, which is usually performed at
48 hours. Coverage with either a local or free vascularized flap should be performed within 72 hours of injury.
In the setting of multiple closed fractures, long bone
fractures of the lower extremity should be stabilized
first. Because of the extensive soft-tissue damage and
blood loss associated with uncontrolled spasm and instability, femur and tibia fractures should be reduced and
stabilized first, whereas fractures of the upper extremity
can be splinted initially, with good results. Pelvic and
spinal fractures may be definitively treated after treating the lower extremity fractures, in most cases, but unstable cervical injuries may warrant earlier treatment,
maintaining the long bone fractures in traction until the
spinal segment is adequately fixed. Upper extremity
fractures may be definitively managed after addressing
the above injuries.
Secondary Stabilization Period
During the secondary stabilization period, the patient is
hemodynamically stable and surgical intervention is performed on a semielective basis. It is at this time that
dbridement of any areas of soft-tissue necrosis is performed. Secondary wound closure and some soft-tissue
reconstructions may be achieved. Intra-articular reconstruction, hand, foot, and upper extremity fracture fixation, and complex spinal and pelvic and acetabular reconstruction may be performed at this time.
Tertiary Stabilization Period
Late reconstructive procedures may be performed in
the tertiary stabilization period, including definitive closure of amputation sites or any procedure that may
have been postponed during the secondary stabilization
period. The prognosis of the patient is usually known.
If the patient is stable and is extubated, rehabilitation may begin. This process should be started on an inpatient basis and should be taken through the outpatient phase if necessary.

Long-Term Outcome
Advanced age and increased severity of injury is associated with increased mortality in the short term. Longterm outcomes of polytrauma patients vary with the severity of injury initially sustained. Severity of injury is
associated with greater disability, higher rate of unemployment after injury, and lower quality of life. Studies
examining both subjective and objective outcomes data

American Academy of Orthopaedic Surgeons

Chapter 15 The Polytrauma Patient

among patients with multiple extremity fractures have
demonstrated that functional disabilities are greatest for
injuries below the knee. Intra-articular injuries to the
foot and ankle, in particular, tend to impair patients who
have had an otherwise satisfactory recovery from
trauma. Additionally, recent evidence indicates that
polytrauma patients sustaining cord or cauda equina injury at the time of spinal fracture have poorer functional outcomes and poorer return to work, even among
those with good neurologic recovery.

Annotated Bibliography
Bhandari M, Guyatt GH, Khera V, Kulkarni AV, Sprague S, Schemitsch EH: Operative management of lower
extremity fractures in patients with head injuries. Clin
Orthop 2003;407:187-198.
The authors compared femoral plating versus intramedullary nailing and tibial plating versus intramedullary nailing in
head-injured patients. The study group included 119 patients
with severe head injuries and lower extremity fractures. There
was no significant difference in mortality rates between patients treated with intramedullary nailing or plating. The strongest predictor of mortality was the severity of the initial head
injury.

Cook RE, Keating JF, Gillespie I: The role of angiography in the management of hemorrhage from major fractures of the pelvis. J Bone Joint Surg Br 2002;84:178-182.
This study examined 150 patients with unstable pelvic fractures and uncontrollable hypotension. In those patients undergoing angiography prior to external fixation or laparotomy,
more than half of them died. The authors recommended that
angiography be used in refractory cases after skeletal stabilization has been attempted in those patients with unstable pelvic
injuries.

Giannoudis PV: Surgical priorities in damage control in

polytrauma. J Bone Joint Surg Br 2003;85:478-483.
The author provides a review of the current trends and
principles in orthopaedic management of polytrauma, including principles of damage control surgery.

Inaba K, Kirkpatrick AW, Finkelstein J, et al: Blunt abdominal aortic trauma in association with thoracolumbar spine fractures. Injury 2001;32:201-207.
The authors report their experience with blunt abdominal
aortic disruption at regional trauma centers. Eight cases were
identified, six of which were associated with thoracolumbar
fractures, with a mean ISS of 42. All spinal fractures were associated with a distractive force pattern. The authors concluded that with all distractive thoracolumbar injuries, aortic
disruption must be considered as this injury may occur as a result of similar distractive forces.

167

The Polytrauma Patient

McCormick JP, Morgan SJ, Smith WR: Clinical effectiveness of the physical examination in diagnosis of posterior pelvic injuries. J Orthop Trauma 2003;17:257-261.
This article presents a prospective study evaluating the
correlation of physical examination to radiographic studies in
the diagnosis of posterior pelvic ring disruptions. Of those patients found to have posterior pelvic ring disruptions as shown
on CT scan, 98% had posterior pelvic pain on examination.
The authors concluded that physical examination was accurate
in detecting injuries of the posterior pelvic ring.

Miller MT, Pasquale MD, Bromberg WJ, Wasser TE,

Cox J: Not so FAST. J Trauma 2003;54:52-59.
The authors performed a study comparing a Focused Assessment with Sonography for Trauma with CT scan to compare accuracy in diagnosis of injury in patients sustaining
blunt abdominal trauma. A total of 372 patients were studied
and it was noted that the Focused Assessment with Sonography for Trauma examination underdiagnosed significant intraabdominal trauma. The authors concluded that patients who
are hemodynamically stable who sustain blunt abdominal
trauma should undergo CT scan for further evaluation for
more accurate diagnosis of intra-abdominal injuries.

Zalavras CG, Patzakis MJ: Open fractures: Evaluation

and management. J Am Acad Orthop Surg 2003;11:212219.
This article is a review of the literature regarding management of open fractures. The authors discuss the controversies
regarding wound closure. Soft-tissue coverage of large and
contaminated wounds is also discussed.

Classic Bibliography

Orthopaedic Knowledge Update 8

Boulanger BR, Stephen D, Brennemann FD: Thoracic
trauma and early intramedullary nailing of femur fractures: Are we doing harm? J Trauma 1997;43:24-28.
Chapman MW: The role of intramedullary fixation in
open fractures. Clin Orthop 1986;212:26-34.
Civil ID, Schwab CW: The Abbreviated Injury Scale,
1985 revision: A condensed chart for clinical use.
J Trauma 1988;28:87-90.
Gustilo RB, Anderson JT: Prevention of infection in the
treatment of one thousand and twenty-five open fractures of long bones: Retrospective and prospective analyses. J Bone Joint Surg Am 1976;58:453-458.
Johnson KD, Cadambi A, Seibert GB: Incidence of
adult respiratory distress syndrome in patients with multiple musculoskeletal injuries: Effect of early operative
stabilization of fractures. J Trauma 1985;25:375-384.
McLain RF, Benson DR: Urgent surgical stabilization of
spinal fractures in polytraumatized patients. Spine 1999;
24:1646-1654.
Schwab CW, Shayne JP, Turner J: Immediate trauma resuscitation with type O uncrossmatched blood: A twoyear prospective experience. J Trauma 1986;26:897-902.
Shackford SR, Hollingsworth-Fridlund P, Cooper GF,
Eastman AB: The effect of regionalization on the quality of trauma care as assessed by concurrent audit before and after institution of a trauma system: A preliminary report. J Trauma 1986;26:812-820.

Bohlman HH: Acute fractures and dislocations of the

cervical spine: An analysis of three hundred hospitalized
patients and review of the literature. J Bone Joint Surg
Am 1979;61:1119-1142.

Tscherne H, Regel G, Pape HC, Pohlemann T, Kretteck

C: Internal fixation of multiple fractures in patients with
polytrauma. Clin Orthop 1998;347:62-78.

Bone L, Bucholz R: The management of fractures in the

patient with multiple trauma. J Bone Joint Surg Am
1986;68:945-949.

Winquist RA, Hansen ST Jr, Clawson DK: Closed intramedullary nailing of femoral fractures: A report of
five hundred and twenty cases. J Bone Joint Surg Am
1984;66:529-539.

Bone LB, Johnson KD, Weigelt J, Scheinbeng R: Early

versus delayed stabilization of fractures: A prospective,
randomized study. J Bone Joint Surg Am 1989;71:336340.

168

American Academy of Orthopaedic Surgeons

Chapter

16

Coagulation and Thromboembolism

in Orthopaedic Surgery
Francis H. Shen, MD
Dino Samartzis, BS, PGCEBHC
Christopher J. DeWald, MD

Introduction
The coagulation pathway is a series of enzymatic processes whose final result is the formation of a thrombus
(Figure 1). This cascade is the result of an equilibrium
between prothrombotic and antithrombotic factors that
occur in the bloodstream. The preferential occurrence of
one process over the other can result in either a bleeding coagulopathy or thromboembolic disease.

Coagulopathies
Preoperative Assessment of Blood Clotting
Preoperative assessment begins with a thorough history,
clinical examination, and appropriate laboratory studies.
A history of easy, excessive, or spontaneous bleeding or
bruising, previous need for blood transfusions, and/or a
family history of bleeding disorders is suggestive of a
clotting disorder. A review of the medical history preoperatively is imperative. Various conditions, such as
chronic liver and renal disease, malnutrition, malabsorption, chronic antibiotic use, hematologic disorders, and
drug or alcohol use may identify patients at risk for coagulopathies and these patients require additional evaluation. Although neither pathognomonic nor exclusionary, signs of potential bleeding or clotting problems
should be investigated, such as bruising not associated
with trauma, petechiae, and stigmata of liver disease and
portal hypertension (for example, spider angiomata or
caput medusa). Routine screening tests may include a
complete blood cell count (CBC) and platelet count,
and determination of prothrombin time (PT) and an activated partial thromboplastin time (PTT).

ders in the coagulation cascade, and coagulation factor

deficiencies or dysfunction. Once a bleeding disorder is
suspected then additional work-up is recommended.
Screening tests of the coagulation cascade include PT,
which reflects deficiencies in factors II, V, VII, and X,
whereas PTT reflects deficiencies of factor II, V, VIII, IX,
X, XI, and XII. Isolated prolongation of the PTT is seen
with factors VIII, IX, or XI deficiencies. Although also
seen in patients with hereditary deficiencies of factor XII,
these disorders are not usually associated with clinical
bleeding. Isolated prolongation of PT usually reflects factor VII deficiency or low fibrinogen levels (< 1 g/L). Because isolated factor II, V, or X deficiencies are rare, prolongation of both PT and PTT are more commonly a
reflection of multiple factor deficiencies as seen in disseminated intravascular coagulation or in hepatocellular dysfunction. Conversely, a normal PT and PTT does not automatically imply normal factor levels, because levels may
have to drop below 30% of normal before elevated values are seen.
Quantitative and qualitative platelet dysfunctions may
be more accurately identified with the use of bleeding
times. Quantitative platelet disorders, such as idiopathic
thrombocytopenic purpura, disseminated intravascular
coagulation, and drug-induced thrombocytopenia are the
result of peripheral platelet destruction and may be seen
as multiple large, young platelets on a peripheral blood
smear. Causes of qualitative platelet dysfunction with normal platelet numbers can occur in renal dialysis patients
and are often drug-induced, as occurs with aspirin and
other newer antiplatelet medications, such as clopidogrel
and ticlopidine.

Work-Up of the Coagulopathic Patient

Most of the time, a CBC with platelet count, PT, and
PTT are the only tests necessary during a routine preoperative screening. Other diagnostic testing, such as
bleeding times, D-dimer, and assessment of specific factor deficiencies are not routinely recommended unless
the patient history or clinical examination is suggestive.
Differential diagnoses of common coagulopathies include quantitative or qualitative platelet disorders, disor-

American Academy of Orthopaedic Surgeons

Treatment of Common Coagulopathies

Although uncommon in the general population (1 in
10,000), patients with hemophilia A (factor VIII deficiency) and B (factor IX deficiency) may be frequently
seen by the orthopaedic surgeon. The gene for both factors is located on the X chromosome and affects males
more than females. Hemarthrosis is one of the most
common complications and first-line therapy is factor

169

Coagulation and Thromboembolism in Orthopaedic Surgery

Orthopaedic Knowledge Update 8

Figure 1 The coagulation pathways. PT measures the function of the extrinsic and common pathways, whereas the PTT measures the function of the intrinsic and common
pathways. FPA = fibrinopeptide A; FPB = fibrinopeptide B. (Reproduced with permission from Stead RB: Regulation of hemostasis, in Goldhaber SZ (ed): Pulmonary Embolism and
Deep Venous Thromboembolism. Philadelphia, PA, WB Saunders, 1985, p 32.)
replacement. In patients with spontaneous bleeds, the
goal is to achieve circulating levels that are approximately 40% to 50% of normal. If surgical intervention is
planned, the deficient factor should be replaced to levels near 100%. Continuous or intermittent boluses of
high-purity plasma derivatives or recombinant factor
concentrates can be given and have greatly reduced the

170

risk of human immunodeficiency virus transmission. Although uncommon, hemophilia patients with factor inhibitors, antibodies that neutralize the coagulation factor, should be identified.
Various congenital bleeding disorders exist that may
affect proper coagulation, the most common being von
Willebrands disease, an autosomal dominant disorder.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 16 Coagulation and Thromboembolism in Orthopaedic Surgery

Table 1 | List of Common Antithrombotic Medications With Mechanism of Action and Laboratory Monitoring Method
Medication
Warfarin (Coumadin)

Low molecular weight heparin

(enoxaparin)
Salicylates (aspirin)
Traditional nonsteroidal
anti-inflammatory drugs
(ibuprofen)
Selective COX-2 inhibitors
Celecoxib (Celebrex)
Thienopyridines
Ticlopidine (Ticlid)
Clopidogrel (Plavix)

Mechanism of Action
Prevents the vitamin K dependent -carboxylation of
factors II, VII, IX, and X, proteins C and S, slowing thrombin
production
Upregulates the inhibitory effect of antithrombin on serine
proteases thrombin, IXa, Xa, XIa, and XIIa with greatest
effect upon Xa
Irreversibly inhibits cyclo-oxgenase activity in platelets and
vascular endothelium
Inhibition of COX-1 and 2, and leukotriene synthesis

Laboratory Monitoring Method

Prothrombin time - International Normalized Ratio

Selective inhibition of COX-2

None necessary, although bleeding times may be

prolonged
None necessary, although bleeding times may be
prolonged

Binds to the GPIIb/IIIa receptor to irreversible blocking

Adenosine diphosphate-induced platelet aggregation

It occurs secondary to a deficiency of von Willebrands

factor, which is important for promoting normal platelet
function and stabilizing factor VIII. Although most patients will respond to desmopressin, a few may require
the use of clotting factor concentrates.
Treatment of the acquired coagulopathies centers on
identifying the cause. Prior to surgery, medications that
inhibit platelet function and the coagulation pathway
should be stopped if possible and blood work rechecked
if necessary (Table 1). Nutrition should be optimized, especially in older patients. If necessary, platelets and
fresh frozen plasma (FFP) should be available for infusion perioperatively. Because FFP contains only the normal amount of clotting factor per milliliter of plasma,
achieving sufficient factor levels can require greater
volumes.
Recombinant factor VII has been used in patients
with a high titer inhibitor directed against factor V, VIII,
or IX. More recent studies focus on the role of various
recombinant factor concentrates to achieve hemostasis
in patients demonstrating massive bleeding refractory to
FFP and platelets. This may be the case in patients who
develop consumptive coagulopathies secondary to massive bleeding from multitrauma or massive blood volume replacement postoperatively, and in patients who
develop dilutional coagulopathies during prolonged surgeries. The lower volumes required with the use of these
recombinant factors may result in less fluid overload,
decreased fluid shifts, and anasarca in critically ill patients; however, additional studies are required to more
clearly define the optimal dosing, safety, and efficacy
of these recombinant factors before routine use
is recommended.
Other areas of investigation include the local use of
fibrin sealant and topical thrombin. Preliminary data

American Academy of Orthopaedic Surgeons

None necessary in uncomplicated cases; Chromogenic

anti-Xa assay available in complicated cases
None necessary, although bleeding times may be
prolonged
None necessary, although bleeding times may be
prolonged

suggest that in patients undergoing primary total knee

arthroplasty, fibrin sealant may safely reduce postoperative wound drain output while maintaining higher postoperative hemoglobin levels.

Thromboembolic Disease
Thromboembolic disease refers to a group of disorders
that includes thrombosis and embolic disorders, which
can occur in either the arterial or venous system. In the
orthopaedic patient, venous thromboembolic disease is
more common, and includes deep venous thrombosis
(DVT) and pulmonary embolism (PE).

Pathophysiology
In the 19th century, Virchow described the three major
pathophysiologic factors (stasis, endothelial injury, and
hypercoagulable states) contributing to thromboembolic
disease. Thrombosis secondary to stasis occurs more
commonly in the low-velocity, high capacitance venous
system (venostasis) and is the result of a disruption of
the normal laminar blood flow. These venous thrombi
are typically composed of fibrin and trapped red cells
and contain few platelets. Endothelial injury leads to
thrombus formation secondary to traumatic exposure of
extracellular matrix components (collagen) within the
vessel walls to circulating platelets. Although this can
occur in the venous system, it more typically is present
in the high-velocity arterial system and is usually composed of platelets with little fibrin. The third cause, hypercoagulability, remains the least understood, and occurs in conditions such as malignancy, pregnancy,
hormone replacement, in smokers, and during the postoperative state. Hereditary causes, such as resistance or
deficiencies in antithrombin III, factor C, factor S, factor

171

Coagulation and Thromboembolism in Orthopaedic Surgery

V Leiden mutations, and antiphospholipid syndrome,
have also been identified. Clinically, these patients are
at increased risk for stroke and vascular and cardiac disease, and may have an increased risk of osteonecrosis,
Legg-Calv-Perthes disease, and other thrombotic disorders.

Epidemiology
Venous thromboembolic disease occurs for the first
time in approximately 100 per 100,000 persons each
year and increases exponentially with age from a negligible rate for those younger than 15 years up to 500 per
100,000 in individuals older than 80 years. Two thirds of
symptomatic venous thromboembolic diseases present
as a DVT, whereas one third present with PE. In large
epidemiologic studies, the incidence is statistically
higher among Caucasians and African Americans than
in Hispanics and Asian-Pacific Islanders. Recurrence
rates, even after successful anticoagulant therapy, are
approximately 7% at 6 months after the initial event.
The risk of venous thromboembolic disease also varies depending on the procedure being performed. The
reported rate of DVT in the total hip arthroplasty patient ranges from 15% to 25% and can be as high as
50% in the patient undergoing a total knee arthroplasty,
whereas the risk of fatal PE without thromboprophylaxis is between 0.1% to 0.5%. The use of either chemical or mechanical prophylaxis has decreased these rates
by half.
In patients with pelvic, acetabular, periacetabular,
and hip fractures, reported DVT rates vary from 20% to
60% and increase when surgery is delayed by 2 days or
longer. Although symptomatic PE rates range from 2%
to 10%, fatal PE occurs in approximately 0.5% to 2%.
In these patients, the occurrence of more proximal
thrombi, particularly those in the pelvis, may result in a
higher risk of embolization (up to 50%).
Identifying rates of thromboembolic disorders in the
patient undergoing spinal surgery has been more difficult. Studies are limited, but reported rates of DVT
range from 0.3% to 26%. In patients undergoing posterior spinal procedures alone there does not appear to be
a difference in DVT rates when compared by sex, length
of procedure, number of levels performed, and/or the
addition of one- or two-level fusions. However, although still not clear, DVT rates may be higher in patients undergoing combined anterior and posterior spinal surgery than in those undergoing posterior-alone
procedures.
In a prospective multicenter study of 2,733 patients
who underwent foot and ankle surgery, the reported
DVT rate was 0.22% with a 0.15% rate of nonfatal PE
without occurrence of fatal PE. The authors did not find
a statistical relationship with tourniquet use or history
of thromboembolic disease. The only statistically signifi-

172

Orthopaedic Knowledge Update 8

cant relationship was the increased use of postoperative
cast immobilization and no weight bearing, which increased the relative risk by 0.04%.

Treatment Strategies
Treatment consists of either thromboembolic disease
prevention (thromboprophylaxis) or management of established thromboembolic disease. Because of high rates
of venous thromboembolic disease in the orthopaedic
patient, initial management should center on thromboprophylaxis. Treatment strategies include either mechanical or pharmacologic methods or a combination of
both. Management techniques for the treatment of established thromboembolic disease is beyond the scope
of this chapter.

Mechanical Approaches
Current mechanical options include pneumatic compression boots, plantar foot compression devices, and
elastic compression stockings. Pneumatic compression
boots reduce stasis by directly increasing the velocity of
venous blood flow. Because these boots also increase
the systemic release of endogenous fibrinolytic activity,
their benefit can be realized even when used on a single
limb or placed on the upper extremity. Plantar foot
compression devices mimic the hemodynamic effects of
ambulation, thus increasing venous return and decreasing venostasis. Although elastic compression stockings
decrease lower extremity venous pooling and stasis,
their use alone as an effective thromboprophylaxis is
unclear. In patients undergoing combined anterior and
posterior spinal procedures and in those who have had
joint arthroplasty, if possible, elastic compressive stockings should be used in conjunction with other prophylactic agents.
The greatest advantages of these mechanical methods are that there is almost no risk of bleeding, minimal
to no side effects, and they do not require laboratory
monitoring. A less common, more invasive mechanical
device is the inferior vena cava filter. This device can be
associated with substantial morbidity; therefore, its routine use as a thromboprophylactic agent is not encouraged. However, it may have a role in management of
patients with known DVT and recurrent PE despite adequate anticoagulant therapy, patients in whom anticoagulant therapy is contraindicated, and potentially in
patients noncompliant with anticoagulant therapy. The
use of temporary inferior vena cava filters (which are
removed typically within 10 days from the time of insertion) is currently being investigated in various clinical
scenarios, such as in the critically ill, multitrauma patient
requiring multiple surgical interventions over a short
time period.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 16 Coagulation and Thromboembolism in Orthopaedic Surgery

Pharmacologic Approaches
At this time, common pharmacologic avenues for the
treatment of thromboembolic disease include the use of
warfarin, low molecular weight heparin (LMWH), and
antiplatelet therapies (such as aspirin or thienopyridines). Warfarin exerts an anticoagulation effect by inhibiting the vitamin K-dependent clotting factors II, VII,
IX, and X. Multiple studies have demonstrated warfarins effectiveness in decreasing the rate of both DVT
and PE in patients undergoing total joint arthroplasty.
Because clinically significant bleeding occurs in 1% to
5% of patients, careful monitoring is necessary to maintain the international normalized ratio in a therapeutic
range. The international normalized ratio was established by the World Health Organization as a way to
standardize coagulation test results by using the international sensitivity index.
Some of the major disadvantages in the use of warfarin include the expense and inconvenience of monitoring, risk of bleeding, and multiple drug interactions. Although awareness has increased about the interaction of
warfarin with many over-the-counter medications such
as aspirin, ibuprofen, and other nonsteroidal antiinflammatory drugs, many physicians and patients are
still unaware of the potential interactions that can occur
with various herbs and supplements such as ginger,
ginkgo, ginseng, and St. Johns Wort. Furthermore, patient education should include identification of foods
high in vitamin K (such as broccoli and kale) that can
reduce the effectiveness of warfarin.
The role of LMWH continues to evolve. LMWH is
created by depolymerization of unfractionated heparin
and contains compounds with molecular weights between 3,000 to 10,000 daltons. Because of the enhanced
affinity for antithrombin III and activated factor X,
LMWH is a more active agent than conventional heparin and intervenes at an earlier point in the clotting
cascade. The dosages required for thromboprophylaxis
do not increase PT or PTT values and therefore do not
require monitoring. Disadvantages of LMWH include
increased cost, parenteral administration, and the increased potential for bleeding. Therefore, LMWH
should be considered with caution in patients undergoing spinal puncture or using epidural catheters as an anesthetic agent. LMWH appears to be effective in the
management of venous thromboembolic disease in total
joint arthroplasty patients and in the multitrauma patient. Ease of use and lack of need for monitoring may
make LMWH an excellent medication for use in the
outpatient setting, especially in patients who continued
to be at risk for venous thromboembolic disease after
being discharged from the hospital.
Currently, the use of aspirin as the sole thromboprophylactic agent is unclear. In doses greater than 100 mg
per day, aspirin irreversibly binds and inactivates cy-

American Academy of Orthopaedic Surgeons

clooxygenase in both immature and circulating platelets,

thereby inhibiting thrombus plug formation. Although
some studies demonstrate that, compared with placebo,
the use of aspirin in total joint arthroplasty patients decreases the rate of both proximal and distal DVTs, the
question remains whether they are as effective as other
methods in preventing fatal PE. In selected patients, the
use of aspirin with or without other forms of prophylaxis may be acceptable; however, larger randomized
controlled trials are necessary.
Newer agents continue to evolve and include medications such as direct thrombin inhibitors (lepirudin), direct and indirect factor Xa inhibitors, and heparinoids
(danaparoid). At present, the role of such agents remains unclear and additional investigation is required
before routine use is recommended.

Summary
Although consultation with an internist or hematologist
may be necessary when complications arise, a basic
knowledge of common bleeding disorders and thromboembolic disease is important for every orthopaedic
surgeon to understand. Identification of possible coagulopathies begins with a careful history, examination, appropriate laboratory studies, and treatment directed toward management of the specific etiology. Newer
recombinant factors that address various factors in the
coagulation pathway are currently being developed.
Without thromboprophylaxis, the rate of thromboembolic disease in the orthopaedic population can be
high. Early ambulation and mobilization should be encouraged and used in conjunction with other established
prophylactic methods. The benefits of thromboprophylaxis should be carefully balanced with the risks of intervention and individualized to the patient being treated
and be procedure-specific.

Annotated Bibliography
Coagulopathies
Eckman MH, Erban JK, Singh SK, Kao GS: Screening
for the risk of bleeding or thrombosis. Ann Intern Med
2003;138:W15-W24.
For nonsurgical and surgical patients without synthetic
liver dysfunction or a history of oral anticoagulant use, routine
testing has no benefit in assessment of bleeding risk.

Hvid I, Rodriguiez-Merchan EC: Orthopaedic surgery

in haemophilic patients with inhibitors: An overview.
Haemophilia 2002;8:288-291.
Recombinant factor VIIa appears to be an efficient hemostasis product for patients with hemophilia A and B with inhibitors undergoing major elective orthopaedic procedures.

Shami VM, Caldwell SH, Hespenheide EE, Arseneau

KO, Bickston SJ, Macik BG: Recombinant activated fac-

173

Coagulation and Thromboembolism in Orthopaedic Surgery

tor VII for coagulopathy in fulminant hepatic failure
compared with conventional therapy. Liver Transpl
2003;9:138-143.
The authors found a statistically significant difference in
correction of coagulopathy and decreased anasarca in patients
with severe coagulopathy from fulminant hepatic failure after
a single dose of recombinant activated factor VII compared
with those receiving fresh frozen plasma alone.

Wang GJ, Hungerford DS, Savory CG, et al: Use of fibrin sealant to reduce bloody drainage and hemoglobin
loss after total knee arthroplasty: A brief note on a randomized prospective trial. J Bone Joint Surg Am 2001;
83:1503-1505.
Preliminary data from a phase III trial of 53 patients undergoing unilateral primary total knee arthroplasty demonstrated a
reduction in bloody drainage and higher postoperative hemoglobin in the fibrin sealant group than in the control group.

Thromboembolic Disease
Edelsberg J, Ollendorf D, Oster G: Venous thromboembolism following major orthopedic surgery: Review of
epidemiology and economics. Am J Health Syst Pharm
2001;58(suppl 2):S4-S13.
A review of epidemiology of venous thromboembolic disease in joint replacement and hip fracture surgery is presented. Costs associated with the diagnosis, treatment, and recovery from venous thromboembolic disease is substantial,
with the initial therapy consuming 90% of the total cost.

Fitzgerald RH Jr, Spiro TE, Trowbridge AA, et al: Prevention of venous thromboembolic disease following
primary total knee arthroplasty: A randomized, multicenter, open-label, parallel-group comparison of enoxaparin and warfarin. J Bone Joint Surg Am 2001;83:900906.
This article discusses a prospective, randomized, multicenter trial that compared the efficacy and safety of enoxaparin and warfarin in 349 patients undergoing total knee arthroplasties. The authors found enoxaparin administered twice
daily to be more effective than warfarin in reducing the occurrence of asymptomatic venous thromboembolic disease.

Hyers TM: Management of venous thromboembolism:

Past, present, and future. Arch Intern Med 2003;163:759768.
A review of the state of current chemoprophylatic methods, including unfractionated heparin and low molecular
weight heparin and their potential adverse effects and theoretical basis for failure is presented. Newer agents recently approved for venous thromboembolic disease, such as heparinoid, and selective Xa inhibitors are discussed.

Kim YH, Oh SH, Kim JS: Incidence and natural history

of deep-vein thrombosis after total hip arthroplasty: A

174

Orthopaedic Knowledge Update 8

prospective and randomized clinical study. J Bone Joint
Surg Br 2003;85:661-665.
The authors prospectively followed 200 patients who underwent primary total hip arthroplasties (THAs). DVT rates
were 26% for bilateral THA and 20% for unilateral THA, but
in the 72 patients with venogram confirmed thrombi, all
thrombi had spontaneously and completely resolved at 6
months regardless of site and size.

Offner PJ, Hawkes A, Madayag R, Seale F, Maines C:

The role of temporary inferior vena cava filters in critically ill surgical patients. Arch Surg 2003;138:591-595.
The authors concluded from a prospective cohort study of
44 patients at a level I trauma center that temporary inferior
vena cava filters are a safe and effective approach to preventing venous thromboembolic disease in the multitrauma, critically ill surgical patient.

Samama CM: Venous thromboembolism deserves your

attention. Critical Care 2001;5:277-279.
The author surveyed intensive care unit directors in Canada and reviewed the use of traditional and advanced compression techniques, as well as unfractionated versus low molecular weight heparin for the prevention and treatment of
DVT and venous thromboembolic disease. The recommendations were that mechanical prophylaxis should be used alone
or in combination with chemoprophylaxis in patients in the intensive care unit and that pharmacologic prophylaxis should
always be combined with mechanical prophylaxis; however,
large randomized controlled trials are needed.

Stannard JP, Riley RS, McClenney MD, Lopez-Ben RR,

Volgas DA, Alonso JE: Mechanical prophylaxis against
deep vein thrombosis after pelvic and acetabular fractures. J Bone Joint Surg Am 2001;83:1047-1051.
This article discusses a prospective, randomized, blinded
study of the use of sequential and pulsatile mechanical compression devices for prophylaxis against DVT in 107 patients
with pelvic or acetabular fractures that required surgical treatment. Pulsatile compression was associated with fewer instances of DVT than standard compression, but results did not
reach significance.

White RH: The epidemiology of venous thromboembolism. Circulation 2003;107(suppl 23):I4-I8.

First-time venous thromboembolic disease occurs in approximately 100 per 100,000 persons each year and a recurrence rate of 7% has been reported. Risk factors identified
from large epidemiologic studies include ethnicity, and advanced age. Gender does not appear to be a factor.

Classic Bibliography
Clagett GP, Anderson FA Jr, Geerts W, et al: Prevention
of venous thromboembolism. Chest 1998;114(suppl 5):
531S-560S.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 16 Coagulation and Thromboembolism in Orthopaedic Surgery

Coventry MB, Nolan DR, Beckenbaugh RD: Delayed

prophylactic anticoagulation: A study of results and
complications in 2,012 total hip arthroplasties. J Bone
Joint Surg Am 1973;55:1487-1492.
Dearborn JT, Hu SS, Tribus CB, Bradford DS: Thromboembolic complications after major throacolumbar
spine surgery. Spine 1999;24:1471-1476.
Freedman KB, Brookenthal KR, Fitzgerald RH Jr, Williams S, Lonner J: A meta-analysis of thromboembolic
prophylaxis following elective total hip arthroplasty.
J Bone Joint Surg Am 2000;82:929-938.
Geerts WH, Code KI, Jay RM, Chen E, Szalai JP: A prospective study of venous thromboembolism after major
trauma. N Engl J Med 1994;331:1601-1606.
Geerts WH, Jay RM, Code KI, et al: A comparison of
low-dose heparin with low-molecular-weight heparin as
prophylaxis against venous thromboembolism after major trauma. N Engl J Med 1996;335:701-707.
Leclerc JR, Geerts WH, Desjardins L, et al: Prevention
of venous thromboembolism after knee arthroplasty: A
randomized, double-blinded trial comparing enoxaparin
with warfarin. Ann Intern Med 1996;124:619-626.
Lieberman JR, Wollaeger J, Dorey F, et al: The efficacy
of prophylaxis with low-dose warfarin for prevention of
pulmonary embolism following total hip arthroplasty.
J Bone Joint Surg Am 1997;79:319-325.
Lotke PA, Palevsky H, Keenan AM, et al: Aspirin and
warfarin for thromboembolic disease after total joint arthroplasty. Clin Orthop 1996;324:251-258.

Montgomery KD, Geerts WH, Potter HG, Helfet DL:

Thromboembolic complications in patients with pelvic
trauma. Clin Orthop 1996;329:68-87.
NIH Consensus Development Conference Statement:
Prevention of Venous Thrombosis and Pulmonary Embolism. Bethesda, MD, US Department of Health and
Human Services, Office of Medical Application of Research, 1986, vol 6, no 2.
Pellegrini VD Jr, Clement D, Lush-Ehmann C, Keller
GS, Evarts CM: Natural history of thromboembolic disease after total hip arthroplasty. Clin Orthop 1996;333:
27-40.
RD Heparin Arthroplasty Group: RD heparin compared with warfarin for prevention of venous thromboembolic disease following total hip or knee arthroplasty. J Bone Joint Surg Am 1994;76:1174-1185.
Salvati EA, Pellegrini VD Jr, Sharrock NE, et al: Recent
advances in venous thromboembolic prophylaxis during
and after total hip replacement. J Bone Joint Surg Am
2000;82:252-270.
Sarmiento A, Goswami AD: Thromboembolic prophylaxis with use of aspirin, exercise, and graded elastic
stockings or intermittent compression devices in patients managed with total hip arthroplasty. J Bone Joint
Surg Am 1999;81:339-346.
Warwick D, Williams MH, Bannister GC: Death and
thromboembolic disease after total hip replacement: A
series of 1162 cases with no routine chemical prophylaxis. J Bone Joint Surg Br 1995;77:6-10.

Mizel MS, Temple HT, Michelson JD, et al: Thromboembolism after foot and ankle surgery: A multicenter
study. Clin Orthop 1998;348:180-185.

American Academy of Orthopaedic Surgeons

175

Chapter

17

Blood Transfusion
John S. Xenos, MD
Andrew G. Yun, MD

Introduction
The perioperative management of blood products in the
orthopaedic patient is becoming more sophisticated as
new data and products become available. The drive for
newer techniques of perioperative blood management is
fueled by the desire to avoid transfusion of allogeneic
blood products and thereby decrease risks associated
with the use of these products, such as transmission of
disease, immunosuppression, infection, and transfusion
reactions.
Also, the supply of allogeneic blood products is expected to exceed demand in the future. According to a
recent study, the domestic supply of blood in the United
States in 1997 was found to be 5.5% less than in 1994,
with the rate of whole blood collections being 12.6%
lower in the population age 18 to 65 years. The red
blood cell (RBC) transfusion rate remained the same,
however, whereas the transfusion rates of platelets and
plasma increased. Although the current margin of allogeneic blood supply is adequate, there is legitimate concern that the future demand for blood products will exceed supply.

Blood Products: Their Role and Indications

Blood products currently used in clinical practice include RBCs, platelets, fresh frozen plasma, and cryoprecipitated antihemophilic factor.

Red Blood Cells

RBCs are concentrated erythrocytes from whole blood
either by an apheresis or by centrifuge. Packed red
blood cells and red cells are terms synonymous with red
blood cells. The typical preparation comprises RBCs
with a hematocrit range of 60% to 70% and approximately 50 mL of acellular plasma, citrate for anticoagulation, and a preservative solution. ABO compatibility
as well as compatibility with other antibodies must be
established before transfusion.
The role of RBC transfusion is to replace deficient
circulating RBC mass in the face of compromised
oxygen-carrying capacity or tissue hypoxia. It is impor-

American Academy of Orthopaedic Surgeons

tant to draw a distinction between the clinical manifestation of hypovolemia and anemia when considering
transfusion of RBCs. Certainly, multiple patient-specific
factors, such as cardiovascular status and other comorbidities that would affect the patients ability to tolerate
tissue hypoxia, come into play in this determination. The
true limit of tolerance of anemia is not known. In
healthy adults, hemoglobin levels as low as 6.0 g/dL may
be tolerated and oxygen delivery is maintained.
One unit of RBCs transfused in a patient of normal
size and not actively bleeding or in hemolysis should result in a 1.0-g/dL increase in hemoglobin. The current
transfusion guidelines recommended by the American
Society of Anesthesiologists are almost always to transfuse when the hemoglobin level is less than 6.0 g/dL and
almost never when greater than 10.0 g/dL (Figure 1). A
National Institutes of Health consensus document recommends that good clinical judgment should be the basis for appropriate perioperative transfusion rather than
the use of a single criterion such as the hemoglobin
level. In a 1999 study of patients undergoing total joint
arthroplasty, a transfusion trigger of 7.0 g/dL was used
unless symptoms were present in the face of higher hemoglobin levels. Most would agree that establishment of
an absolute transfusion trigger based on hemoglobin is
inappropriate and that avoidance of unnecessary transfusion is desirable.
The specific indications for transfusion in patients
with hemoglobin levels between 6.0 g/dL and 10.0 g/dL
should be based on the patients risk associated with
anemia after other measures, such as volume repletion,
have been attempted. Patients should have adequate
volume resuscitation before RBC transfusion is considered. Blood loss of less than 15% of total volume is well
tolerated by most patients with minimal symptoms. In
patients with 30% blood loss, tachycardia is present, and
30% to 40% loss of blood volume is associated with
signs of severe shock. Invasive monitoring is the most
effective method to assess tissue perfusion. Other measures to consider include heart rate, blood pressure, hemoglobin, and status of bleeding (active, controlled, or
uncontrolled.)

177

Blood Transfusion

Figure 1 General transfusion guidelines. Use of an absolute transfusion trigger

should be avoided. PRBC = packed red blood cell; EKG = electrocardiogram.

Orthopaedic Knowledge Update 8

mented clinically and/or with laboratory parameters. A
prothrombin time greater than 18 seconds or 1.5 times
normal, an activated partial prothrombin time above
55 to 60 seconds, or a coagulation factor assay demonstrating less than 25% activity are suggestive of coagulopathy. Other indications for fresh frozen plasma infusion include massive blood transfusion of greater than
one blood volume with evidence of coagulopathy or
documented coagulation factor deficiency before surgery, such as factor II, V, VII, X, XI, or XIII, von Willebrands disease, or factor VIII deficiency when cryoprecipitate is not available. Conditions with multiple
coagulation factor deficiencies as seen in liver disease or
vitamin K deficiency may also require fresh frozen
plasma infusion. The effects of warfarin may be reversed
before an invasive procedure when the prothrombin
time is greater than 18 seconds or the international normalized ratio is greater than 1.5. Fresh frozen plasma is
not indicated for volume augmentation alone in the absence of coagulopathy. The usual dosage is 2 units, although correction of severe coagulopathy may require
significantly greater doses of fresh frozen plasma. Monitoring should be performed with repeated laboratory
parameters following fresh frozen plasma infusion.

Platelets
Platelets, cells required for hemostasis, are prepared either as random donor platelets or as platelet pheresis
also known as single donor platelets. Random donor
platelets are derived from whole blood and usually contain approximately 7.5 1010 platelets per bag. Greater
platelet quantities are obtained in single donor platelets
preparations with approximately 4.5 x 1011 platelets per
bag.
The primary indication for platelet infusion is when
platelet counts are below 5,000/mm3 regardless of
bleeding. General guidelines call for platelet transfusion
with counts less than 50,000/mm3 before a major surgical intervention. Platelet infusions are not recommended for counts greater than 100,000/mm3. The usual
dosage of platelets is 6 units or 1 unit per 10 kg of body
weight for random donor platelets or one apheresis
unit. Reinfusion is generally required every 3 to 5 days
in the absence of platelet consumption and more often
when platelet consumption exists. Daily platelet counts
should be monitored. Platelets should be administered
to ABO-compatible patients as with RBC transfusions.
Alloimmunized patients may require platelet apheresis
units that have an HLA match.

Fresh Frozen Plasma

Fresh frozen plasma is the acellular component of blood
and contains all of the peptide coagulation factors. It is
separated from the RBCs and platelets and frozen at
18C within 8 hours of collection. Use of fresh frozen
plasma is indicated to address coagulopathy, docu-

178

Cryoprecipitated Antihemophilic Factor

Cryoprecipitated antihemophilic factor is the precipitable protein fraction derived from fresh frozen plasma.
The typical preparation contains: factor VIII C, 80 to
120 units, von Willebrands factor, 80 units, fibrinogen,
200 to 300 mg, and factor XIII, 40 to 60 units. Cryoprecipitated antihemophilic factor is considered to be acellular and does not require compatibility testing. Its use
is indicated to address hypofibrinogenemia (rare),vonWillebrands disease, and hemophilia A. General indications are for use in patients with von Willebrands disease not responsive to desmopressin, bleeding with von
Willebrands disease, or fibrinogen levels less than 80 to
100 mg/dL.
The dosage of cryoprecipitated antihemophilic factor required varies depending on the indications for usage. The recommended dosage for hypofibrinogenemia
is one bag per 5 kg of body weight, whereas for von
Willebrands disease one bag per 10 kg body weight is
usually required daily. Monitoring should be performed
using measurement of ristocetin cofactor, factor VIII antigen, or von Willebrands factor multimer levels. Cryoprecipitated antihemophilic factor is used in patients
with hemophilia A to replace factor VIII C. Fifteen percent of hemophiliacs have inhibitors to factor VIII activity. The formula used to calculate dosage of the number
of bags of antihemophilic factor for infusion is [(plasma
volume in mL % increase needed in factor VIII)/100]/
80. Factor VIII activity is used to monitor therapeutic
effect of the infused antihemophilic factor.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Assessment of the Patient During and After

the Administration of a Blood Product
The transfusion of a specific blood product is indicated
only when that blood product is deficient and the risk of
transfusion is outweighed by the potential benefit.
RBCs, fresh frozen plasma, cryoprecipitated antihemophilic factor, or platelets should never be used for volume expansion alone. Although a specific laboratory
value as the sole basis for a transfusion decision is not
appropriate, some general principles should be followed.
The patients vital signs should be monitored closely
during transfusion of a blood product, with close attention paid to temperature elevation, tachycardia, respiratory rate, blood pressure, and mental status. A significant temperature elevation of more than 2C may
indicate bacterial infection or an acute hemolytic reaction. Dyspnea, chest pain, or tachypnea may reflect fluid
overload with subsequent pulmonary edema, which may
occur especially in at-risk patients or patients given
large transfusions over a short period of time. Pulmonary edema may not manifest clinically for up to 24
hours after transfusion. Hypothermia after rapid infusion of large blood volumes may result if the blood is
chilled and an approved blood warmer is not used. Finally, metabolic complications may occur when the volume of transfused blood exceeds the patients normal
blood volume. These metabolic complications include
hypokalemia or hyperkalemia, citrate toxicity, and ammonia toxicity with possible acidosis or alkalosis. In addition to measurement of circulating electrolytes and
renal function, ionized calcium testing and electrocardiogram monitoring may be indicated.

Safety in Banking of Blood and

Blood Products
The risks associated with blood transfusions include
transmission of viral disease, hemolysis caused by ABO
incompatibility, acute lung injury, transmission of bacteria or endotoxins, transmission of parasites, graft versus
host disease, alloimmunization, allergic reaction, and immunosuppression.

Infectious Disease
Transmission of infectious disease via transfusion of human blood may be caused by known or unknown organisms. Current screening criteria exclude donors with human immunodeficiency virus (HIV), human T cell
leukemia virus (HTLV), hepatitis, and other known
agents. Blood is also screened for hepatitis B and C,
HIV-1 and -2 including HIV antigen, and HTLV-I and
HTLV-II. Although the risk of viral transmission has not
been totally eliminated, the risk of transmission of hepatitis C and HIV has been decreased to less than 1:2 to 4

American Academy of Orthopaedic Surgeons

Chapter 17 Blood Transfusion

million using newer screening techniques such as nucleic
acid testing.
Blood components containing white blood cells are
unpredictably prone to contain cytomegalovirus, with up
to 70% of donors positive for the disease. Transmission
of cytomegalovirus is a concern primarily for immunocompromised patients. Some infectious agents, such as
Babesia sp., Bartonella sp., Borrelia sp., Brucella sp., Colorado tick fever, Leishmania sp., Parvovirus sp., plasmodia, rickettsia, Toxoplasma sp., malaria, and some trypanosomes, are not screened for because pertinent tests
are not available. Theoretically, Creutzfeldt-Jakob disease may also be transmitted through blood products.
Although rare, bacterial contamination may occur
and may cause severe effects, including death. In the
United States, septic reactions from bacterial contamination of blood components resulted in 77 of 694
transfusion-related deaths reported to the Food and
Drug Administration (FDA) from 1985 to 1999. Platelets are the most likely among blood components to be
contaminated. Skin flora are the most common pathogens. Endotoxemia may occur as a result of Yersinia enterocolitica contamination of RBC products.

Immunologic Complications
Immunologic complications can be acute or delayed and
may vary significantly in clinical presentation. Although
current cross-matching techniques minimize risk for immune reactions, the incidence has not been completely
eliminated.
Hemolytic transfusion reactions usually occur as a
result of antigen or ABO incompatibility of transfused
RBCs, although nonimmunologic reactions may occur.
Patient symptoms include elevated temperature and
pulse, chest or back pain, and dyspnea. In unresponsive
patients or patients under general anesthesia, disseminated intravascular coagulopathy may be the initial clinical presentation. Laboratory studies helpful to confirm
the diagnosis include hemoglobin, serum bilirubin, and
direct antiglobulin test. Delayed hemolytic reactions are
possible in patients who are alloimmunized, with presentation of symptoms as late as 14 days after transfusion. In contrast to the potentially fatal course of acute
hemolytic reactions, delayed hemolytic reactions do not
require treatment and are usually benign.
Febrile nonhemolytic reactions and allergic reactions
may also occur and are common. Febrile reactions occur
in 1% of transfusions and require only antipyretics for
palliative treatment. Most allergic reactions are minor
with the primary report of urticaria and are adequately
treated with antihistamines. Anaphylactic reactions, although rare, may also occur and require aggressive immediate treatment.
A potentially fatal immunologic complication is
transfusion-related acute lung injury, which occurs

179

Blood Transfusion
through an unknown mechanism. This condition manifests within 6 hours (most often occurring within
2 hours) of transfusion and requires aggressive pulmonary treatment and maintenance of the hemodynamic
status of the patient. Clinical manifestation of
transfusion-related acute lung injury is the sudden onset
of pulmonary edema, shortness of breath, hypotension
unresponsive to intravenous fluids, and hypovolemia
with or without fever. Rapid progression of severe hypoxia follows. Results from radiographic evaluation of
the chest mimic adult respiratory distress syndrome. Although adult respiratory distress syndrome and
transfusion-related acute lung injury are clinically identical, other causes of pulmonary edema should be excluded to confirm the diagnosis of transfusion-related
acute lung injury. Risk factors for this condition are unknown and most patients have not been reported to
have a previous history of transfusion reaction.
Other delayed immunologic complications are alloimmunization and graft versus host disease. Alloimmunization, which may occur after transfusion of any
blood component and is asymptomatic, may result in
shortened survival of subsequent blood products transfused, as in immune-mediated platelet destruction. Graft
versus host disease may occur when T lymphocytes in
the transfused component react to recipient tissue antigens.
Immunosuppression is associated with transfusion of
allogeneic blood and is considered to be one of the
greatest risks to the recipient. Although the mechanism
is not well defined in humans, several studies have demonstrated transfusion of allogeneic blood to be an independent risk factor for development of postoperative
infection in patients undergoing uncontaminated orthopaedic surgery. In a multicenter study of patients undergoing total joint arthroplasty, the infection rate in patients who had allogeneic blood transfusions was more
than twice that of patients who did not undergo transfusion. Other studies suggest a range of risk of infection to
be zero to 10 times greater in patients receiving allogeneic blood.

Trends in Transfusion Medicine

Preoperative assessment of the patient is important in
determining the risk for postoperative transfusion. The
most significant factor associated with risk of postoperative transfusion is preoperative hemoglobin. Most recent data indicate that a preoperative hemoglobin level
of less than 13.0 g/dL carries a substantial increased risk
of transfusion. Conversely, a preoperative hemoglobin
level of more than 15.0 g/dL has a significantly diminished risk. Duration of surgery, smaller patient size, and
female gender have also been associated with increased
risk of transfusion in several studies. Preoperative patient assessment for transfusion risk is essential for im-

180

Orthopaedic Knowledge Update 8

proved planning and minimizing the use of allogeneic
blood. Effective perioperative blood management strategy may include use of agents that increase the preoperative hemoglobin level, agents or techniques that minimize intraoperative and postoperative blood loss, or
protocols that allow for reinfusion of salvaged blood.
Recombinant human erythropoietin (EPO), a glycoprotein hormone secreted by the kidneys with a minor
hepatic contribution in conditions of extreme stress, is
responsible for regulation of erythropoiesis. Data suggest that there is a dose-related relationship between allogeneic transfusion and preoperative administration of
EPO such that patients given higher doses of EPO are
at lower risk of transfusion when undergoing major orthopaedic operations. Use of EPO is indicated for patients with a preoperative hemoglobin level between
10.0 g/dL and 13.0 g/dL and an expected blood loss of
two units of blood. No significant adverse reactions
have been associated with preoperative use of EPO in
appropriate patients.
Reported use of EPO in athletes as a form of blood
doping is of concern. The term blood doping refers to
the use of blood transfusion as an ergogenic aid but has
also been applied to use of EPO. Although banned by
the International Olympic Committee, illicit use of EPO
is difficult to detect. Urine tests that can differentiate
recombinant human EPO and endogenous EPO have
been described but have not yet been determined to be
a reliable screening tool.
Perioperative use of pharmacologic therapies to reduce blood loss is an effective strategy to reduce risk of
blood transfusion. Topical use of fibrin tissue sealants
has been studied extensively. Fibrin tissue adhesive is
composed of variable amounts of thrombin and fibrinogen. According to a 1999 study, use of fibrin tissue adhesive in total knee arthroplasty reduces postoperative
blood loss and requirement for blood transfusions from
28% to 17%. Antifibrinolytic agents such as aprotinin,
tranexamic acid, and aminocaproic acid have been
shown to reduce surgical blood loss. Aprotinin is a naturally occurring broad-spectrum protease peptide that inhibits trypsin, plasmin, tissue kallikreins, and plasmin
kallikreins. Although the exact mechanism of action is
not known, aprotinin is thought to regulate fibrinolysis,
stabilize platelet function, and modulate the intrinsic coagulation pathway. Aprotinin, used commonly in cardiothoracic surgery and liver transplantation, has been
shown to have slightly better efficacy than the other antifibrinolytic agents. Aprotinin is also effective in orthopaedic surgery and has been determined to be a safe
agent for maintaining hemostasis in total hip arthroplasty. There is a theoretical increased risk of deep
venous thrombosis with this agent, however. Desmopressin is a valuable agent to minimize blood loss in patients with hemophilia A and von Willebrands disease,
but has not been shown beneficial in patients without

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

coagulopathy undergoing elective surgery. Recombinant
factor VIIa (eptacog alfa) is currently under investigation but preliminary data suggest that it may be effective in reduction of blood loss in surgical patients without coagulopathy.
Acute normovolemic hemodilution is a technique
during which whole blood is drawn from the patient before or early in the surgical procedure, usually after initiation of anesthesia, while intravascular volume is
maintained with a crystalloid and solutions. Thus, the
blood loss during the procedure is diluted and the whole
blood removed at the beginning of the procedure is reinfused. Acute normovolemic hemodilution has been
shown to be as effective as preoperative autologous
blood donation in patients undergoing total hip arthroplasty in the reduction of risk of allogeneic blood transfusion. Hypervolemic hemodilution has been found to
be as beneficial as normovolemic hemodilution in one
study and has the advantage of being simpler to perform without the need for blood withdrawal. Patients
with cardiac disease may not be candidates for either
technique.
Intraoperative salvage of shed blood that is spun and
washed of contaminants and debris is well established as
a means of avoiding blood transfusion. Comparison of
different techniques of intraoperative cell salvage such
as centrifuge, direct transfusion, and ultrafiltration has
shown no difference in transfusion risk in each of these
modalities. Cell salvage is generally considered in patients in whom intraoperative blood loss is expected to
exceed 1,000 mL. Postoperative blood salvage requires
transfusion of unwashed blood from a blood collection
device. Several studies have indicated that postoperative
blood salvage significantly reduces the risk for postoperative blood transfusion. The safety of this technique is
somewhat controversial because there are increased
risks of infection by contaminated blood, as well as reports of hyperthermia and hypotension.

Blood Substitutes
The ideal blood substitute would be readily available
and without the associated risks of transfusion and
transfusion-related infections. Blood alternatives are either plasma expanders or oxygen-based carrier substitutes. Blood volume expanders include crystalloid and
colloid replacement. Crystalloid replacement with combinations of saline, lactated Ringers solution, and glucose solutions are the first line of volume supplementation. Patients who are at risk of excessive intravascular
volume overload may tolerate colloid replacement with
albumin, dextran, or gelatins.
Oxygen-based carrier substitutes encompass a novel
class of therapeutic agents. Hemoglobin-based preparations are either cell free or liposome enclosed as a substrate carrier. The molecules are genetically engineered

American Academy of Orthopaedic Surgeons

Chapter 17 Blood Transfusion

human hemoglobin from Escherichia coli, salvaged from
expired human blood, or purified from bovine blood.
The success of first-generation, cell-free preparations
was marred by nephrotoxicity and excessive oxygen
binding affinity. Some preparations have been discontinued after clinical trials demonstrated increased mortality in trauma patients with hemoglobin substitutes compared with placebo. The second-generation, liposomebased hemoglobin carriers currently demonstrate some
promise in preclinical animal studies and phase 2 and 3
clinical trials.
Perfluorocarbon emulsions are another alternative
to hemoglobin-based blood products and substitutes.
These agents are hydrocarbon analogs substituting fluorine for hydrogen. Perfluorocarbons have 20 to 30 times
the oxygen solubility of plasma. Fluosol-DA-20 is currently the only FDA-approved blood substitute. Its use,
however, has been associated with hepatosplenomegaly,
complement-induced hypertension, and anaphylactoid
reaction.

Religious Practices and Transfusion

Alternatives
The Jehovahs Witnesses is a growing religious group
that originated during the Adventist movement of the
19th century. The Watchtower Bible and Tract Society is
the controlling theological organization and has outlined rigid religious doctrine with specific relevance to
medical care. Jehovahs Witnesses freely accept medical
and surgical care, but most refuse the acceptance of
blood and blood products based on their interpretation
of the Old Testament. Deviation from this doctrine not
only is believed to be a direct spiritual betrayal, but also
carries profound social implications with shunning
and excommunication from church and family.
Judicial precedent protects the sacrosanct position of
Jehovahs Witnesses to refuse blood willingly against
medical advice. Under the protection of informed consent, the Supreme Court and state courts have ruled repeatedly that the integrity of the medical profession
cannot supersede the wishes or religious beliefs of the
patient. The American Medical Association has likewise
supported physicians and hospitals that provide care of
Jehovahs Witnesses who provide informed consent.
Most Jehovahs Witnesses will not accept blood or
most blood products, including whole blood, red blood
cells, autologous blood donation, platelets, and plasma.
Under specific circumstances outlined by the Watchtower Society, however, Jehovahs Witnesses may accept
clotting factors, stem cells, bone marrow transplants, and
separated plasma proteins of albumin, fibrin, and globulin. Jehovahs Witnesses may also accept dialysis, plasmapheresis, and intraoperative blood salvage where the
extracorporeal circulation is uninterrupted. The medical
team needs to consult with each patient to determine

181

Blood Transfusion
acceptable alternatives. The Watchtower Society readily
provides patient resources through its Hospital Information Society and its regional Hospital Liaison Committees.
The surgical care of Jehovahs Witnesses can be successfully managed using anesthetic and pharmacologic
alternatives. Many Jehovahs Witnesses will accept preoperatively the red cell production stimulating factor
EPO and white cell stimulators granulocyte-colony
stimulating factor and granulocyte-macrophage colony
stimulating factor. Jehovahs Witnesses accept the use of
fibrin glue as a hemostatic agent intraoperatively and
the pharmacologic antifibrinolytics such as aprotinin
and tranexamic acid. Proven anesthetic techniques in
these patients involve hypotension, hypothermia, and
acute normovolemic hemodilution. Intravascular volume can be acceptably maintained with colloid and
crystalloid replacement.
With such protocols, major surgery on Jehovahs
Witnesses has been successfully performed. The combined literature from cardiac, urologic, and obstetric surgery has demonstrated successful outcomes after major
surgery without blood transfusion. Safe outcomes have
also been reported after total joint surgery and major
spine surgery. A study of 100 Jehovahs Witnesses undergoing total hip replacement reported no deaths, and
smaller studies of spinal fusion for scoliosis have also reported success without mortality.
Controversy remains regarding the emergent care of
the exsanguinating and obtunded patient and in the care
of minors. A 1998 report determined that during a lifethreatening emergency when the patient cannot provide
informed consent and where there is no valid advance
directive, emergency blood transfusions may be given
until the patients condition stabilizes. Interestingly, patients who are unconsciously transfused are not subject
to religious sanction or the threat of excommunication.
Regarding children, the American Academy of Pediatrics in 1997 stated that the constitutional guarantees of
freedom of religion do not permit children to be
harmed through religious practice. Furthermore, an
anonymous group of Jehovahs Witnesses called the Associated Jehovahs Witnesses for Reform on Blood Policy have voiced dissent over the religions official blood
policy. Consultation with the hospital ombudsman or
ethics committee may be necessary to obtain consent in
cases where life-saving transfusion is essential in a juvenile or incompetent patient whose guardian is a Jehovahs Witness.

Autologous Blood Donation: Current Trends

The impetus for the use of preoperative autologous
blood donation in perioperative blood management in
orthopaedic surgery is to avoid the major risks of allogeneic transfusions, especially the transmission of dis-

182

Orthopaedic Knowledge Update 8

ease and immune-mediated transfusion reactions. The
potential risk of transfusion error caused by clerical issues still exists with the use of preoperative autologous
blood donation, but the major disadvantage is the cost
related to this method. The high cost of preoperative autologous blood donation is magnified by the significant
wastage of blood that occurs with preoperative autologous blood donation. Rates of discarded blood have
been reported to be as high as 83% of units, with most
studies demonstrating wastage in the 50% range. Better
identification of patients likely to require transfusion
before embarking on preoperative autologous blood donation is required to reduce the waste associated with
this technique.
Most studies indicate that preoperative autologous
blood donation is an effective way to minimize risk of
allogeneic transfusion and lend support for its continued
use. One concern related to preoperative autologous
blood donation is donation-induced preoperative anemia. Blood donations within 2 weeks of surgery are
more likely to result in preoperative anemia because of
an insufficient erythropoietic response. A recent study
of total joint arthroplasty patients presented convincing
data that the use of EPO preoperatively in patients also
undergoing preoperative autologous blood donation significantly decreased the allogeneic transfusion rate in
patients either receiving EPO or using preoperative autologous blood donation alone. Patients with a hemoglobin level less than 10.0 g/dL are not considered candidates for preoperative autologous blood donation.
Transfusion of autologous blood continues to be the
most commonly used alternative to allogeneic transfusion.

Annotated Bibliography
Allain JP: Transfusion risks of yesterday and of today.
Transfus Clin Biol 2003;10:1-5.
Improved screening techniques decrease risk of disease
transmission with transfusion but may not be cost effective.

Bezwada HP, Nazarian DG, Henry DH, Booth RE: Preoperative use of recombinant human erythropoietin before total joint arthroplasty. J Bone Joint Surg Am 2003;
85:1795-1800.
In this prospective randomized study, the efficacy of erythropoietin and autologous blood in combination was demonstrated to have the lowest risk of allogeneic blood transfusion.

Cable R, Carlson B, Chambers L, et al (eds): Practice

Guidelines for Blood Transfusion: A Compilation From
Recent Peer-Reviewed Literature. Washington, DC,
American National Red Cross, 2002, pp 1-48.
A thorough review of blood products, indications, and
risks associated with transfusion is presented.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Dodd RY: Bacterial contamination and transfusion safety: Experience in the United States. Transfus Clin Biol
2003;10:6-9.
Septic transfusion reactions from 1985-1999 are reviewed.

Salido JA, Marin LA, Gomez LA, Zorrilla P, Martinez

C: Preoperative hemoglobin levels and the need for
transfusion after prosthetic hip and knee surgery: Analysis of predictive factors. J Bone Joint Surg Am 2002;84:
216-220.
The most important predictive factor related to risk of
postoperative transfusion is preoperative hemoglobin. Duration of surgery, gender, and patient size were also predictive
factors.

Sullivan MT, McCullough J, Schreiber GB, Wallace EL:

Blood collection and transfusion in the United States in
1997. Transfusion 2002;42:1253-1260.
A review of supply and demand of blood products in 1994
and 1997 for comparison.

Classic Bibliography
Anders MJ, Lifeso RM, Landis M, Mikulsky J, Meinking
C, McCracken KS: Effect of preoperative donation of
autologous blood on deep-vein thrombosis following total joint arthroplasty of the hip or knee. J Bone Joint
Surg Am 1996;78:574-580.
Bierbaum BE, Callaghan JJ, Galante JO, Rubash HE,
Tooms RE, Welch RB: An analysis of blood management in patients having a total hip or knee arthroplasty.
J Bone Joint Surg Am 1999;81:2-10.
Faris PM, Ritter MA, Abels RI: The effects of recombinant human erythropoietin on perioperative transfusion
requirements in patients having a major orthopaedic operation: The American Erythropoietin Study Group.
J Bone Joint Surg Am 1996;78:62-72.

Chapter 17 Blood Transfusion

patients with and without deposited autologous units.
J Bone Joint Surg Am 2000;82:951-954.
Hatzidakis AM, Mendlick R, McKillip T, Reddy RL,
Garvin KL: Preoperative autologous donation for total
joint arthroplasty: An analysis of risk factors for allogeneic transfusion. J Bone Joint Surg Am 2000;82:89-100.
Levy O, Martinowitz U, Oran A, Tauber C, Horoszowski
H: The use of fibrin tissue adhesive to reduce blood loss
and the need for blood transfusion after total knee arthroplasty. J Bone Joint Surg Am 1999;81:1580-1588.
Lundberg GD: Practice parameter for the use of freshfrozen plasma, cryoprecipitate, and platelets. JAMA
1994;271:777-781.
Migden DR. Braen GR: The Jehovahs Witness blood
refusal card: Ethical and medicolegal considerations for
emergency physicians. Acad Emerg Med 1998;5:815-824.
Murkin JM, Haig GM, Beer KJ, et al: Aprotinin decreases exposure to allogeneic blood during primary
unilateral total hip arthroplasty. J Bone Joint Surg Am
2000;82:675-684.
NIH Consensus Statement available on perioperative
red cell transfusion. Am J Public Health 1988;78:1588.
Practice guidelines for blood component therapy: A report by the American Society of Anesthesiologists Task
Force on Blood Component Therary. Anesthesiology
1996;84:732-747.
Religious objections to medical care: American Academy of Pediatrics Committee on Bioethics. Pediatrics
1997;99:279-281.

Feagin BG, Wong CJ, Kirkley A, et al: Erythropoietin

with iron supplementation to prevent allogeneic blood
transfusion in total hip joint arthroplasty: A randomized, controlled trial. Ann Intern Med 2000;133:845-854.

Simon TL, Alverson DC: AuBuchon J, et al: Practice parameter for the use of red blood cell transfusions: Developed by the Red Blood Cell Administration Practice
Guideline Development Task Force of the College of
American Pathologists. Arch Pathol Lab Med 1998;122:
130-138.

Goodnough LT, Despotis GJ, Merkel K, Monk TG: A

randomized trial comparing acute normovolemic hemodilution and preoperative autologous blood donation
in total hip arthroplasty. Transfusion 2000;40:1054-1057.

Sparling EA, Nelson CL, Lavender R, Smith J: The use

of erythropoietin in the management of Jehovahs Witnesses who have revision total hip arthroplasty. J Bone
Joint Surg Am 1996;78:1548-1552.

Grosvenor D, Goyal V, Goodman S: Efficacy of postoperative blood salvage following total hip arthroplasty in

Spence RK: Surgical red blood cell transfusion practice

policies. Am J Surg 1995;170(suppl 6A):3S-15S.

American Academy of Orthopaedic Surgeons

183

Chapter

18

Bone Metabolism and Metabolic

Bone Disease
Mary A. Murray, MD

Introduction
Osteoporosis is a major public health care issue. In the
United States, the cost of fracture treatment is estimated at $10 to $15 billion annually. Although bone
mineral density (BMD) is the current measure of fracture risk, it is becoming evident that there is much that
remains to be learned about predicting fracture occurrence. BMD is not the sole determinant of fracture
riskanatomic considerations, ethnicity, and lifestyle all
are contributing factors.
Age-related bone loss is universal; therefore, factors
that compromise the ability to attain peak bone mass in
young adult life increase the likelihood of osteoporosis
with aging. Osteoporosis may represent one of several
adult diseases in which intervention during the pediatric
and adolescent years may alter the course of the disease. Strategies to increase peak bone mass may include
maximizing calcium and vitamin D intake, minimizing
the use of drugs that inhibit normal bone growth and
formation, encouraging weight-bearing activities, and
ensuring normal sex steroid exposure. Various professional societies have recently established working
groups whose goals are to better characterize and understand pediatric and adolescent bone health and formation and their effects on adult bone health.

Basic Science Aspects of Bone Metabolism

The three basic functions of bone are (1) mechanical
support and locomotion, (2) protection of the internal
organs, and (3) metabolic activities to maintain calcium
homeostasis. Bone is responsive to hormonal signals at
all stages of life. The cellular constituents that mediate
hormonal signals within the bone are osteoblasts and
osteoclasts.
Osteoblasts are derived from a mesenchymal cell lineage and are mainly responsible for synthesizing the organic, nonmineralized bone matrix. They are cuboidal
cells aligned in layers along the osteoid and possess features consistent with protein-producing and -secreting
cells. Osteoblasts differentiate under the influence of
growth factors including fibroblast growth factor (FGF)

American Academy of Orthopaedic Surgeons

and bone morphogenetic proteins. They synthesize and

secrete type I collagen and other matrix proteins. Osteoblasts contain receptors for most of the systemic mediators of bone remodeling (parathyroid hormone [PTH],
sex steroids, glucocorticoids, 1, 25-dihydroxyvitamin D3
[1,25(OH)2D3], insulin, and thyroid hormone) as well as
local factors such as cytokines and interleukins. Osteoblasts can stimulate osteoclastogenesis by secretion of
factors that act in a paracrine manner, such as colonystimulating factor-1, osteoprotegerin, and receptor activator of nuclear factor-kappaB ligand (RANKL). A
membrane rich in alkaline phosphatase is the hallmark
of the osteoblast cell. Alkaline phosphatase is released
into the serum and is elevated at times of increased
bone formation (such as pubertal growth acceleration).
At the end of its life cycle, an osteoblast becomes either
a flat lining cell or an osteocyte.
Osteocytes are differentiated osteoblasts that are incorporated into the bone matrix during bone formation.
They possess long cytoplasmic arms that provide a
mechanism for intercellular communication. Osteocytes
possess cell surface receptors for PTH and respond to
receptor binding with a release of calcium from the
bone into the circulation. Osteocytes also respond to
mechanical signals and transfer stress and strain indices
to surrounding cells to effect bone remodeling. Mechanical stimulation has been shown to increase the production of prostaglandin E2 by cells of the osteoblastic lineage; this substance stimulates both bone formation and
bone resorption.
Osteoclasts are large, multinucleated giant cells
formed by fusion of mononuclear cells; they are responsible for bone resorption. They are derived from hematopoietic cells related to monocyte/macrophage lineages
and are found within Howships lacunae in contact with
calcified bone. Osteoclasts have a ruffled border surrounded by a clear zone (ring of actin) that attaches the
cell to the bone surface and seals off the resorption compartment. Attachment is accomplished through integrin
receptors that bind to specific matrix proteins in response to specific signaling molecules. Osteoclasts express high levels of tartrate-resistant acid phosphate and

187

Bone Metabolism and Metabolic Bone Disease

Table 1 | Factors That Influence Renal Resorption of

Calcium
Factors That Increase Renal
Calcium Resorption

Factors That Decrease Renal

Calcium Resorption

PTH
PTH-related peptide
1,25-dihydroxyvitamin D
Calcitonin
Increased phosphate intake
Chronic thiazide diuretic use

Increased sodium intake

Increased calcium intake
Metabolic acidosis
Phosphate depletion
Glucocorticoids
Furosemide (loop diuretics)
Cyclosporin A

cathepsin K through the ruffled border. The bone resorption compartment contains lysosomal enzymes and
the bony substrate in acidic pH. The acidic pH causes
dissolution of hydroxyapatite crystals; the lysosomal enzymes digest matrix proteins resulting in resorption of
bone within the attachment site of the osteoclast.

Hormonal Regulation of Cell Differentiation

Calcium and Phosphorus Metabolism
Calcium is important in regulating muscle contraction,
coagulation, intracellular signal transduction, and control of cell membrane potentials. Calcium balance is
regulated by intestinal absorption of dietary intake, excretion by the kidney, and storage in the skeleton. Normal serum calcium levels are tightly regulated with a
normal range of 8.5 to 10 mg/dL (2.2 to 2.5 mmol/L).
In the pediatric population calcium levels reach a nadir
at 1 week of age (range, 7.5 to 10 mg/dL, 1.9 to 2.5
mmol/L) and then increase to a slightly higher level
than the normal adult range (range, 9.0 to 11.0 mg/dL,
2.3 to 2.8 mmol/L). In the circulation, calcium exists approximately equally in the free (ionized) form and the
protein-bound form (the major binding protein is albumin). A small portion is bound to anions, such as phosphate. The skeleton acts as a reservoir for calcium that
can be metabolized to maintain circulating concentrations.
Dietary calcium absorption in the intestines occurs
in the duodenum and the jejunum and is a function of
active, saturable transport regulated by vitamin D and
passive, gradient dependent transport. Calcium absorption is regulated by 1,25(OH)2D concentration and by
calcium intake. Both increased calcium intake and increased 1,25(OH)2D serum concentrations act to increase intestinal calcium absorption. In an average
adult, calcium loss in the stool exceeds calcium absorption if calcium intake is less than 200 mg/day (5 mmol/
day). To maintain calcium balance, an average adult
must consume more than 400 mg/day (10 mmol/day). As
calcium intake increases, the efficiency of absorption de-

188

Orthopaedic Knowledge Update 8

creases such that net intestinal absorption tends to plateau at about 300 mg/day when intake exceeds about
1,000 mg/day (25 mmol/day); however, there is wide
variability in net absorption at higher calcium intake
that probably reflects differences in active transport
rather than passive diffusion. The activated form of vitamin D, 1,25(OH)2D, regulates active transport of calcium in the intestine. Elevated concentrations act to increase the intestinal absorption at all calcium intake
levels.
The kidney also regulates calcium balance. To maintain net calcium balance, the renal tubules must resorb
98% of the filtered calcium. Approximately 70% of the
filtered calcium is resorbed in the proximal tubule primarily via passive diffusion. About 20% of calcium resorption takes place in the thick ascending loop of
Henle driven by positive voltage in the lumen generated
by the Na-K-2Cl transporter. The absorption in this portion of the kidney is impaired by loop diuretics such as
furosemide. A calcium-sensing receptor (CaSR) along
the basolateral surface of the cells responds to alterations in the peritubular concentration of calcium. Some
paracellular transport takes place in the tight junctions
via the protein paracellin-1. Remaining resorption
occurs in the distal convoluted tubule, which is the
major site for regulation of calcium excretion; here
calcium enters the cytosol via the renal epithelial calcium channel down an electrical and chemical gradient.
It is then shuttled through the cytoplasm bound to
calbindin-D28K, a vitamin D-dependent protein, without
altering the intracellular free calcium concentration. At
the basolateral membrane, calcium is actively transported across the large chemical gradient to the interstitium by a magnesium-dependent adenosine triphosphatase.
Various factors influence the renal resorption of calcium (Table 1). Genetic mutations that alter renal tubular handling of calcium have been described and some
have been localized to the proteins involved in the
transport of calcium through the renal cells.
Phosphorus also is necessary for adequate mineralization of bone; approximately 85% is present in crystallized form in the skeleton. The remaining 15% is
present largely in extracellular fluid and intracellular
phosphorylated intermediates. Phosphorylated intermediates are important in multiple processes of intermediary metabolism including the production of intracellular
energy and intracellular signaling. Absorption is dependent on passive diffusion and active transport; the serum concentration is less tightly regulated than is calcium concentration. Serum concentrations range
between 2.7 to 4.5 mg/dL (0.81 to 1.45 mmol/L) in
adults; concentrations are higher in infants and young
children and decrease by late adolescence to normal
adult levels. Passive transport is directly related to the
intraluminal concentration of phosphate after eating.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Phosphate absorption is decreased in the vitamin Ddeficient individual. Phosphorus resorption takes place
primarily in the proximal tubule (85%) against an electrochemical gradient. Three sodium gradient-dependent
phosphate transporters have been described (Npt1,
Npt2, and Npt3). Npt2 is highly regulated and responsible for most phosphorus transport. FGF-23 has been
identified as a regulator of Npt2. The PHEX gene product metabolizes FGF-23 and mutations in the PHEX
gene can cause hypophosphatemia because of the inability to metabolize FGF-23 (X-linked hypophosphatemic rickets).

Hormonal Regulation of Calcium and Phosphorus

Metabolism
Parathyroid Hormone
The principal regulators of calcium metabolism are
PTH and vitamin D. PTH acts in the bone to stimulate
the release of calcium and phosphorus into the circulation and in the renal tubules to increase the resorption
of calcium and the excretion of phosphorus. It also increases the 1- hydroxylase activity, thereby raising the
concentration of 1,25(OH)2D, which acts to enhance the
intestinal absorption of calcium and phosphorus.
The major physiologic regulator of PTH secretion is
the extracellular calcium concentration. Low calcium
concentrations increase, and high calcium concentrations decrease the PTH secretory rate, PTH gene expression, and parathyroid cell proliferation. The CaSR
appears to be the principal mediator of calcium concentration and PTH response. The CaSR is a G-protein
coupled receptor; binding activates several intracellular
signaling pathways including protein kinase C and inositol triphosphate, which increase cytosolic calcium concentrations and several mitogen-activated protein kinases. These changes modulate the expression of genes
controlling cell proliferation, differentiation, and apoptosis. CaSR also stimulates the influx of extracellular
calcium ions through a mechanism that is not yet fully
understood. In the parathyroid gland, CaSR has three
key regulatory actions: PTH secretion, PTH gene expression, and cell proliferation. Mouse gene knockout
studies and human gene mutations both have shown the
key role of CaSR in the regulation of PTH. A drop in
serum calcium concentration, perceived by the CaSR, is
translated into a rapid release (in seconds) of PTH, a
decreased intracellular PTH degradation rate that increases the amount of bioavailable PTH for release (in
minutes), increased transcription of the PTH gene (in
hours to days) and eventually increased proliferation of
parathyroid cells (in days to weeks).
PTH has complex actions in bone with a net result
that may be either anabolic or catabolic. Chronic PTH
exposure leads to a catabolic state; increased numbers
of osteoclasts and osteoclast activity increase degrada-

American Academy of Orthopaedic Surgeons

Chapter 18 Bone Metabolism and Metabolic Bone Disease

tion of bone and the release of calcium, phosphate, and
matrix proteins into the circulation. PTH stimulates expression of RANKL on the cell surfaces of osteoblasts,
and stimulates synthesis of macrophage colonystimulating factor, which act together to stimulate maturation of osteoclasts. PTH also inhibits the expression of
osteoprotegerin, a decoy protein that blocks the effect
of RANKL; this action further increases osteoclastogenesis and the activity of mature osteoclasts. Intermittent
PTH exposure, however, appears to result in increased
bone remodeling and a net increase in trabecular bone.
The mechanisms by which PTH increases bone formation are not fully understood. PTH increases the number of osteoblasts and may act by increasing local concentrations of growth factors, such as insulin-like growth
factor-I (IGF-I) and FGFs, which then exert paracrine
actions on the bone to stimulate growth and new bone
formation. This observation of the anabolic actions of
PTH may lead to new therapies for the treatment of osteoporosis.
Vitamin D
Vitamin D is a classic steroid hormone that is derived
from cholesterol and undergoes hydroxylation in the
liver and kidney to its biologically active form,
1,25(OH)2D. The primary role of vitamin D is to maintain calcium levels within the normal range by increasing intestinal absorption of calcium and by increasing
maturation of osteoclasts to mobilize calcium from the
bone when needed. In addition, vitamin D has effects on
cell differentiation and proliferation and on the immune
system. It enhances insulin secretion and downregulates
the renin-angiotensin system. Vitamin D is produced in
the skin during exposure to sunlight (ultraviolet B, 290
to 315 nm) and absorbed from dietary intake. The production of vitamin D by sunlight exposure is severely
impaired by the use of sunscreens. Natural dietary
sources of vitamin D are limited to oily fish such as
mackerel and salmon. Dietary vitamin D is also available from fortified foods such as milk and orange juice,
and from multivitamins. A recent increase in the incidence of rickets in children has led the American Academy of Pediatrics to release new recommendations for
the intake of vitamin D in infants.

Sex Steroids/Estrogen
Adolescence is a time of rapid increase in skeletal mass
resulting from the concurrent release of sex steroids and
an increase in the growth hormone secretory rate. Up to
25% of peak bone mass is accumulated during the peak
of pubertal growth. At peak height velocity, adolescents
have reached 90% of adult height but only 57% of peak
bone mass. Over the next several years, the bone density
increases.

189

Bone Metabolism and Metabolic Bone Disease

Adult women, at the end of the reproductive life cycle, experience rapid loss of bone mass associated with
menopause and declining estrogen production. Bone
loss also occurs in men, although at a slower rate, as testosterone production declines. Evidence suggests that it
is the circulating estradiol concentration in men that is
most correlated with bone loss. Both osteoblasts and osteoclasts possess estrogen receptors. Estrogen promotes
the differentiation of osteoblasts and suppresses osteoclast activity in part by increasing osteoclast apoptosis.
Estrogen may also act through the stromal cells in the
bone marrow to decrease cytokine production, which
decreases osteoclast recruitment.
In the elderly population, growth hormone secretion
also declines and may contribute to decreased bone
mineralization. In response to decreased growth hormone levels, IGF-I concentration is reduced. The magnitude of the decrease of human growth hormone and
IGF-I on bone loss is unclear. Glucocorticoids inhibit
bone formation by shortening osteoblast lifespan via a
combination of decreased replication and increased apoptosis and decreased transcription of type I collagen,
osteocalcin, and other matrix proteins. Glucocorticoids
suppress IGF-I and insulin-like growth factor binding
protein expression in osteoblasts. IGF-I and insulin-like
growth factor binding protein act in a paracrine manner
to increase bone formation. Glucocorticoids increase
bone resorption, decrease sex steroid concentrations,
and inhibit calcium absorption in the intestine. All of
these factors result in increased bone resorption and
subsequent osteoporosis.

Metabolic Bone Disease

Bone formation and bone resorption pathways are
tightly coupled. When there is a disruption in either
component of bone formation or bone resorption, the
result is metabolic bone disease. Depending on the nature of the disturbance, either bone formation or bone
resorption will predominate, resulting in increased or
decreased bone formation.

Osteoporosis
Osteoporosis is a global problem that is becoming more
prevalent with the increasing age of the population. The
major clinical outcome of osteoporosis is fracture, but
multiple factors contribute to fracture risk. The most
readily available and best standarized clinical measure
of osteoporosis is the BMD measurement; BMD is
strongly correlated with bone strength and is a predictor
of fracture risk. The World Health Organization has defined osteopenia and osteoporosis based on BMD values, which are defined in adult women by a T score. A
T score of 0 is equal to the mean for age. A T score of
1.0 to 2.5 defines osteopenia, and a T score of less
than 2.5 defines osteoporosis. Osteoporosis and os-

190

Orthopaedic Knowledge Update 8

teopenia can be defined for both the lumbar spine and
for the femoral neck. There are inherent limitations in
BMD measurements that limit their applicability in defining standards of osteopenia and osteoporosis in
adults, and even more so in pediatric patients. Dualenergy x-ray absorptiometry (DEXA) scans interpret
the bone density unicompartmentally with no ability to
evaluate the cortical density separately from the trabecular density, although each may respond differently to
disease state or therapy. Response of one compartment
rather than another could potentially alter outcomes. A
DEXA scan provides limited information on bone integrity and is unlikely to impart the complete status of
bone health.
In postmenopausal women, the mainstay of treatment of osteoporosis has been hormone replacement
therapy (HRT) using an estrogen preparation. Because
bone loss resumes when estrogen therapy is stopped, it
was believed by some physicians that therapy should be
lifelong. Recently that recommendation has been questioned by the results of the Womens Health Initiative.
This long-term study was designed to compare treatment of postmenopausal women with an estrogen and
progestin combination arm, an unopposed estrogen arm,
and a placebo arm; outcome measures were incidence of
colorectal cancer, breast cancer, heart disease, and fractures. After 5 years, the study arm using the estrogen
and progestin combination was terminated early when
an interim analysis indicated evidence of an increased
risk for breast cancer. This finding, along with evidence
of some increased risk for cardiovascular disease, stroke,
and pulmonary embolism, outweighed the benefits of a
reduced risk for fractures and the possible benefits in
preventing colorectal cancer. The study arm using estrogen alone is continuing. The study has significant limitations, but it is likely to discourage some women from using postmenopausal HRT.
Authors of a recent meta-analysis of pooled results
from randomized studies concluded that multiple therapies, including vitamin D, calcitonin, raloxifene, and alendronate decrease the incidence of vertebral fractures.
Based on the quality of the available studies, the magnitude of the treatment effects, the narrow confidence intervals, and the consistent positive results throughout
each study, the authors of this meta-analysis concluded
that the evidence most strongly supports the effect of alendronate and risedronate for reducing vertebral fractures; the effects of other agents remain unclear. The
data on HRT indicate a confidence interval that includes a possible increase in risk for vertebral fractures.
It has been shown that HRT increases BMD; studies of
longer duration may ultimately prove that it also decreases the incidence of vertebral fractures. Large randomized controlled studies, such as the Womens Health
Initiative, are needed to definitively determine the ben-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

efit and risk/benefit ratio. Most studies that have shown
an impact for HRT have been case controlled or cohort
studies.

Pediatric Bone Health

Osteoporosis has been considered to be a disease of
adulthood; however, there is increasing evidence that
the roots of osteoporosis begin much earlier in life. Osteoporosis is often assumed to be the result of bone loss
in the older patient, however, a patient who does not attain optimal bone mass is at increased risk for accelerated bone loss and premature osteoporosis. Peak bone
mass is attained as a result of bone mass acquisition during childhood and adolescence. Therefore, any factors
that inhibit the maximal acquisition of bone mass can
contribute to lower peak bone mass and the subsequent
development of osteoporosis and its inherent health
consequences. Attainment of peak bone mass is a function of intrinsic, unalterable factors such race, gender,
and genetics, and extrinsic or alterable factors such as
diet, participation in weight-bearing exercise, the effects
of illness, and the use of certain medications.
It is more difficult to assess bone health in children
than in adults because there is limited information
about normal or average bone mass acquisition and virtually no data correlating fracture incidence with BMD.
Fracture incidence during childhood may not be the correct end point to assess bone health. The DEXA scan is
more difficult to interpret in children and adolescents
because diagnostic standards developed by the World
Health Organization are based on adult bone mass and
are not applicable to pediatric and adolescent patients,
who are still acquiring bone. For the pediatric and adolescent patient, it is crucial to have the DEXA scan
done on a machine that has a pediatric software database and to have the scan interpreted by a health professional experienced with pediatric scans. Pediatric
BMD measurements are subject to many variables not
encountered in the adult population. During puberty,
there is a substantial increase in skeletal mass and mineralization exerted by the growth stimulating effects of
the sex steroids (estrogen and testosterone) and by
growth hormone. The timing of peak growth velocity
and peak bone mass is different in males and females
with peak growth of the skeleton preceding the attainment of peak bone mass. Pubertal status has a major effect on bone mineralization; it may be more appropriate
to compare patients by pubertal stage (Tanner stage)
rather than by chronologic age. Some centers with large
databases can provide that information.
All pediatric DEXA scans must be interpreted with
a Z score rather than a T score. The T score relates the
patients BMD measurement with an adult reference
standard. The Z score relates the patients BMD to ageand gender-matched controls. The use of adult reference

American Academy of Orthopaedic Surgeons

Chapter 18 Bone Metabolism and Metabolic Bone Disease

data to interpret pediatric scans will result in an underestimate of BMD, leading to misdiagnosis of osteoporosis and osteopenia. Incorrect diagnoses of these conditions in children and teens may lead to excessive
parental concern, increased health care expenditures,
and unnecessary therapy. Unnecessary restrictions on
patient activity could prove detrimental rather than
helpful.
The lack of diagnostic criteria for osteoporosis is a
major limitation to the usefulness of DEXA data in the
pediatric population. Currently, BMD data will allow
comparison of a pediatric patient with an age-matched,
gender-matched, and possibly a pubertal stage-matched
average; however, this comparison does not allow for
the prediction of fracture risk. Studies that correlate
BMD with fracture incidence in the pediatric population are ongoing, but data are not yet available.
Serum and urine markers of bone formation and resorption have been used in adult patients to determine
the etiology of bone loss and to follow up on treatment
effectiveness (Table 2); however, the data are of limited
use in the pediatric and adolescent populations. Reference ranges must be determined for age and pubertal
status and any values must be interpreted with knowledge of the patients pubertal status. The fact that peak
growth and bone mineral acquisition correlates with increased bone turnover and net bone formation, should
be reflected in the markers of bone metabolism and will
alter the reference range for a child of any age. As better standards are developed, bone markers in conjunction with an imaging modality may help to identify children with bone mass acquisition deficiencies and assist
in determining therapeutic plans.

Rickets
Rickets is a disease of growing bone resulting in decreased mineralization of new bone, which can ultimately limit peak bone mass if not recognized and
treated. The deficiencies that cause rickets are present
long before physical signs are evident. Rickets easily can
be diagnosed with a radiograph of the wrist or knee and
laboratory evaluation, which is essential to determine
the etiology of the disease (Tables 3 and 4).
In the past several years, a resurgence of nutritional
rickets has become evident. In response to this resurgence, the American Academy of Pediatrics published
new guidelines in 2003 for the intake of vitamin D in infants and children (Table 5). The increase in the incidence of rickets is probably a consequence of several
factors. (1) There has been an increased emphasis on
breastfeeding infants. Human breast milk typically has a
low concentration of vitamin D (< 136 IU/L), which is
limited even with maternal supplementation. Also, many
women do not maintain an adequate intake of calcium
and vitamin D during lactation. (2) Over the years, it

191

Orthopaedic Knowledge Update 8

Bone Metabolism and Metabolic Bone Disease

Table 2 | Markers of Bone Metabolism Measurable in Serum and Urine

Bone Formation Markers
Alkaline phosphatase
Osteocalcin

Serum
Serum

Bone-specific alkaline phosphatase

Procollagen type I carboxyterminal
propeptide
Procollagen type I amino-terminal
propeptide
Bone Resorption Markers

Serum
Serum

Tartrate-resistant acid phosphatase

Serum

Collagen type I cross-linked C-telopeptide

Serum

Hydroxyproline

Urine

Pyridinoline/deoxypyridinoline

Urine

Serum

Collagen type I cross-linked N-telopeptide Urine

Most extensively used; 80% of activity is derived from bone

Available from specialized reference laboratories; must use age-appropriate reference
ranges for the pediatric population
Available from specialized reference laboratories
Cleavage product from formation of type I collagen; other sources of type I collagen
besides bone hamper test specificity
Cleavage product from formation of type I collagen; other sources of type I collagen
besides bone hamper test specificity
Produced by osteoclasts; assay available in reference laboratories; data on normal
values are scarce
Fragment from cross-linking peptides within collagen; assay available in reference
laboratories, data on normal values are scarce
Amino acid component of collagen; specificity is hindered by wide variation in
day-to-day values and effects from dietary sources
Cross-linking peptides within collagen; good specificity; standard is to assay on first
morning void
Fragment from cross-linking peptides within collagen; good specificity; values strongly
affected by puberty, standard is to assay on second morning void

Table 3 | Laboratory Diagnosis of Rickets

Nutritional
Hypophosphatemic (x-linked)
VDDR I
(1- hydroxylase deficiency)
VDDR II
(1,25 D receptor defect)

Ca++

Phosphorus

Alkaline
Phosphorus

PTH

25-D

1,25-D

normal/low
normal
low

normal/low
low
normal

high
high
high

high
normal
high

low
normal
normal

normal/low
normal
low

low

normal

high

high

normal

high

VDDR = vitamin D-dependent rickets

was assumed that normal casual sunlight exposure in

older children and adolescents ensured adequate vitamin D production (especially in areas of the country
that enjoy many days of bright sunlight throughout the
year). However, it is difficult to quantify the amount of
sunlight exposure of any individual and to determine
the amount of sunlight necessary to ensure adequate vitamin D production. This situation is further complicated by the American Academy of Pediatrics recommendation for sunscreen use on children and for the
limitation of sunlight exposure in very young children to
decrease the risk of skin cancer. The use of sunscreens
markedly decreases skin production of vitamin D; application of a 5% p-aminobenzoic acid solution (sun protection factor 8) blocks over 95% of the conversion of

192

7-dehydrocholesterol to previtamin D in the skin. Even

in regions with high amounts of sunlight, it is likely that
the proper use of sunscreens will markedly reduce the
endogenous production of vitamin D in children. (3) Dietary calcium deficiency may contribute to the increase
in the number of observed cases of rickets. Many children who are breast fed for most of the first year of life
are being weaned from breast milk to nondairy drinks;
many children drink little or no milk and some consume
virtually no dairy products. Older toddlers are at risk for
developing rickets because of inadequate calcium intake
compounded by marginal vitamin D status. Surveys of
children and teens have repeatedly documented calcium
intake of less than 50% of the American Dietetic Association guidelines (Table 6).

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Table 4 | Laboratory Tests to Evaluate Suspected Rickets

Assay/Test

Purpose

Calcium, phosphorus
Alkaline phosphatase

Assess for deficiency

High level indicates increased bone
turnover
Evaluate renal function

Blood urea nitrogen, creatinine,

bicarbonate
25(OH)D, 1,25(OH)2D
PTH
Urine calcium: creatinine ratio

Tubular resorption of phosphorus

Assess for deficiency and vitamin D

metabolism
Evaluate PTH response
Rule out hypercalciuria; young child
has higher levels than does older
child or adult (normal values
< 0.25 in older child or adult)
Assess for renal phosphorus
wasting

The American Academy of Pediatrics guidelines on

vitamin D intake in infants and children (Table 5) are
designed for these patients to achieve a serum concentration of vitamin D greater than 11 ng/mL. This is a
conservative target and was determined from studies
that evaluated the dosage required to prevent rickets.
Traditionally, serum levels or 10 to 12 ng/mL have been
considered adequate for sufficient vitamin D stores;
however, most endocrinologists now recommend serum
levels of at least 15 to 20 ng/mL. Based on serum vitamin D levels and studies on the use of supplemental vitamin D, an intake of 300 to 400 IU/day is probably necessary to maximize bone health in the absence of
sunlight. In the presence of sunlight, the requirement for
dietary vitamin D may be lower. Nutritional rickets is a
preventable disease. The cost of prevention is low and
the safety profile of vitamin D is good.

Renal Osteodystrophy
Impaired renal function results in disordered mineral
metabolism that is manifested in multiple tissues. Renal
osteodystrophy includes multiple disorders of bone metabolism associated with chronic renal disease. Chronic
renal disease leads to altered PTH synthesis and secretion, parathyroid gland hyperplasia, abnormal calcium
and vitamin D metabolism, chronic acidosis, and inhibition of growth hormone action. All of these effects alter
bone metabolism and manifest in renal bone diseases.
The spectrum of renal bone disease includes both increased and decreased turnover states. In general, high
turnover states are associated with increased PTH levels
and low turnover states with normal or low PTH levels.
The definitive method for determining the type of renal
disease is bone histology.
High turnover renal bone disease is a consequence
of sustained high PTH levels. Secondary hyperparathy-

American Academy of Orthopaedic Surgeons

Chapter 18 Bone Metabolism and Metabolic Bone Disease

Table 5 | The 2003 American Academy of Pediatrics Guidelines for Vitamin D Intake for Infants and Children
The ingestion of 200 IU of vitamin D per day is recommended for healthy
infants and children
A supplement of 200 IU of vitamin D per day is recommended for:
All breastfed infants unless they are weaned to at least 500 mL per
day of vitamin-D fortified formula or milk
All nonbreastfed infants who are ingesting less than 500 mL per day of
vitamin-D fortified formula or milk
Children and adolescents who do not get regular exposure to sunlight
and who do not ingest at least 500 mL per day of vitaminD- fortified
milk

Table 6 | Recommendations for Daily Calcium Intake

Adequate
Age Range (years) Intake* (mg/d)
0.0
0.5
1.0
4.0
9.0
14.0

to 0.5
to 1.0
to 3.0
to 8.0
to 13.0
to 18.0

210
270
500
800
1,300
1,300

Age Range
(years)
0.0
0.5
1.0
6.0
11.0

to 0.5
to 1.0
to 5.0
to 10.0
to 18.0

Recommended

Intake (mg/d)
400
600
800
800 to 1,200
1,200 to 1,500

*Institute of Medicine

National Institutes of Health Guidelines

roidism is a result of several factors including hypocalcemia and hyperphosphatemia, decreased 1,25(OH)2D
production by the kidney, skeletal resistance to PTH action, altered PTH gene transcription, and reduced expression of CaSR and vitamin D receptors on the parathyroid cells. With sustained elevation of serum PTH
levels, osteoblastic and osteoclastic activity in bone increases and results in increased bone turnover.
Low turnover bone disease occurs when the PTH
level is normal or minimally elevated. This disease had
previously been associated with aluminum toxicity. Currently, low turnover bone disease is more likely to be associated with diabetes, corticosteroid use, or advanced
age. The recent use of large doses of oral calcium and
active vitamin D analogs can lead to lowered PTH levels and decreased stimulus for bone turnover; this situation probably accounts for the increase in adynamic disease, which is histologically evidenced by increased
osteoid and undermineralized bone collagen. The longterm consequences of low turnover renal osteodystrophy are unknown.

Hyperparathyroidism
Hyperparathyroidism is characterized by hypercalcemia
resulting from excessive PTH secretion. The normal

193

Orthopaedic Knowledge Update 8

Bone Metabolism and Metabolic Bone Disease

feedback control of PTH secretion is disrupted, leading
to increased secretion that is not normally responsive to
an increased calcium concentration. Most patients
(80%) with primary hyperparathyroidism have solitary
adenoma in one parathyroid gland. In about 15% of patients, hyperparathyroidism is a consequence of fourgland hyperplasia, which can occur in conjunction with
multiple endocrine neoplasia type 1 or type 2. Very
rarely (in < 1% of patients), it is a result of parathyroid
carcinoma. The exact molecular basis for primary hyperparathyroidism is unknown. Familial causes of hyperparathyroidism include multiple endocrine neoplasias
(1 and 2A). Germline mutations in a nuclear protein tumor suppressor gene encoded on chromosome 11 have
been described in patients with both familial and sporadic multiple endocrine neoplasm 1. Multiple endocrine neoplasm 2 is caused by mutations in the RET
proto-oncogene.
Primary hyperparathyroidism is treated by removal
of the abnormal parathyroid tissue. Medical management is more difficult but bisphosphonates can be used
to acutely lower serum calcium levels. Over the long
course, bisphosphonates may help protect bone density
in patients with hyperparathyroidism who cannot have
or choose not to have surgery.

Osteopetrosis
Osteopetrosis (marble bone disease) is a disorder characterized by increased bone mass. It can result from a
defect in a variety of genes or gene products, but the final common pathway is a failure of osteoclast mediated
bone resorption.
Infantile osteopetrosis is evident in infancy and is
rapidly fatal. Cranial nerve deficits, delayed dental eruption, hypersplenism, and hemolysis may occur. Treatment may include bone marrow transplant, high-dose
glucocorticoids, limitation of dietary calcium, and highdose calcitriol.
Intermediate osteopetrosis causes short stature, recurrent fractures, and possible cranial nerve deficits.
One variant of intermediate osteopetrosis is caused by
carbonic anhydrase II deficiency; it is characterized by
osteopetrosis, renal tubular acidosis, and cerebral calcifications and is inherited in an autosomal recessive pattern.
Some individuals with adult osteopetrosis are asymptomatic, whereas others have facial cranial nerve involvement, carpal tunnel syndrome, slipped capital femoral epiphysis, and osteoarthritis.
Iatrogenic suppression of osteoclast activity is another potential etiology of osteopetrosis. Because of
promising results seen with the use of bisphosphonates
in the treatment of children with osteogenesis imperfecta, the use of bisphosphonates in treating children
with a variety of conditions has increased. However, few

194

data are available from randomized clinical trials for the

use of bisphosphonate treatment in children and teenagers. Although bisphosphonates may prove beneficial,
they should be used with extreme caution. Growing
bone requires coordination of bone resorption and bone
formation. Bone formation predominates in children
and teens, and thus the pharmacologic suppression of
resorption may have different consequences than seen
in adult patients. Also, the half-life of the bisphosphonates is nearly a decade. Treating a young woman before
or during childbearing years may have unforeseen consequences for the fetus and mother. It is important to
select patients for treatment in whom the risk to benefit
ratio is apparent; this information should be clearly
communicated with the parents and the child or teen.

Pagets Disease
Pagets disease is a localized disorder of bone remodeling; it has a familial pattern of aggregation, but no single
gene abnormality has been implicated in the etiology.
Infectious agents have also been implicated as a possible cause. The initial defect in Pagets disease is increased bone resorption associated with abnormalities
of osteoclasts found in the lesion. Osteoclasts in a Pagets lesion are more numerous and contain more nuclei
than normal. Early lesions usually contain abundant osteoblasts and increased new bone formation. The increased bone turnover is manifested by increased urinary excretion of biomarkers indicative of increased
bone resorption. Pagets disease is slightly more predominant in men than in women and occurs most often
in patients older than 40 years of age. The most common
presenting symptom is bone pain. The diagnosis is usually evident from the radiographic appearance of a localized lesion consisting of cortical thickening, cortical
expansion, mixed areas of lucency and sclerosis, and laboratory evaluation showing increased markers of bone
resorption in the urine and increased alkaline phosphatase in the serum.
The treatment of Pagets disease is directed at decreasing the activity of the pagetic osteoclasts. Approved pharmacologic therapies include the bisphosphonates and calcitonin. Medical treatment is indicated
for patients with painful lesions in the long bones, skull,
and vertebrae, which may cause neurologic damage or
secondary arthritis. Although intended to alleviate pain
and to prevent complications, few data exist to show
that pharmacologic intervention actually prevents complications from this metabolic bone disease.

Future Directions
Bone Mineral Density
BMD has been a useful tool in the understanding of
bone health and in identifying patients with osteoporosis. It is also increasingly clear that there are limitations

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

to the use of DEXA as a device to measure bone
health. Better studies correlating DEXA data with fracture risk in adults and especially in children and teens
are necessary. Alternate imaging modalities may become more useful in understanding the determinants of
fracture risk and in improving the understanding of how
different therapies alter bone characteristics and
strength. One of the major weaknesses of DEXA is its
inability to separate and assess cortical bone and trabecular bone compartments. Imaging modalities such as peripheral quantitative CT can be used to evaluate both
compartments individually and may prove useful in the
next few years to determine the effects of medications,
growth, and aging on bone and to help evaluate various
treatment modalities.

New Medications
The current predominant therapy for altering bone metabolism is directed at bone resorption (for example,
bisphosphonate therapy). Developing medications that
increase bone formation also may prove to be a means
to improve systemic therapy for those with osteoporosis
or genetic syndromes. Recently, recombinant human
parathyroid hormone (1-34) [rhPTH(1-34)] has been approved by the Food and Drug Administration for the
treatment of osteoporosis. When present continuously,
PTH causes demineralization of bone; when present intermittently or episodically, it increases bone mineralization. In animal studies, intermittent rhPTH exposure increases osteoblast numbers without a concomitant
increase in osteoclasts and has been shown to increase
bone mass, vertebral strength, and trabecular number in
monkey studies. In a fracture prevention study of
women with postmenopausal osteoporosis, rhPTH increased the BMD of patients in the spine and hip and
decreased vertebral fractures; there are no data available regarding use in children. During the next several
years, more information, including data on optimal dosage and timing will become available and may offer new
alternatives for the treatment of osteoporosis.
The effect of bisphosphonates on postoperative
healing needs to be studied. The healing of bone after
fractures or surgical intervention (such as cervical spine
fusion) is a tightly coupled balance of bone resorption
and formation. Treatment with bisphosphonates has the
potential to alter that balance. Fracture healing does not
seem to be impaired in children with osteogenesis imperfecta who are being treated with bisphosphonates;
however, this finding may not be applicable for bisphosphonate treatment in elderly patients. It is imperative
that fracture and postoperative healing in patients
treated with bisphosphonates be monitored over the
next several years because these findings may significantly impact the decision on how to treat some patients.

American Academy of Orthopaedic Surgeons

Chapter 18 Bone Metabolism and Metabolic Bone Disease

Pediatric Bone Health

Pediatric bone metabolism is very different than adult
bone metabolism, most notably because bone formation
is predominant. Impaired bone formation during childhood has long-term effects that may set the stage for
premature osteoporosis in adulthood. Better definitions
of normal pediatric bone health and the effects of
chronic illness on bone health are needed to ensure that
all children achieve peak bone mass and minimize the
risks for the bone diseases of adulthood.

Annotated Bibliography
General Reference
Proceedings of the Surgeon Generals Workshop on Osteoporosis and Bone Health, December 12-13, 2002.
Washington, DC, Department of Health and Human
Services.
A summary of a meeting held to provide an opportunity
for key stakeholders to provide input as to the most important
priorities for the Surgeon Generals report on osteoporosis
and bone health is presented.

Basic Science Aspects of Bone Metabolism

Favus MJ (ed): Primer on the Metabolic Bone Diseases
and Disorders of Mineral Metabolism, ed 5. Washington,
DC, American Society for Bone and Mineral Research,
2003.
A comprehensive review of bone metabolism and bone
disease is presented.

Ma YL, Bryant HU, Zeng Q, et al: New bone formation

with teriparatide (human parathyroid hormone-(1-34) is
not retarded by long-term pretreatment with alendronate, estrogen, or raloxifene in ovariectomized rats.
Endocrinology 2003;144:2008-2015.
This article discusses the ability of teriparatide to induce
bone formation.

Vahle JL, Sato M, Long GG, et al: Skeletal changes in

rats given daily subcutaneous injections of recombinant
human parathyroid hormone (1-34) for 2 years and relevance to human safety. Toxicol Pathol 2002;30:312-321.
This article discusses effects of daily subcutaneous injections of rhPTH(1-34). Results indicated increased bone mass
and BMD.

Hormonal Regulation of Cell Differentiation

Leonard MB, Zemel BS: Current concepts in pediatric
bone disease. Pediatr Clin North Am 2002;49:143-173.
A description of expected gains in bone size and mass during childhood and adolescence is presented. The article discusses modifiable determinants of bone mineralization during
growth and the impact of osteopenia on childhood and lifetime fracture risk.

195

Orthopaedic Knowledge Update 8

Bone Metabolism and Metabolic Bone Disease

Metabolic Bone Disease

Anderson JB: Calcium requirements during adolescence
to maximize bone health. J Am Coll Nutr 2001;20(suppl
2):186S-191S.
A review of information on the calcium needs of children
and young adults from approximately 9 to 20 years of age is
presented. It was concluded that children and teens are not
getting the recommended daily amounts of calcium and vitamin D in their diets and that it is increasingly important to
provide counseling to families about the appropriate calcium
and vitamin D intake required to maximize bone mineralization.

Bilezikian BP (ed): Endocrinology and Metabolism

Clinics of North America. Philadelphia, PA, WB Saunders, 2003.
This book summarizes important developments in the detection and treatment of osteoporosis. The areas with the most
significant advances are identified.

Cranney A, Guyatt G, Griffith L, et al: Meta-analyses of

therapies for postmenopausal osteoporosis: IX. Summary of meta-analyses of therapies for postmenopausal
osteoporosis. Endocr Rev 2002;23:570-578.
Cranney A, Tugwell P, Adachi J, et al: Meta-analyses of
therapies for postmenopausal osteoporosis: III. Metaanalysis of risedronate for the treatment of postmenopausal osteoporosis. Endocr Rev 2002;23:517-523.
Cranney A, Tugwell P, Wells G, Guyatt G: Osteoporosis
Methodology Group and The Osteoporosis Research
Advisory Group: Meta-analyses of therapies for postmenopausal osteoporosis: I. Systematic reviews of randomized trials in osteoporosis: Introduction and methodology. Endocr Rev 2002;23:496-507.
Cranney A, Tugwell P, Zytaruk N, et al: Meta-analyses of
therapies for postmenopausal osteoporosis: IV: Metaanalysis of raloxifene for the prevention and treatment
of postmenopausal osteoporosis. Endocr Rev 2002;23:
524-528.
Cranney A, Tugwell P, Zytaruk N, et al: Meta-analyses of
therapies for postmenopausal osteoporosis: VI. Metaanalysis of calcitonin for the treatment of postmenopausal osteoporosis. Endocr Rev 2002;23:540-551.
Cranney A, Wells G, Willan A, et al: Meta-analyses of
therapies for postmenopausal osteoporosis: II. Metaanalysis of alendronate for the treatment of postmenopausal women. Endocr Rev 2002;23:508-516.
Papadimitropoulos E, Wells G, Shea B, et al: Metaanalyses of therapies for postmenopausal osteoporosis:
VIII. Meta-analysis of the efficacy of vitamin D treat-

196

ment in preventing osteoporosis in postmenopausal

women. Endocr Rev 2002;23:560-569.
Shea B, Wells G, Cranney A, et al: Meta-analyses of
therapies for postmenopausal osteoporosis: VII. Metaanalysis of calcium supplementation for the prevention
of postmenopausal oseoporosis. Endocr Rev 2002;23:
552-559.
Wells G, Tugwell P, Shea B, et al: Meta-analyses of therapies for postmenopausal osteoporosis: V. Meta-analysis
of the efficacy of hormone replacement therapy in treating and preventing osteoporosis in postmenopausal
women. Endocr Rev 2002;23:529-539.
The previous nine bibliographic entries comprise a series
of articles describing the principles of a meta-analysis (Section
I) and the results of a meta-analysis of many large published
studies and some unpublished observations that were designed
to evaluate therapies for osteoporosis. This series compared
placebo controlled studies and showed that the results of
large, randomized studies may be different from case controlled studies.

Rossouw JE, Anderson GL, Prentice RL, et al: Risks

and benefits of estrogen plus progestin in healthy postmenopausal women: Principal results from the Womens
Health Initiative randomized controlled trial. JAMA
2002;288:321-333.
An interim analysis of the Womens Health Initiative led
to early termination of one arm of the study and has had a significant impact on the treatment of osteoporosis. The results
are summarized in this article.

Whyte MP, Wenkert D, Clements K, McAlister WH,

Mumm S: Bisphosphonate-induced osteopetrosis.
N Engl J Med 2003;349:457-463.
A case report of bisphosphonate-induced osteopetrosis in
a 12-year-old boy is presented. This article shows the consequences of overtreatment with bisphosphonates in the pediatric population.

Future Directions
Deal C, Gideon J: Recombinant human PTH 1-34 (Forteo): An anabolic drug for osteoporosis. Cleve Clin J
Med 2003;70:584.
This article presents a discussion of the use of forteo, a genetically engineered fragment of parathyroid hormone, in the
treatment of osteoporosis.

Steelman J, Zeitler P: Osteoporosis in pediatrics. Pediatr

Rev 2001;22:56-65.
A description of determinants of bone mass, pediatric populations at risk for osteoporosis, test procedures for assessing
bone quality and quantity, and available treatments are
presented.

American Academy of Orthopaedic Surgeons

Chapter

19

Musculoskeletal Oncology
Richard D. Lackman, MD

Introduction
The evaluation of a patient with a bone tumor involves
the collection of data from several sources that can impact the differential diagnosis; these sources include patient history, physical examination, and imaging studies.
Ultimately, if it is determined that histologic confirmation is required, careful evaluation of lesional tissue will
confirm a specific diagnosis.
The history associated with the presence of a musculoskeletal tumor defines the clinical context of the lesion. Patient age, sex, duration of symptoms, presence
and quality of pain, and history of trauma, weight loss,
smoking, and prior malignancy all are important factors.
The knowledge that the early symptoms associated with
skeletal neoplasms mimic all types of ordinary musculoskeletal disorders is critical to the early diagnosis of a
skeletal tumor. Any pain that extends beyond the expected duration associated with a provisional, nonmalignant diagnosis may indicate an underlying tumor. Night
pain is a red flag that may lead to the supposition of an
occult lesion; however, it should be recognized that
many nonneoplastic conditions may also result in pain
at night.
A history of trauma in a patient with an occult tumor may delay an accurate diagnosis. Patients will often
experience some mild trauma to the affected area and
then notice pain, which would probably not have occurred in the absence of an underlying lesion. The patient may directly attribute the local symptoms and findings to the traumatic event, which then influences the
treating physician to initially concentrate on the local
symptoms. Eventually it will become obvious that the
true nature of the lesion is more involved than a minor
trauma. An example of this scenario is that of a waiter
who kicked a kitchen door to open it while carrying a
heavy tray. The door was stuck and did not move, resulting in an apparent calf injury. When the pain did not resolve, a compartment syndrome was suspected; it was
not until several months later that tissue was obtained
revealing an underlying lymphoma. Similar is the example of an elderly woman taking full-dose warfarin after

American Academy of Orthopaedic Surgeons

implantation of a mechanical heart valve who bumps

her thigh on a kitchen table and discovers months later
that the resulting large anterior thigh mass is a softtissue sarcoma and not a simple hematoma. It is only
throughintellectualdisciplineanddiligencethatearlydiagnoses of occult tumors can be accomplished in patients
with a history of trauma involving the affected area.
To assemble a complete differential diagnosis it is
helpful to memorize or to have readily available a reasonable list of common lesions to review when a set of
radiographs is examined. Without such mental organization, it is difficult or impossible to assemble a comprehensive differential diagnosis of a particular lesion. Table 1 lists three categories of lesions that should be
individually considered each time a radiograph is reviewed to ensure the inclusion of the most relevant lesions in a specific differential diagnosis.

Table 1 | Tumors and Common Lesions to Consider When

Reviewing Radiographs for a Differential Diagnosis
Bone-Forming
Tumors

Cartilage-Forming
Tumors

Osteoid osteoma

Osteochondroma

Osteoblastoma
Osteochondroma
Osteosarcoma
Blastic metastasis
Pagets disease

Other Tumors and

Lesions

Lesions caused by
infection
Chondroblastoma
Metastatic lesions
Chondromyxoid fibroma Round cell tumors
Enchondroma
Unicameral (simple)
bone cyst
Chondrosarcoma
Aneurysmal bone cyst
Nonossifying fibroma
Fibrous dysplasia
Giant cell tumor
Langerhans cell
histiocytosis
Stress fracture
Lesions caused by
metabolic condition

197

Orthopaedic Knowledge Update 8

Musculoskeletal Oncology

Figure 1 A, Radiograph of osteoid osteoma

showing lytic nidus and surrounding sclerosis.
B, CT scan of osteoid osteoma showing lytic
nidus.

Bone-Forming Tumors
Osteoid Osteoma
Osteoid osteoma most commonly occurs during the first
two decades of life and appears as a small lytic nidus, often with a target appearance surrounded by significant
sclerosis (Figure 1). The nidus may be very tiny and difficult to find on a radiograph. MRI scans will show extensive edema, which may be mistaken for a marrow replacing neoplasm. A CT scan with fine cuts (for
example, 1 mm) is the study of choice for locating the
lesion. Bone scintigraphy shows focal intense uptake.
Osteoid osteomas are associated with a classic pattern
of constant pain secondary to prostaglandin secretion.
For this reason, pain resulting from these lesions is relieved significantly for short periods by drugs such as aspirin or ibuprofen that inhibit prostaglandin synthesis.

Osteoblastoma
Osteoblastoma is a rare neoplasm most often occurring
in the posterior elements of the spine or in the metadiaphyseal region of long bones. This tumor has a variable
appearance, may be blastic or lytic, and is rarely diagnosed correctly before histologic material is reviewed.
The classic appearance is that of a calcified lesion in the
posterior elements of the spine (Figure 2).

Osteochondroma
Osteochondromas are formed by radial growth of bone
during childhood and are characterized by a lesion that
grows away from the bone at an angle from the adjacent
growth plate. The hallmark of an osteochondroma is
that, because it grows away from the underlying bone,
the cortex of the lesion is confluent with the cortex of
the bone of origin and it pulls medullary bone up into
itself (Figure 3). A tumor that is located on an intact
cortex is never an osteochondroma. Osteochondromas
can be sessile or pedunculated and many that grow out

198

of the flat bones or the proximal femur can be very

large and take on a cauliflower appearance. Secondary
chondrosarcomatous degeneration occurs in less than
1% of patients and should be suspected in any osteochondroma that grows after puberty or has a cartilage
cap of greater than 3 cm during adulthood.

Osteosarcoma
Osteosarcoma usually occurs during the first through
third decades of life with a second peak in occurrence
after the sixth decade. These tumors present as permeative metaphyseal lesions with soft-tissue extension and
new bone formation (Figure 4). Periosteal reaction is
common and frequently takes on a sunburst or hair on
end appearance. Osteosarcoma needs to be considered
in the differential diagnosis of every aggressive lesion
seen in bone in patients of all ages. It may appear as a
purely lytic lesion with no radiographically apparent
bone formation.

Blastic Metastasis and Pagets Disease

Blastic metastasis is most frequently seen with prostate
and breast carcinoma and typically presents as a permeative lesion with infrequent soft-tissue extension. Pagets disease may mimic other conditions and can exhibit
a variety of radiographic appearances. Early stage Pagets disease is characterized by lytic bone, whereas in
late stage Pagets disease, bone is blastic with coarse trabeculae and thickened cortices (Figure 5).

Cartilage-Forming Tumors
Chondroblastoma
Chondroblastoma usually appears as a painful lytic lesion in the epiphysis of a child; significant edema is seen
on an MRI scan. In older adolescents, it can occasionally
grow across an old epiphyseal line to involve the adjacent metaphysis. This classic picture of a painful epiphy-

American Academy of Orthopaedic Surgeons

Chapter 19 Musculoskeletal Oncology

Orthopaedic Knowledge Update 8

Figure 2 CT scan of the spine showing osteoblastoma

(arrow) in the posterior elements.

Figure 3 Radiograph showing a typical osteochondroma.

seal lytic lesion with abundant edema may cause this lesion to be confused with infection or osteochondritis
dissecans.

Chondromyxoid Fibroma
Chondromyxoid fibroma is a rare tumor usually presenting as a lytic metaphyseal lesion with cortical thinning but no periosteal reaction (Figure 6). This tumor
usually occurs during the first through third decades of
life and is most often located in the proximal tibia. It
frequently has the appearance of a very large nonossifying fibroma.

Enchondroma
An enchondroma is a nest of cartilage tissue typically
found in a metaphyseal region and is usually discovered
as an incidental finding in adult patients. These lesions
are commonly found when a radiograph of the adjacent
joint is obtained for reasons unrelated to the enchondroma itself. Enchondromas tend to be noncalcified or
minimally calcified in young adults and usually show an
increase in calcification but not an increase in size with
long-term follow-up. The calcification has a typical stippled or popcorn pattern. These lesions usually do not
cause pain but typically appear hot on a technetium
bone scan. Unlike chondrosarcomas, enchondromas do
not damage the host bone because they do not cause
cortical thinning or expansion. They may cause some

American Academy of Orthopaedic Surgeons

Figure 4 Radiograph showing a typical osteoblastic

osteosarcoma.

slight scalloping; however, the scallops seen on radiographs are usually short and less than 1 cm each. When
enchondromas occur in thin or small bones such as the
fibular head or scapula, some expansion may be seen in
the radiographic appearance. Because enchondromas
do not know what bone they are in, some cortical expansion in small or thin bones may be present when the
lesion grows to a typical size.

Chondrosarcoma
Unlike enchondromas, chondrosarcomas are active lesions that grow and alter the host bone over time (Figure 7). Working from the inside of the bone to the outside, the changes associated with chondrosarcomas
include intralesional lysis, endosteal scalloping, cortical
thinning, and expansion. Most chondrosarcomas show
chondroid calcification but some may appear purely lytic; they also cause pain.

Other Tumors
Lesions Caused by Infection
Infection can also mimic the appearance and symptoms
of other lesions and can exhibit a variety of radiographic appearances from geographic to permeative and
from lytic to blastic. Periosteal reaction is commonly
seen on radiographs; localized heat and erythema are
often found on physical examination.

199

Orthopaedic Knowledge Update 8

Musculoskeletal Oncology

Figure 5 Radiograph showing typical findings of late

Pagets disease including cortical thickening and
coarse trabeculae.

Figure 6 Radiograph showing chondromyxoid fibroma

of the proximal tibia.

Metastatic Lesions
Metastatic lesions are the most common, aggressive, destructive lesions in adults and can exhibit a variety of radiographic appearances from lytic to blastic. Metastasis
frequently results in multiple lesions and may be the
first presentation of the underlying neoplasm. Although
some metastatic bone lesions, especially those from kidney and lung tumors, may grow beyond the confines of
the bone involved, the presence of an associated softtissue mass should increase the suspicion of a sarcoma.

Round Cell Tumors

Pediatric round cell tumors include Ewings sarcoma
and metastatic neuroblastoma. Ewings sarcoma classically presents as a diaphyseal permeative lesion with
onion skin periosteal reaction and a large associated
soft-tissue mass. Alternatively, Ewings sarcoma may be
more metaphyseal and destructive, and it should be considered in the differential diagnosis of such a lesion
(Figure 8). Twenty percent of patients with Ewings sarcoma will have associated systemic symptoms such as
fever, chills, and a high white blood cell count, which
may cause the lesion to mimic osteomyelitis. Neuroblastoma (Figure 9) appears as a permeative lesion with
medullary bone replacement and varying degrees of osteolysis. Adult round cell lesions include myeloma and
lymphoma. Myeloma occurs as solitary or multiple medullary lytic lesions with sharp margins but little reaction
(Figure 10). Whereas most myeloma patients present
with multiple lesions, a small percentage will have a sol-

200

Figure 7 Radiograph of chondrosarcoma of the proximal femur.

itary bone lesion. If no other bone lesions are found and

a random bone marrow aspirate from another bone
shows fewer than 5% plasma cells, then a solitary plasmacytoma exists and the prognosis is much better than
for a patient with multiple myeloma. Only 25% of patients with solitary plasmacytoma will have a positive
serum or urine myeloma protein (m-protein) whereas
most patients with multiple myeloma will have measurable m-proteins detected by serum and urine protein
electrophoresis. The levels of these m-proteins are useful for tumor staging and also for assessing response to
treatment. MRI scans, especially of the spine, are useful
screening studies; technetium bone scans are usually
normal in patients with multiple myeloma. Lymphoma
is typically a very permeative but minimally destructive
lesion, which usually progresses by filling the medullary
canal and then growing into the surrounding soft tissues
while causing little destruction of the bone itself (Figure
11). Radiographs may be remarkably normal, whereas
MRI scans show marrow replacement and often an associated soft-tissue mass.

Unicameral (Simple) Bone Cyst

Unicameral bone cysts are usually found in the proximal humerus; are always central, full width lesions; may
cause cortical thinning and minimal expansion; and usually have no associated periosteal reaction. The widest
portion of these lesions is usually no wider than the widest portion of the adjacent metaphysis (Figure 12).

American Academy of Orthopaedic Surgeons

Chapter 19 Musculoskeletal Oncology

Orthopaedic Knowledge Update 8

Figure 8 Radiograph of Ewings sarcoma of the proximal fibula.

Figure 9 Radiograph of metastatic neuroblastoma in

the proximal humerus of a child.

Aneurysmal Bone Cyst

Aneurysmal bone cysts (ABCs) are usually very expansile and occur as eccentric metaphyseal lesions in patients during the first 3 decades of life (Figure 13).
These lesions frequently have a delicate rim of expanded cortical bone, which may be seen best on a CT
scan. Fluid-fluid levels are frequently present but are
not diagnostic of primary ABC because areas of secondary ABC formation may exist in other lesions such as
osteosarcomas and giant cell tumors.

Nonossifying Fibroma
Nonossifying fibromas that occur in children are usually
eccentric metaphyseal lesions that grow to varying sizes
(Figure 14). As skeletal growth continues and external
remodeling occurs, lesions that were previously intramedullary in the metaphysis become intracortical in
the metadiaphysis. This cortical disruption creates a mechanical insufficiency and causes the bone to replace the
lesion with new bone formation as the lesion heals.

Fibrous Dysplasia
The classic fibrous dysplastic lesion presents as a long
lesion in a long bone with ground-glass textured medullary calcification and cortical thinning but no periosteal
reaction (Figure 15). Fibrous dysplasia can have a variable appearance and should be included in the differential diagnosis of every benign-appearing lesion in bone.

Giant Cell Tumor

A giant cell tumor presents as a juxta-articular lytic lesion and frequently has a moth-eaten or irregular margin, cortical thinning, and expansion; and may have soft-

American Academy of Orthopaedic Surgeons

Figure 10 Radiograph of myeloma of the proximal

humerus.

tissue extension, but usually without a periosteal

reaction (Figure 16). It very rarely occurs before the
growth plates are closed or after the age of 50 years.

Langerhans Cell Histiocytosis

Langerhans cell histiocytosis is an inflammatory condition that usually presents as intramedullary lysis in patients during the first three decades of life. This condition includes three separate clinical entities: eosinophilic
granuloma, Hand-Schller-Christian disease, and
Letterer-Siwe disease. These diseases have different clinical courses and prognoses but share identical histologic
findings. Eosinophilic granuloma is the most common
and mildest form of this disease and usually occurs as a
solitary lesion in bone. The larger the lesion grows and
the closer to a cortex it becomes, the more likely it is to
show some associated sclerosis. These lesions also may
occur on the surface of a bone where they occasionally
elicit an aggressive-appearing periosteal reaction, which
can be mistaken for a malignant tumor on radiographic
evaluation. Histiocytosis must be considered in the differential diagnosis of any intramedullary lesion in a patient younger than 30 years of age.

Factors to Consider in the Differential

Diagnosis
Metabolic Conditions and Trauma
Metabolic conditions include a wide variety of underlying diseases such as osteoporosis, osteomalacia, and renal osteodystrophy that affect bone formation and healing. Such conditions should be considered when forming
a differential diagnosis.

201

Orthopaedic Knowledge Update 8

Musculoskeletal Oncology

Figure 11 A, Radiograph of lymphoma of the

proximal tibia. B, MRI scan of lymphoma of
the proximal tibia.

Radiographic Findings
Certain radiographic findings are important to consider
when making a specific differential diagnosis. By using
the information described in Table 2, a very complete
differential diagnosis of most lesions should be possible.
For example, the radiograph in Figure 17 shows a juxtaarticular lytic lesion in the proximal tibia with a motheaten margin, cortical thinning, and no periosteal reaction in a 19-year-old man with a 6-week history of
progressive knee pain that is worse at night and with
weight bearing. A reasonable differential diagnosis using the lists in Table 1, and taking into consideration the
patients age, includes osteosarcoma, infection, Ewings
sarcoma, giant cell tumor, and ABC. Biopsy proved the
lesion to be a giant cell tumor. If the same radiograph
had been taken of a 50-year-old man, the differential diagnosis would include osteosarcoma, chondrosarcoma,
metastasis, myeloma, lymphoma, and giant cell tumor. In
either case, having a list of lesions to mentally review
greatly increases the completeness of a radiographic differential diagnosis and is helpful in ensuring that relevant lesions are considered in the diagnosis.

Imaging Studies
Computed Tomography
Figure 12 Radiograph of a unicameral bone cyst of the proximal humerus.

Obvious blunt, acute trauma does not usually mimic

tumors, whereas patients with chronic stress factures
may present with bony reaction that can mimic and be
confused with lesions such as lymphoma, osteoid osteoma, metastasis, or infection.

202

The major role of a CT scan is to show bony detail including bone formation as well as bone destruction. CT
scans are the best study for determining the amount of
bone destruction and the presence of soft-tissue calcification. The ability of CT scans to detect the extent of a
permeative lesion in bone, soft-tissue extension from a
bone lesion, or the extent of a lesion in soft tissue is less
than optimal.

American Academy of Orthopaedic Surgeons

Chapter 19 Musculoskeletal Oncology

Orthopaedic Knowledge Update 8

Figure 14 Radiograph of a nonossifying fibroma of the

Figure 13 A and B, Radiographs of an aneurysmal bone cyst of the proximal ulna.

distal tibia that is spontaneously resolving.

Magnetic Resonance Imaging

MRI scans are excellent for showing the extent (starting
and stopping points) of a lesion in bone. They are also
excellent for showing the presence and extent of edema
within bone and the presence or absence of an associated soft-tissue extension. MRI is the study of choice for
any soft-tissue lesion. The addition of contrast to an
MRI scan can help distinguish areas of cyst formation
(which do not enhance with contrast but may show rim
enhancement) from areas of solid tumor (which do enhance). One exception to this generalization are chondroid lesions, such as low grade chondrosarcomas, which
may also show rim enhancement with little internal enhancement and, in this regard, may mimic a cyst. Care
must be taken to differentiate edema in bone and tumor
in bone. Lymphoma frequently presents as high signal in
marrow and must be included in the differential diagnosis of traumatic marrow lesions such as stress fractures
or bone bruises.

Technetium Bone Scan

Technetium bone scans are most useful as a skeletal survey tool to determine the total number of lesions
present or to locate a single lesion that is suspected but
not seen on the initial radiograph. All active boneforming lesions appear hot, whereas some lytic lesions,
which engender no bone reaction, may be normal or
cold (for example, myeloma). Sclerotic lesions, which

American Academy of Orthopaedic Surgeons

are normal on bone scan, are usually old and inactive.

Most metastatic carcinomas and sarcomas in bone are
hot on bone scans, although this finding is not a constant.

Biopsy
If after careful clinical evaluation and imaging studies a
diagnosis has not been confirmed, a biopsy is the next step.
The ideal biopsy provides all of the tissue needed to establish a histologic diagnosis without affecting subsequent
treatment options. Biopsy has caused much concern
among clinicians and several studies have reported the
problems caused by poorly planned and executed biopsies. There is disagreement on whether a community orthopaedic surgeon or a specialized orthopaedic oncologist
should perform a biopsy. If the diagnosing physician
deems that referral to a tumor specialist for treatment
would be appropriate, the physician should defer performing the biopsy. The physician who will ultimately treat the
lesion should determine if and how a biopsy is performed.
For example, in a 17-year old boy with a large painful mass
about the distal femur and radiographic findings indicative of osteosarcoma, most orthopaedic surgeons would
not perform the resection and reconstruction of such a lesion. Therefore, the patient should be referred to a subspecialist who will perform the biopsy in a manner that
will not compromise the definitive treatment. It is understood that the treating surgeon should be familiar with the

203

Orthopaedic Knowledge Update 8

Musculoskeletal Oncology

Figure 15 Radiograph of fibrous dysplasia of the distal humerus showing the ground glass appearance.

Figure 16 Radiograph of a giant cell tumor of the

distal femur.

guidelines concerning the type of biopsy to be performed

and the appropriate biopsy techniques.
Current biopsy options include both open and needle techniques. The advantage of percutaneous needle
biopsies is that they cause little soft-tissue contamination and require little or no anesthesia. They are frequently performed under CT scan or ultrasound guidance, which can direct a biopsy needle into an
underlying lesion. The disadvantage of percutaneous
needle biopsy is that only a small amount of tissue is obtained for the pathologist to review. Such biopsies can
frequently be nondiagnostic. More importantly, primary
bone tumors are notoriously heterogeneous, thereby
creating a great potential for sampling error with closed
techniques. Needle biopsies are optimal for initial sampling of lesions in anatomically inaccessible areas such
as the spine or pelvis. In most locations, however, the
carefully performed open biopsy is still the gold standard. Important rules that should be considered when
performing an open biopsy are outlined in Table 3.

Histologic Evaluation
The histologic diagnosis of musculoskeletal tumors is
difficult and complicated; one of the difficulties encountered in musculoskeletal pathology is the large number
of potential diagnoses. If all lesions are considered separately, it may be very difficult for the orthopaedist to
place a specific lesion into a reasonable differential diagnosis. By considering trends and patterns of groups of

204

Figure 17 Radiograph of a giant cell tumor of the

proximal tibia.

histologic lesions and as the relationship of histologic

findings to a differential diagnosis based on available
imaging studies, the whole spectrum of histologic diagnoses, at least at an elementary level, is more understandable.

Bone-Forming Tumors
Bone forming tumors, including fracture callus, myositis
ossificans, osteoid osteoma, osteoblastoma, fibrous dysplasia, parosteal osteosarcoma, and osteosarcoma, typically contain woven bone and a spindle cell stroma,
each of which needs to be scrutinized to understand the
nature of the lesion. Histologic evaluation of boneforming lesions requires an understanding that the bone
differentiates reactive from neoplastic lesions and the
stromal cells differentiate benign from malignant neoplasms. First, the bone produced by a specific lesion
should be examined. Figure 18, A shows lamellar bone
with a surrounding round cell infiltrate (Ewings sarcoma) in which the lines of individual bone lamellae can
be seen. Lamellar bone is rarely produced by tumors
and is usually just native host bone entrapped in a lesion or part of a mature bone reaction. However, woven
bone can be either neoplastic or reactive, which is determined by the presence or absence of osteoblastic rimming. Typically, woven bone with significant osteoblastic
rimming is reactive and indicative of fracture callus, periosteal reaction, or myositis ossificans (Figure 18, B). In
such lesions, spindled stromal cells also may be present
and have some mild atypia (especially in early myositis

American Academy of Orthopaedic Surgeons

Chapter 19 Musculoskeletal Oncology

Orthopaedic Knowledge Update 8

or early fracture callus). In contrast, woven bone that
shows no osteoblastic rimming is usually neoplastic in
origin and is seen with both benign and malignant boneforming neoplasms (Figure 19, A). Neoplastic woven
bone (bone with no osteoblastic rimming) found in association with a benign spindle cell stroma is indicative
of a benign bone-forming neoplasm. Such lesions include osteoid osteoma, osteoblastoma, and fibrous dysplasia. Also included in this histologic differential diagnosis is parosteal osteosarcoma, a grade 1 malignant
tumor, which presents with a stroma with little overt
atypia. The final pattern seen in bone-forming lesions is
the presence of neoplastic woven bone in association
with a malignant spindle cell stroma that constitutes osteosarcoma (Figure 19, B). The findings that make a
spindle cell stroma appear malignant include increased
cellularity, spontaneous necrosis, the presence of significant atypia or pleomorphism, and a high mitotic rate
with many abnormal mitoses.
These bone-forming lesions tend to look very different radiographically; therefore, a radiographic differential diagnosis integrated into the histologic differential
diagnosis helps to determine the final diagnosis.

Cartilage-Forming Tumors
There are three histologic patterns of cartilage tumors:
(1) benign cartilage (enchondroma) merging into lowgrade cartilage which merges into intermediate and high
grade chondrosarcoma; (2) chondroblastoma that is
characterized by cobblestone chondroblasts and intervening chicken-wire calcification with immature cartilage; (3) chondromyxoid fibroma that is a benign spindle cell lesion with some areas of immature cartilage.
In the first histologic pattern, it is essential to understand the spectrum of cartilage appearances that match
the spectrum of histologic grading. Normal cartilage has
two components: cells and matrix. Both are important
when evaluating cartilage histologically. Normal cartilage is sparsely cellular (Figure 20, A). The cells have
small oval (pyknotic) nuclei and only one nucleus per
cell. There is only one cell per lacuna and there are
rarely cells outside of the lacunae. The matrix is wellformed and regular with no areas that are loose or falling apart (myxoid change).
As cartilage changes from benign to low grade the
following changes occurincreased cellularity, the presence of plump nuclei, occasional binucleate cells, more
than one cell in some lacunae, some cells outside the lacunae, and myxoid change in the matrix.
The finding of low-grade cartilage includes the spectrum of lesions ranging from cellular or active enchondromas to what are sometimes called grade one-half
chondrosarcoma, to grade 1 chondrosarcoma. It is critical to appreciate that tumors in this range have a variable biologic potential in terms of their propensity for

American Academy of Orthopaedic Surgeons

Table 2 | Important Radiographic Findings for Making a

Specific Differential Diagnosis
Finding

Differential Diagnosis

Sclerotic soap bubble lesion in the

anterior cortex of the tibial shaft
Small sclerotic lesion with a
central lytic nidus

Adamantinoma
Cortical fibrous dysplasia
Osteoid Osteoma
Stress fracture
Infection
Changes that indicate edema in
Intramedullary changes caused by
bone*
lymphoma
Intramedullary edema caused by a
bone bruise or stress reaction
Cauliflower osteochondroma
Cauliflower exophytic lesion
Secondary chondrosarcoma arising
in an osteochondroma
Multiple bone lesions
Metastases
Myeloma
Enchondromas
Histiocytosis
Fibrous dysplasia
Nonossifying fibromas
Lytic lesion in the humeral shaft of Simple bone cyst
a child with no periosteal
reaction
Lytic lesion in the sacrum
Chordoma
Chondrosarcoma
Giant cell tumor
Metastasis
Myeloma
Calcified lesion on the bone
Osteochondroma
surface
Periosteal osteosarcoma
Parosteal osteosarcoma
Myositis ossificans
Periosteal chondroma
Periosteal chondrosarcoma
Aggressive metaepiphyseal lesion in Osteosarcoma
patients younger than age 30
Ewings sarcoma
years
Infection
Aneurysmal bone cyst
Giant cell tumor
Aggressive metaepiphyseal lesion in Osteosarcoma
patients 30 years of age or older Chondrosarcoma
Metastasis
Adult round cell tumor
Giant cell tumor
Lytic lesion in the epiphysis of a
Chondroblastoma
child with edema seen on MRI
Infection
*Sequential MRI scans of lymphomas show stable or progressive marrow replacement,
whereas sequential MRI scans of stress reaction or a bone bruise show a decrease in
bone edema over time

It is important to measure the thickness of the cartilage cap

continued growth. Throughout this range of histologic

appearance, pathologists have no effective means of
predicting growth; therefore, an attempt to artificially

205

Orthopaedic Knowledge Update 8

Musculoskeletal Oncology

Table 3 | Important Rules to Consider When Performing an Open Biopsy

Rules Regarding the Incision
The smallest possible incision should be made over the lesion
The incision on the extremities should be longitudinal; the incision on the trunk should be planned to be part of a resection incision
The use of an Esmarch tourniquet wrapped over a tumor may rupture the tumor into the surrounding tissues and should be avoided; it is reasonable to use
an extremity tourniquet proximal to a tumor of the extremity (particularly to minimize blood loss)
A small incision should be made into the capsule of a tumor so that it can be easily closed; a small incision is especially important when no tourniquet is
used because a large incision can cause appreciable bleeding that is difficult to control
The size of the incision into the tumor should be no larger than the surgeon can fill with a fingertipthis will allow a rapid, temporary hemostasis; if
excessive bleeding does not occur, a larger incision can then be made
Rules Regarding Avoidance of Contamination
Care should be taken to prevent direct contamination of the neurovascular bundle
Care should be taken to prevent the violation of major flap structures (eg, gluteus maximus) or functionally important structures (eg, rectus femoris)
Minimal retraction should be used to limit soft-tissue contamination
It is preferable to go through a single muscle belly (if it is large enough) than to go between two structures, which could contaminate both
Good hemostasis must be obtained by meticulous, multilayered watertight closure; large tumors may put the point of closure under pressure and the
vascularity that is present may predispose the area to subsequent drainage and breakdown
Rules Regarding Conditions During and After Open Biopsy
Wound complications should be avoided because they increase the risk of secondary infection and may delay subsequent chemotherapy or radiation
Drains should not be used routinely; if needed, they should be thin and should exit 1 to 2 cm beyond one end of the incision so that the drain track can be
resected easily with the biopsy track
A frozen section should be obtained when feasible to ensure that diagnostic material is present; the necrotic nature of tumors may require a large volume of
tissue to ensure that diagnostic material is obtained
Sending biopsy material for culture and sensitivity should be considered if there is suspicion that the lesion is neoplastic or infectious
When performing a diagnostic open biopsy, the operating surgeon should accompany the specimen to the pathology department if feasible; the pathologist
and surgeon should then review imaging studies, the specimen, frozen section slides, and the patients clinical history

determine where cellular enchondroma ends and lowgrade chondrosarcoma begins would be arbitrary. The
term low-grade cartilage tumor has become popular, denoting the fact that the pathologist is acknowledging the
position of the tumor somewhere within this spectrum
of behavior. The key for the treating physician is to
evaluate the clinical history and imaging studies and to
determine a reasonable course of treatment. Options
may range from careful observation, to curettage, to resection depending on a host of factors including the
presence or absence of pain, the age of the patient, the
bone involved, and the presence or absence of radiographic changes typical of chondrosarcomas.
Chondrocytes look less normal and behave in a less
normal fashion with increasing grades of malignancy;
the chondrocytes develop dark, plump nuclei, cellularity
increases appreciably, and mitoses become more common. The cells make less or no matrix, or may produce
an abnormal matrix, which may look nothing like recognizable chondroid. Associated with this histologic progression is a progressive increase in biologic potential in
terms of local aggressiveness and metastatic behavior.
The second histologic pattern in chondroid lesions is
seen in chondroblastoma. This lesion is often suspected
even before histologic material is examined because of
its characteristic radiographic and clinical features (for

206

example, a painful lytic lesion in the epiphysis of a

child). Histologically, a pattern of cobblestone chondroblasts (polygonal cells with well-defined cell borders)
are seen (Figure 20, B). In association with these cells, a
branching pattern of calcification referred to as chicken
wire is also common, as is the presence of immature
chondroid matrix. These histologic findings in association with an epiphyseal lytic lesion in a child yield a
fairly straightforward diagnosis of chondroblastoma.
The third histologic pattern seen in cartilage tumors
is found in chondromyxoid fibroma. This is a rare lesion
that is composed of benign spindle cells in a collagenous
matrix with varying amounts of immature chondroid
(Figure 20, C). This histologic pattern in association with
the typical radiographic findings helps in determining
the diagnosis.
Many bone- and cartilage-forming lesions can be accurately diagnosed by first assembling a radiographic
differential diagnosis via the radiology lists and then assessing the histologic spectrum of bone and cartilage
forming lesions using a knowledge of simple trends and
pattern recognition. For example, in a 40-year-old
woman with a recent history of knee pain who has otherwise been in good health, radiographs show an aggressive lytic lesion in the distal femur with cortical bone
destruction and soft-tissue extension. The radiographic

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 19 Musculoskeletal Oncology

Figure 18 A, Lamellar bone with a surrounding round cell tumor. B, Reactive woven bone showing abundant osteoblastic rimming (arrow) of trabecular surfaces.

Figure 19 A, Neoplastic woven bone showing the absence of osteoblastic rimming and the presence of an associated benign spindle cell stroma typical of a benign bone
forming neoplasm. B, Osteosarcoma showing neoplastic woven bone in association with a malignant spindle cell stroma.
differential of this lesion includes osteosarcoma, chondrosarcoma, metastasis, myeloma, lymphoma, and giant
cell tumor. The histology shows woven bone with no osteoblastic rimming compatible with neoplastic woven
bone (Figure 19, B). A malignant spindle cell stroma is
also seen histologically. The final diagnosis, osteosarcoma, is compatible with the radiograph. Although this
system cannot encompass and diagnose every musculoskeletal lesion, it can provide a framework for developing an approach to these diagnoses.

Other Tumors
Round Cell Tumors
Round cell tumors are another group of bone lesions
and include Ewings sarcoma and neuroblastoma in children and myeloma, lymphoma, and small round cell
metastatic carcinomas in adults. Also included in the differential diagnosis of this group of tumors are nonmalignant lesions such as Langerhans cell histiocytosis and infection. The common thread between all of these lesions
is that they are composed in part or in whole of small
round cells. When round cell infiltrate is present in a histologic slide from a bone lesion, the following classes of

American Academy of Orthopaedic Surgeons

disorders should be considered: infection (ie, osteomyelitis), Langerhans cell histiocytosis, primary round cell
tumors, and small round cell metastatic carcinomas.
The histology of bone infection includes the presence of acute and chronic inflammatory cells. Whereas
lymphocytes, which tend to be associated with chronic
inflammatory conditions, may resemble lymphoid cells
seen in lymphoma, polymorphonuclear leukocytes are
usually easy to see and, when present in large numbers,
indicate a diagnosis of infection. When a round cell infiltrate is seen and most of the cells can be shown to be
polymorphonuclear leukocytes, then infection is the
likely diagnosis. Confirmation requires culture of an appropriate pathogenic microorganism.
Langerhans cell histiocytosis is also a type of inflammatory condition in bone and can present as a round
cell infiltrate. As the name implies it is composed of foci
of proliferating histiocytes; varying numbers of small
round cells including lymphocytes, neutrophils and, most
notably, eosinophils are also present. Histiocytes are
large cells with ill-defined cytoplasmic borders and an
oval or indented nucleus and are often difficult for a
nonpathologist to recognize. Eosinophils are distinctive

207

Musculoskeletal Oncology

Orthopaedic Knowledge Update 8

cogen are typically positive. A neuroblastoma occurring
in bone is usually a metastasis from a primary tumor in
the midline. The cells of a neuroblastoma look similar to
those of Ewings sarcoma but show the presence of pseudorosettes, which appear as circles of round cells surrounding a pink ground substance. Thus, in pediatric
round cell tumors, small round blue cells with no pseudorosette formation is typical of Ewings sarcoma whereas
similar appearing cells that also show pseudorosette formation are typical of neuroblastoma.
Adult round cell tumors include myeloma, lymphoma, and small round cell metastatic carcinomas.
Usually, myeloma is easily recognizable (when well differentiated) because it is composed of sheets of plasma
cells. Plasma cells have a single round or oval nucleus
with eccentric cytoplasm. The cell outlines are distinct
and the nuclei often show prominent clumping of chromatin, which produces a clock face or wheel spoke appearance.
Lymphoma and small round cell carcinoma metastases are much more difficult to differentiate by routine light microscopy and may require immunohistochemical stains for final verification. As such,
metastases usually stain positive for cytokeratin,
whereas lymphomas will stain positive for lymphoid
markers such as leukocyte common antigen or B or T
cell markers. Thus, with adult round cell tumors, sheets
of plasma cells indicate myeloma whereas round cells,
which are not plasma cells and look more lymphocytic,
are either lymphoma or a small round cell carcinoma.

Figure 20 A, Normal cartilage showing sparse cellularity and good matrix production.
B, Chondroblastoma showing cobblestone chondroblasts and a chondroid matrix.
C, Chondromyxoid fibroma showing bland spindle cells in a chondroid matrix.

small round cells with bilobed nuclei and red cytoplasm

on hematoxylin-eosin staining. When diagnosing histiocytosis, pathologists tend to look for histiocytes whereas
orthopaedists look for eosinophils.
Pediatric round cell tumors occurring in bone include
Ewings sarcoma and metastatic neuroblastoma. Ewings
sarcoma is a malignant tumor of unknown histogenesis
composed of uniform sheets of small round blue cells.
These cells show round or oval nuclei of uniform size with
poor delineation of cytoplasm. Stains for intracellular gly-

208

Unicameral and Aneurysmal Bone Cysts

Bone cysts include unicameral or simple bone cysts and
aneurysmal bone cysts. As previously noted, these lesions appear different radiographically. Histologically,
unicameral bone cysts appear as a thin layer of fibrous
tissue lining large empty spaces. Benign giant cells, hemosiderin pigment, and a few chronic inflammatory
cells also may be present. Aneurysmal bone cysts have a
different histologic appearance and are characterized by
the presence of blood-filled cavernous spaces with walls
that lack the normal features of blood vessels. There is
some overlap between these two lesions however, and
some cysts will have characteristics of both. This situation is most evident in cysts directly abutting the growth
plate in skeletally immature patients, especially in young
children. In these instances, the cyst is often central as
opposed to eccentric and often minimally expansile.
These lesions can have a very aggressive course, particularly the aneurysmal bone cysts, which may be unresponsive to percutaneous treatments and often require
open treatments such as curettage, bone grafting, and
embolization. Great care needs to be directed toward
preservation of the adjacent growth plate.

American Academy of Orthopaedic Surgeons

Chapter 19 Musculoskeletal Oncology

Orthopaedic Knowledge Update 8

Giant Cell Tumor
Giant cell tumor of bone is a lesion with a variable biologic potential. Although this is typically an aggressive
benign lesion, 2% of patients with histologically benign
giant cell tumor of bone will have pulmonary metastases; several of these patients will die from progressive
metastatic disease. In other patients, however, the metastases will not have an aggressive course. The diagnosis of benign giant cell tumor is usually fairly simple
when the histologic findings are coupled with typical radiographic changes. The radiographs usually show a
juxta-articular lytic lesion with a moth-eaten margin,
cortical thinning or erosion, and no periosteal reaction.
Histologically, the lesion is composed of multinucleate
giant cells and mononuclear stromal cells. The characteristic finding is that the nuclei of the giant cells look
identical to the nuclei of the stromal cells. Mitotic figures may be found in all of these lesions and may be
prominent in some. Small areas of woven bone also may
be seen along with areas of spindle cells with spindled
nuclei. Whereas the diagnosis of most of these lesions is
straightforward, there are some giant cell rich osteosarcomas that are indistinguishable from benign giant cell
tumor. In these lesions, it is often only a malignant pattern of subsequent growth and metastasis that elucidates the true nature of the neoplasm. When evaluating
a probable benign giant cell tumor of bone, it is recommended that the possibility of giant cell rich osteosarcoma be considered, realizing that the ultimate differentiation may be difficult or impossible.
Nonossifying Fibroma (Metaphyseal Fibrous Defect)
A nonossifying fibroma is a common lesion that only
rarely requires surgical intervention. Most of these lesions
are incidental findings and require no specific treatment
other than occasional radiographs to document stability
and healing; however, they occasionally may be large
enough to cause mechanical pain and limit the ability of
the child or adolescent to participate in desired activities.
When this occurs or when the specific diagnosis remains
elusive, open biopsy with curettage and some form of
bone grafting is reasonable. Histologically, these lesions
have a characteristic appearance showing benign spindle
cells with frequent benign giant cells. After surgical excision and grafting, these lesions will usually heal completely in about 8 weeks, whereas such healing would commonly take 2 years or more without surgical intervention.
Adamantinoma
Adamantinoma is a characteristic lesion usually found in
the anterior cortex of the midshaft of the tibia. It usually
appears as a soap bubble sclerotic lesion and can mimic
cortical fibrous dysplasia. Cortical fibrous dysplasia occurs
predominately in males in the first and second decades of
life, whereas adamantinoma occurs in both genders and
all age groups usually occurring during the first through

American Academy of Orthopaedic Surgeons

fourth decades of life. Histologically, adamantinoma is a

low-grade spindle cell sarcoma with islands of epithelial
cells that may resemble cutaneous basal cells. A second
histologic pattern for this lesion consists of islands of neoplastic cells surrounded by columnar cells in a palisading
fashion. These findings are usually fairly diagnostic when
seen with the usual radiographic presentation.

Immunohistochemistry
One of the major advances in diagnostic pathology that
has occurred over the past two decades has been the development of sophisticated immunohistochemical techniques. These stains have greatly improved the diagnostic acumen of tissues seen via light microscopy and have
added a new area of classification based on specific cell
proteins. Although a detailed knowledge of this field is
not essential, these stains are commonly referred to in
pathology reports of musculoskeletal lesions and should
be familiar at a basic level to all orthopaedic surgeons.
Immunohistochemical stains are available to identify
specific intermediate filament proteins, which are basic
structural components of all human cells. These proteins
include distinct moieties that are separated biochemically and include vimentin, desmin, keratins, and neurofilament, glial fibrillary, and lamin filament proteins
(nuclear envelope proteins). In terms of cell function,
the intermediate filament proteins serve a nucleic acid
binding function and may also act as modulators of nuclear function at a translational or transcriptional level.
It should be emphasized that whereas certain tumors
have typical immunohistochemical profiles, the profiles
vary from tumor to tumor as individual tumors exhibit
specific genotypes and phenotypes.

Keratins
Keratins are usually seen in epithelial tissues and cells.
As a diagnostic marker in bone tumors, keratins are the
classic markers of metastatic carcinomas. In rare instances, however, they may be seen in almost any form
of sarcoma. Keratins are also commonly seen in the epithelial component of those sarcomas that show some
epithelial differentiation, including synovial sarcoma, adamantinoma, and epithelioid sarcoma.

Vimentin
Vimentin is a protein typically found in tumors with a
mesothelial origin. Immunohistochemical staining of vimentin is positive in almost all sarcomas but usually
negative in carcinomas. Because its presence is widespread among sarcomas, it is not a useful marker to distinguish between specific sarcoma types.

Desmin and Actin

Desmin is typically found in muscle cells and in tumors
with myodifferentiation. Such tumors most commonly

209

Orthopaedic Knowledge Update 8

Musculoskeletal Oncology

with even trivial local trauma. It is also remarkable that

soft-tissue masses including sarcomas can grow to a very
large size and yet cause minimal or no symptoms. Many
patients falsely assume that because the lesion is painless it must also be harmless. Although this assumption
is incorrect, it is often responsible for long delays in diagnosis. Ironically, the lesions in soft tissue that are usually painful are the benign soft-tissue tumors including
desmoid tumors, hemangiomas, benign nerve sheath tumors, and soft-tissue infections.

Radiographic Evaluation

Figure 21 Radiograph of a densely calcified synovial sarcoma mimicking myositis

ossificans.

present in soft tissue and include rhabdomyomas, rhabdomyosarcomas, leiomyomas, and leiomyosarcomas.
Desmin is also occasionally present in desmoid tumors
and in primitive neuroectodermal tumors. Like desmin,
actin is indicative of myogenous differentiation and its
tissue-specificity parallels that of desmin.

S-100
S-100 is a protein that derives its name from the fact
that it is soluble in 100% solution of ammonium sulfate.
It has a wide distribution in human tissues and the
stains indicating the presence of this protein are most
commonly associated with neural, chondroid, or melanocytic differentiation.

Factor VIII Related Antigen

Factor VIII related antigens (also called von Willebrands factor) are found in lesions with vascular differentiation. These antigens are typically seen in benign
and low-grade vascular lesions such as hemangiomas
and hemangioendotheliomas; they usually are not
present in high-grade angiosarcomas.

Soft-Tissue Tumors
Soft-tissue tumors, like bone lesions, require a systematic approach for diagnosis. These lesions have a limited
number of clinical presentations. Histologically, however, they form a large and diverse group with fewer
trends in their histologic appearance than those found
in bone tumors.

Clinical Presentation
Most soft-tissue tumors present with pain and/or a mass.
As noted previously, there is often a history of trauma
as patients tend to relate the emergence of a mass lesion

210

Most soft-tissue masses are seen poorly or not at all on

plain radiographs; however, those with calcification will
be more apparent. Although myositis ossificans is the
most common lesion characterized by soft-tissue calcification, synovial sarcoma may present in this manner
also. The calcification seen in synovial sarcomas can
vary considerably from minute, almost imperceptible
calcifications to very dense calcification, which may
mimic a benign lesion (Figure 21).

MRI Findings
MRI is the gold standard for the evaluation of lesions in
soft tissue. MRI is quite useful in locating a lesion but
less useful in delineating the nature of the lesion. There
are, however, some notable exceptions to this generalization. The classic MRI finding in most soft-tissue tumors is that of a lesion that is well circumscribed and
showing dark signal on T1, and high signal on T2, fatsuppressed T2, or short-tau inversion recovery (STIR)
views (Figure 22). The possible etiology of such a lesion
includes benign tumor, malignant tumor, abscess, cyst,
and hematoma. It is often incorrectly believed that softtissue sarcomas are grossly invasive whereas benign lesions are radiographically distinct and encapsulated.
Most soft-tissue sarcomas are very distinct and often
show some edema in the compartment in which they occur, whereas many benign lesions including desmoid tumors, hemangiomas, inflammation, injury, and infection
are poorly marginated on MRI scans. Lesions with a
characteristic MRI appearance include lipomas, atypical
lipomas, myositis ossificans, and hemangiomas.
Lipoma
Lipoma is one of the few histologic diagnoses that can
be made confidently on the basis of MRI and clinical
findings alone. Benign lipomas appear as masses of uniform fat density and parallel the appearance of normal
subcutaneous fat on all sequences. They are bright on T1
and T2 views (Figure 23) but suppress, as does normal
fat, on fat-suppressed T2 and STIR views. A mass seen
on MRI as a uniform fat density with no interstitial
markings is diagnostic of benign lipoma.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 19 Musculoskeletal Oncology

Figure 22 A soft-tissue tumor showing the

typical findings of low signal on T1 (A) and
high signal on T2 (B). This lesion is a benign
schwannoma.
Atypical Lipoma
Atypical lipoma is also called well-differentiated liposarcoma and lipoma-like well-differentiated liposarcoma. It
is a fat-containing lesion characterized by lobules of fat
signal on MRI with surrounding layers of fibrous tissue,
which appear as thin layers of high signal (Figure 24).
The critical difference between this lesion and highergrade liposarcoma is that the lobules in this lesion appear fatty on MRI. Ordinary liposarcoma looks like the
typical nonlipomatous lesion, which appear dark on T1
and bright on T2 and fat-suppressed T2 views, and on
STIR views. Although atypical lipomas do not metastasize, they present a 10% risk of malignant transformation, usually to high-grade liposarcoma.
Myositis Ossificans
Late inactive myositis ossificans lesions present radiographically as uniform, well marginated, and benign appearing calcification in soft tissue. The calcification often
is more prominent peripherally and is termed eggshell calcification (Figure 25). Early lesions show little or no calcification but do show tremendous inflammation and
edema in the adjacent soft tissues. In such cases, the area
of edema is much larger in volume than the area of the
lesion itself.
Hemangioma
Hemangiomas are typically diffuse heterogeneous lesions with serpiginous borders (Figure 26). The classic
findings of a hemangioma include the presence of a
painful lesion in soft tissue which presents with a softtissue mass seen on MRI scan but with no mass effect.
Atrophy, underlying pain, and smooth soft-tissue calcifications (phleboliths) are common.

Histology of Soft-Tissue Tumors

The histology of soft-tissue tumors has fewer common
trends compared with the histology of bone lesions.

American Academy of Orthopaedic Surgeons

Table 4 | Surgical Steps for Effective Intralesional Surgery

Extensive soft-tissue exposure
Create a large bone window to completely expose all surfaces of the
underlying cavity
Wide resection of any area of soft-tissue extension
Complete curettage of all gross tumor in the cavity
Burring of the perimeter of the bony cavity to extend the curettage
beyond all visible areas of tumor extension
Pulsatile lavage to expose clean surfaces of adjacent bone
Cauterization of the bony surface of the cavity using 90% phenol or
liquid nitrogen
Reconstruction with bone cement or bone graft with or without internal
fixation as needed for bony stability and joint surface support

There are many specific tumors to consider and the

range of lesions goes beyond the scope of this chapter.

Treatment of Musculoskeletal Tumors

Several treatment options are available for musculoskeletal tumors. Lesions in bone may be treated with options ranging from simple curettage, to aggressive intralesional excision, to wide and radical resection.
Simple curettage is sufficient for lesions that tend to be
self-limited such as nonossifying fibroma and eosinophilic granuloma. Aggressive benign lesions such as giant cell tumor, aneurysmal bone cyst, chondromyxoid fibroma, and osteoblastoma require a more aggressive
surgical technique entailing several steps, which together allow the surgeon to control the tumor bed and
result in a high likelihood of local control. These steps
for aggressive intralesional excision are shown in Table
4. It is only through the meticulous and methodic application of these steps that eradication of the underlying
lesion can be reliably achieved.
Soft-tissue tumors may be treated with a full range of
surgical options including debulking, marginal excision,

211

Orthopaedic Knowledge Update 8

Musculoskeletal Oncology

Figure 23 A benign lipoma (arrow) of the thigh showing a uniform fat density with no interstitial markings.

Figure 24 An atypical lipoma (well-differentiated liposarcoma) showing lobules of fat with surrounding fibrous strands.

Figure 25 Myositis ossificans of the elbow with typical eggshell calcification.

Figure 26 A benign hemangioma showing extensive infiltration within the forearm. A, Axial
view. B, Sagittal view.
wide resection, and radical resection. Most benign softtissue masses such as lipomas, schwannomas, and myxomas can be easily excised. Soft-tissue tumors with a
greater tendency for local invasion and infiltration can
be very challenging to treat and may require a variety of
treatment modalities beyond simple surgical techniques.
Benign vascular lesions, for example, have a high rate of
recurrence following surgical excision and often involve
such large areas of tissue that excision is not reasonable. These lesions may be treated with a variety of nonsurgical techniques such as embolization or direct alcohol injection. Many patients with benign vascular lesions
such as hemangiomas and arteriovenous malformations
are difficult to cure; many will have prolonged symptoms
and disability despite therapeutic intervention. Desmoid
tumors are another example of benign yet problematic

212

lesions in soft tissue. Like hemangiomas, these lesions

have a very high rate of recurrence following even aggressive attempts at local resection and are often large
and may involve significant portions of major functional
structures. The idea of a major and morbid resection for
a benign lesion is often unacceptable. Nonsurgical options exist and have replaced surgery in many centers.
Treatment using low-dose chemotherapy with methotrexate, vinblastine, or vinorelbine has been reported by
several authors and achieves a high rate of local control
with minimal morbidity. Radiation is also a reasonable
option but is a cause for concern for subsequent malignant transformation because of the young age of most
patients with desmoid tumors.
The treatment options for sarcomas of bone and soft
tissue are summarized in Table 5. The therapeutic strat-

American Academy of Orthopaedic Surgeons

Chapter 19 Musculoskeletal Oncology

Orthopaedic Knowledge Update 8

egy is to apply treatments to specific lesions that respond to that type of treatment and to avoid options
that are not effective. The differentiation of local versus
systemic treatments is also important. Local treatment
options include surgery and radiation. In the context of
sarcoma treatment, surgery usually refers to procedures
that achieve wide margins around the tumor. Current
musculoskeletal oncology surgery infrequently requires
amputation or radical margin surgery for extremity sarcomas. The primary indication for such procedures
would involve long-neglected lesions or lesions with the
direct involvement of neurovascular structures, whose
resection would preclude the maintenance of a functional limb. Adjuvant treatments such as chemotherapy
and radiation have permitted wide margins to become
increasingly thin, especially along neurovascular structures. These surgical procedures allow functional preservation and still achieve good rates of local control that
are not statistically different from those obtained with
amputation. Both surgery and radiation are aimed at local control and do not affect the systemic tumor burden.
Chemotherapy is a systemic treatment that has the potential to act effectively against tumor cells throughout
the body even if they cannot be detected by current imaging techniques. Chemotherapy may affect the local tumor as well as any distant disease. Low-grade malignant
tumors tend to metastasize at a lower rate than do highgrade lesions; therefore, it is reasonable to approach the
former with local treatments only. High-grade lesions,
which carry a high likelihood of metastasis, are best
treated with a systemic treatment designed to cure systemic spread and increase the resectability of the primary lesion, followed by local treatment designed to remove the primary tumor.
Low-grade bone sarcomas such as chondrosarcoma,
chordoma, and adamantinoma respond poorly to radiation and chemotherapy, making surgery the best treatment option. High-grade bone sarcomas such as osteosarcoma and Ewings sarcoma respond well to
chemotherapy and it is therefore a consistent part of
treatment protocols. Other high-grade bone sarcomas
such as malignant fibrous histiocytoma in bone, highgrade chondrosarcoma, and dedifferentiated chondrosarcoma present a challenge because of their predilection
to metastasize. Although less data are available than for
osteosarcoma, chemotherapy is often used in an attempt
to both increase resectability and decrease the appearance or progression of systemic spread. Radiation is still
a possible local treatment choice for Ewings sarcoma,
but usually has been replaced by surgery because of the
potential for postradiation morbidity and malignant
transformation. Despite these concerns, radiation is still
used and reasonable in patients where resection carries
the risk for unacceptable morbidity or when surgical
margins are closer than desired. These principles also
apply to other high-grade bone sarcomas; a treatment

American Academy of Orthopaedic Surgeons

Table 5 | Sarcoma Treatment Modalities

Radioresistant Tumors
Low Grade

High Grade

Radiosensitive Tumors

Surgery
Surgery + Radiation
Chondrosarcoma
Low-grade soft-tissue
Chordoma
sarcomas
Adamantinoma
Surgery + Chemotherapy Surgery + Radiation +
Chemotherapy
Osteosarcoma
High-grade soft-tissue
Ewings sarcoma
sarcomas
Other high grade bone
sarcomas

plan should be individualized for each patient with consideration for the potential benefits of each treatment
option.
Low-grade soft-tissue sarcomas respond well to a
combination of wide resection surgery and local radiation. The radiation can be given either preoperatively or
postoperatively because there is no difference in local
control if the treatments are well planned. The major
advantage to preoperative radiation is that the presence
of the lesion in situ allows the radiation oncologist to
concentrate the field of radiation on the lesion, which
usually allows treatment of a smaller field size than in
patients where the lesion has been removed. Radiation
also has beneficial effects on the local tumor, which usually will decrease in size and vascularity and increase in
firmness, thereby greatly facilitating resection. The disadvantage of preoperative radiation treatment is that it
necessitates working with tissues that have been damaged by radiation (usually 50 cGy [5,000 rads] of radiation); this carries a 20% risk of major wound complications in most series.
High-grade soft-tissue sarcomas still present a major
challenge to the oncology team. Local control can be
achieved with radiation and wide resection as with the
low-grade malignant lesions. Ideally, the high-grade lesions should also receive systemic treatment in the form
of chemotherapy because of the higher rate of metastasis associated with high-grade histology. Although these
patients are usually treated with protocols that include
multidrug chemotherapy, the results of this treatment in
terms of improved survival are not as dramatic as would
be hoped. Chemotherapy for high-grade soft-tissue sarcomas continues to be a controversial topic and the subject of much research. It is hoped that new findings regarding the molecular and genetic bases of these
diseases will spawn new drug treatments for all types of
sarcomas and further improve the survival of these patients.

213

Musculoskeletal Oncology

Orthopaedic Knowledge Update 8

Emerging Concepts and Future Directions

Treatment of Musculoskeletal Tumors

There are several areas in orthopaedic oncology where

promising new discoveries may advance current treatment and diagnosis. From a surgical perspective, one of
the practical problems that has impeded the use of prostheses for segmental replacement has been the inability
of current designs to optimize soft-tissue attachment,
thus presenting a major challenge for treatment of the
greater trochanter, the patellar tendon, and the shoulder
capsule. On the horizon are promising new fiber metals
that promote scar invasion and perhaps even bone ingrowth, especially when combined with biologic stimulants of bone formation. These new materials may allow
for much firmer and more consistent attachment of important tendons and ligaments, which in turn will promote joint stability and function.
Advances in several fields will facilitate a deeper understanding of our present classification of tumors and
will aid in tumor diagnosis. Newer immunohistochemical stains simplify the job of classifying tumors, which
may be difficult to label based on light microscopy
alone. In addition, ongoing research into the genetic
makeup of musculoskeletal tumors is proceeding. This
work, aided by microarray and comparative genetic hybridization techniques, is beginning to identify genetic
defects as well as individual genes that are upregulated
and downregulated in different tumor types. This increased knowledge will aid in subclassifying tumors
based on their genetic makeup and should suggest new
treatment options directed against specific genetic targets.

Aboud JA, Patel RV, Donthineni-Rao R, Lackman RD:

Proximal tibial segmental prosthetic replacement without the use of muscle flaps. Clin Orthop 2003;414:189196.

Annotated Bibliography
Factors to Consider in the Differential Diagnosis
Muscolo DL, Ayerza MA, Makino A, Costa-Paz M,
Aponte-Tinao LA: Tumors about the knee misdiagnosed as athletic injuries. J Bone Joint Surg Am 2003;85:
1209-1214.
Of 667 tumors evaluated, 25 (3.7%) were initially misdiagnosed as athletic injuries. Oncologic surgical treatment was affected in 15 of the 25 patients.

Imaging Studies
Aboulafia AJ, Levin AM, Blum J: Preferential evaluation of patients with suspected bone and soft tissue tumors. Clin Orthop 2002;397:83-88.
This article concludes that many unnecessary imaging
studies are often obtained in the initial work up of suspected
musculoskeletal tumors especially in those that are found to
be benign.

214

Twenty-two patients with tibial bone tumors were treated

with proximal tibial segmental replacement prostheses by direct reattachment of the patellar tendon to the prosthesis
without the use of muscle flaps. Functional results were comparable to patients treated with the use of flaps.

Bickels J, Wittig JC, Kollender Y, et al: Distal femoral

resection with endoprosthetic reconstruction. Clin
Orthop 2002;400:225-235.
A review done at one center of distal femoral replacement
prostheses that included 110 reconstructions with 7.8-year
follow-up is presented. Important statistics include a 5.4% infection rate, 5.4% local recurrence rate, and 10.8% revision
rate for aseptic loosening or polyethylene failure.

Jung ST, Ghert MA, Harrelson JM, Scully SP: Treatment

of osseous metastases in patients with renal cell carcinoma. Clin Orthop 2003;409:223-231.
A good review, including survival statistics, of 99 patients
with metastatic renal cell tumors is presented. A reasonable
argument is presented for wide resection of solitary bone lesions in patients with no other metastatic locations.

Les KA, Nicholas RW, Rougraff B, et al: Local progression after operative treatment of metastatic kidney cancer. Clin Orthop 2001;390:206-211.
Twenty-two of 41 patients (53%) treated with intralesional
procedures required reoperation whereas only 1 of 37 patients
(3%) treated with wide or marginal resection required repeat
surgery.

Malo M, Davis AM, Wunder J, et al: Functional evaluation in distal femoral endoprosthetic replacement for
bone sarcoma. Clin Orthop 2001;389:173-180.
A multicenter review of 56 patients compared outcomes
for 31 patients treated with an uncemented Kotz prostheses
with outcomes for 25 patients treated with cemented modular
replacement prostheses. Functional scores were significantly
better with the cemented modular replacement system prostheses.

Rougraff BT, Kling TJ: Treatment of active unicameral

bone cysts with percutaneous injection of demineralized
bone matrix and autogenous bone marrow. J Bone Joint
Surg Am 2002;84:921-929.
Twenty-three patients with active unicameral bone cysts
were treated with percutaneous injections of demineralized
bone matrix and autogenous bone marrow. Only 5 of 23 required a second injection and patients returned to full activity
at an average follow-up of 6 months.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 19 Musculoskeletal Oncology

Turcotte RE, Wunder JS, Isler MH, et al: Giant cell tumor of long bone: A Canadian Sarcoma Group study.
Clin Orthop 2002;397:248-258.

Enneking WF, Spanier SS, Goodman MA: A system for

the surgical staging of musculoskeletal sarcoma. Clin
Orthop 1980;153:106-120.

A multicenter study of 186 patients with giant cell tumor

of long bone showed a local recurrence rate of 18% after
curettage and 16% after resection. The nature of the filling
material or adjuvant uses did not alter recurrence risk.

Glasser DB, Lane JM: Stage IIB osteogenic sarcoma.

Clin Orthop 1991;270:29-39.

Zeegen EN, Aponte-Tinao LA, Hornicek FJ, Gebhardt

MC, Mankin HJ: Survivorship analysis of 141 modular
metallic endoprostheses at early followup. Clin Orthop
2004;420:239-250.
A review done at one center of endoprostheses about the
hip, knee, and shoulder, which included a short mean
follow-up of less than 2 years, is presented. Statistical data and
Kaplan-Meier analyses are presented and provide a good indication of short-term complications and short-term prosthetic
survival.

Mankin HJ, Gebhardt MC, Tomford WW: The use of

frozen cadaveric allografts in the management of patients with bone tumors of the extremities. Orthop Clin
North Am 1987;18:275-289.
Mankin HJ, Lange TA, Spanier SS: The hazards of biopsy in patients with malignant primary bone and softtissue tumors. J Bone Joint Surg Am 1982;64:1121-1127.
Mankin HJ, Mankin CJ, Simon MA: The hazards of the
biopsy, revisited: Members of the Musculoskeletal Tumor Society. J Bone Joint Surg Am 1996;78:656-663.

Emerging Concepts and Future Directions

Bacci G, Ferrarri S, Bertoni F, et al: Histologic response
of high-grade nonmetastatic osteosarcoma of the extremity to chemotherapy. Clin Orthop 2001;386:186-196.
In 510 consecutive patients with high-grade osteosarcoma
of the extremity, the histologic response to chemotherapy depended on the number of drugs given and the histologic subtype of the tumor. Four drug regimens and telangiectatic tumors each correlated with an improved prognosis.

Classic Bibliography

Nelson TE, Enneking WF: Staging of bone and softtissue sarcomas revisited, in Stauffer RN (ed): Advances
in Operative Orthopedics. St. Louis, MO, Mosby YearBook, 1994, vol 2, pp 379-391.
Pisters PW, Leung DHY, Woodruff J, Shi W, Brennan
MF: Analysis of prognostic factors in 1041 patients with
localized soft tissue sarcomas of the extremities. J Clin
Oncol 1996;14:1679-1689.

Consensus Conference: Limb-sparing treatment of adult

soft-tissue sarcomas and osteosarcomas. JAMA 1985;
254:1791-1794.

Simon MA, Aschliman M, Thomas N, Mankin HJ: Limbsalvage treatment versus amputation for osteosarcoma
of the distal end of the femur. J Bone Joint Surg Am
1986;68:1331-1337.

Enneking WF, Dunham W, Gebhardt MC, Malawar M,

Pritchard DJ: A system for the functional evaluation of
reconstructive procedures after surgical treatment of tumors of the musculoskeletal system. Clin Orthop 1993;
286:241-246.

Winkler K, Beron G, Delling G, et al: Neoadjuvant chemotherapy of osteosarcomas: Results of a randomized

cooperative trial (COSS-82) with salvage chemotherapy
based on histological tumor response. J Clin Oncol
1988;6:329-337.

American Academy of Orthopaedic Surgeons

215

Chapter

20

Infection
Michael J. Patzakis, MD
Charalampos Zalavras, MD, PhD

Introduction
Infections of the musculoskeletal system are associated
with considerable morbidity and can be challenging to
treat. Prompt diagnosis, aggressive eradication of infection using appropriate antibiotic administration and surgical dbridement, and the restoration of function constitute the principles and goals of treatment. The
resistance of pathogens and the tissue loss resulting
from infection or from surgical dbridement are common clinical problems that complicate the achievement
of these goals.
New diagnostic techniques and new classes of antibiotics have been introduced and serve as a useful adjunct
to surgical management; however, the importance of
prevention cannot be overemphasized. Minimization of
nosocomial contamination, judicious use of antibiotic
prophylaxis, and proper surgical technique are key factors in preventing the development of orthopaedic infections.

Pathogenesis of Musculoskeletal Infections

The pathogenesis of musculoskeletal infections involves
inoculation of the microorganism in musculoskeletal tissues and interaction of the microorganism with the host
environment, resulting in the clinical picture of infection.
Staphylococcus aureus is the most common organism
responsible for musculoskeletal infections; however, any
organism is capable of causing infection depending on
the source of inoculation and the host environment. The
microorganism may gain access to musculoskeletal tissues through three principal mechanisms: hematogenous
spread, spread from a contiguous source of infection, or
direct inoculation during trauma or surgical procedures.
The complex interaction of the inoculated microorganisms with the local and systemic host environment
will determine the occurrence and severity of infection.
Microorganisms adhere to host tissues, begin proliferating and colonizing the involved area, and result in an inflammatory response and host tissue damage. The ability
of the organism to overcome the host defenses and

American Academy of Orthopaedic Surgeons

cause infection is termed virulence. Virulence is a key

mechanism leading to infection, and varies among and
within organism species. S aureus has a multitude of
mechanisms that enable the organism to survive and
cause infection. S aureus initially expresses receptors for
host extracellular matrix proteins such as fibronectin,
thereby facilitating adhesion of the pathogens on host
tissues and colonization of bone, joints, and implants.
Subsequently, the organism releases toxins that damage
host cells and stimulate a systemic inflammatory reaction. S aureus is protected from host immune defenses
by several mechanisms including the excretion of protein A, which inactivates immunoglobulin G; production
of a capsular polysaccharide, which reduces opsonization and phagocytosis of the organism; and the formation of a biofilm, which secludes the organism from host
defense mechanisms. The biofilm is an aggregation of
microbe colonies embedded within a glycocalyx matrix
that usually develop on implants or devitalized bone
surfaces.
Local and systemic host factors play a major role in
the outcome of the microorganism and host interaction.
Local host factors that facilitate infection include reduced vascularity (resulting from arterial disease,
venous stasis, irradiation, scarring, and smoking), neuropathy, trauma, and the presence of implants. Trauma
predisposes a patient to infection by compromising the
soft tissues, by creating a dead space with hematoma accumulation, and in patients with open fractures, by direct inoculation of tissues. The presence of implants promotes adherence of microbes, biofilm formation, and
adversely affects phagocytosis, thereby facilitating development of infection. Early postoperative infections in
the presence of implants usually result from contamination of the surgical wound. Late infection usually results
from hematogenous seeding of the implant; S aureus
and Staphylococcus epidermidis are the most common
pathogens.
Systemic host factors that may reduce the ability of
the immune system to respond to the pathogen include
renal and liver disease, malignancy, diabetes mellitus, alcoholism, malnutrition, rheumatologic diseases, and an

217

Infection
immunocompromised status (such as in patients with acquired immunodeficiency syndrome or those receiving
immunosuppressive therapy). Intravenous drug users
experience multiple episodes of bacteremia and are at
increased risk for hematogenous infections.

Antibiotic Therapy and Antibiotic Resistance

The selection of appropriate antibiotic therapy from the
many available agents is dependent on knowledge of
the microbiology of musculoskeletal system infections
(Table 1). Antibiotic therapy should initially cover the
most probable pathogens until culture and sensitivity results are available; the antibiotic regimen then should be
reevaluated and modified if necessary. Table 2 summarizes the microbiology of common pathogens and the
suggested empiric antibiotic therapy.
Although a pathogen may initially be susceptible to
an antibiotic, resistance may gradually develop by either
spontaneous mutation or by acquisition of new DNA in
the form of plasmids. A plasmid is a construct of autonomously replicating DNA that is distinct from the normal genome of bacteria, spreads from organism to organism, and can be incorporated in the organisms
genome. The emergence of resistance is facilitated by
the suboptimal use of antibiotics for both the prophylaxis and treatment of infection; this would include unnecessary use of antibiotics, empiric administration of
wide-spectrum agents when a narrow spectrum agent is
appropriate, and inadequate dosage or duration of treatment. Resistance is also promoted by prolonged antibiotic therapy in patients with immunosuppression or
multiple comorbidities, and in patients who had inadequate dbridement leading to recurrence of infection.
Noncompliance with measures to prevent infection
(such as hand washing and isolation of patients with resistant pathogens) spreads these organisms from one patient to another. The use of antibiotics in livestock to
promote growth is another factor contributing to antibiotic resistance.
In 1999 in the United States, 52% of infections that
occurred in intensive care units were caused by strains
of methicillin-resistant S aureus (MRSA), and 25% of
enterococci infections resulted from vancomycinresistant enterococci (VRE). These resistance rates represent a 37% and 43% increase, respectively, compared
with the period from 1994 to 1998. Infection with a resistant organism requires modification of antimicrobial
therapy, patient isolation, and implementation of contact precautions to prevent nosocomial spread. MRSA
infections can be treated with vancomycin; however, S
aureus with intermediate resistance to vancomycin has
been reported. Two newer antimicrobial agents, linezolid
and quinupristin-dalfopristin, appear promising in the
treatment of MRSA and VRE infections.

218

Orthopaedic Knowledge Update 8

Diagnostic Modalities
Diagnosis of musculoskeletal infection is facilitated by
laboratory tests, imaging modalities, histology, Gram
stain and culture of specimens, and molecular techniques. The erythrocyte sedimentation rate (ESR) and
the C-reactive protein (CRP) are markers of the acute
phase response secondary to infection or the noninfectious inflammatory processes. The ESR is elevated in
approximately 92% of pediatric patients with osteomyelitis. It rises within 2 days from the onset of infection,
continues to rise for 3 to 5 days after appropriate antibiotic treatment is instituted, and returns to normal after
approximately 3 weeks. In contrast, the CRP is elevated
in approximately 98% of pediatric patients with osteomyelitis, begins rising within 6 hours, reaches a peak
within 36 to 50 hours, and returns to normal approximately 1 week after successful therapy. Surgical treatment prolongs the peak and normalization times of both
the ESR and CRP. The CRP shows a closer temporal relationship to the course of infection; therefore, it is the
preferred marker for early diagnosis and for monitoring
the response to treatment. In periprosthetic infections,
studies have shown that the sensitivity and specificity of
the ESR was 82% and 85%, respectively, whereas that
of the CRP was 96% and 92%, respectively. An elevated
peripheral blood white blood cell (WBC) count with increased polymorphonuclear cells is indicative of infection, but is only elevated in up to 50% of patients; therefore, its absence does not rule out infection.
Radiographs may show soft-tissue swelling, bone
changes (resorption, periosteal new bone formation),
and may disclose the presence of a fracture or tumor
mimicking infection. In arthroplastic infections, radiographs may show lucency around the implants; however,
this may also result from aseptic loosening. Bone scintigraphy using technetium Tc 99m evaluates the perfusion and osteoblastic activity of the skeleton, and is especially useful in localizing the pathologic process to an
anatomic area. Indium-111-labeled leukocyte scans help
distinguish between an infectious and noninfectious etiology and have 83% to 85% sensitivity and 75% to 94%
specificity. Bone scintigraphy with indium-111-labeled
immunoglobulin has 90% to 93% sensitivity and 85% to
89% specificity. In patients who have had a hip arthroplasty, bone scans may be positive in the presence of
aseptic loosening and may also be positive for up to
2 years postoperatively in a well-fixed prosthesis. However, the combination of results from technetium
Tc 99m and indium-111-labeled leukocyte scans show
88% sensitivity and 95% specificity in the diagnosis of
infection around hip and knee arthroplasties. Positron
emission tomography with F-18 fluorodeoxyglucose is a
new modality that has shown 100% sensitivity and 88%
specificity for chronic musculoskeletal infection. MRI
can detect marrow changes secondary to infection at a

American Academy of Orthopaedic Surgeons

Chapter 20 Infection

Orthopaedic Knowledge Update 8

Table 1 | Antibiotics Commonly Used in Musculoskeletal Infections

Antibiotics

Adult Dosage

Spectrum of Activity

Penicillin G

2,000,000 IU every 4 hours, IV

S pyogenes, S pneumoniae, Anaerobes

Penicillinase-resistant penicillins
Oxacillin, nafcillin

1 to 2 g every 4 hours, IV

S aureus (-lactamase producing)

Other gram-positive cocci

Aminopenicillins
Ampicillin

1 to 2 g every 4 to 6 hours, IV

Non -lactamase producing gram-positive cocci

Gram-negative organisms (not Pseudomonas), Anaerobes

Aminopenicillins and -lactamase inhibitors

Ampicillin-sulbactam
Amoxicillin-clavulanate

1.5 to 3 g every 6 hours, IV

250 to 500 mg every 8 hours, PO

Antipseudomonal penicillins
Piperacillin
Ticarcillin

3 g every 6 hours, IV
3 g every 6 hours, IV

Beta-Lactam Antibiotics (active against cell wall)

Coverage expanded to -lactamase producing organisms

Non -lactamase producing gram-positive cocci

Gram-negative organisms (including Pseudomonas)
Anaerobes

Antipseudomonal penicillins and -lactamase inhibitors

Piperacillin-tazobactam
3.75 g every 6 hours, IV
Ticarcillin-clavulanate
3 g every 6 hours, IV

Coverage expanded to -lactamase producing organisms

Cephalosporins: first generation

Cefazolin

Gram-positive cocci (-lactamase producing)

1 to 2 g every 6 to 8 hours, IV

Cephalosporins: second generation

Cefuroxime

1.5 g every 8 hours, IV

Gram-positive cocci (-lactamase producing)

Gram-negative organisms (not Pseudomonas)

Cephalosporins: third generation

Ceftriaxone

2 g every 24 hours, IV

Gram-negative organisms
Gram-positive cocci (-lactamase producing)

Cephalosporins: fourth generation

Cefepime

2 g every 8 to 12 hours, IV

Gram-positive cocci (-lactamase producing)

Gram-negative organism (including Pseudomonas)

Carbapenems
Imipenem-Cilastatin

500 mg every 6 hours, IV

Gram-positive and gram-negative organisms

Anaerobes

Monobactam
Aztreonam

1 to 2 g every 8 to 12 hours, IV

Aminoglycosides (active against ribosomes)

Gentamicin
Tobramycin
Fluoroquinolones (active against DNA)
Ciprofloxacin
Levofloxacin
Glycopeptides (active against cell wall)
Vancomycin
Lincosamides (active against ribosomes)
Clindamycin
Streptogramins (active against ribosomes)
Quinupristin-dalfopristin
Oxazolidinones (active against ribosomes)
Linezolid

Gram-negative organisms (including Pseudomonas)

3 to 5 mg/kg/day (in a single dose

or divided into three doses), IV
3 to 5 mg/kg/day in a single dose
or divided into three doses), IV

Gram-negative organisms (including Pseudomonas),

Gram-positive cocci

500 to 750 mg every 12 hours, PO

Gram-positive cocci
500 to 750 mg every 24 hours, PO/IV Gram-negative organisms (including Pseudomonas)
1 g every 12 hours, IV

Gram-positive cocci (including MRSA and Enterococci),

Clostridium difficile

900 mg every 8 hours, IV

Gram-positive cocci, anaerobes

7.5 mg/kg every 8 to 12 hours, IV

Gram-positive cocci (including MRSA, VRE)

600 mg every 12 hours, PO/IV

Gram-positive cocci (including MRSA, VRE)

IV = intravenously; PO = by mouth; MRSA = methicillin-resistant enterococci S aureus; VRE = vancomycin-resistant enterococci

American Academy of Orthopaedic Surgeons

219

Orthopaedic Knowledge Update 8

Infection

Table 2 | Most Common Pathogens and Suggested Empiric Antibiotic Therapy in Musculoskeletal Infections
Infection and Clinical Setting

Most Common Pathogens

Empiric Antibiotic Therapy

Infant

S aureus
S pyogenes
S pneumoniae
Gram-negative organisms

Penicillinase-resistant penicillin and aminoglycoside or

ceftriaxone

Child younger than age 3 years

S aureus
S pneumoniae
H influenzae (if nonimmunized)

Ceftriaxone

Older child

S aureus

Cefazolin or Penicillinase-resistant penicillin

Child with sickle cell disease

Salmonella species
S aureus

Ceftriaxone

Adult

S aureus
Suspected MRSA

Penicillinase-resistant penicillin
Vancomycin or clindamycin

Immunocompromised adult or child

Gram-positive cocci
Gram-negative organisms

Penicillinase-resistant penicillin and aminoglycoside

Septic arthritis in sexually active patients

S aureus
N gonorrhoeae

Ceftriaxone

Diskitis

S aureus

Penicillinase-resistant penicillin

Lyme disease

B burgdorferi

Amoxicillin-doxycycline

Clenched-fist bite wounds

E corrodens
P multocida
Anaerobes

Ampicillin-sulbactam or piperacillin-tazobactam

Nail puncture wounds

S aureus
P aeruginosa

Penicillinase-resistant penicillin and aminoglycoside or

piperacillin-tazobactam

Necrotizing fasciitis

Streptococcus group A
beta-hemolytic
Gram-positive cocci, anaerobes
Gram-negative organisms

Penicillin and Clindamycin Aminoglycoside

Osteomyelitis and septic arthritis

very early stage and is a highly sensitive modality for

detecting osteomyelitis with a sensitivity approaching
100%. The increased water content secondary to edema
and hyperemia results in a decreased marrow signal in
T1-weighted images, and an increased signal in T2weighted images, respectively. In the diagnosis of septic
arthritis, studies have shown that MRI has 97% sensitivity and 92% specificity. It allows for the detection of increased intra-articular fluid, and for the evaluation of
potential spread of infection into adjacent bone or soft
tissue.
Identification of organisms following Gram stain of
specimens from the involved area occurs in approximately one third of cases. Despite its low yield, the
Gram stain examination can be useful because it is
highly specific and may help the physician to determine
the most appropriate antibiotic for initial therapy. For
infections occurring after hip arthroplasty, evaluation of
inflamed tissue specimens with Gram staining has a

220

19% sensitivity and a 98% specificity; frozen section histologic examination for the presence of more than five
polymorphonuclear cells per high-power field has an
80% sensitivity and a 94% specificity.
Cultures remain the gold standard for establishing
the diagnosis of infection. However, the prior administration of antibiotics, inadequate specimen sampling, or
improper handling of specimens may preclude the
growth of pathogens.
Molecular diagnostic techniques are available and
may improve diagnostic efficacy in the future. A recent
study indicates that polymerase chain reaction (PCR)
can amplify and detect bacterial DNA, potentially leading to an earlier diagnosis compared with cultures. Another advantage of this technique is that the results are
not affected by the concurrent use of antibiotics because
PCR does not depend on in vitro growth of the organism. However, concerns about false-positive results secondary to contamination still exist.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Osteomyelitis in Adults
Osteomyelitis (bone inflammation secondary to the
presence of microbial pathogens) can be classified based
on the pathogenesis (hematogenous, traumatic, contiguous spread), the duration of the process (acute, subacute, chronic), and the age of the patient (adult versus
pediatric).
Osteomyelitis in adults usually results from trauma
(open fractures), surgical procedures (postoperative infections after open reduction and internal fixation of
fractures), or contiguous spread from adjacent infections. The tibia is the most common site of adult osteomyelitis. Hematogenous osteomyelitis is uncommon in
adults but may occur in intravenous drug users. The
most common pathogen is S aureus; however, a variety
of organisms may be involved depending on the clinical
setting. Pseudomonas aeruginosa or other gramnegative organisms may be responsible for the infection
in intravenous drug users, and less virulent microbes or
fungi in immunocompromised patients. Adult osteomyelitis can be staged with the Cierny-Mader classification
system, which evaluates the anatomic type of bone involvement (medullary, superficial, localized, diffuse) and
the physiologic class of the host (A: normal; B: systemic,
local, or combined compromise of the host; C: morbidity
of treatment worse than that of disease).
A diagnosis is based on clinical findings (pain,
erythema, draining sinuses, systemic symptoms), laboratory tests (elevated CRP, ESR), imaging modalities (radiographs, MRI, scintigraphy), Gram stain, and cultures.
It should be noted that osteomyelitis may be clinically
silent, so a high index of suspicion is warranted in situations such as atrophic nonunions after an open fracture
or internal fixation of a closed fracture. Chronic draining sinuses may be complicated by malignant transformation and development of squamous cell carcinoma
(Marjolins ulcer) in approximately 1% of patients.
Osteomyelitis can be a limb-threatening condition.
Treatment may be prolonged and financially and socially demanding for the patient; therefore, amputation
may be a reasonable option in some complex cases.
Treatment of osteomyelitis with a limb-salvage protocol
consists of dbridement, systemic and local antibiotic
treatment, skeletal stabilization, soft-tissue coverage,
and treatment of bone defects and nonunited fractures.
These principles can be incorporated in a staged protocol.
The first stage of a limb salvage protocol includes
radical dbridement of all nonviable tissues and skeletal
stabilization. Dbridement should proceed until bleeding, definitively viable tissue is present at the resection
margins. Inadequate dbridement leads to recurrence of
infection despite antibiotic therapy, because pathogens
form biofilms on nonviable tissue and escape antibiotic
therapy and host defense mechanisms. Specimens of pu-

American Academy of Orthopaedic Surgeons

Chapter 20 Infection
rulent fluid, soft tissue, and bone from the affected area
require aerobic, anaerobic, mycobacterial, and fungal
cultures. The latter two cultures are especially important
in immunocompromised patients or those with chronic
osteomyelitis. The wound is copiously irrigated with saline; antibiotics may be added to the irrigation fluid.
The dead space that results from dbridement is
filled with physician-made polymethylmethacrylate
beads impregnated with antibiotics such as tobramycin,
vancomycin, cefepime, or other microbial-specific antibiotics available in powder form. Elution of antibiotics
depends on the surface area and characteristics of the
antibiotic delivery vehicle, the type and concentration of
the antibiotic(s) used, the presence of fluid, and the rate
of fluid turnover. Local antibiotic delivery is a useful option resulting in high local concentration and low systemic side effects and can supplement systemic therapy,
provided the pathogen is susceptible to the eluted antibiotic. Nonabsorbable antibiotic delivery vehicles may
require revision for removal. Intravenous administration
of antibiotics for 4 to 6 weeks is the recommended therapy and can be accomplished on an outpatient basis.
Oral administration of linezolid or quinolones can
achieve adequate blood and tissue concentration levels
and may be a useful alternative to intravenous therapy.
Antibiotic administration is a key part of the treatment
but will not be effective without adequate dbridement.
Development of resistant organisms may occur.
Skeletal stabilization in the presence of fractures
that have not yet united is necessary for infection control. The optimal method of stabilization depends on the
involved bone and the condition of its soft-tissue envelope. The need for careful consideration of the soft tissue when planning fixation constructs cannot be overemphasized.
The second stage of the protocol consists of wound
management. If the soft-tissue envelope is adequate, delayed closure can be performed. In the presence of compromised soft tissues, coverage should be achieved by
local or free muscle flaps, depending on the location and
extent of the soft-tissue defect. Muscle flaps eliminate
dead space, provide soft-tissue coverage, prevent contamination with new pathogens, improve the local vascularity and biologic environment, assist the host defense mechanisms, enhance antibiotic delivery, and
promote the healing process. Flap coverage is usually
done 3 to 7 days after the initial dbridement. Local
muscle flaps used for coverage of the tibia include the
gastrocnemius for proximal third defects and the soleus
for middle third defects. It is important to consider the
biologic status of the local muscle to be transferred to
avoid using muscle that has itself been compromised by
the injury or by ischemic changes. In patients with distal
third tibial defects, a free muscle flap is necessary.
The third stage of the treatment regimen consists of
management of existing bone defects, usually by autoge-

221

Infection
nous bone grafting, which is performed when the softtissue envelope has healed (usually 6 to 8 weeks after
the muscle transfer). At this stage, viability of the flap
and control of infection have been determined. For anterior defects and most nonunions of the tibia, the muscle flap is elevated and the graft is placed at the site of
the nonunion or defect. Posterolateral tibia bone grafting is an alternative if there is no anterior sequestrum or
need for a soft-tissue procedure anteriorly. The usefulness of bone graft substitutes as void fillers in a defect
of infectious etiology is still being evaluated. Bone defects greater than 6 cm require specialized reconstructive procedures such as vascularized bone grafts or distraction osteogenesis. Limb salvage can result in a
satisfactory functional outcome. An outcomes study of
patients with chronic osteomyelitis of the tibia done in
2000 showed that at a mean follow-up of 5 years, 39 of
46 patients (85%) were able to ambulate independently
without pain. Patient age (advanced age) and a history
of smoking adversely affected the outcome.

Septic Arthritis in Adults

The portals of pathogen entry in patients with septic arthritis include hematogenous inoculation, spread from
an adjacent area of infection, or inoculation by a penetrating or surgical wound. Susceptibility to septic arthritis is increased in immunocompromised hosts, patients
with preexisting joint pathology, and patients with frequent bacteremic episodes. Microorganisms trigger an
inflammatory response that recruits polymorphonuclear
cells. Enzymes released by bacteria and polymorphonuclear and synovial cells promote the degradation of glycosaminoglycans and the subsequent loss of collagen,
which results in gross damage to the articular cartilage.
Neisseria gonorrhoeae is the most common pathogen
causing septic arthritis in otherwise healthy adults;
women are more frequently affected than men. Clinical
symptoms of disseminated disease include migratory
polyarthritis, rash, and tenosynovitis of the dorsal aspect
of the wrist and hand. Septic involvement of a single
joint also may occur with the knee joint most commonly
affected. S aureus is the second most common pathogen
causing adult septic arthritis. Immunocompromised
hosts may have infections with gram-negative organisms, or unusual pathogens, such as mycobacteria or
fungi. Intravenous drug users, in addition to being susceptible to S aureus, are prone to infections with
Pseudomonas aeruginosa or Serratia marcescens.
Septic arthritis is more common in joints of the
lower extremity, with the knee most frequently involved. The affected joint is painful, swollen, has limited
range of motion, and is often erythematous. The CRP
and ESR are generally elevated, and the WBC count is
elevated in approximately 50% of patients. Radiographs
help assess bone involvement and ultrasound is useful in

222

Orthopaedic Knowledge Update 8

detecting a joint effusion. MRI, in addition to detecting
a joint effusion, delineates any bone and soft-tissue involvement.
Aspiration of the involved joint always should be
done to establish the diagnosis and to identify the
pathogen. The joint aspirate should be sent for Gram
stain, cultures, sensitivity testing, WBC count with differential, and crystal analysis. A WBC count greater than
50,000 cells/mm3 and a differential with polymorphonuclear leukocytes exceeding 75% indicate septic arthritis;
however, lower values do not preclude the diagnosis.
Synovial fluid cultures are positive in approximately
90% of patients with nongonococcal arthritis, compared
with 25% of those with gonococcal arthritis. Blood cultures are positive in approximately 50% of patients with
nongonococcal arthritis, whereas the yield is only 10%
in those with gonococcal arthritis.
Surgical decompression by arthrotomy of a septic
joint relieves pressure and evacuates enzymes, inflammatory mediators, and bacteria from the joint to minimize cartilage damage. Arthroscopic irrigation may be
an alternative. Repeated joint aspirations may be satisfactory in some knee infections but should not be used
for infections of the hip or small joints. Synovial biopsy
for culture and histology is recommended for joints undergoing arthrotomy. In chronic or recurrent infections,
a complete synovectomy is warranted. Septic arthritis
after anterior cruciate ligament reconstruction occurs in
approximately 0.2% to 0.5% of patients. According to
one recent study, the rate of infection after anterior cruciate ligament reconstruction was 0.14% in 3,500 consecutive procedures. In early (occurring from up to 4 to
6 weeks) postoperative infections, graft retention may
be possible with prompt irrigation, dbridement, and antibiotic therapy.
Empiric systemic antibiotic therapy should be
started immediately after cultures have been collected
and should cover the most likely pathogens based on
the clinical setting. In otherwise healthy adults, both
N gonorrhoeae and S aureus should be covered, usually
with a third generation cephalosporin. Immunocompromised hosts and intravenous drug users should receive
coverage for both S aureus and P aeruginosa. The duration of antibiotic therapy is usually 4 weeks for nongonococcal arthritis and 1 week in gonococcal arthritis
that responds well to therapy.
Lyme disease is a multisystem spirochetal disorder
caused by Borrelia burgdorferi, which is transmitted by
the bite of an infected tick. It usually occurs in the
Northeast, Midwest, and Northwest regions of the
United States. The disease occurs in three clinical stages.
In the first stage (early localized disease) a pathognomonic skin lesion (erythema migrans) emerges and progressively increases in size. The second stage (early disseminated disease) is characterized by neurologic and
cardiac manifestations. In the third stage (late disease)

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

musculoskeletal symptoms develop in 80% of untreated
patients and include arthralgia, intermittent episodes of
arthritis, and subsequently chronic monoarthritis, usually
of the knee. Acute monoarthritis may clinically resemble septic arthritis. The synovial fluid in acute noninfectious monoarthritis usually has a WBC count of 10,000
to 25,000 cells/mm3 with an increased number of polymorphonuclear cells; a higher count may be present, creating a resemblance to a more typical bacterial infection. Chronic Lyme arthritis is rarely destructive and has
been associated with HLA-DRB1*040, thereby suggesting an autoimmune mechanism. Enzyme-linked immunosorbent assay establishes the diagnosis. Oral antibiotic
therapy with amoxicillin or doxycycline for approximately 4 weeks is recommended. Intravenous therapy
with ceftriaxone may be used for patients with recurrent
episodes of arthritis and for those with neurologic involvement. A single 200-mg dose of doxycycline given
within 72 hours after a tick bite that occurs in an endemic area can prevent the development of Lyme disease.
Tuberculosis should be considered in the differential
diagnosis of arthritis and chronic osteomyelitis, especially in immunocompromised patients and in patients
from areas endemic for tuberculosis outside of the
United States. Tubercular arthritis usually affects the hip
and the knee, and is characterized by an insidious onset,
subtle inflammatory signs, and disproportionately extensive bone involvement relative to the symptoms present.
Radiographic changes include subchondral bone erosions and joint space narrowing. Synovial biopsy, acidfast stains, and cultures help to establish the diagnosis.
Growth of the organism in solid medium culture requires up to 8 weeks and sensitivity testing should always be performed. Therapy for tuberculosis should include a multidrug regimen with isoniazid, rifampin,
pyrazinamide, and ethambutol. When sensitivity results
are available, ethambutol can be discontinued if the organism is fully susceptible. The duration of therapy
ranges from 6 to 9 months.

Pediatric Musculoskeletal Infections

Osteomyelitis in Children
Osteomyelitis in children is usually hematogenous in origin. Osteomyelitis is more common in males, the lower
extremity, and increases in incidence during the warmer
months. The most common organism causing pediatric
osteomyelitis is S aureus. Neonates may also have infections with group B Streptococci or gram-negative pathogens. Haemophilus influenzae type B was previously a
common musculoskeletal pathogen in children between
1 and 4 years of age, but vaccination against this organism has almost eliminated its role in pediatric osteomyelitis. In children younger than 3 years of age, the most
common pathogens include S aureus, Streptococcus

American Academy of Orthopaedic Surgeons

Chapter 20 Infection
pneumoniae, and Streptococcus pyogenes. Children with
sickle cell disease are more prone to infections from the
Salmonella species compared with the normal pediatric
population. Similarly, immunocompromised children are
more susceptible to osteomyelitis from less virulent microbes or fungi than patients with uncompromised immune systems.
Pediatric acute hematogenous osteomyelitis has a
predilection for the metaphysis of long bones. The metaphysis is perfused by end-arteries that enter large
venous sinusoids. The sluggish circulation and defective
phagocytosis in the capillary loops allow bacteria to inoculate in the metaphyseal area by the physeal plate.
The infection subsequently may spread through Volkmanns canals of the metaphyseal bone to the subperiosteal region where a resultant abscess may elevate the
periosteum and devitalize cortical bone. Osteomyelitis
may spread to the epiphysis, especially in infants because of their distinct vascular pattern with metaphyseal
vessels traversing into the epiphyseal area. Septic involvement of the adjacent joint occurs in 33% of patients with metaphyseal osteomyelitis. The most commonly affected joint is the knee. Joint involvement is
facilitated if the metaphysis is intra-articular (proximal
femur, proximal humerus, proximal radius, distal lateral
tibia). Therefore, careful evaluation of the adjacent joint
should be an important part of the evaluation for any
child with osteomyelitis. The infectious process may also
spread to the surrounding soft tissues and to the medullary canal. Acute hematogenous osteomyelitis in older
children usually involves a single site. However, in neonates, polyostotic involvement occurs in 30% of patients, and may be identified with a bone scan.
Clinical findings of acute hematogenous pediatric osteomyelitis include pain, refusal to bear weight, inability
to use the affected extremity, and fever. Previous trauma
to the affected site is present in 30% to 50% of patients.
The patient history should include the presence of medical conditions (useful for assessing the likely pathogen)
and any recent antibiotic administration (likely to alter
the clinical presentation). Diagnosis can be challenging
in neonates because the clinical picture is often subtle.
The proximal femur and hip joint are most commonly
involved and a high index of suspicion is needed for
early diagnosis; pseudoparalysis in a neonate should be
carefully evaluated.
Laboratory tests include a WBC count and differential, ESR, and CRP. Bone aspiration and blood cultures
are essential. Bone aspiration or intraoperative cultures
identify the pathogen in 48% to 85% of patients,
whereas blood cultures are positive in 30% to 60%. The
combination of data from bone aspiration and blood
cultures is important to achieve accurate diagnosis. Radiographs show soft-tissue swelling but are of limited
value in identifying osseous changes, which typically do
not occur until 7 to 14 days later. Bone scanning is use-

223

Infection
ful when the location of the pathology is uncertain or
multiple locations are suspected, such as in the neonate.
In children with sickle cell anemia, osteomyelitis can be
differentiated from bone infarction with acute bone
pain by a combination of sequential bone marrow and
bone scintigraphy.
Treatment of pediatric acute hematogenous osteomyelitis consists of prompt systemic antibiotic administration and close monitoring of the patient. Surgical
treatment is based on the presence of an abscess. In approximately 50% of patients, surgery is not necessary
because early antibiotic therapy contains the infectious
process before an abscess can form. Surgical intervention is warranted if an abscess is present (diagnosed by
aspiration, or MRI), if the adjacent joint is septically involved, or if changes are present on plain radiographs,
indicating late presentation.
Antibiotics should be given immediately after bone
aspiration and blood cultures have been obtained. Empirical antibiotic administration while the culture results
are pending should target the most likely pathogens
based on the age of the child and should always cover
S aureus, which is the most frequent pathogen in all age
groups. Systemic antibiotics can be substituted with oral
therapy provided the patient is afebrile, shows considerable clinical improvement, and in whom CRP levels
have normalized or considerably decreased. Antibiotics
are usually given for 4 to 6 weeks.
In some patients subacute osteomyelitis may insidiously develop over a period of months until the patient
develops symptoms, such a pain and limp. The radiographic appearance of subacute osteomyelitis may resemble a tumor, and biopsy is frequently necessary to
establish the diagnosis.

Chronic Recurrent Multifocal Osteomyelitis

Chronic recurrent multifocal osteomyelitis is characterized by bilateral, usually symmetric bone involvement.
The onset is insidious and inflammatory symptoms may
wax and wane over time. The clavicle and the metaphyses of long bones usually are affected; lytic and sclerotic
changes are seen on radiographs. A pustular rash on the
palms and soles may be present. Cultures are negative
and the diagnosis is one of exclusion after microbial osteomyelitis or bone tumors are ruled out. No effective
treatment exists; however, anti-inflammatory medication
may relieve symptoms. Antibiotics are not needed if cultures are negative. Although symptoms tend to recur
over a 2-year period, the long-term prognosis appears to
be good.

Septic Arthritis
Septic arthritis is usually hematogenous in origin, is
caused by pathogens similar to those involved in hematogenous osteomyelitis, and commonly occurs in the

224

Orthopaedic Knowledge Update 8

knee and hip joints. Septic arthritis may originate from
contiguous spread of adjacent infection or osteomyelitis.
In neonates, osteomyelitis in the metaphysis can spread
to the epiphysis via blood vessels traversing the physis,
and then to the joint space via the thin bone cortex.
Pathogens also may be directly inoculated into the joint
by a penetrating injury.
Clinical symptoms include fever, pain, inability to
use the upper extremity (pseudoparalysis), or refusal to
walk when the lower extremity is affected. The range of
motion of the involved joint is markedly decreased and
painful. In septic arthritis of the hip, the joint is positioned in external rotation, abduction, and mild flexion
to increase joint volume and release tension on the capsule.
An elevated CRP and ESR are present in at least
90% of patients with septic arthritis, whereas the WBC
count is elevated in approximately 50%. MRI shows
fluid collection in the joint and also is useful in elevating
any adjacent bone or soft-tissue involvement. Ultrasonography can detect a joint effusion. Scintigraphy is
useful for the evaluation of multiple joints when the
pathologic process cannot be localized to a single joint,
or when the joint involved is not clinically apparent.
Joint aspiration should always be performed. Fluoroscopic or ultrasonographic guidance is useful for hip aspiration. The joint aspirate should be sent for a Gram
stain, cultures, sensitivity testing, and WBC count and
differential. A WBC count greater than 50,000/mm3 is
found in approximately 50% of patients.
Transient synovitis of the hip in a young child (age
3 to 8 years) may present as an acutely irritable hip with
an effusion; the clinical presentation may be indistinguishable from septic arthritis. The presence of a fever,
the inability to bear weight, an ESR greater than
40 mm/h, and a peripheral WBC count greater than
12,000 cells/mm3 are independent variables that can
help distinguish the two conditions. According to a 1999
study, the probability of a correct diagnosis of septic arthritis was found to be 99.6% for children with all four
factors, 93.1% for those with three factors, 40% for
those with two factors, and 3% for those with only one
factor.
Inflammatory arthritides, such as juvenile rheumatoid arthritis, poststreptococcal arthritis, and rheumatic
fever should be included in the differential diagnosis of
pediatric arthritis with joint effusion. Juvenile rheumatoid arthritis may present as a single acutely swollen and
painful joint. However, in juvenile rheumatoid arthritis
the onset is usually gradual, the patient may be able to
walk, systemic symptoms are milder, the joint fluid
WBC count is usually less than 50,000 cells/mm3, and
the Gram stain and cultures are negative.
Surgical decompression of a septic joint may be accomplished by open arthrotomy or arthroscopy. Repeated joint aspiration may be an alternative in some

American Academy of Orthopaedic Surgeons

Chapter 20 Infection

Orthopaedic Knowledge Update 8

patients in whom the affected joints are easily accessible; however, failure to respond to nonsurgical treatment within 24 to 48 hours warrants surgical intervention. In patients with a septic hip, prompt surgical
arthrotomy, irrigation, and dbridement are mandatory
because delayed or incomplete decompression impairs
perfusion of the femoral head and may cause further
morbidity, such as osteonecrosis, hip dislocation, and osteomyelitis. The anterior approach is preferred to avoid
any damage to the vascularity of the femoral head.
Antibiotic therapy should be started immediately after cultures have been collected and should target the
most probable pathogens based on the childs age and
the clinical setting. In sexually active adolescents N gonorrhoeae is a potential pathogen. Systemic antibiotics
can be converted to oral therapy as the patient clinically
improves. The total duration of therapy is usually 3
weeks. Contiguous osteomyelitis should be suspected if
the clinical course is not dramatically improved after
drainage of the septic joint and the CRP remains elevated.

Diskitis
Diskitis is a hematogenous infection involving the disk
and vertebral body. S aureus is the most common pathogen. Clinical symptoms may be insidious and may include back pain, abdominal pain, or inability to walk.
Tenderness is present over the involved vertebrae. Systemic symptoms and signs of infection may be absent;
ESR and CRP are usually elevated but the WBC count
may be normal. Radiographs are initially noncontributory but subsequently show disk space narrowing. MRI
reveals early changes in the disk and the adjacent vertebral bodies. Bone scintigraphy may be helpful when the
location of the pathologic process is uncertain. Blood
cultures should be obtained. Needle or open biopsy and
cultures may be helpful if the diagnosis is uncertain;
however, the yield (as low as 27%), the potential complications of biopsy, and the preponderance of S aureus
infections do not justify routine biopsy. Although the infectious nature of diskitis has been questioned based on
the low yield of biopsy cultures and the resolution of
symptoms with rest alone, treatment should include rest
and antibiotic therapy with an antistaphylococcal agent.

Sacroiliac Joint Infection

Sacroiliac joint infection may present with a variety of
symptoms including back pain, abdominal pain, pain in
the gluteal area, limp, fever, malaise, and tenderness
over the involved area. The FABER test (flexion, abduction, external rotation) and compression of the pelvis
elicit pain. MRI is useful in determining the location of
the infectious process and the extent of bone and softtissue involvement. S aureus is the most common pathogen and should be covered by initial antibiotic therapy.

American Academy of Orthopaedic Surgeons

Surgical dbridement is seldom necessary, except in the

presence of an abscess.

Wound Infections
Wound infections involve a wide spectrum of anatomic
locations and etiologic mechanisms. An awareness of
the potential for infection in particular wounds, such as
clenched-fist injuries and nail puncture wounds, is
needed for proper treatment. Identification and prompt
treatment of severe life-threatening infections, such as
necrotizing fasciitis, are essential to reduce mortality
and morbidity.
Clenched-fist injuries represent human bite wounds
with a high potential for damage of underlying tissues
and infection. Violation of the joint capsule of the
metacarpophalangeal joints occurs in 68% of patients.
Eikenella corrodens and Pasteurella multocida are gramnegative, facultative anaerobes, often found in human
mouth flora. All patients with clenched-fist lacerations
or puncture wounds over joints should be treated with
surgical dbridement and exploration of the deep structures, including the joint and the extensor tendon, at the
time of initial medical care. The wound should not be
closed primarily. Antibiotic therapy consisting of widespectrum antibiotics active against anaerobes should be
administered.
Nail puncture wounds to the foot may lead to bone
or joint penetration and predispose the patient to osteomyelitis and/or septic arthritis. The area overlying the
metatarsal neck region and extending distally to the
toes carries the highest risk for infection, because the
metatarsal heads are a major weight-bearing area with a
limited amount of overlying tissue. Approximately 97%
of patients requiring hospitalization for septic complications had sustained a puncture wound through this region. P aeruginosa is the most common pathogen involved, especially if the nail enters through an athletic
shoe; the exact reason for this is unknown. If increasing
tenderness is present over a puncture wound area, admission to the hospital is needed for antibiotic therapy
and surgical dbridement.
Necrotizing fasciitis involves the fascia and overlying
tissues but spares the muscles. Although group A streptococcus is often involved, the disease may be polymicrobial. Immunocompromised hosts and children with
varicella infections are at increased risk for developing
the disease. Necrotizing fasciitis initially may resemble
cellulitis, but the edema and induration extend beyond
the area of erythema. The infection rapidly spreads
along fascial planes, results in septic shock, and does not
respond to antibiotic therapy alone. Emergent surgical
dbridement is warranted. The disease carries a high
mortality rate ranging from 6% to 76%, especially if
surgical treatment is delayed.

225

Infection

Infection in the Setting of Internal Fixation

In the presence of internal fixation, microorganisms
grow in a biofilm that adheres to the implant surface
and protects them from host defense mechanisms and
antibiotics. The choice of retaining or removing infected
implants depends on several factors including time since
fracture fixation, stability provided by the hardware, and
bone healing status. In the early postoperative period,
before the fracture is united, internal fixation is necessary to maintain reduction of the fracture. Treatment
should include irrigation, dbridement, intraoperative
cultures, brushing of exposed implants, and antibiotic
therapy for 6 weeks. If the fracture is healed, internal
fixation is removed. Loose hardware that does not provide stability should be removed regardless of the time
frame. If the fracture is not healed, it should be stabilized with another device, preferably an external fixator
for the tibia and an intramedullary rod for the femur. In
intra-articular united fractures, internal fixation should
not be removed or replaced unless it is loose and not
providing stability.

Infection in the Setting of Arthroplasty

Infection occurs in approximately 1% of patients who
have undergone total joint arthroplasty, with S epidermidis and S aureus the most common infecting organisms.
The pathogenesis of infection involves either direct inoculation of the organisms (during the procedure and in
the early postoperative period) or hematogenous spread
(at any point during the life of the implant).
The clinical indicators of periprosthetic infection
may be straightforward and consist of pain, decreased
range of motion, drainage, and systemic symptoms.
However, the clinical picture is often subtle with pain as
the only symptom. Laboratory tests (ESR and CRP),
bone scintigraphy, and joint aspiration are useful. The
combination of normal ESR and CRP levels reliably
predicts the absence of infection. Aspiration has 86%
sensitivity and 94% specificity in the absence of existing
antibiotic therapy and should be used when the ESR or
the CRP level is elevated or when a clinical suspicion of
infection remains. In addition, intraoperative tests, such
as cultures (94% sensitivity and 97% specificity), and
frozen sections evaluating the number of polymorphonuclear leukocytes per high-power field (80% sensitivity
and 94% specificity) can be helpful in equivocal cases.
The Gram stain is unreliable (19% sensitivity and 98%
specificity) for determining the presence of periprosthetic infection.
Treatment options, in addition to administration of
systemic antibiotics for 4 to 6 weeks, include irrigation
and dbridement with retention of components, or exchange arthroplasty (one-stage versus two-stage). Salvage procedures include permanent resection or arthrodesis. Chronic antibiotic suppression or amputation may

226

Orthopaedic Knowledge Update 8

be considered in selected patients. Treatment choice depends on duration of infection, virulence of the organism, implant stability, the patients immune and medical
status, and the condition of the local soft-tissue envelope. Staging systems may be useful to evaluate these
factors, compare results among treatment centers, and to
establish guidelines for care.
Irrigation, dbridement, exchange of the polyethylene liner, and retention of the components are options
for patients who develop infections in the early postoperative period (less than 1 month), and in patients with
acute hematogenous infections (symptoms of duration
less than 2 to 4 weeks in a previously well-functioning
patient) with well-fixed components. Component retention is associated with lower rates of infection control
(50% to 71%), possibly because of concurrent hematogenous seeding and osteomyelitis of adjacent bone, biofilm formation, and limited dbridement.
Exchange arthroplasty is the preferred option for
patients with chronic late-developing infections (developing more than 1 month after the procedure with indolent course and duration greater than 2 to 4 weeks), and
for those with acute hematogenous infections with implant loosening (provided that the patient is in good
condition and the bone stock is adequate or reconstructible). Flap coverage may be warranted for treatment of
an infected arthroplasty with a poor soft-tissue envelope. Following irrigation and dbridement, the exchange arthroplasty can be performed at the same time
(one stage), or preferably as a subsequent procedure
with an interval of 6 weeks or longer (two stage). Local
antibiotic delivery during the reimplantation interval
can be accomplished by using antibiotic-impregnated
spacers. Two-stage exchange arthroplasty has a higher
rate of eradication of infection (88% to 95%) compared
with the one-stage technique (70% to 85%).
Reimplantation of a prosthesis may not be possible
because of a severely compromised host, deficient bone
stock, poor condition of the soft tissues, the presence of
infection with resistant organisms, or repeated failed exchanged arthroplasties. In this situation, resection arthroplasty is an option. Arthrodesis may be a preferable
choice for use in the lower extremity of active patients,
because it offers improved function compared with resection arthroplasty. Chronic antibiotic suppression is an
option for severely compromised patients, or for patients with limited life expectancy, provided the infecting organism is sensitive to oral antibiotics. Amputation
of the extremity may be needed for persistent uncontrolled infection in a severely compromised patient.
Episodes of bacteremia may cause hematogenous
seeding of implants; therefore, patients should be in
good dental health before a total joint arthroplasty is
done and should maintain good oral hygiene after the
procedure. Single-dose antibiotic prophylaxis with cephalosporin or amoxicillin (clindamycin in patients allergic

American Academy of Orthopaedic Surgeons

Chapter 20 Infection

Orthopaedic Knowledge Update 8

to penicillin) is recommended when a dental procedure
with high risk for bacteremia, such as dental extraction,
is performed on a patient with increased risk for hematogenous infection (such as immunocompromised patients, or those with comorbidities) within 2 years from
the date of arthroplasty.

Future Directions
The goal of future research is improved prevention, diagnosis, and treatment of musculoskeletal infections. Immunization against pathogens and development of
implant materials resistant to infection may aid prevention. Refinement of diagnostic methods, such as PCR,
will lead to early and accurate diagnoses. Development
of new antibiotics, biodegradable materials for local antibiotic delivery, and pharmacologic improvement of
host defenses (such as by granulocyte stimulating hormone), may make treatment more effective. Genetic interventions may help fight infection by the alteration of
resistant organisms and, in combination with tissue engineering, may find application in the restoration of damaged host tissue.

Annotated Bibliography
Pathogenesis of Musculoskeletal Infections
Elasri MO, Thomas JR, Skinner RA, et al: Staphylococcus aureus collagen adhesin contributes to the pathogenesis of osteomyelitis. Bone 2002;30:275-280.
The authors created an S aureus strain that was mutant for
collagen-binding adhesin. The mutant strain, which was able to
bind fibronectin but not collagen, resulted in hematogenous
osteomyelitis in 5% of injected mice, compared with 70%
when the nonmutant strain was used.

Antibiotic Therapy and Antibiotic Resistance

Birmingham MC, Rayner CR, Meagher AK, Flavin SM,
Batts DH, Schentag JJ: Linezolid for the treatment of
multidrug-resistant, gram-positive infections: Experience
from a compassionate-use program. Clin Infect Dis
2003;36:159-168.

Septic Arthritis in Adults

Indelli PF, Dillingham M, Fanton G, Schurman DJ: Septic arthritis in postoperative anterior cruciate ligament
reconstruction. Clin Orthop 2002;398:182-188.
The rate of infection after anterior cruciate ligament reconstructions in 3,500 consecutive procedures was found to be
0.14%. All patients had arthroscopic dbridement followed by
6 weeks of intravenous antibiotics. In four patients the grafts
were retained and full range of motion was achieved.

Pediatric Musculoskeletal Infections

Khachatourians AG, Patzakis MJ, Roidis N, Holtom PD:
Laboratory monitoring in pediatric acute osteomyelitis
and septic arthritis. Clin Orthop 2003;409:186-194.
A review of 50 children with osteomyelitis, septic arthritis,
or both showed that surgical intervention resulted in postoperatively increased levels of ESR and CRP and an increase in
the amount of time to peak and normalization of ESR and
CRP values.

Skaggs DL, Kim SK, Greene NW, Harris D, Miller JH:

Differentiation between bone infarction and acute osteomyelitis in children with sickle-cell disease with use
of sequential radionuclide bone-marrow and bone scans.
J Bone Joint Surg Am 2001;83:1810-1813.
The combination of sequential bone marrow scans and
bone scintigraphy was used in 79 episodes of acute bone pain
in children with sickle cell anemia and reliably differentiated
osteomyelitis from a bone infarction.

Wound Infections
Wong CH, Chang HC, Pasupathy S, Khin LW, Tan JL,
Low CO: Necrotizing fasciitis: Clinical presentation, microbiology, and determinants of mortality. J Bone Joint
Surg Am 2003;85:1454-1460.
The authors reviewed 89 patients with necrotizing fasciitis.
Only 13 patients had an admitting diagnosis of necrotizing fasciitis. A delay in surgery of more than 24 hours resulted in increased mortality. A high index of suspicion and prompt treatment are essential for this condition.

Infection in the Setting of Arthroplasty

The authors evaluated the role of linezolid in 796 patients

with multidrug-resistant, gram-positive infections, mostly from
vancomycin-resistant enterococci and methicillin-resistant staphylococci. Linezolid use resulted in high rates of clinical and
microbiological cure with good overall tolerance.

American Academy of Orthopaedic Surgeons Website.

Advisory Statement: Antibiotic Prophylaxis for Dental
Patients with Total Joint Replacements: American Dental Association and American Academy of Orthopaedic
Surgeons. Available at: http://www.aaos.org/wordhtml/
papers/advistmt/1014.htm. Accessed February, 2004.

Diagnostic Modalities

This advisory statement updates recommendations for antibiotic prophylaxis in patients with total joint replacements
who are undergoing dental procedures.

Tarkin IS, Henry TJ, Fey PI, Iwen PC, Hinrichs SH,
Garvin KL: PCR rapidly detects methicillin-resistant
staphylococci periprosthetic infection. Clin Orthop 2003;
414:89-94.
PCR successfully predicted the presence of methicillinresistant staphylococci infections in a septic arthritis model
and gave results concordant with culture results in 34 of 35
samples obtained during revision arthroplasty.

American Academy of Orthopaedic Surgeons

Kilgus DJ, Howe DJ, Strang A: Results of periprosthetic

hip and knee infections caused by resistant bacteria.
Clin Orthop 2002;404:116-124.
The authors reviewed 70 periprosthetic infections and reported successful retention of the prosthesis or reimplantation

227

Infection
in 48% of hips and 18% of knees infected with methicillinresistant S aureus or S epidermidis, compared with 81% of
hips and 89% of knees infected with antibiotic-sensitive organisms.

Lehman CR, Ries MD, Paiement GD, Davidson AB: Infection after total joint arthroplasty in patients with human immunodeficiency virus or intravenous drug use.
J Arthroplasty 2001;16:330-335.
Patients infected with the human immunodeficiency virus
and/or intravenous drug users are susceptible to deep
periprosthetic infection. The infection rate was 14% (4 of 28)
in human immunodeficiency virus-positive patients undergoing total joint arthroplasty and 25% (2 of 8) in intravenous
drug users.

Classic Bibliography
Cierny G III, Mader JT, Penninck JJ: A clinical staging
system for adult osteomyelitis. Contemp Orthop 1985;10:
17.
Costerton JW, Stewart PS, Greenberg EP: Bacterial biofilms: A common cause of persistent infections. Science
1999;284:1318-1322.
Kocher MS, Zurakowski D, Kasser JR: Differentiating
between septic arthritis and transient synovitis of the
hip in children: An evidence-based clinical prediction algorithm. J Bone Joint Surg Am 1999;81:1662-1670.
Mazur JM, Ross G, Cummings J, Hahn GA Jr, McCluskey WP: Usefulness of magnetic resonance imaging for
te diagnosis of acute musculoskeletal infections in children. J Pediatr Orthop 1995;15:144-147.

Orthopaedic Knowledge Update 8

Perlman MH, Patzakis MJ, Kumar PJ, Holtom P: The incidence of joint involvement with adjacent osteomyelitis
in pediatric patients. J Pediatr Orthop 2000;20:40-43.
Scott RJ, Christofersen MR, Robertson WW Jr, Davidson RS, Rankin L, Drummond DS: Acute osteomyelitis
in children: A review of 116 cases. J Pediatr Orthop
1990;10:649-652.
Siegel HJ, Patzakis MJ, Holtom PD, Sherman R, Shepherd L: Limb salvage for chronic tibial osteomyelitis: An
outcomes study. J Trauma 2000;48:484-489.
Spangehl MJ, Masri BA, OConnell JX, Duncan CP:
Prospective analysis of preoperative and intraoperative
investigations for the diagnosis of infection at the sites
of two hundred and two revision total hip arthroplasties.
J Bone Joint Surg Am 1999;81:672-683.
Tsukayama DT, Estrada R, Gustilo RB: Infection after
total hip arthroplasty: A study of the treatment of one
hundred and six infections. J Bone Joint Surg Am 1996;
78:512-523.
Unkila-Kallio L, Kallio MJ, Eskola J, Peltola H: Serum
C-reactive protein, erythrocyte sedimentation rate, and
white blood cell count in acute hematogenous osteomyelitis of children. Pediatrics 1994 ;93:59-62 .
US Department of Health and Human Services, Public
Health Service, Centers for Disease Control and Prevention (CDC): Semiannual Report: Aggregated data
from the National Nosocomial Infections Surveillance
(NNIS) System, March 2000.

McPherson EJ, Tontz W Jr, Patzakis M, et al: Outcome

of infected total knee utilizing a staging system for prosthetic joint infection. Am J Orthop 1999;28:161-165.

228

American Academy of Orthopaedic Surgeons

Chapter

21

Arthritis
Kam Shojania, MD, FRCPC
John M. Esdaile, MD, MPH, FRCPC
Nelson Greidanus, MD, MPH, FRCSC

Introduction
Arthritis encompasses a heterogeneous group of more
than 100 diseases that involve the synovial joints and
the periarticular structures. The exact pathoetiologic
mechanisms underlying most of these disorders are uncertain; however, pathology of the synovium, articular
cartilage, and their subcomponents are believed to be
the primary cause of the most common types of arthritis. Correct diagnosis relies primarily on clinical features
that may be both musculoskeletal and nonmusculoskeletal in nature. Patients with arthritic involvement of
their joints have significant pain, loss of motion, deformity, and instability. The mainstay of contemporary
treatment is nonsurgical and includes patient education,
lifestyle and activity modifications, and pharmacologic
agents. Surgical intervention, particularly in the hip and
knee, may be necessary for the treatment of severe
symptoms, and occasionally may be indicated as a preventive measure. Pain relief remains the most predictable result of reconstructive surgery and represents the
primary indication for most surgical interventions. Restoration of motion and function is less predictable;
therefore, preoperative assessments need to be individualized and patients should be counseled concerning
functional expectations. The development of novel pharmacologic strategies is underway to treat and possibly
alter the natural history of these disorders. Innovations
in prosthetic design and surgical technique are improving the outcomes for patients requiring surgical intervention.

Cartilage Structure and Physiology

The two major types of articular cartilage are fibrocartilage and hyaline cartilage. Menisci are typical examples
of fibrocartilage. Hyaline cartilage, which is the predominant form of articular cartilage, covers the cortical bone
ends and protects them by absorbing force, providing an
extremely low coefficient of friction, and improving
joint stability (Figure 1). Articular hyaline cartilage is
securely fastened to cortical bone by a layer of calcified
cartilage. The low friction joint surface helps to with-

American Academy of Orthopaedic Surgeons

Figure 1 High power photomicrograph of normal cartilage from a rat knee. (Courtesy
of Dr. Michael Nimmo.)

stand the high shear forces created by the muscles that

typically insert near the joint with short lever arms.

Chondrocytes
Chondrocytes are isolated cells within cartilage that
function to produce, maintain, and remodel the extracellular matrix constituents. Chondrocyte activity is regulated by mechanical factors, cytokines, and growth factors. For example, interleukin (IL)-1 can stimulate
chondrocytes to produce proinflammatory mediators
such as matrix metalloproteinases (MMPs), while transforming growth factor beta stimulates chondrocytes to
differentiate and produce type II collagen and proteoglycans. IL-4 is a chondroprotective cytokine that is
activated with mechanical loading of the chondrocyte.
Mechanical forces across the cartilage stimulate chondrocyte synthesis of proteoglycan and collagen; conversely, prolonged inactivity contributes to degenerative
changes in the cartilage through reduction in extracellular matrix synthesis by chondrocytes.

229

Orthopaedic Knowledge Update 8

Arthritis

limited to about 20% of its potential volume because

the network of collagen fibers in the articular cartilage
restricts absorption of water. Pressure on the articular
cartilage causes aggrecan to release water from the cartilage into a thin film on the articular surface. This film
of water provides a surface of minimal friction for
smooth articular motion. The release of the pressure on
the articular surface allows fluid to be drawn back into
the cartilage, bringing nutrients to the chondrocytes. Because cartilage is avascular, this fluid is its only source of
nutrition.

Synovial Fluid
Synovial fluid is an acellular plasma ultrafiltrate that
protects the subchondral structures and lubricates the
joint. Its high viscosity gives it important mechanical
properties and is related to large amounts of polymerized hyaluronic acid.

Cartilage Homeostasis
Figure 2 Diagram of aggrecan, collagen. (Courtesy of Dr. Andrew Thompson.)

Collagen
Collagen provides tensile strength to the cartilage and
provides the extracellular matrix architecture for the
proteoglycans and chondrocytes to fill (Figure 2). Type
II collagen comprises 90% of the articular cartilage with
additional small amounts of collagen type IX and XI.
Type II collagen is oriented tangentially to the articular
surface in the superficial zone of the articular cartilage.
This orientation allows the cartilage to withstand the
high-intensity shearing forces in this area. Collagen fibers curve downward to form vertical sheets through
the middle and deep zones of articular cartilage to provide further vertical tensile strength. Type IX collagen
helps to maintain the orientation of the ubiquitous type
II collagen network.

Proteoglycans
Aggrecan, biglycan, and decorin are the three main proteoglycans in the extracellular matrix. Aggrecan, the
predominant proteoglycan and thereby the best studied,
is a large aggregation of long-chain, negatively charged
glycosaminoglycan molecules attached covalently to a
protein core that itself attaches via a link protein to hyaluronic acid (Figure 2). The glycosaminoglycan molecules are composed of chondroitin sulfate and keratan
sulfate. A single hyaluronic acid chain will have many
aggrecan molecules linked to it. Biglycan and decorin
are nonaggregating proteoglycans that have roles in the
cartilage matrix structure.
Aggrecan provides elastic strength to articular cartilage, and as a polyanionic molecule attracts water, which
allows the cartilage to swell. The swelling, however, is

230

In the normal joint, there is a homeostasis between the

breakdown of cartilage matrix and the formation of
newly synthesized matrix. Growth factors stimulate
chondrocytes to produce matrix constituents whereas
cytokines such as IL-1 and tumor necrosis factor alpha
(TNF-) stimulate matrix degradation. When this balance is disturbed, there is a shift toward excessive degradation, which causes disruption of the structural and
functional integrity of the cartilage.
Collagenase, stromelysin, and gelatinase are MMPs
that act as degradative enzymes. They contain zinc at
the active site and are produced by the chondrocyte as
part of the cartilage development process. The cytokine
IL-1 stimulates production of MMPs by infiltrating leukocytes and connective tissue cells. MMP activity is inhibited by tissue inhibitor of metalloproteinase located
in the cartilage. The complex interplay between MMPs,
tissue inhibitor of metalloproteinase, and the various cytokines that mediate chondrocyte activity will influence
the overall structural integrity of the articular cartilage.

Osteoarthritis
Epidemiology
Osteoarthritis (OA) is the most common form of arthritis and a leading cause of disability in the developed
world. Hip and knee OA rarely occur before the age of
50 years. The prevalence of clinically apparent knee OA
is 30% in the population older than 75 years; OA of at
least one joint occurs in 80% of this population. Autopsy and radiographic studies show a much higher rate
of OA than do epidemiologic studies. The fact that the
incidence of OA increases significantly with age has led
to the erroneous conclusion that OA is simply an agerelated degenerative condition. The prevalence of OA
increases with age because of ligamentous laxity, a fail-

American Academy of Orthopaedic Surgeons

Chapter 21 Arthritis

Orthopaedic Knowledge Update 8

ure of the periarticular structures such as muscles and
proprioceptors to function appropriately, a reduction in
matrix production by chondrocytes, or decreased responsiveness of chondrocytes to growth factors. Other
risk factors for OA include gender, genetic predisposition, obesity, higher than normal bone mineral density,
and joint trauma. The incidence of OA is higher in
women than in men, especially after the age of 50 years.

Etiology
OA is likely a genetically heterogeneous disease where
local factors (such as excessive load) act within the context of a systemic susceptibility (such as a genetic predisposition). OA can be classified as primary (idiopathic) or
secondary (Table 1). Secondary causes of OA can result
from highly intense or abnormal chronic forces across
the joint (for example, from trauma or occupational
overuse) or from a disorder in a joint constituent (for example, hemochromatosis, chronic inflammatory arthritis,
chronic crystal joint disease, septic arthritis). The incidence of OA is increased with certain occupations. Farmers and miners have an increased incidence of hip OA.
Occupations that require regular kneeling and squatting
are correlated with an increased incidence of knee OA.
Elite athletes, but not recreational athletes, have an increased incidence of knee OA.

Pathogenesis
In the common forms of OA such as polyarticular small
joint nodal arthritis in women, there is a familial component. However, in rare subtypes of OA (such as osteochondrodysplastic syndromes), genetic susceptibility
may follow mendelian inheritance modes, and often involve the genes that code type II collagen. Mutations in
the COL2A1 gene (type II procollagen) were found in
rare types of OA in certain families. Mutations in this
same gene have been found to cause skeletal dysplasias
that clinically resemble OA. However, in families with
common forms of OA there were no mutations in the
COL2A1 gene.
Obesity is an important, modifiable risk factor for
bilateral knee OA in both the patellofemoral joint and
tibiofemoral joint. Fortunately, weight loss reduces the
risk of knee OA. The association of obesity and knee
OA is stronger in women than in men. Obesity may increase the incidence of knee OA by increasing force
across the knee, by increasing bone density (which may
be independently associated with OA), or by the adipose tissue increasing production of OA-causing growth
factors or hormones. Obesity is also associated with OA
in the hand, which suggests that weight is not the sole
explanation linking obesity and knee OA. Obesity has
not been consistently associated with hip OA.
Higher bone density has been linked to OA of the hip
and knee, but not the hand. The relationship of higher

American Academy of Orthopaedic Surgeons

TABLE 1 | Classification of Osteoarthritis

Primary (idiopathic)
Peripheral joints
Nodal (at PIP and DIP joints)
First CMC and first MTP
Large joints (hip and knee)
Spine
Variant OA
Inflammatory OA
Generalized OA
Diffuse idiopathic skeletal hyperostosis
Secondary
Traumatic (occupational, sports)
Local joint disorders
Osteonecrosis
Postinfectious
Diffuse joint disorders
Rheumatoid arthritis
Hypermobility syndrome
Endocrine disorders
Diabetes
Acromegaly
Systemic metabolic diseases
Hemochromatosis
Ochronosis
Wilsons disease
Crystal joint disease
Gout
Calcium crystal diseases
Calcium pyrophosphate dihydrate
Calcium apatite
Neuropathic diseases (Charcot joint)
Tabes dorsalis
Diabetes mellitus
PIP = proximal interphalangeal; DIP = distal interphalangeal; CMC = carpometacarpal;
MTP = metatarsophalangeal

bone density with OA of weight-bearing joints may be the

result of the presence of stronger, stiffer subchondral
bone, which deforms less under loading, therefore increasing stress to the cartilage.
Several local factors may play a role in the incidence
or progression of OA including knee injury, which has
been associated with a subsequent increase in the incidence of both hip and knee OA. Abnormal knee alignment also increases the incidence of OA. A varus or valgus deformity of 4 to 5 results in a fourfold increase in
the risk of medial and lateral knee OA. Weakness of the
quadriceps plays a small role in OA in women, but not
in men. Conversely, increased hand muscle strength is
associated with hand OA in men. Development abnormalities, such as acetabular dysplasia and slipped capital
femoral epiphysis, frequently lead to premature OA of
the hip.

231

Arthritis

Figure 3 Low power photomicrograph of OA cartilage shows fibrillation of the cartilage surface. (Courtesy of Dr. Michael Nimmo.)
The chondrocyte is key in the development of OA.
In response to mechanical stress and to cytokines such
as IL-1 and TNF-, chondrocytes release MMPs that degrade the extracellular matrix. The cytokines stimulate
prostaglandin release, which may cause the pain and
stiffness of OA. Microscopic examination of the cartilage shows fibrillation (Figure 3). Subcortical bony sclerosis and osteophytosis reduce bone elasticity and transfer more force across the articular cartilage.
Chondrocytes attempt to repair the damaged type II
cartilage; however, the repair is inadequate and the new
cartilage contains type I collagen and increased fibronectin. This new cartilage lacks the low friction and
elastic properties of healthy cartilage. Further breakdown ushers the beginning of clinical OA.

Biomarkers
Biomarkers, including markers for type II collagen and
aggrecan turnover, cartilage oligomeric matrix protein,
MMPs, and tissue cytokines, monitor the progression of
OA on cartilage, adjacent bone, and synovium. Changes
in certain biomarkers occur within weeks of surgical or
nonsurgical trauma to articular cartilage; such injuries
are associated with subsequent OA. It has not yet been
established which markers are best suited to provide an
early diagnosis of OA and which will correlate with
treatment response or disease progression. The study of
such biomarkers is currently an area of intense research.

Treatment
Nonsurgical and Nonpharmacologic Treatment
In most patients, OA has usually been present for several
years with resultant muscle deconditioning, increased
weight, and radiographic changes before a physician is
consulted. The goals of treatment are to reduce pain, improve function, and delay progression of the disease. Non-

232

Orthopaedic Knowledge Update 8

pharmacologic interventions have been shown to be beneficial for patients with OA; the most important is patient
education. Several studies that have evaluated the effectiveness of arthritis self-management programs have
shown improvements in pain scores, function, therapy
compliance, and quality of life. Weight loss reduces the
symptoms of OA in weight-bearing joints. Physical therapy and regular exercise improve function and pain scores
and prevent disability caused by muscular deconditioning.
Exercise may also reduce the progression of OA in joints
that are presymptomatic. Reduced-load and low-impact
exercises (such as cycling, swimming, or tai chi preceded
by an aerobic warm-up if thought to be safe from a cardiovascular standpoint) are most beneficial. To improve
symptoms of medial joint compartment knee OA, lateral
wedge orthotics or knee bracing for more muscular,
younger patients may provide symptomatic improvement;
knee braces often are not tolerated in the elderly. The use
of a cane is beneficial.
Therapies that have little or conflicting supportive
data include dietary supplements and passive modalities
such as ultrasound and transcutaneous electrical nerve
stimulation. There is conflicting data on the use of oral
glucosamine to treat OA pain or reduce OA joint damage. Commercial preparations of glucosamine are not
standardized and may not contain the amount of glucosamine indicated on the package. Few data exist to
support the use of chondroitin sulfate in patients with
OA. Additional studies are necessary before glucosamine or other dietary supplements can be routinely
recommended.
Pharmacologic Treatment
The goal of pharmacologic therapy is to reduce pain
and improve function; the ability to significantly alter
the progression of OA has not been clearly shown for
any agent. Initial treatment is with acetaminophen (up
to 4 g/day). If this treatment is unsuccessful, or if there
are features of local inflammation such as joint warmth
or effusion, a nonsteroidal anti-inflammatory drug
(NSAID) would be appropriate. Whereas OA is typically considered to be a noninflammatory arthropathy,
the induction of prostaglandin E2 in OA provides some
rationale for the use of NSAIDs in this context.
NSAIDs exert a peripheral role in the reduction of OA
pain; they also may contribute to pain reduction by their
action on the central nervous system.
NSAIDs are superior to acetaminophen in the symptomatic treatment of knee OA, but can cause increased
gastrointestinal (GI) side effects such as pain and dyspepsia. More serious GI side effects include peptic ulcer, perforation, obstruction, or hemorrhage which occur in 2% to 4% of chronic NSAID users each year and
in a higher percentage for those patients with several
risk factors (Table 2). Unfortunately, half of GI ulcers
are silent. In high-risk patients, it is prudent to avoid

American Academy of Orthopaedic Surgeons

Chapter 21 Arthritis

Orthopaedic Knowledge Update 8

TABLE 2 | Relative Risk of Traditional NSAID-Induced GI

Complications

TABLE 3 | Recommended Dosage for Coxibs Available in

the United States

Risk Factor

95% CI

Coxib

OA dose

RA dose

4.6 to 6.6
4.1 to 5.6
1.1 to 4.7
2.0 to 9.7
6.3 to 25.7

Celecoxib
Valdecoxib

100 to 200 mg/day

10 mg/day

200 mg/bid
20 mg/day

Age (> 60 years)

Prior GI event
Severe RA
Concurrent corticosteroids
Concurrent anticoagulants

Relative Risk
5.5
4.8
2.3
4.4
12.7

CI = confidence interval

NSAIDs; however, if an NSAID is required, the newer

selective cyclooxygenase-2 (COX-2) inhibitors will reduce the incidence of ulcers and upper GI bleeding by
approximately 50%. Alternatively, an NSAID can be
given with a proton pump inhibitor.
Coxibs In the early 1990s, it was recognized that two
separate proteins, cyclooxygenase enzymes (COX-1 and
COX-2) were responsible for both the benefits and the
toxicity of traditional NSAIDs (Figure 4). It was found
that the COX-1 enzyme protected the upper GI tract
and promoted platelet aggregation. The inducible
COX-2 enzyme was found to be primarily responsible
for promoting pain and inflammation. Traditional
NSAIDs inhibit both COX-1 and COX-2. NSAIDs that
inhibit COX-2 with little or no effect on COX-1 are
called coxibs or COX-2 specific inhibitors. The selective
inhibition of COX-2 still provides an anti-inflammatory
benefit in joints while reducing the upper GI complications (caused by the inhibition of COX-1) seen with traditional NSAIDs. Coxibs that are currently available in
the United States include celecoxib and valdecoxib (Table 3). Etoricoxib, lumiracoxib, and paracoxib are other
coxibs that may be available soon.
Coxibs have no appreciable effect on platelet function, which allows the use of these medications preoperatively and perioperatively with improved postoperative
pain control. All NSAID and COX-2 inhibitors may
cause cardiovascular and renal side effects in higher risk
patients such as the elderly, those on diuretics, or those
with preexisting cardiovascular or renal dysfunction. In
October 2004, Merck withdrew the coxib rofecoxib from
the worldwide market. This decision was based on data
from a randomized controlled trial looking at the use of
25 mg of rofecoxib in patients with gastrointestinal polyps. Results of the study showed a significant increase in
myocardial infarction in patients after 18 months
(1.48% in the rofecoxib group compared with 0.75% in
the placebo group). Other rofecoxib studies have shown
an increase in serious cardiovascular events. The issue of
cadiovascular safety for the currently available coxibs
(celecoxib and valdecoxib) especially in comparison to

American Academy of Orthopaedic Surgeons

Figure 4 Diagram of COX-1 and -2.

naproxen, is unclear. Therefore, for patients who require
anti-inflammatory medications, it would be reasonable
to use coxibs in patients with higher GI risk but lower
cardiovascular risks. Animal and in vitro studies have
shown an impairment of bone healing with the use of
NSAIDs and COX-2 inhibitors, probably caused by the
inhibition of prostaglandins. Prostaglandins are an integral component in the stages of bone healing from the
initial inflammatory phase, followed by bone resorption
(osteoclasts), and new bone formation (osteoblasts). Although there are data from animal studies suggesting
that the effects of COX-2 inhibitors are both dosedependent and reversible, it is recommended that shortterm administration or other analgesics be used in the
treatment of these patients until sound clinical evidence
becomes available.
Intra-Articular Injections If acetaminophen or
NSAIDs are not beneficial or are contraindicated, an
intra-articular corticosteroid injection may help up to
half of affected patients. The risk of sepsis is about 1 in
10,000. Intra-articular injections of a hyaluronic acid
preparation have been used in knee OA. Although side
effects are uncommon, the data on efficacy are conflicting and the magnitude of clinical improvement is small.
Other Analgesics Tramadol is an analgesic medication
that works through several mechanisms. It has activity at
the Mu opioid receptors but also inhibits serotonin and
norepinephrine uptake. Opioid analgesics are useful for
OA flare-ups if other therapies have not been adequate.

233

Orthopaedic Knowledge Update 8

Arthritis

TABLE 4 | 1987 Revised Criteria for the Classification of RA*

Criteria

Definition

(1) Morning stiffness

(2) Arthritis of three or more joint
areas
(3) Arthritis of the hand joints
(4) Symmetric arthritis

Morning stiffness in and around the joints, lasting at least 1 hour before maximal improvement
At least three joint areas simultaneously have had soft-tissue swelling or fluid (not bony overgrowth alone) observed by
a physician; The 14 possible areas are right or left PIP, MCP, wrist, elbow, knee, ankle, and MTP joints
At least one area swollen (as defined above) in a wrist, MCP, or PIP joint
Simultaneous involvement of the same joint areas (as defined in [2] ) on both sides of the body (bilateral involvement
of PIPs, MCPs, or MTPs is acceptable without absolute symmetry)
Subcutaneous nodules, over bony prominences, or extensor surfaces, or in juxta-articular regions, observed by a
physician
Demonstration of abnormal amounts of serum rheumatoid factor by any method for which the result has been positive
in < 5% of normal subjects
Radiographic changes typical of rheumatoid arthritis on PA hand and wrist radiographs, which must include erosions or
unequivocal bony decalcification localized in or most marked adjacent to the involved joints (OA changes alone do
not qualify)

(5) Rheumatoid nodules

(6) Serum rheumatoid factor
(7) Radiographic changes

*Requirements: A patient shall be said to have RA if he/she has satisfied at least four of the seven criteria. Criteria 1 through 4 must have been present for at least 6 weeks. Patients with two
clinical diagnoses are not excluded. Designation as classic, definite, or probable RA is not to be made.
PIP = proximal interphalangeal; MCP = metacarpophalangeal; MTP = metatarsophalangeal
(Reproduced with permission from Arnett FG, Edworthy SM, Bloch DA, McShane OJ, et al: The American Rhematism Association 1987 Revised Criteria for the Classification of Rheumatoid
Arthritis. Arthritis Rheum 1988;31:315-324.)

Short- or long-acting codeine, oxycodone, or propoxyphene can be used, preferably on a fixed interval basis for less than 2 weeks at a time. Patients with severe OA
who are not surgical candidates and have severe pain that
is unresponsive to other agents should be treated with appropriate long-term opioid analgesics.Although there are
limited data to suggest that combination medications containing caffeine may have additional analgesic properties,
caffeine has significant side effects including sleep disturbances and withdrawal headache.
Surgical Management
The goals of surgical intervention in OA are to decrease
or eliminate pain and to improve function. The success
of joint arthroplasty in achieving these goals has resulted in substantial investment in and development of
prosthetic joints and techniques for their implantation.
However, it is important to consider that other procedures that decrease pain and improve function by restoring, resecting, or replacing the joint also have a role
and are attractive alternatives for young patients who
do not desire a prosthetic joint or for patients with less
advanced joint degeneration who want to maintain a
high level of activity.
Procedures to treat OA by preserving or restoring
articular cartilage surfaces include osteotomies and
muscle releases, joint dbridement, partial resection or
perforation of subchondral bone to stimulate fibrocartilage healing, resection arthroplasty, and the use of various autografts and allografts. Joints compromised by adjacent bony deformity may be appropriate for

234

osteotomy. Arthrodesis may be most appropriate for

joints that are severely compromised or in patients too
young for successful prosthetic arthroplasty. Prosthetic
joint arthroplasty is often recommended for patients
with advanced OA and/or advanced symptoms of OA
that are no longer responding to maximal nonsurgical
modalities. Although complication rates are low in prosthetic hip and knee joints, prosthetic elbow and ankle
joints have higher complication rates.
The ability of the prosthetic arthroplasty intervention
to alleviate pain and improve function has been described
as one of the great advances in 20th century health care.
Whereas traditional indications for prosthetic arthroplasty included incapacitating symptoms, old age, and sedentary lifestyle, contemporary advances in arthroplasty
technique and prosthetic design have expanded the indications to younger patients with intractable symptoms.

Rheumatoid Arthritis
Epidemiology
Rheumatoid arthritis (RA) is an autoimmune disease
that affects 1% of the white population. All racial
groups are affected but aboriginal North Americans
have a higher prevalence. RA is a chronic, fluctuating
polyarthritis that is symmetric, erosive, and eventually
deforming (Table 4).

Clinical Features
In most patients, RA manifests over weeks to months
with an insidious onset of fatigue and joint pain, which

American Academy of Orthopaedic Surgeons

Chapter 21 Arthritis

Orthopaedic Knowledge Update 8

Figure 5 Photograph showing early RA of the hand.

TABLE 5 | Extra-Articular Manifestations of RA

Organ System

Manifestations

Pulmonary

Cricoarytenoid cartilage, interstitial lung disease,

bronchiolitis obliterans, pleural effusion, rheumatoid nodules
Rheumatoid nodules, cutaneous vasculitis, pyoderma gangrenosum
Pericardial effusions
Myelopathy resulting from atlantoaxial instability,
entrapment neuropathies, mononeuritis multiplex,
peripheral sensory neuropathy
Feltys syndrome (neutropenia, splenomegaly),
anemia, thrombocytosis
Episcleritis, scleritis, corneal melt,
keratoconjunctivitis sicca

Skin
Cardiac
Neurologic

Hematologic
Ocular

may be transient early in the course of the disease.

However, 10% of patients will experience an acute onset of symptoms that occurs within days. Eventually,
joint inflammation becomes more persistent with
warmth, effusion, tenderness, and loss of function. Early
morning joint stiffness occurs in most patients and improves with physical activity. Morning stiffness usually
exceeds 60 minutes and its duration correlates with the
extent of synovial inflammation. Classic areas of symmetric polyarthritis are the wrists, metacarpophalangeal
(MCP), metatarsophalangeal (MTP), and proximal interphalangeal (PIP) joints of the hands and feet (Figures 5 through 7). Structural damage to the joints, once
believed to occur only after 1 or more years, has been
shown by MRI to occur within months of RA onset.

American Academy of Orthopaedic Surgeons

Figure 6 Rheumatoid nodules are shown.

TABLE 6 | Infectious Agents That May Be An Environmental

Trigger for RA
Viral
Epstein-Barr virus
Parvovirus
Rubella
Cytomegalovirus
Bacteria
Escherichia coli
Proteus mirabilis
Mycoplasma
Mycobacteria

Extra-articular manifestations of RA are summarized in

Table 5.

Rheumatoid Factor
Rheumatoid factors are immunoglobulin M antibodies
directed against the constant fragment region of the immunoglobulin G molecule. Rheumatoid factor is present
in as few as half the patients with early RA, is found in
5% of healthy individuals, and is found in a host of
chronic inflammatory and infectious conditions. Therefore, as a diagnostic test, rheumatoid factor has been
overvalued; however, its presence has some prognostic
value because a positive test portends a more aggressive
course of RA with a higher morbidity and mortality.

Etiology
Although the exact etiology of RA is unknown, genetic
susceptibility and environmental factors have some demonstrable importance. Compared with a prevalence of

235

Orthopaedic Knowledge Update 8

Arthritis

Figure 7 Medium power photomicrograph of rheumatoid nodule depicting palisading

histiocytes and central fibrinoid necrosis. (Courtesy of Dr. Michael Nimmo.)

Figure 8 High power photomicrograph of normal synovium showing a thin layer of

cells. (Courtesy of Dr. Michael Nimmo.)
Geographic clusters of erosive arthritis have been
found in ancient archaeological sites in North America;
however, no convincing findings of RA in Western Europe have been found that predate the settlement of the
New World. This finding suggests an infectious etiology
for RA. Several potential agents have been proposed,
but none has yet been proven (Table 6). There is no
consistent evidence of a specific microorganism in RA
synovium, which may indicate that the infectious trigger
is subclinical, outside the synovium, or precedes the RA
by a long latent period before clinically evident joint inflammation.

Pathogenesis

Figure 9 Medium power photomicrograph of pannus. Thick villi of synovium with

neovascularization and lymphoid aggregates are seen. (Courtesy of Dr. Michael
Nimmo.)

1% in the general population, the concordance rate in

monozygotic twins is 12% to 15%, and is 2% to 5% for
dizygotic twins and siblings. About half of the genetic
susceptibility resides within the major histocompatibility
complex class II loci, in which the DR4 chains with the
greatest association with RA are DRB*0401 and
DRB*0404. The remaining genetic predisposition is believed to lie within other loci that are currently being investigated. Such loci include the major histocompatibility complex class III genes for TNF-, heat shock
protein 70 (HSP70), and complement C4, as well as
genes not included in the major histocompatibility complex such as T cell and immunoglobulin genes.

236

RA probably begins with an antigen-mediated activation of T cells in an immunogenetically susceptible host,

which then initiates a cascade of events including endothelial cell activation, proliferation of type A and B synoviocytes, and recruitment of additional inflammatory
cells from the bone marrow. TNF- and IL-1 are the
apex of a cytokine cascade leading to the production of
proteases and autoantibodies. When the immune response is triggered by the antigen, additional antigens
may be created and recognized by the T cell to create
an ongoing immune response. The synovium is transformed from a thin layer only several cells deep (Figure
8), to a proliferative and invasive synovium (pannus)
composed of synoviocytes, T cells, and macrophages
(Figure 9). The synovitis and vascular proliferation invade adjacent cartilage and bone. TNF- and IL-1 stimulate chondrocytes to degrade cartilage and osteoclasts
to promote periarticular bony erosion and osteopenia.
A significant humoral component may be involved
in the pathogenesis of RA. Whereas T cells outnumber
B cells in the synovium, there are active plasma cells in

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

the synovium and the expanded B lymphocyte pool,
which exhibits an origin from a limited number of B cell
clones. It is intriguing that the B cell-depleting therapy,
rituximab, has shown remarkable preliminary benefits in
the treatment of RA and has provided further evidence
that B cells are important in the pathogenesis of RA.
The improved understanding of RA pathogenesis has
resulted in important therapeutic breakthroughs.

Pharmacologic Therapy
The traditional approach to the treatment of early RA
has been the use of analgesics and NSAIDs to reduce
symptoms. Such pharmacotherapies, however, do not
prevent joint erosion and articular damage. MRI has
shown that joint erosions can occur within months of
symptom onset, and the rate of progression of erosions
is highest in the first 3 years after RA onset. The recognition of these early changes has resulted in a dramatic
shift in treatment philosophy for patients with RA. RA
must be treated early and aggressively. The evidence is
overwhelming that a delay in instituting diseasemodifying antirheumatic drug (DMARD) therapy or a
failure to maintain disease suppression with DMARDs
can result in irreversible joint destruction and disability.
DMARDs, previously considered second-line drugs,
are currently the mainstay of the pharmacologic treatment of RA.This heterogeneous group of medications reduces inflamed joint counts, acute phase reactants, and
erosion scores and stabilizes or even improves functional
status and reduces long-term disability (Table 7). Methotrexate is the most commonly used DMARD in North
America. It is relatively potent compared with the other
DMARDs and more patients will continue to use methotrexate over the long term than other traditional
DMARDs. About 50% of patients who begin treatment
with methotrexate will remain on the medication after
5 years compared with 20% of patients who are initially
treated with intramuscular gold or sulfasalazine. Newer
DMARDs include leflunomide and anticytokine agents.
In addition to beginning early treatment with
DMARDs, a new but well-established development is
the early use of combination drug therapy. Established
examples of drugs used in combination with methotrexate are hydroxychloroquine, sulfasalazine, gold, cyclosporine, and anti-TNF- agents.
Understanding the cytokine cascade in RA has
helped in the development of novel new therapies that
target these factors. TNF- and IL-1 are key proinflammatory cytokines in RA. In clinical studies, inhibition of
TNF- has resulted in a dramatic reduction in the
symptoms of RA, an improvement in function, and a retardation of radiographic progression. Etanercept, infliximab, and adalimumab are the available anti-TNF-
agents. The mode of action and route of administration
of these anticytokine DMARDs are shown in Table 8.

American Academy of Orthopaedic Surgeons

Chapter 21 Arthritis
Compared with traditional DMARDs, these agents have
remarkable efficacy with minimal adverse events in the
treatment of RA. Immunosuppression and the risk of
infection (including reactivation of mycobacterium tuberculosis) are potential complications with all antiTNF- drugs. There have been rare reports of demyelinating diseases with the use of etanercept. Infliximab
must be given concomitantly with methotrexate or azathioprine (in the methotrexate intolerant patient),
whereas etanercept and adalimumab can be given as
monotherapy or in combination with methotrexate. Infliximab has been shown to exacerbate congestive heart
failure.
Anakinra is the only IL-1 inhibitor currently available. There is less reduction of inflammation with anakinra than with the anti-TNF- agents; however, anakinra
has a good safety profile. Injection site reactions with
anakinra are frequent, but are generally mild and selflimiting. Anakinra use is also associated with a small increase in the incidence of pneumonia and urinary tract
infections.
There is no perioperative standard of practice to
guide the use of anticytokine DMARDs. It is recommended that surgery be performed when presumed levels of the therapies are at their lowest2 weeks prior to
the infliximab infusion; 3 days after the etanercept injection; 10 days after the adalimumab injection. Some physicians will postpone resumption of the medications for
1 to 2 weeks postoperatively, in the hope that the reduced immunosuppression therapy will reduce the risk
of subsequent postoperative infection. However, this approach may necessitate the use of perioperative corticosteroids or cause a flare of synovitis, which could contribute to a poor postoperative recovery and an
increased risk of infection. The other approach is to continue the use of DMARDs unless there are contraindications (such as an active infection). There is no consensus on which of these two methods is safer; few data are
currently available.
Glucocorticoids
The DMARDs generally have a slow onset of action,
and although they are effective in reducing pain and inflammation, patients usually require ongoing analgesic
and anti-inflammatory medication. While waiting for
DMARDs to take effect, some rheumatologists prescribe a course of bridging glucocorticoid (GC) medication such as prednisone (10 mg daily) on a tapering
dose over several weeks to months. GCs are powerful
suppressors of inflammation, but long-term side effects
and an absence of disease modifying benefits preclude
their long-term use. GCs can be given orally, parenterally, and intra-articularly. High-dose parenteral GC can
be used to treat severe extra-articular manifestations
such as interstitial lung disease, ocular inflammation, or
vasculitis. Intra-articular therapy causes few systemic

237

Orthopaedic Knowledge Update 8

Arthritis

TABLE 7 | Common Disease-Modifying Antirheumatic Drugs

Drug

Onset of Action

Relative Effectiveness

Half Life

Monitoring

Usual Dose

Hydroxychloroquine

2 to 6 months

weak

Up to 50 days

Minocycline
Sulfasalazine

2 to 6 months
2 to 3 months

weak
moderate

15 hours
10 hours

Azathioprine

2 to 3 months

moderate

3 hours

Methotrexate

2 to 3 months

strong

3 to 10 hours

Parenteral gold salts

2 to 6 months

strong

Up to 27 days

Leflunomide

2 to 3 months

strong

15 days, but
enteropathic
circulation may
increase half life

Cyclosporine

2 to 3 months

strong

Up to 18 hours

Visual changes,
< 6.5 mg/kg/day of
funduscopic
ideal body weight
examination and
visual fields every 6 to
12 months
200 mg/day
CBC, liver enzymes every Gradual increase to
2 weeks for first
1 to 1.5 g/bid
2 months then every
3 months
Monthly CBC, liver
Start at 1 mg/kg/day
enzymes, but more
and slowly increase
frequent when dose
as needed to
increased
2.5 mg/kg/day
Monthly CBC, liver
7.5 mg to 15 mg by
enzymes, albumin,
mouth or
creatinine
subcutaneously
weekly
Increase to response or
to 25 mg/wk
CBC, urinalysis and
50 mg/wk
serum creatinine every intramuscularly for
second injection
first 6 months, then
slow taper if response
occurs
Monthly CBC, liver
Loading dose of
enzymes, creatinine
100 mg/day for
If toxicity or pregnancy
3 days then 10 to
occurs, must wash out 20 mg/day
with cholestyramine
CBC, creatinine, blood Start at 1.5 mg/kg/day
pressure, liver enzymes and increase to
every 2 weeks until
4 mg/kg/day
steady dose then
monthly

CBC = complete blood count

side effects and can be particularly useful for patients

with a few resistant, inflamed joints.

not be the only therapeutic agent used.

Surgical Management of Rheumatoid Arthritis

Nonsteroidal Anti-Inflammatory Drugs
NSAIDs are useful for reduction of pain and inflammation in RA; however, their use in RA as in OA is complicated by GI and other toxicities. Patients with RA are
more prone to developing symptomatic GI ulcers than
patients with OA. GCs also increase the incidence of GI
ulcers independently. Therefore, if anti-inflammatory
medications are indicated in a patient with RA, a coxib
is the preferential therapy. Although anti-inflammatory
medications reduce the symptoms of RA, they do not
prevent the progression of damage and therefore should

238

Because patients with RA often have special needs because of multiple joint involvement, it is important that
a multidisciplinary team (surgeon, rheumatologist,
nurse, physiotherapist, occupational therapist, and social
worker) be consulted before surgical intervention. The
team should ensure that planning, staging, and rehabilitation after surgery is appropriate and that the patient is
provided with optimal information on goals, expectations, and opportunities to improve outcome and minimize complications.

American Academy of Orthopaedic Surgeons

Chapter 21 Arthritis

Orthopaedic Knowledge Update 8

TABLE 8 | Anticytokine DMRDs

Anticytokine
DMARDs
Etanercept

Infliximab

Adalimumab

Anakinra

Mechanism of Action

Route and Dosage

Human soluble p75

25 mg subcutaneously
TNF- receptor fusion
twice weekly
protein
Chimeric (human/
3mg/kg intravenously at
murine) IgG1
0, 2, 6, then every
monoclonal antibody to 8 weeks
Can be increased up to
TNF-
10 mg/kg every
4 weeks
Human IgG1 monoclonal 40 mg subcutaneously
antibody to TNF-
every 2 weeks
Can be increased to
weekly
Human IL-1 receptor
100 mg/daily
antagonist
subcutaneously

Before any surgical intervention, it is imperative that

the patient receive a comprehensive preoperative evaluation. If acute joint inflammation is apparent, it is prudent
to first optimize medical treatment and resolve the acute
synovitis. Surgical, chemical, or radioactive synovectomy
can be considered if the synovitis does not adequately respond to medication and there is no structural damage to
the joint. It is unlikely that synovectomy will alleviate all
symptoms if significant structural damage to the joint has
already occurred. If structural joint damage exists and patient symptoms of synovitis do not respond to conventional medical treatment,surgical intervention with arthrodesis or arthroplasty may be indicated. Patients with
multiple joint involvement should receive staged surgical
reconstruction to avoid prolonged periods of hospitalization and immobility.
Extensive RA involvement of the shoulder, elbow, or
wrist may compromise any anticipated postoperative
use of crutches or a walker for patients needing lower
extremity surgery. Therefore, it may be necessary to
treat the involved joints of the upper extremity before
proceeding with lower extremity surgery (for example,
wrist arthrodesis may facilitate the use of crutches for
lower extremity reconstruction). However, the completion of lower extremity reconstruction before treatment
of upper extremity joints may be most appropriate if it
is anticipated that the patients need for a walking device in the postoperative period could overload delicate
arthroplasties in the elbow and/or hand. Patients unable
to attain at least 100 of flexion of the knee require
maximal assistance from the upper extremities in rising
from a seated position. If the upper extremities cannot

American Academy of Orthopaedic Surgeons

withstand these forces, then maximal hip and knee flexion must be a goal of surgery.
Prior to surgery, the patient should be in optimal
medical condition. The patient should be taking the lowest possible dose or should discontinue taking corticosteroids and certain DMARDs before surgery to minimize any increased risk of postoperative infection from
use of these medications. Any other sites, or potential
sites, of infection in the patient should be resolved before surgical intervention. For example, coexisting dental infections, skin ulcers, and urinary tract infections
can all increase the risk of infection in the joint targeted
for surgery. The rehabilitation team should perform a
comprehensive preoperative assessment to ensure that
the patient is motivated and able to participate in the
expected postoperative program. Instructing the patient
in the use of a walker, crutches, or a cane before surgery
may help to facilitate an uneventful transition to these
devices in the immediate postoperative period.
Significant cervical spine involvement exists in up to
40% of patients with RA. Because this involvement is
frequently asymptomatic, preoperative evaluation with
lateral cervical spine flexion and extension radiographs
are necessary to document the existence of C1-C2 or
subaxial instability that might compromise the spinal
cord during routine intubation.
The perioperative use of antirheumatic drugs should
be managed to minimize the risk of bleeding or infection while optimizing patient comfort and pain relief.
The rheumatologist, surgeon, and patient should work
together to determine the ideal individualized drug
treatment plan. The use of anti-inflammatory medication in the perioperative period is also important. Most
NSAIDs should be discontinued a minimum of at least
five half-lives before surgery, and aspirin should be discontinued at least 7 to 10 days before surgery to decrease bleeding associated with the surgical procedure.
Coxibs such as celecoxib and valdecoxib have no significant effect on platelet activity and can be continued.
Antimalarial agents, gold salt injections, auranofin, sulfasalazine, and penicillamine may be continued during
the preoperative and immediate postoperative periods.
It is sometimes recommended that methotrexate be
temporarily discontinued for 1 to 2 weeks during the
immediate preoperative period and during hospitalization because of fluid balance alterations and immunesuppressant effects. Azathioprine may be continued in
the preoperative period if the patients leukocyte count
remains greater than 3,500 mm3; however, it should be
discontinued while the patient is in the hospital after
surgery. GCs can be reduced in the preoperative period,
but there is a potential for adrenal suppression in longstanding use. It is currently recommended that most patients taking oral corticosteroids receive stress coverage
the day of surgery and for the following 24 hours.

239

Orthopaedic Knowledge Update 8

Arthritis

TABLE 9 | Juvenile Idiopathic Arthritis: 1997 Durban Classification Compared With the ARA/ACR Classification
ILAR Classification of
JIA
Type

ARA/ACR 1972 Classification of JRA

Subtype

Systemic
Polyarthritis
Oligoarthritis

RF positive
RF negative
Persistent
Extended

Systemic onset Stills

disease
Polyarticular JRA
(RF positive)
Pauciarticular JRA

Psoriatic arthritis
Enthesitis-related
arthritis
Other (does not meet
criteria for any of the
above categories)
ARA = American Rheumatism Association; ACR = American College of Rheumatology

Juvenile Idiopathic Arthritis

Juvenile idiopathic arthritis (JIA) is a new term that was
first proposed at the 1997 International League of Associations for Rheumatology (ILAR) meeting. The classification was revised at the ILAR meeting in 2001. JIA
has replaced the older term juvenile rheumatoid arthritis (JRA) that was used primarily in North America and
the term juvenile chronic arthritis that had been used in
Europe. The new classification also incorporates other
idiopathic arthritides. These heterogeneous diseases
have diverse prognoses, treatments, and long-term outcomes. Table 9 compares the ILAR classification with
the American College of Rheumatology classification.
The onset of JIA must be before the age of 16 years and
must persist for at least 6 weeks. The new classification
system specifically excludes arthritis of a known cause,
such as reactive arthritis.
Systemic JIA (formerly, systemic onset JRA or Stills
disease) includes children with rash, daily high fever,
and any number of inflamed joints. These children appear quite ill with anemia and a high white cell count
and may have serositis, hepatosplenomegaly, and lymphadenopathy. An infection or a hematologic malignancy usually must be ruled out when making this diagnosis. Children with this subtype of arthritis have the
highest likelihood of systemic complications and internal organ involvement.
Childhood polyarthritis includes children with the
previous diagnosis of polyarticular JRA with some modifications. This disease has been divided into rheumatoid
factor (RF) positive and RF negative polyarthritis. A
family history of psoriasis is a specific exclusion. Child-

240

hood polyarthritis is characterized by the involvement

of more than four joints in the first 6 months of the disease. Both large and small joints are involved. Patients
who are older than 10 years of age at onset and who
have a positive RF, have disease characteristics much
like adult patients with RA.
Childhood oligoarthritis includes children with the
previous diagnosis of oligoarticular (pauciarticular)
JRA. It is the most common type of JIA and affects approximately 50% of JIA patients. This type of arthritis is
characterized by the involvement of four or fewer joints
in the first 6 months of the disease; the extended subtype encompasses children who develop more inflamed
joints after 6 months of the disease. The peak incidence
occurs in the second or third years of life and affects
more girls than boys. Childhood oligoarthritis usually involves large joints other than the hip. The child typically
presents with a limp that improves during the day.
Other than uveitis, which occurs in about 20% of patients with childhood oligoarthritis, there are few systemic manifestations. The uveitis is more frequently associated with a positive antinuclear antibody (ANA)
and is clinically silent in the early, reversible stages.
Thus, children with childhood oligoarthritis who are
ANA positive should have ocular screening every 4
months and those who are ANA negative should be
screened every 6 months to prevent long-term ocular
damage. A leg-length discrepancy is another common
complication of childhood oligoarthritis, which is caused
by the increased blood flow to the inflamed joint that
results in increased length and width of the bone. A single inflamed knee can result in a significant leg-length
discrepancy.
Childhood psoriatic arthritis includes children with
arthritis and psoriasis and also includes children without
psoriasis who have arthritis with dactylitis or characteristic fingernail abnormalities (onycholysis or nail pitting), or a first degree relative with psoriasis.
Childhood enthesitis-related arthritis was previously
called childhood spondyloarthropathy. Children with
this disease have arthritis and enthesitis, or enthesitis
with at least two of the following other symptoms or
characteristics: sacroiliac or inflammatory lumbosacral
pain; acute, symptomatic anterior uveitis; a positive
HLA-B27; a family history of spondyloarthropathy.
The monitoring and treatment of children with JIA
depends on the severity and type of disease and extraarticular involvement. Some types of JIA may go into
remission and some may recur in adulthood. Medical
therapy, such as DMARD use, does not always mirror
the same use as in adult therapy. For example,
DMARDs are used more often in patients with the polyarticular type of JIA than in those with the oligoarticular type. Also, some drugs commonly used to treat adult
arthritis, such as the coxibs, have not been adequately
studied for use in children (yet are frequently pre-

American Academy of Orthopaedic Surgeons

Chapter 21 Arthritis

Orthopaedic Knowledge Update 8

scribed). Corticosteroids may cause significant side effects in both adults and children, but children may also
be affected by marked growth retardation. A pediatric
rheumatologist is the best resource in treating children
with JIA.
Patients with JIA often have unique perioperative
challenges. The temporomandibular joint is frequently
involved and, when combined with micrognathia, intubation and postextubation ventilation can be difficult.
Fiberoptic intubation may be necessary.

Seronegative Spondyloarthropathies

Figure 10 Photograph of nail pits in a

patient with psoriatic arthritis.

Clinical Features
The seronegative spondyloarthropathies are a group of
four conditions: ankylosing spondylitis (AS), psoriatic arthritis (PsA), reactive arthritis (ReA), and arthritis associated with inflammatory bowel disease. AS is a chronic
inflammatory disease of the axial skeleton. The major
symptoms of AS are back pain with prolonged (longer
than 1 hour) morning stiffness and progressive loss of motion of the axial spine. Other clinical features that help to
distinguish AS from the much more common mechanical
back pain are back pain that awakens the patient at night
and pain that is relieved by light exercise. The hallmark
radiologic findings are sacroiliitis and progressive fusion
of the spine (called a bamboo spine), which occurs in a
later stage of the disease.AS typically affects young adults
between 15 and 35 years of age. The onset of axial symptoms is insidious. Other symptoms include frequent peripheral joint inflammation (usually oligoarticular and
asymmetric), acute, unilateral anterior uveitis, and rarely,
cardiac and pulmonary disease.
There are five types of PsA. The most common is an
oligoarticular, asymmetric pattern. Other types include a
polyarticular symmetric arthritis similar to RA, a pattern of axial involvement similar to AS, a distal interphalangeal joint involvement pattern, and arthritis mutilans with telescoping digits. PsA usually is preceded by
cutaneous psoriasis, but in about 20% of patients, the arthritis precedes the psoriasis. PsA usually is associated
with nail pitting (Figure 10) and patients may have dactylitis (sausage digit) with a fusiform swelling of the entire digit.
ReA was previously called Reiters syndrome and is
triggered by a sexually transmitted disease (infecting
bacteriaChlamydia) or a bowel infection (infecting
bacteriaYersinia, Salmonella, Campylobacter, Shigella). Typical features of ReA include asymmetric
lower limb oligoarthritis, conjunctivitis, and dysuria. The
dysuria can be either caused by a Chlamydia pyuria or
by a sterile pyuria that can be triggered by a bowel infection.
Approximately 20% of patients with inflammatory
bowel disease will develop an inflammatory arthritis
with axial and/or peripheral joint involvement. The pe-

American Academy of Orthopaedic Surgeons

ripheral joint inflammation often can flare with bowel

disease activity, whereas the axial inflammation does not
flare in conjunction with the inflammatory bowel disease.
These four spondyloarthropathies often have overlapping symptoms of axial inflammation including sacroiliitis and enthesitis. The enthesis is the structure that
includes the ligament and tendon attachment to the
bone as well as the extension into the bone cortex. Enthesitis is a characteristic finding.

Human Leukocyte Antigen-B27

The major factor in the pathogenesis of the spondyloarthropathies is the class I major histocompatibility antigen, HLA-B27. HLA-B27 is present in 5% to 8% of the
general population, but is present in 95% of white patients and 50% of black patients with AS. This antigen
also is found in 50% to 80% of patients with other seronegative spondyloarthropathies. The presence of
HLA-B27 is not needed to confirm a diagnosis of AS;
however, it can be used as a diagnostic test in children
and young women who may have normal radiographs
and in patients who have an atypical presentation.

Treatment
Treatment protocols of the spondyloarthropathies include maintenance of posture, back range-of-motion exercises, and the use of NSAIDs. If there is chronic peripheral
joint
involvement, methotrexate
and
sulfasalazine are frequently used to reduce joint erosions and damage, similar to the therapy used for RA
(Table 7). Local GC injections may be useful for large
joint involvement, but systemic steroids are not often
used. For patients with ReA, the triggering infection
should be treated. If the joint inflammation persists after the infection has been eradicated, long-term antibiotic use has not been shown to be effective. Studies of
the use of anti-TNF- therapies for patients with PsA
have found remarkable improvement in joint pain, damage control, and function, as well as in the psoriasis le-

241

Orthopaedic Knowledge Update 8

Arthritis
sions. In resistant patients with AS, anti-TNF- can
cause dramatic clinical improvement. No treatment has
been shown to prevent the ankylosis of the axial spine;
however, current studies show that the anti-TNF-
agents used in the treatment of AS are promising.

Surgical Management
The protocols for the perioperative use of DMARDs
and NSAIDs in RA also apply to their use in treatment
of the spondyloarthropathies. Additional surgical issues
need to be considered for AS patients. These patients
have diminished chest excursion, which creates a greater
risk for postoperative pulmonary complications. Intubation also may be extremely difficult because of cervical
spine rigidity; fiberoptic intubation may be necessary.
Ossification of the anulus fibrosus and spinal ligaments
may make administration of spinal anesthetic difficult.
When such patients are positioned on the operating table, great care must be exercised to ensure that the
spine is adequately padded and supported, particularly
if a significant kyphotic deformity exists. Patients with
AS often fail to fully regain the same range of motion
after joint reconstruction in comparison with other
groups of patients. Although postoperative range of motion may be less, it is still often adequate to significantly
improve the patients ability to perform activities of
daily living. Patients with AS may have an increased risk
of heterotopic ossification. Indomethacin or perioperative radiation may be used to decrease this risk.
Patients with PsA may have skin involvement in the
area of the proposed surgical incision. Because local
bacterial contamination may increase the risk of infection, it is recommended that the skin be treated aggressively with topical agents or ultraviolet light before any
surgical procedure. In addition, patients should use antimicrobial soaps before surgery.

Emerging Concepts and Future Directions

There is increasing knowledge about the systemic risk factors of OA and the modification of these risk factors to
prevent OA. The search for the first disease-modifying
OA drug (similar to a DMARD in RA) continues; this
process will be facilitated by the increased understanding
of the molecular basis of articular cartilage degradation.
Possible candidates for such a disease-modifying agent include the combination cyclooxygenase/lipoxygenase inhibitors, MMP inhibitors, IL-1, and nitric oxide blockade.
Although the anti-TNF- therapies have dramatically improved function and quality of life for patients
with RA, the use of these agents has not led to a permanent remission of inflammation in RA patients. There
are probably other factors contributing to the persistent
systemic inflammation in RA. Other cytokine inhibitors
are currently in clinical trials and are showing good preliminary results. B cell inhibition is a promising treat-

242

ment of RA that is currently under investigation. Although the treatment of RA has advanced greatly in
recent years, the possible causes of RA are still unknown. Understanding the causes of RA offers the best
chance for finding a cure or a preventive therapy.
Treatment of the spondyloarthropathies also will
benefit from the anti-TNF- research in RA. These
agents may be the first to actually reduce long-term progression of axial disease. Investigating the trigger for the
spondyloarthropathies may lead to the identification of
other infectious organisms, similar to those found in
ReA.
Stem cell transplant has been tested in patients with
RA, JIA, and other autoimmune diseases with the goal
of replacing most of the activated cells with uncommitted stem cells (based on the assumption that the original
triggers for the disease are no longer present). A recent
study noted that positive responses were found with
59% remission for up to 4 years in patients with JIA.
Transplant mortality was 1.4% for RA patients and
12.5% for those with JIA. Most RA patients who did
not experience remission regained sensitivity to antirheumatic medications, allowing for better disease control. There is one report of the development of RA after
autologous peripheral blood stem cell transplantation.

Annotated Bibliography
Osteoarthritis
Brown KN, Saunders MM, Kirsch T, Donahue HJ, Reid
JS: Effect of COX-2-specific inhibition on fracturehealing in the rat femur. J Bone Joint Surg Am 2004;86:
116-123.
The results of this study on rats with nondisplaced fractures of the femur showed that a traditional NSAID (indomethacin) delayed fracture healing more than a coxib (celecoxib) or placebo.

DeGroot J, Bank RA, Tchetverikov I, Verzijl N, TeKoppele JM: Molecular markers for osteoarthritis: The road
ahead. Curr Opin Rheumatol 2002;14:585-589.
This is an excellent recent summary of the developments
made in the study of molecular markers for osteoarthritis.

DeMaria AN, Weir MR: Coxibs: Beyond the GI tract:

Renal and cardiovascular issues. J Pain Symptom
Manage 2003;25(suppl 2):S41-S49.
This article presents an important review of the clinical
implications of the use of coxibs on the renal and cardiovascular systems. The article also reviews the lack of platelet effect
of the coxibs and the role of supplemental acetylsalicylic acid
in those patients who require an antiplatelet agent for cardiovascular prophylaxis.

Fries JF, Lorig K, Holman HR: Patient self-management

in arthritis: Yes! J Rheumatol 2003;30:1130-1132.

American Academy of Orthopaedic Surgeons

Chapter 21 Arthritis

Orthopaedic Knowledge Update 8

This article reviews the importance of patient selfmanagement in arthritis. Patients who learn about their arthritis and who take an active role in the management of this
chronic condition have better long-term outcomes.

Jordan JM, Kraus VB, Hochberg MC: Genetics of osteoarthritis. Curr Rheumatol Rep 2004;6:7-13.
This article provides a perspective on the various genetic
studies of OA and the limitations of these studies given the
variation in study populations and diagnostic criteria.

Rheumatoid Arthritis
Alderman AK, Ubel PA, Kim HM, Fox DA, Chung KC:
Surgical management of the rheumatoid hand: Consensus and controversy among rheumatologists and hand
surgeons. J Rheumatol 2003;30:1464-1472.
This article describes the large discrepancy in communication between rheumatologists and hand surgeons and knowledge of timing and options in hand surgery.

Cannella AC, ODell JR: Is there still a role for traditional

disease-modifying antirheumatic drugs (DMARDs) in
rheumatoid arthritis? Curr Opin Rheumatol 2003;15:185192.
This article reviews recent and relevant trials in the treatment of RA, suggests a treatment algorithm, and argues that traditional disease-modifying antirheumatic drugs continue to play
a pivotal role in treatment. A more aggressive approach using
combination DMARDs and biologics in the treatment of early
RA is discussed.

Laine L, Bombardier C, Hawkey CJ, et al: Stratifying

the risk of NSAID-related upper gastrointestinal clinical
events: Results of a double-blind outcomes study in patients with rheumatoid arthritis. Gastroenterology 2002;
123:1006-1012.
The risks of traditional NSAIDs in the treatment of RA is
investigated with a review of the recent literature.

Mohan AK, Cote TR, Siegel JN, Braun MM: Infectious

complications of biologic treatments of rheumatoid arthritis. Curr Opin Rheumatol 2003;15:179-184.
Authors from the United States Food and Drug Administration present a review of the infectious complications from
the use of the biologic therapies as reported in the medical literature from November 2001 through October 2002.

Juvenile Idiopathic Arthritis

Petty RE, Southwood TR, Manners P, et al: International League of Associations for Rheumatology classification of juvenile idiopathic arthritis: Second revision,
Edmonton, 2001. J Rheumatol 2004;31:390-392.
An excellent review of the classification of JIA that includes the rationale for the choice of subtypes is presented.

American Academy of Orthopaedic Surgeons

Ravelli A: Toward an understanding of the long-term

outcome of juvenile idiopathic arthritis. Clin Exp
Rheumatol 2004;22:271-275.
This article presents an analysis of the outcome of JIA in
terms of clinical remission, physical disability, and radiographic
damage. The extensive prognostic indicators and the subtypes
that persist into adulthood are reviewed.

Seronegative Spondyloarthropathies
Davis JC Jr, Van Der Heijde D, Braun J, et al: Recombinant human tumor necrosis factor receptor (etanercept)
for treating ankylosing spondylitis: A randomized, controlled trial. Arthritis Rheum 2003;48:3230-3231.
This study of patients with moderate to severe ankylosing
spondylitis showed that etanercept was highly effective and
well tolerated. Patients showed a dramatic improvement in
symptoms and measures of spinal mobility. It also had a good
safety profile.

Emerging Concepts and Future Directions

Edwards JC, Leandro MJ, Cambridge G: B lymphocyte
depletion therapy with rituximab in rheumatoid arthritis. Rheum Dis Clin North Am 2004;30:393-403.
Most of the current DMARD therapy for RA focuses on
T cell inhibition. This article discusses the concept of B-cell inhibition with rituximab as a promising new therapy for RA.

Furst DE: The status of stem cell transplantation for

rheumatoid arthritis: A rheumatologists view.
J Rheumatol Suppl 2001;64:60-61.
Stem cell transplantation may become an important therapy for the treatment of certain subtypes of RA. This review
discusses specific selection criteria and outcome measurements.

Wulffraat NM, Brinkman D, Ferster A, et al: Long-term

follow-up of autologous stem cell transplantation for refractory juvenile idiopathic arthritis. Bone Marrow
Transplant 2003;32(suppl 1):S61-S64.
This review discusses the outcome of autologous stem cell
transplantation in 31 children with JIA from eight pediatric
European transplant centers.

Classic Bibliography
Bathon JM, Martin RW, Fleischmann RM, et al: A comparison of etanercept and methotrexate in patients with
early rheumatoid arthritis. N Engl J Med 2000;343:15861593.
Bresnihan B, Alvaro-Gracia JM, Cobby M, et al: Treatment of rheumatoid arthritis with recombinant human
interleukin-1 receptor antagonist. Arthritis Rheum 1998;
41:2196-2204.
Maini R, St Clair EW, Breedveld F, et al: Infliximab
(chimeric anti-tumour necrosis factor alpha monoclonal
antibody) versus placebo in rheumatoid arthritis pa-

243

Arthritis

Orthopaedic Knowledge Update 8

tients receiving concomitant methotrexate: A randomised phase III trial. Lancet 1999;354:1932-1939.

College of Rheumatology Subcommittee on Osteoarthritis Guidelines. Arthritis Rheum 2000;43:1905-1915.

Oliveria SA, Felson DT, Reed JI, Cirillo PA, Walker

AM: Incidence of symptomatic hand, hip, and knee osteoarthritis among patients in a health maintenance organization. Arthritis Rheum 1995;38:1134-1141.

Shojania K: Rheumatology: 2. What laboratory tests are

needed? CMAJ 2000;162:1157-1163.

Recommendations for the medical management of osteoarthritis of the hip and knee: 2000 update: American

244

Watterson JR, Esdaile JM: Viscosupplementation: Therapeutic mechanisms and clinical potential in osteoarthritis of the knee. J Am Acad Orthop Surg 2000;8:277-284.

American Academy of Orthopaedic Surgeons

Chapter

22

Connective Tissue Diseases

Jacques L. DAstous, MD, FRCSC
Kristen L. Carroll, MD

Introduction

Hypermobility

Connective tissue disorders such as Ehlers-Danlos syndrome (EDS), osteogenesis imperfecta (OI), and
Marfan syndrome are phenotypic expressions of inherited collagen and/or fibrillin anomalies. The errors of the
extracellular matrix of the involved tissues in these diseases lead to pathology within the bone, vascular system, and viscera. Although the genetics and pathophysiology are different, the underlying similarity of
abnormal genetic expression leading to systemic disease
is unmistakable.

The hypermobility subtype (formerly type III) is the most

orthopaedically debilitating of all forms of EDS, and the
most likely to require orthopaedic surgical intervention.
It is an autosomal dominant variant of unknown molecular basis; therefore, no diagnostic test presently exists.
Clinically, these patients present with soft skin and both
small and large joint hypermobility; recurrent and/or
chronic dislocations may also be present. Up to 20% of
patients may have aortic root anomalies. Multidirectional
instability of the shoulder, patella subluxation, and chronic
ankle instability are common. A snapping of the iliotibial
band is often misinterpreted as a dislocated hip, although
true dislocations can occur (albeit rarely) even in the presence of radiographically normal bony architecture (Figure 2). Surgical interventions such as capsular shift or plication should be undertaken only if physical therapy is
ineffective. Excellent results are elusive. Bracing is frequently used, especially of the hands because of finger instability. Nearly 90% of patients have debilitating pain, often with abnormal gait or the need for assistive devices.

Ehlers-Danlos Syndrome
EDS includes a group of the most common heritable
disorders of the connective tissue. Although all have
features of skin and joint hypermobility, they are a heterogenous group presently classified based on genetic
transmission, biochemical anomaly, and major and minor clinical findings. The previous classification system
of 11 different types has been reduced to 6 in the
present Villefranche classification system (Table 1).

Classic
Classic EDS includes the former type I (gravis) and
type II (mitis) subsets. The classic type combined with
hypermobility and vascular types are by far the most
common, comprising more than 90% of all types of
EDS. The classic form is an autosomal dominant condition with 40% to 50% of patients manifesting a
COL5A1 or COL5A2 gene mutation of type V collagen. The major clinical diagnostic criteria include skin
hypermobility, widened atrophic scars, and joint hypermobility. Up to 33% of patients will have aortic root dilatations and therefore, echocardiography is suggested.
Other minor criteria include velvety skin, spheroids, hypotonia, and tissue fragility. Nearly 30% of patients will
have scoliosis, with most having the thoracic or thoracolumbar type. Tower vertebra may be seen (Figure 1).
More than 50% of these patients may have chronic
musculoskeletal pain.

American Academy of Orthopaedic Surgeons

Vascular
The vascular subtype (formerly type IV) is most often autosomal dominant and occasionally autosomal recessive.
Biochemically, a defect in the COL3A1 gene for type III
collagen is present in more than 90% of the patients.As in
the other EDS subtypes, hypermobility of the small joints
and clubfoot may be present. However, the hallmark of
the vascular subtype is malignant involvement of the viscera and arteries. These patients have thin, translucent
skin, and may experience spontaneous rupture of the
bowel, uterus, or large arteries. Aortic root dilatation is
present in more than 75% of patients.Twenty-five percent
of women die during pregnancy because of complications,
most often uterine rupture. Life expectancy is 45 to
50 years.

Kyphoscoliosis
This rare subtype (previously type VI) is an autosomal
recessive disorder with a biochemical deficiency in lysyl

245

Orthopaedic Knowledge Update 8

Connective Tissue Diseases

TABLE 1 | Ehlers-Danlos Classification

Villefranche
Classification
(1998)

Berlin Classification
(1988)

Genetics

Major Symptomatic Criteria

Biochemical Defects (Minor Criteria)

COL5A1, COL5A2 mutations (40% to
50% of families)
Mutations in type V collagen
Unknown

Classic

Type I (gravis)
Type II (mitis)

AD

Hyperextensible skin, atrophic scars, joint

hypermobility

Hypermobility

Type III (hypermobile)

AD

Vascular

Type IV (vascular)

AD
(rarely) AR

Kyphoscoliosis

Type VI (ocular-scoliotic)

AR

Arthrochalasis

Type VIIA, VIIB

AD

Dermatosparaxis

Type VIIIC

AR

Velvety soft skin, small and large joint

hypermobility and tendency for
dislocation, chronic pain, scoliosis
COL3A1 mutation, abnormal type III
Arterial, intestinal and uterine fragility,
rupture, thin translucent skin, extensive
collagen structure of synthesis
bruising
Severe hypotonia at birth, progressive
Lysyl hydroxylase deficiency, mutations
infantile scoliosis, generalized joint laxity, in the PLOD gene
scleral fragility, globe rupture
Congenital bilateral hip dislocation,
Deletion of type I collagen exons that
hypermobility, soft skin
encode for N-terminal propeptide
(COL1A1, COL1A2)
Severe sagging or redundant skin
Mutations in type I collagen N-terminal
peptidase

AD = autosomal dominant; AR = autosomal recessive

Figure 2 Radiograph of an 18-year-old woman who required three closed reductions

of the hip following ground level falls. As seen here and confirmed by MRI, the bony
(and labral) architecture of the hip appear normal.

Figure 1 Radiograph of the lateral lumbar spine of a 16-year-old girl with classic
(previously type II) EDS. Note the high vertebral height known as tower vertebra.

246

hydroxylase, an enzyme that modifies collagen. Major

diagnostic criteria include muscle hypotonia at birth and
a progressive scoliosis; most of the curves are double
thoracic in pattern. There is generalized joint laxity. Ocular findings such as scleral fragility and globe rupture
are found in 50% of patients. Minor criteria include
bruising, tissue fragility, osteopenia, and arterial rupture.
Kyphoscoliosis EDS may be confused with Marfan syn-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 22 Connective Tissue Diseases

Figure 3 A and B, Radiographs of a 3-year-old girl with bilateral developmental dysplasia of the hip, a family history of EDS, and arthrochalasis. After bilateral reduction of the
hip through an anterior approach and capsulorrhaphy, the left hip has resubluxated.
drome as patients have scoliotic, cardiac, and ocular involvement and often a tall, thin body habitus.

Arthrochalasis
This extremely rare form of EDS was previously classified as types VIIA and VIIB. This autosomal dominant
form is characterized by a deficiency in the pro- I or
pro- 2 collagen type I chains at their N-terminal end.
Children with arthrochalasis type EDS are born with bilateral developmental dysplasia of the hip that is often
recalcitrant to surgical intervention (Figure 3). They
may also, as minor criteria, display skin hyperextensibility, muscle hypotonia, osteopenia, and kyphoscoliosis.
Easy bruising and tissue fragility also exist.

Dermatosparaxis
Dermatosparaxis is a rare, autosomal recessive form of
EDS (formally known as type VIIC) notable for a deficiency of procollagen I N-terminal peptidase. Patients
have redundant, severely fragile, and often sagging skin.
Premature rupture of fetal membranes, large hernias,
and easy bruising also may be seen.

Other Forms
In addition to the six major forms of the Villefranche
classification, other forms of EDS exist encompassing
previous types V, VIII, and X and include those with periodontal friability (VIII), and the poor clotting/
fibronectin deficient type (X). In addition, a form of
EDS with symptoms similar to the classic form exists,
but with an X-linked inheritance pattern. This was formerly type V EDS and has been described in only a single family.

American Academy of Orthopaedic Surgeons

Osteogenesis Imperfecta
OI is a hereditary condition resulting from an abnormality in type I collagen that is manifested by an increased fragility of bones and low bone mass (osteopenia). It is estimated that in the United States alone,
20,000 to 50,000 people are affected with OI.
Eighty percent to 90% of patients with OI can be
grouped into the Sillence type I to IV categories and
have mutations of one of the two type I collagen genes.
(Recently, types V, VI, and VII have also been added).
The COL1A gene encodes the pro-1(I) protein chain
and the COL2A gene encodes pro-2(I) protein chain
of type I procollagen. The etiologies of the remaining
10% to 20% remain unclear.
Iliac crest biopsies of patients with OI show a decrease in cortical widths and cancellous bone volume,
with increased bone remodeling. There is a direct relationship between the increase in bone turnover and the
severity of the disease.
The clinical features of OI are osteopenia, bone fragility, and fractures and may include some or all of the
following: joint laxity, gray-blue sclerae, dentogenesis
imperfecta, premature deafness, kyphoscoliosis, and
basilar invagination (prolapse of the upper cervical
spine into the base of the skull). In OI, the bones can be
biologically soft, producing deformities such as protrusio acetabuli in the pelvis and basilar invagination at
the craniovertebral junction. It is believed that repeated
microfractures in OI bone combined with healing and
remodeling is responsible for the soft bones.
Symptoms of basilar invagination in OI typically occur in the third and fourth decades but may be present
during the teenage years. These symptoms include brainstem dysfunction such as apnea, altered consciousness,

247

Orthopaedic Knowledge Update 8

Connective Tissue Diseases

TABLE 2 | Types of OI Modified From Sillence

Type

Transmission

Biochemistry

Orthopaedic Manifestation

Miscellaneous

IA
IB

AD

50% amount of type I

collagen

Mild to moderate bone fragility,

osteoporosis, late fractures

II
II A
II B
II C
II D

AD, AR and mosaic

Unstable triple helix

III

AD and AR

Abnormal type I collagen

IV A
IV B

AD
AD

Shortened pro-1 chain

Multiple intrauterine fractures, extreme

bone fragility, beaded ribs, broad bones,
platyspondyly
Severely osteopenic with generally well
formed skeleton, normally shaped vertebrae
and pelvis
Progressive deforming phenotype, severe
bone fragility with fractures at birth,
popcorn epiphyses/metaphyses, scoliosis
and platyspondyly, severe osteoporosis,
extreme short stature
Mild-moderate bone fragility, osteoporosis,
bowing of long bones, scoliosis

Blue sclerae, hearing loss, easy bruising,

normal life expectancy
More severe than IA with dentinogenesis
imperfecta, slightly decreased life
expectancy
Usually lethal in perinatal period, delayed
ossification of skull, blue sclerae

AD

Irregular mesh-like lamellar

bone, lack of parallel
organization of collagen
fibers
Mineralization defect,
abundance of osteoid,
fish-scale organization of
lamellae

VI

VII

AR

Hearing loss, blue sclerae becoming less

blue with age, shortened life expectancy,
dentinogenesis imperfecta, relative
macrocephaly with triangular faces
Light sclerae, normal hearing, normal
dentition
Same as IVA plus dentogenesis imperfecta
present, slight decreased life expectancy

Hypertrophic callus after fractures,

calcification of interosseous membrane

Low bone mineral density, frequent

fractures, vertebral compressions, long
bone deformities

Normal vitamin D levels, absence of

hypocalcemia and hypophosphatemia

Rhizomelic limb shortening

AD = autosomal dominant; AR = autosomal recessive

lower cranial nerve deficits, myelopathy, and ataxia. Neurologic abnormalities are consequences of direct neural
compression, altered cerebrospinal fluid flow, or vascular
compromise. The recommended treatment of basilar invagination in OI consists of extensive removal of bony
compression by a transoral approach followed by a posterior fusion and posterior rigid fixation that transfers the
weight of the head to the thoracic spine.
Although it may be difficult, determining the clinical
distinction between children with OI and those with
nonaccidental trauma has obviously important implications. Skin biopsies and fibroblast cultures may be helpful, but are only positive in 80% of patients with
type IV OI (the most commonly confused with nonaccidental trauma) (Table 2).

Nonsurgical Treatment
The medical treatment of OI involves strategies to improve bone mass. Recombinant human growth hormone

248

has been used in the past because of its anabolic effects

on bone; however, clinical studies showed no increase in
bone mass or change in natural history. For the past
10 years, bisphosphonates have been used to treat patients with types III and IV OI. Bisphosphonates decrease the resorption of bone by suppressing the activity
of osteoclasts. Pamidronate, an injectable bisphosphonate, increases cortical bone thickness and, in severe
forms of OI, improves overall bone mass. Pamidronate
therapy decreases the incidence of fractures, relieves
chronic bone pain, increases activity levels, decreases
the reliance on mobility aids, and increases the height of
the collapsed vertebral bodies. Unfortunately, there has
been no decrease in the incidence of scoliosis. Radiographically, pamidronate therapy creates growth lines in
the bone (Figure 4). These radiodense areas of bone
probably represent the inhibition of osteoclastic resorption, whereas the clear areas between the lines represent the interval growth between treatment cycles.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 22 Connective Tissue Diseases

Figure 4 Growth lines secondary to pamidronate therapy.

Newer, more potent forms of bisphosphonates (for example, zoledronate) that require fewer injections are being evaluated. There is concern, however, that bisphosphonates may make bone more brittle. Bone marrow
transplantation, which introduces normal marrow stem
cells that could potentially differentiate into normal osteoblasts, has been used with some success for severely
affected patients with OI. Problems of graft rejection
and graft versus host reactions limit this approach.
Orthotic devices are indicated to stabilize lax joints,
prevent progression of deformity and fractures, and allow early weight bearing following surgical intervention.

Surgical Treatment
The mainstay of surgical treatment of patients with OI
is realignment osteotomy, which is performed to improve the mechanical axis of appendicular bones. In addition to decreasing the risk of fracture, the goals and
principles of realignment surgery are to allow for early
weight bearing and to achieve union. The realignment
osteotomies should be performed through small incisions in an attempt to preserve the blood supply, and
then stabilization should be achieved with intramedullary devices, of which there are both telescopic and nontelescopic forms. The most commonly used nontelescopic devices include Rush rods and Williams rods. The
use of telescopic rods appears to decrease the incidence

American Academy of Orthopaedic Surgeons

Figure 5 A and B, The Fassier-Duval telescopic nail obtains purchase in the greater
trochanter and femoral epiphysis without requiring an arthrotomy. A, The limb is shown
shortly after corrective osteotomy and instrumentation. B, Note interval extension of
rod with growth.

of re-rodding to accommodate for growth, (51% for

simple rods versus 27% for telescopic rods) and therefore are preferred for use when possible. The Sheffield
rod and the newly developed Fassier-Duval rod represent an improvement over the older Bailey-Dubow rod
(Figure 5).
The treatment of spinal deformities in OI can be
challenging. It is impossible to effectively push on the
spine through the ribs because of the fragility of the rib
cage and the possibility of further aggravating the thoracic deformity. When the curves approach 45 in the
mild forms of OI and 30 to 35 in the severe forms, a
posterior spinal fusion with segmental instrumentation
to prevent progression of the curve should be considered (Figure 6). Because of bone fragility, it may be necessary to reinforce the upper and lower hook claws with
bone cement or use Mersilene tape instead of sublaminar wires. Because of the paucity of bone in the iliac
crest, allograft bone should be used. Anterior spinal fu-

249

Connective Tissue Diseases

Orthopaedic Knowledge Update 8

Figure 6 Preoperative (A) and postoperative (B)

radiographs showing scoliosis in a patient with
type IV OI treated by posterior spinal fusion.
sion may be considered in the very young patient to
prevent crankshafting of the spine. Surgery for basilar
invagination is indicated for patients with radiographic
progression or with neurologic deficits resulting from
brainstem and high cervical cord compression.

Marfan Syndrome
Marfan syndrome is an inherited connective tissue disorder that is usually transmitted as an autosomal dominant trait. Approximately 25% of cases arise from new
mutations. The incidence is roughly 1:10,000 with approximately 200,000 people in the United States having
Marfan syndrome. There is no ethnic or gender predilection.
The genetic mutation of Marfan syndrome is on the
fibrillin-1 (FBN1) gene located on chromosome 15q21.
More than 135 mutations in the FBN1 gene have been
identified. The genetic heterogenicity explains the pleiotropic manifestations of Marfan syndrome with variable
phenotypic expression. FBN1 is the main component of
the 10 to 20 nm extracellular microfibrils that are important for elastogenesis, elasticity, and homeostasis of elastic fibers.
The clinical features include increased height, disproportionately long, gracile limbs (dolichostenomelia),
arachnodactyly with a positive wrist (Walker) and
thumb (Steinberg) sign, an arm span greater than height
(span to height ratio > 1.05), anterior chest wall defor-

250

mity (pectus excavatum > pectus carinatum), generalized ligamentous laxity, severe planovalgus, and/or long
thin feet with a disproportionately long great toe. Scoliosis is seen in 60% to 70% of patients (right thoracic
lordotic curves are the most common); males and females are equally affected. The head and neck reveal a
high arched palate, myopia, corneal flatness, dislocation
of lenses (ectopia lentis) and iridodonesis (tremor of the
iris secondary to lens dislocation). Cardiac manifestations include mitral valve prolapse, mitral regurgitation,
dilatation of aortic root, aortic regurgitation, aortic dissection, and aortic aneurysm. Marfan syndrome may
lead to spontaneous pneumothorax secondary to lung
bullae and striae distensae.
A severe neonatal form of Marfan syndrome exists.
These children are identified within the first few months
of life by serious cardiac abnormalities and congenital
contractures. This form of the syndrome is believed to
result from a spontaneous mutation in the FBN1 gene.
The diagnosis of Marfan syndrome is based on family history, evaluation of six organ systems, and molecular data. The Ghent system outlines diagnostic criteria
for both the index patient (the first case in a family) and
for subsequent relatives (Table 3). To establish the diagnosis of Marfan syndrome in the index case, the patient
must fulfill one major criteria in at least two different
organ systems and have involvement of a third system.
To establish the diagnosis in a relative of an index case,

American Academy of Orthopaedic Surgeons

Chapter 22 Connective Tissue Diseases

Orthopaedic Knowledge Update 8

TABLE 3 | Marfan Syndrome

System

Major Criteria

Minor Criteria

Musculoskeletal*

Pectus carinatum; pectus excavatum requiring surgery;

dolichostenomelia; wrist and thumb signs; scoliosis > 20
or spondylolisthesis; reduced elbow extension; pes planus;
protrusio acetabuli
Ectopia lentis

Moderately severe pectus excavatum; joint hypermobility;

highly arched palate with crowding of teeth; facies
(dolichocephaly, malar hypoplasia, enophthalmos,
retrognathia, down-slanting palpebral fissures)
Abnormally flat cornea; increased axial length of globe;
hypoplastic iris or hypoplastic ciliary muscle causing
decreased miosis
Mitral valve prolapse regurgitation

Ocular

Cardiovascular
Family/Genetic history

Skin and integument||

Dura
Pulmonary||

Dilatation of ascending aorta aortic regurgitation,

involving sinuses of Valsalva; or dissection of ascending
aorta
Parent, child, or sibling meets diagnostic criteria; mutation None
in FBN1 known to cause Marfan syndrome; or inherited
haplotype around FBN1 associated with Marfan syndrome
in family
None
Stretch marks not associated with pregnancy, weight gain,
or repetitive stress; or recurrent incisional hernias
Lumbosacral dural ectasia
None
None
Spontaneous pneumothorax or apical blebs

*Two or more major or one major plus two minor criteria required for involvement
At least two minor criteria required for involvement
One major or minor criterion required for involvement
One major criterion required for involvement
||One minor criterion required for involvement
(Adapted from Miller NH: Connective tissue disorders, in Koval KJ (ed): Orthopaedic Knowledge Update: 7. Rosemont, IL, American Academy of Orthopaedic Surgeons, 2002, pp 201-207.)

a major criterion in the family history needs to be

present, the individual must have one major criterion in
any organ system and involvement of a second organ
system.
Marfan syndrome must be differentiated from the
MASS phenotype, an acronym assigned to the following
clinical features: mitral valve prolapse, aortic root diameter at the upper limits of normal, stretch marks, and
skeletal features of Marfan syndrome (including scoliosis, pectus excavatum or carinatum, and joint hypermobility). The aortic root diameter may be at the upper
limits of normal for body size, but there is no progression to aneurysm or predisposition to dissection. Ectopia lentis does not occur. MASS patients therefore have
a better prognosis than those with classic Marfan syndrome. This condition can be inherited within families
and has been shown to result from mutations in the
FBN1 gene.
In addition to the MASS phenotype, the differential
diagnosis of Marfan syndrome includes congenital contractural arachnodactyly (Beals syndrome), Sticklers
syndrome, arthrochalasis type EDS, hypermobility syndrome, and homocystinuria.

American Academy of Orthopaedic Surgeons

There are no specific laboratory investigations for

Marfan syndrome, although genetic testing is now available in some centers. Prenatal molecular diagnosis is
possible only if the familys mutation is known or several affected family members are available for genetic
linkage analysis.
Radiographic analysis of Marfan syndrome may reveal the presence of scoliosis, protrusio acetabuli, chondrolysis, and high-grade spondylolisthesis (Figure 7).
MRI reveals dural ectasia in more than 60% of patients.
Other important investigations include transesophageal
echocardiogram, electrocardiogram, slit lamp examination, and keratometry.
The medical treatment of Marfan syndrome includes
the use of -blockers to slow the progression of aortic
dissection. Aortic replacement is indicated if aortic diameter is greater than 5 cm. Pregnant women with aortic dilatation should have a cesarean section at 38
weeks. The success rate for bracing of spinal deformities
is much lower than in adolescent idiopathic scoliosis.
Curves greater than 25 in children with Risser grade II
skeletal maturity will likely require surgery despite
bracing. Preoperative imaging of the spine with MRI to
identify dural ectasia and CT to assess fixation points

251

Connective Tissue Diseases

Orthopaedic Knowledge Update 8

creased pain, increased ease of drug administration, and
a reduction in fractures and deformity. For patients with
Marfan syndrome, advances in cardiothoracic surgery
have doubled the lifespan, and new drugs such as deoxycycline and losartan may decrease the incidence of aortic aneurysms. Ophthalmologic interventions and rigorous slit lamp screening have decreased the incidence of
blindness. As the knowledge of the genetics and biochemical anomalies of connective tissue disorders expands, patient care improves.

Annotated Bibliography
Ehlers-Danlos Syndrome

Figure 7 Radiograph of the hip of a 13-year-old girl with Marfan syndrome showing
classic protrusio acetabuli.

are essential. Preoperative cardiopulmonary evaluation

is mandatory. There is an increased incidence of pseudarthrosis in posterior spinal fusion. Closure of the triradiate cartilage to treat protrusio acetabuli also has
been proposed.
If untreated, the life expectancy of a patient with
Marfan syndrome is 30 to 40 years, with death resulting
from aortic dissection, aortic rupture, or valvular induced cardiac failure. With modern cardiovascular treatment regimens, the average lifespan is now 70 years. It is
important to remember that the tall, agile, athletic patient with Marfan syndrome is at risk for aortic dissection and sudden death. It is imperative to educate patients about the signs and symptoms of aortic and
cardiac disease, and to counsel them on activity restriction and the possibility of genetic transmission.

Future Directions
As recent studies indicate, an advanced understanding
of the genetic basis behind inherited connective tissue
disorders will potentially allow for intervention at a molecular level with gene therapy. Although a safe mode of
introducing genetically modified collagen for patients
with EDS, osteoblasts for those with OI, or fibrillin for
those with Marfan syndrome is presently elusive, it
should become a reality. Several new strains of Marfan
mice have recently been developed with mutant
fibrillin-1 proteins to further clarify the mechanism of
disease and eventually its treatment.
While awaiting interventions that address the molecular causes of connective tissue disease in these patients,
the treatment of their symptoms has become more comprehensive. For patients with OI, advances in the pharmacologic treatment has and will continue to lead to de-

252

Wenstrup RJ, Meyer RA, Lyle JS, et al: Prevalence of

aortic root dilatation in the Ehlers-Danlos syndrome.
Genet Med 2002;4:112-117.
Twenty-eight percent of all EDS patients have aortic root
dilatation (25% in classic form, 17% in the hypermobile
form). Patients without the vascular form of EDS also should
be carefully monitored for this complication.

Osteogenesis Imperfecta
Cole WG: Advances in osteogenesis imperfecta. Clin
Orthop 2002;401:6-16.
This article presents a review of the current state of
knowledge regarding collagen mutations in OI.

Glorieux FH: The use of bisphosphonates in children

with osteogenesis imperfecta. J Pediatr Endocrinol
Metab 2001;14(suppl 6):1491-1495.
One of the more recent reviews on the use of bisphosphonates for patients with OI is presented by the group who first
studied the use of bisphosphonates for this condition.

Osteogenesis imperfecta. NIH Medline Plus Website.

Available
at:
http://www.nlm.nih.gov/medlineplus/
osteogenesisimperfecta.html. Accessed June, 2004.
A complete, up-to-date review of all aspects of OI is presented, with numerous links to related websites.

Osteogenesis Imperfecta Foundation Website. Available

at: http:/www.oif.org/. Accessed June, 2004.
The OI foundations official site presents a wealth of information for health professionals and families dealing with all
aspects of OI.

Pattekar M, Caccrarelli A: Osteogenesis imperfecta.

eMedicine
Website.
Available
at:
http:/
www.emedicine.com/ped/topic1674.htm. Accessed June,
2004.
This thorough review of OI is available online and includes excellent photographs.

American Academy of Orthopaedic Surgeons

Chapter 22 Connective Tissue Diseases

Orthopaedic Knowledge Update 8

Zeitlin L, Fassier F, Glorieux FH: Modern approach to
children with osteogenesis imperfecta. J Pediatr Orthop B
2003;12:77-87.

Robinson PN, Booms P: The molecular pathogenesis of

the Marfan syndrome. Cell Mol Life Sci 2001;58:16981707.

This article reviews the molecular pathogenesis of OI. It also

summarizes the most recent medical and surgical management
of patients with OI and describes three new types of OI.

This article provides an overview of the clinical aspects of

Marfan syndrome and current knowledge concerning the
pathogenesis of this disorder.

Marfan Syndrome

Future Directions

Channell K, Washington E: Marfan syndrome. eMedicine website. Available at: http://www.emedicine.com/

orthoped/topic414.htm. Accessed June, 2004.

Turgeman G, Aslan H, Gazit Z, Gazit D: Cell-mediated

gene therapy for bone formation and regeneration. Curr
Opin Mol Ther 2002;4:390-394.

A complete review of all aspects of Marfan syndrome is

presented.

This article reviews the most recent studies related to cellmediated gene therapy for bone formation and regeneration.

Chen H: Marfan syndrome. eMedicine Website. Availabe at: http://www.emedicine.com/PED/topic1372.htm.

Accessed June, 2004.

Classic Bibliography

All aspects of Marfan syndrome are reviewed.

Giampietro PF, Raggio C, Davis JG: Marfan syndrome:

Orthopedic and genetic review. Curr Opin Pediatr 2002;
14:35-41.
This review summarizes recent developments in the genetic and orthopaedic aspects of Marfan syndrome.

Jones KB, Erkula G, Sponseller PD, Dormans JP: Spine

deformity correction in Marfan syndrome. Spine 2002;
27:2003-2012.
A recent review of 39 patients with Marfan syndrome
treated for spinal deformities with a discussion on preoperative planning, complications, and outcomes is presented.

Beighton P, De Paepe A, Steinman B, Tsipouras P, Wenstrup RJ: Ehlers-Danlos syndromes: Revised Nosology:
Ehlers-Danlos National Foundation (USA) and EhlersDanlos Support Group (UK), Villefranche, 1997. Am J
Med Genet 1998;77:31-37.
Byers PH: Osteogenesis imperfecta: Perspectives and
opportunities. Curr Opin Pediatr 2000;12:603-609.
Collod-Beroud G, Le Bourdelles S, Ades L, et al: Update of the UMD-FBN1 mutation database and creation
of an FBN1 polymorphism database. Hum Mutat 2003;
22:199-208.
De Paepe A, Devereux RB, Dietz HC, Hennekam RC,
Pyeritz RE: Revised diagnostic criteria for the Marfan
syndrome. Am J Med Genet 1996;62:417-426.

Loeys BL, Matthys DM, de Paepe AM: Genetic fibrillinopathies: New insights in molecular diagnosis and clinical management. Acta Clin Belg 2003;58:3-11.

Dietz HC, Mecham RP: Mouse models of genetic diseases resulting from mutations in elastic fiber proteins.
Matrix Biol 2000;19:481-488.

This article provides an overview of the current diagnostic

criteria and medical management of fibrillinopathies, estimates the role of fibrillin-1 mutation analysis, sheds new light
on genotype-phenotype correlations, and summarizes new insights on the pathogenesis of this disorder based on mouse
models.

Harkey HL, Capel WT: Bone softening diseases and disorders of bone metabolism, in Dickman CA, Spetzler
RF, Sonntag VKH (eds): Surgery of the Craniovertebral
Junction. New York, NY, Thieme, 1998, pp 197-202.

The National Marfan Foundation Website. Available at:

http://www.marfan.org/. Accessed June 2004.
The latest information concerning the genetic aspects of
Marfan syndrome is presented.

Nollen GJ, Groenink M, van der Wall EE, Mulder BJ:

Current insights in diagnosis and management of the
cardiovascular complications of Marfan syndrome.
Cardiol Young 2002;12:320-327.
This article presents a review of new advances in surgical
and medical treatment for cardiovascular complications of
Marfan syndrome that have dramatically improved the prognosis.

American Academy of Orthopaedic Surgeons

Harkey HL, Crockard HA, Stevens JM, Smith R, Ransford AO: The operative management of basilar impression in osteogenesis imperfecta. Neurosurgery 1990;27:
782-786.
Joseph KN, Kane HA, Milner RS, Steg NL, Williamson
MB, Bowen JR: Orthopaedic aspects of the Marfan phenotype. Clin Orthop 1992;277:251-261.
Kocher MS, Kasser JR: Orthopaedic aspects of child
abuse. J Am Acad Orthop Surg 2000;8:10-20.
Lubicky JP: The spine in osteogenesis imperfecta, in
Weinstein SL (ed): The Pediatric Spine: Principles and
Practice, ed 2. New York, NY, Raven Press, 2001.

253

Connective Tissue Diseases

Orthopaedic Knowledge Update 8

Niyibizi C, Smith P, Mi Z, Robbins P, Evans C: Potential

of gene therapy for treating osteogenesis imperfecta.
Clin Orthop 2000;379(suppl):S126-S138.

Sponseller PD, Bhimani M, Solacoff D, Dormans JP: Results of brace treatment of scoliosis in Marfan syndrome. Spine 2000;25:2350-2354.

Pyeritz RE: The Marfan syndrome. Annu Rev Med 2000;

51:481-510.

Stanitski DF, Nadjarian R, Stanitski CL, Bawle E, Tsipouras P: Orthopaedic manifestations of Ehlers-Danlos
syndrome. Clin Orthop 2000;376:213-221.

Sillence DO, Senn A, Danks DM: Genetic heterogeneity

in osteogenesis imperfecta. J Med Genet 1979;16:101116.

254

American Academy of Orthopaedic Surgeons

Chapter

23

Shoulder and Elbow Injuries in the

Throwing Athlete
Patrick J. McMahon, MD
Michael Gilbart, MD
Brian S. Cohen, MD
Anthony A. Romeo, MD

Introduction
Athletic injuries to the upper extremity are much less
common and receive significantly less attention in the
literature than lower extremity injuries. Unfortunately,
compared with lower extremity injuries, upper extremity
injuries can be just as devastating to the overhead athlete. The shoulder and elbow of the overhead athlete are
subject to a tremendous amount of force during overhead throwing that an average patients shoulder and
elbow never experience. Although completely different
in their anatomic structure (shoulder: ball and socket
joint; elbow: hinged joint), both areas are most susceptible to injury during the late cocking and early acceleration phases of throwing. It is during these phases of
throwing that the tensile and compressive forces peak,
placing potentially pathologic stresses on both areas. As
a result of the repetitive nature of overhead sports,
overuse disease can exacerbate an essentially benign
problem into a full-blown disabling condition.

Shoulder
History and Physical Examination
In overhead throwing athletes, grouping symptoms according to a specific phase of throwing, as well as localizing the pain to the front or back of the shoulder, may
help in determining the etiology of the injury (Table 1).
Physical examination of the shoulder, as with all areas,
should begin with visual inspection and should include a
comparison of the contralateral or normal side. Examination of the joint above (cervical spine) and the joint
below (elbow) should also be included in a comprehensive evaluation of the patient with a shoulder disorder.
The general posture of the shoulder should be assessed.
It is not uncommon for the overhead athlete to have
overdevelopment of the shoulder musculature and humeral head in the throwing extremity. Evaluation of
both active and passive motion is critical in assessing
shoulder function; specifically, observing the patient
from behind to assess scapular motion particularly looking for scapular winging, which can be the result of
weakness of the scapular stabilizers. Increased external

American Academy of Orthopaedic Surgeons

rotation and decreased internal rotation can be a normal finding in an athletes throwing shoulder. This increase in external rotation may be related to greater
laxity of the anterior ligaments from the deforming
forces of overhead throwing, or an increase in the
amount of humeral retroversion (little leaguers shoulder) that represents an adaptive change that probably
occurs through the physis. This change has two benefits:
first, it allows for increased external rotation during
throwing; and second, it acts as a protective mechanism
against impingement of the greater tuberosity on the
posterior superior glenoid rim during throwing. A
greater loss of active motion compared with passive motion within the same shoulder may be the result of pain,
rotator cuff pathology, or neurologic compromise. Palpation of specific areas, such as the acromioclavicular
(AC) joint and biceps, may help to further localize contributing pathology, whereas specific provocative testing
such as the apprehension and relocation tests, the Neer
and Hawkins impingement signs, and the OBriens and
Speeds tests can also contribute to the establishment of
the correct diagnosis.

Internal Impingement
Internal impingement is defined as abnormal contact
between the rotator cuff undersurface and the posterosuperior glenoid rim, resulting in tearing of the rotator
cuff and labrum. Presently, there is no consensus on the
treatment of this condition. The etiology of posterosuperior glenoid impingement has also been the source of
much debate. Internal impingement has been attributed
to anterior microinstability and tightness of the capsule
posteriorly. Others argue that posteroinferior capsular
contracture results in posterosuperior instability and a
peel-back to the superior labrum (a posterior subtype
[B] of a type II superior labrum anterior and posterior
[SLAP] lesion) and a partial-thickness tearing of the rotator cuff.
On physical examination, the patients pain is usually posterior and will be reproduced with the arm in
the abducted, externally rotated, extended position. Pa-

257

Orthopaedic Knowledge Update 8

Shoulder and Elbow Injuries in the Throwing Athlete

Table 1 | Most Common Causes of Shoulder Pain

Anterior Shoulder Pain
Rotator cuff tendinitis
Anterior shoulder instability
Biceps tendinitis/subluxation
Anterior deltoid strain
Labral pathology
Posterior Shoulder Pain
Rotator cuff tendinitis
Posterior shoulder laxity
Posterior labral pathology
Posterior capsular tightness
Teres minor strain
Internal impingement
Quadrilateral space syndrome
Throwers exostosis
Long thoracic nerve injury
Supracapular nerve injury
Osteochondritis dissecans

tients may experience pain during an apprehension maneuver, which may be relieved by the relocation test.
Radiographic evaluation will show cystic and sclerotic changes in the greater tuberosity of nearly half of
the patients with internal impingement and there may
be evidence of rounding of the posterior glenoid rim in
one third of patients; both are nonspecific findings. MRI
evaluation may show partial undersurface rotator cuff
tears and often reveal a pathologic insertion of the supraspinatus tendon. A magnetic resonance arthrogram
with an abduction-external rotation view is a useful diagnostic test, revealing posterior-superior labrum abnormalities with an associated kissing lesion of articularsided rotator cuff fraying.
Initial treatment focuses on rehabilitation protocols
that improve the function of the dynamic stabilizers
and, at the same time, stretch the posterior capsule. Another approach, based on study of baseball pitchers, focused on adjusting throwing mechanics to avoid shoulder extension beyond the plane of the scapula during
the cocking phase and through most of the acceleration
phase. This phenomenon was termed hyperangulation.
The correction of this abnormality in the throwing mechanics of these patients resulted in elimination of their
symptoms.
Surgical treatment is reserved for those patients who
fail to improve following a well-executed rehabilitation
program. Arthroscopic treatment of this disorder focuses on dbridement of the partial-thickness rotator
cuff tear and associated posterior glenoid labral lesion.
Tightening of the anterior capsuloligamentous structures is also performed either by thermal capsulorrhaphy or with a suture plication technique. If internal
rotation of the abducted shoulder is diminished, a con-

258

comitant posterior capsular release may be necessary.

Following this procedure, a well-planned rehabilitation
program is needed that allows for tissue healing (6 to 8
weeks), followed by restoration of motion and strength
(8 to 12 weeks), and concluding with proprioception
training and a gradual return to sport-specific activities,
with a return to competitive throwing at about 4 to 6
months.
Another procedure that is available, but is used as a
last resort, is a humeral derotational osteotomy. This
procedure is reserved for patients in whom conservative
management and/or arthroscopic treatment have failed
and who are faced with either additional surgery or permanent cessation of throwing. This procedure has a high
complication and reoperation rate, and, therefore, requires that the patient be thoroughly educated on expected outcomes before surgery. The goal of the surgery
is to obtain a postoperative humeral retroversion of 30.
The amount of correction needed is determined from a
preoperative CT arthrogram. Overcorrection can cause
impingement of the lesser tuberosity on the anterior
glenoid with consequent loss of internal rotation. Postoperatively, external rotation is limited for the first 4
weeks to allow for subscapularis healing. Active range
of motion is permitted at 4 weeks, and resistive exercises are permitted at 8 weeks. Return to overhead
throwing activity is accomplished at about 6 months. Results following this procedure have been mixed. The
procedure was performed in 20 throwing athletes; 11 patients returned to their sport at their preinjury level.
However, the complication rate was high, with 16 patients requiring removal of hardware for pain in the biceps area or problems associated with anterior forward
elevation.

Instability
Anterior Glenohumeral Joint Instability
When the shoulder is repetitively forced beyond the
limit of its normal range of motion, displacement of the
articular surface of the humeral head from the glenoid
may occur. Anterior instability occurs in athletes when
the abducted shoulder is repetitively placed in the anterior apprehension position of external rotation and horizontal abduction. Two lesions may occur to the anteroinferior capsulolabrum with anterior instability. The first
is the Bankart lesion, which is an avulsion of the anteroinferior capsulolabrum from the anteroinferior glenoid
rim. The other lesion associated with recurrent anterior
instability in athletes results from stretching of the anteroinferior capsulolabrum.
The history given by the athlete as to when they experience their symptoms may help to determine the direction of their instability. Throwers with anterior instability will report a sensation that the shoulder wants to
slide out the front during the late cocking phase of

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

throwing, whereas athletes with posterior laxity will report shoulder discomfort and instability symptoms on
the follow-through phase of throwing. The dead arm
syndrome may occur after anterior joint instability. An
overhead throwing athlete with transient anterior instability may experience sudden, sharp pain associated
with loss of control of the extremity, numbness, and an
inability to throw the ball. These symptoms are transient, and usually resolve within a few seconds to minutes.
The apprehension test is useful in assessing for anterior instability. The examiner applies an anteriorly directed force to the humeral head with the shoulder in
abduction and external rotation. A positive test results
with the patient having apprehension that the joint will
dislocate. This maneuver mimics the position of instability and causes reflex guarding. The relocation test is positive if relief occurs when a posteriorly directed force is
applied to the humeral head.
Treatment, especially in patients who have symptomatic laxity as the result of repetitive microtrauma,
should focus on a rehabilitation program that avoids the
provocative position during the first 6 to 8 weeks and
incorporates a strengthening program to enhance the
function of the dynamic stabilizers. Following a period
of tissue healing, proprioception training followed by a
slow return to sport-specific activities should occur.
Surgical intervention should be reserved for patients
in whom conservative treatment is unsuccessful, or who
represent a population with a high risk of failure, such
as young patients with a history of a traumatic anterior
dislocation. Identification of the pathoanatomy is essential for relieving the athletes instability; a patientspecific selective capsular repair can be done, along with
refraining from addressing all instability problems with
a single type of surgical procedure. Arthroscopically, patients with a history of a traumatic anterior event can
have their Bankart lesion repaired anatomically with suture anchors. Treatment of specific capsuloligamentous
structures for predetermined directional instabilities can
be accomplished with arthroscopic techniques of suture
plication, thermal capsulorrhaphy, and, when necessary,
closure of the rotator interval.
Posterior Glenohumeral Joint Instability
Posterior instability results either from avulsion of the
posterior glenoid labrum from the posterior glenoid, or
stretching of the posterior capsuloligamentous structures. Posterior instability is often difficult to diagnose.
No single test has high sensitivity and specificity. During
the circumduction test, the patient actively rotates the
shoulder from the side to an internally rotated and
cross-body position, to a fully elevated position, to an
abducted and externally rotated position, and then returns the arm to the side. The examiner stands behind
the patient and palpates the posterior shoulder. If posi-

American Academy of Orthopaedic Surgeons

Chapter 23 Shoulder and Elbow Injuries in the Throwing Athlete

tive, the joint subluxates in the internally rotated, crossbody position, and reduction occurs with further rotation. The posterior apprehension test is performed by
applying a posterior force to the shoulder with the arm
in a forward flexed and internally rotated position. In
the Jahnke test, a posterior directed force is applied to
the forward flexed shoulder. The shoulder is then
moved into the coronal plane as an anterior-directed
force is applied to the humeral head. A clunk occurs as
the humural head reduces from the subluxated position.
The sulcus sign is used to evaluate inferior laxity and
associated inferior instability. The test is performed with
the athlete in a sitting position with the arm at the side.
A distraction force is applied longitudinally along the
humerus. The magnitude of displacement and the presence of discomfort or apprehension is compared with
that of the contralateral limb.
Initial treatment should be nonsurgical, consisting of
rehabilitation of the glenohumeral and scapular muscles, and surgical treatment should be considered if nonsurgical treatment fails. Posterior stabilization may be
performed with arthroscopic or open techniques, and includes repair of the posterior labral avulsion and plication of the stretched posterior capsuloligamentous structures. Special attention should be sought for associated
inferior instability and rotator internal lesions should be
repaired if present. Patient selection is important, because those with a traumatic etiology and no evidence
of generalized ligamentous laxity have the best prognosis.

Superior Labral Anterior and Posterior

Lesions
The SLAP lesion was first described in 1985. A more extensive injury pattern was subsequently identified that
involved the superior labrum; four basic types were described (Figure 1). Type II lesions were further subdivided into a type II lesion with an anterior component
(A), a posterior component (B), or a combined anteroposterior lesion (C). A peel-back phenomenon was described that is part of the pathoanatomy associated with
type II SLAP lesions that have a posterior component
(B or C), which is observed arthroscopically in the abducted and externally rotated position and may contribute to posterosuperior glenohumeral instability. Anatomic repair of the SLAP lesion has been shown to
eliminate the dynamic peel-back of the posterior superior labrum.
The diagnosis of a SLAP lesion is difficult because
the injury may be caused by either the repetitive nature
of overhead throwing, or as the result of a traumatic
event. SLAP lesions are commonly associated with
other injuries and usually have overlapping, nonspecific
symptoms. The patient may have difficulty describing
their pain. The discomfort is typically associated with

259

Orthopaedic Knowledge Update 8

Shoulder and Elbow Injuries in the Throwing Athlete

Figure 1 SLAP classification. A, Type I: fraying and degeneration of the superior labrum with a normal biceps tendon anchor. B, Type II: pathologic detachment of the labrum and
biceps anchor from the superior glenoid. C, Type III: vertical tear of the superior labrum analogous to a bucket-handle tear. D, Type IV: tear extending into biceps tendon.
(Reproduced with permission from Snyder SJ, Karzel RP, Del Pizzo W, et al: SLAP lesions of the shoulder. Arthroscopy 1990;6:274.)

the late cocking phase of throwing as the arm begins to

accelerate forward. The patient may feel a sudden, sharp
pain in the extreme abducted and externally rotated position followed by a dead arm sensation. These symptoms usually result in an inability to throw with preinjury velocity.
No finding on physical examination has been found
to be specific for identifying patients with a SLAP lesion. Provocative tests that have been suggestive of the
existence of superior labral pathology include Speeds
test, the OBrien test, and the Jobe relocation test. The
OBrien test (active compression test) may be the most
useful. The internally rotated shoulder is forward flexed
to 90 and brought across the body in horizontal abduction about 10. The test is positive if the patient has pain
with resisted forward flexion that is relieved by external
rotation of the shoulder. It should be noted that a positive relocation test for patients with superior labral pathology is highlighted by posterosuperior glenohumeral
joint pain with the arm in the abducted and externally
rotated position that is relieved by the relocation maneuver. Imaging studies including plain radiographs and
MRI evaluation should be done to rule out coexisting
pathologies. A magnetic resonance arthrogram may help
in the diagnosis of a SLAP lesion. Diagnostic arthroscopy remains the best means to definitively diagnose
SLAP lesions.
Because patient complaints can be nonspecific with
no preceding traumatic event, treatment generally begins with a physical therapy program focusing on
strengthening of the dynamic stabilizers with the goal of
improving joint stability during overhead throwing. Surgical intervention is usually reserved for patients in
whom conservative management has failed and is accomplished through an arthroscopic approach. Meaningful detachment of the capsulolabral structures requires

260

repair to the bony glenoid rim. Lesions producing significant defects extending into the biceps tendon may require biceps tenotomy, with or without tenodesis.
Rehabilitation following a SLAP repair allows for a
6-week interval for tissue healing. During this time, patients are in a gentle range-of-motion program with set
limitations, followed by a progressive strengthening program. Sport-specific activities are begun at about 3 to 4
months with a return to sports activity at approximately
6 months. Success following this procedure has been reported to be as high as an 87% return to preinjury level
of competition.

Rotator Cuff Tears

Rotator cuff pathology in the overhead throwing athlete
is seen primarily as a partial-thickness articular-sided
tear of the supraspinatus and, to a lesser extent, the infraspinatus tendons. Factors that predispose the rotator
cuff to this injury include not only the abnormal contact
between the greater tuberosity and posterosuperior glenoid rim, but also a less vascular articular cuff surface, a
higher modulus of elasticity, higher eccentric forces, and
a less favorable stress-strain curve than on the bursal
side. Full-thickness tears are less common in the overhead throwing athlete and can be the result of a partialthickness tear progressing to a full-thickness tear or the
result of a traumatic event.

Acromioclavicular Joint
Repetitive overhead throwing can result in microtraumatic wear and tear of the AC joint resulting in osteolysis of the distal end of the clavicle. Patients report pain
localized to the AC joint that becomes worse with both
overhead motion and cross-body adduction. Physical examination reveals tenderness at the AC joint, and pain
with a cross-body adduction compression test. Radio-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

graphic evaluation of the AC joint with a Zanca view
may show osteolysis or arthritic changes. Diagnosis and
treatment can be accomplished by performing a local injection into the AC joint with a combination of a local
anesthetic and a corticosteroid.
Patients who fail to improve with conservative treatment will benefit from a distal clavicle resection, which
can be done either through an open or arthroscopic approach. If there is concern about additional shoulder pathology, arthroscopic evaluation of the glenohumeral
joint and the subacromial space is recommended before
the distal clavicle is resected. Although the best amount
of resection is unknown, between 5 and 10 mm is acceptable. Postoperatively, patients are placed in a physical therapy program with the goal of regaining motion
and strength; cross-body adduction should be avoided
for 6 weeks. A slow progression to sport specific activities is followed by a return to sports activity in about 2
to 3 months.

Suprascapular Neuropathy
Suprascapular neuropathy is a rare condition in overhead throwing athletes. In elite volleyball players, it may
result from a stretch injury or direct compression of the
nerve. A paralabral ganglion cyst compression at the
transverse scapular ligament or the spinoglenoid notch
is possible. Proximal compression at the suprascapular
notch will result in denervation of the supraspinatus and
infraspinatus muscles. Distal compression at the spinoglenoid notch results in denervation of the infraspinatus
only. In the late cocking phase of throwing, the medial
tendinous margin between the supraspinatus and infraspinatus may compress the infraspinatus branch of
the suprascapular nerve against the scapular spine.
Patients with suprascapular neuropathy present with
a history of weakness and dull shoulder pain. On physical examination, weakness of the supraspinatus and infraspinatus muscles correlates with the location of nerve
compression. Chronic injuries are associated with muscular atrophy. MRI will help to identify a ganglion cyst,
which will appear as high signal intensity on a T2 image,
and aids in ruling out a rotator cuff tear as the etiology
of weakness. Electrodiagnostic studies confirm the diagnosis. Increased latency on the affected side is consistent
with nerve impairment and can be used to localize the
suprascapular nerve compression.
When an anatomic reason cannot be identified, conservative treatment with a well-organized physical therapy program is helpful for most patients. When compression is localized to an anatomic reason, surgical
decompression is recommended. Ganglion cysts in the
spinoglenoid notch can be treated with an arthroscopic
technique. Because the cyst usually originates from the
glenohumeral joint, a repair of the labrum may be required to prevent recurrence. For patients with compres-

American Academy of Orthopaedic Surgeons

Chapter 23 Shoulder and Elbow Injuries in the Throwing Athlete

sion at the suprascapular notch, possibly from a hypertrophied transverse scapular ligament, decompression is
best with an open technique.

Vascular Injuries
Vascular injuries in the overhead throwing athlete are
rare, and include digital vessel thrombosis, proximal
thrombosis with distal embolization, aneurysms, and
vessel compression such as thoracic outlet syndrome
and quadrilateral space syndrome.
Aneurysms in overhead throwing athletes have been
reported in the subclavian artery, axillary artery, and the
posterior humeral circumflex artery. It is believed that
these injuries occur as a result of repetitive trauma or
impingement during the throwing motion. The pectoralis muscle and humeral head have both been implicated
as a source of the traumatic impingement when the arm
is in the abducted and externally rotated position.
Quadrilateral space syndrome is another neurovascular compression syndrome that appears to be unique
to overhead throwing athletes. The quadilateral space is
the area bordered by the teres minor superiorly, the humeral shaft laterally, the teres major inferiorly, and the
long head of the triceps medially. Arteriograms have
documented compression of the posterior humeral circumflex artery with abduction and external rotation, but
it is believed that the symptoms are caused by compression of the axillary nerve, which runs with the artery.
Conservative management is the standard of care focusing on stretching of both the posterior capsule and
the teres minor. Patients in whom conservative management has failed can undergo open decompression of the
quadrilateral space, which has proved to be a successful
procedure for resistant cases.

Elbow
In the overhead throwing athlete, most injuries seen in
the elbow are related to the repetitive nature of throwing, and are best characterized as medial tension and
lateral compression injuries. During the late cocking and
early acceleration phases of throwing, the medial elbow
is subject to significant tensile forces, whereas the lateral
elbow experiences considerable compressive forces. Unlike the shoulder, the elbow derives most of its stability
from its bony configuration. The ligaments surrounding
the elbow help to further stabilize the joint. It is the ulnar or medial collateral ligament (MCL) on the medial
side that functions as the primary stabilizer to valgus
stress, whereas the lateral collateral ligament resists
both varus and external rotation stresses. Dynamic muscle contraction also contributes to elbow stability by increasing the joint compression forces during muscle
contraction. Repetitive trauma to the MCL allows for
increased secondary compression at the radiocapitellar
joint on the lateral side, which can ultimately lead to

261

Shoulder and Elbow Injuries in the Throwing Athlete

Orthopaedic Knowledge Update 8

Figure 2 Tests for medial UCL injury. A,

Valgus stress test. B, Milking maneuver.
The patients elbow is flexed beyond 90
while a valgus stress is applied. (Reproduced from Chen FS, et al: Medial elbow
problems in the overhead throwing
athlete. J Am Acad Orthop Surg 2001;9:
102.)
chondromalacia about the elbow and possibly loose
body formation.

Medial Elbow Pain

The MCL is divided into three segments: the anterior
bundle, the posterior bundle, and the transverse segment. It is the primary stabilizing structure to valgus
stresses about the elbow and is most often injured during the late cocking and acceleration phases of throwing. About one third of patients will not report a preceding acute event, and in those who do, a subset of them
will describe a period of nondescript elbow discomfort
before the traumatic event. Pain is usually localized to
the medial side of the elbow. Athletes with chronic
MCL incompetence may develop concomitant ulnar
nerve neuritis, and this neuritis may be accompanied by
posterior elbow pain caused by posteromedial olecranon impingement.
On physical examination, patients with an acute injury may develop medial-sided ecchymosis. Tenderness
to palpation on the medial side will lend support to the
diagnosis, but positive results on provocative testing are
also required. The valgus stress test is positive if the applied valgus stress reproduces the symptoms of medial
elbow pain, and the valgus laxity is greater on the affected side than the uninjured side. This test must be
done with the elbow flexed 20 to unlock the olecranon
from within the olecranon fossa. A gravity stress test is
performed with the athlete supine, the shoulder externally rotated to 90, and the elbow flexed 20 so that
gravity is used to elicit the valgus stress. If the ulnar collateral ligament (UCL) has been injured but remains intact, stress testing may elicit pain but no instability. The
milking maneuver will also elicit pain along the medial
side of the elbow (Figure 2). An additional test recently
described, called the moving valgus stress test, reproduces the patients medial elbow pain and symptoms by
rapidly extending the elbow from a starting position of
full flexion while maintaining a constant valgus torque.

262

Radiographs may be normal. In patients with

chronic MCL injuries, radiographs may show calcification of the MCL, medial spurs on the humerus and ulna,
posterior spurs on the olecranon tip, and loose bodies
within the joint. AP stress radiographs of the elbow are
useful when a wider medial opening in the unstable elbow is found compared with the contralateral normal
side. When the history and physical examination support
the diagnosis of an MCL injury, a magnetic resonance
arthrogram is especially useful if a T sign of ulnar avulsion or if humeral avulsion or a midsubstance tear is
found.
MCL injuries initially can be treated with a conservative approach focusing on a rehabilitation program
that incorporates modality treatments such as phonophoresis, electrical stimulation, and iontophoresis, a
stretching program, and a strengthening protocol focusing on the medial side flexor-pronator muscle mass.
Sport-specific throwing, beginning with soft tossing of
the ball, usually starts 3 months after initiation of treatment. Adjustments in the athletes mechanics during
throwing may further support a return to previous levels
of competition.
Overhead throwing athletes with MCL injuries who
continue to throw run the risk of developing posteromedial olecranon impingement and subsequent ulnohumeral arthritis. Surgical treatment is recommended
for athletes who wish to continue sports participation
and in whom conservative management has failed, and
for elite athletes who are unable to throw at their normal level. Surgical treatment consists of reconstruction
of the MCL with a tendon graft. Direct repair should be
considered only in those patients who sustain an acute
traumatic avulsion injury. Graft options include the ipsilateral or contralateral palmaris longus tendon, the
fourth toe extensor tendon, the plantaris tendon, and
the gracilis tendon.
Approximately 40% of athletes requiring an MCL
reconstruction reportedly have ulnar nerve symptoms,

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

yet surgical treatment is rarely needed. Only patients
who demonstrate muscle involvement on physical examination or electrodiagnostic testing, or experience pain
from ulnar nerve subluxation require treatment. More
than 70% of throwers who have undergone an MCL reconstruction are able to return to their previous level of
competition.

Chapter 23 Shoulder and Elbow Injuries in the Throwing Athlete

closure is usually completed within 1 year. Avulsion of
the medial epicondyle is confirmed on radiographic examination. Patients are usually treated with surgery if
the epicondyle is displaced greater than 1 cm, ulnar neuropathy exists, valgus instability is documented, or the
medial epicondyle displaces into the elbow joint. The
surgical technique is open reduction with internal fixation with Kirschner wires or a single screw.

Posterior Elbow Impingement

Impingement may result from mechanical abutment of
bone and soft tissues in the posterior elbow. This condition may or may not be associated with injury of the
MCL. Pure hyperextension injuries associated with an
intact MCL occur in gymnasts, football linemen, and
weightlifters. The lesion is usually located in the center
of the posterior elbow and is reproduced by forcible extension of the elbow. In athletes with MCL insufficiency,
as often occurs with overhead throwing athletes, the
posterior elbow impingement occurs posteromedially.
Then, the impingement is located between the medial
aspect of the olecranon and the lateral side of the medial wall of the olecranon fossa. Pain may be reproduced with the valgus stress test as previously described
for valgus instability, but the pain is posteromedial as
well as medial. Clinically, a flexion contracture is commonly present. Radiographs, including AP, lateral, and
oblique views, may show osteophytes of the olecranon
fossa.

Elbow Injuries in the Skeletally Immature Throwing

Athlete
Little Leaguers Elbow
Overhead throwing exposes the elbow of skeletally immature athletes to specific pathology. As in the mature
thrower, the elbow is exposed to tension injuries on the
medial side and compression injuries on the lateral side.
On the medial side, the medial epicondyle growth plate
(apophysis) is weaker and therefore more susceptible to
trauma than the MCL. Patients may experience medialsided elbow pain from apophysitis, or little leaguers elbow that occurs during the cocking and acceleration
phases of throwing, partly from tension in the MCL, and
partly from the pronator flexor muscle pull that occurs
during throwing. The medial epicondyle growth plate is
vulnerable to injury in athletes between the ages of 11
and 13 years when the growth plate is beginning to
close. Avulsion of the medial epicondyle may also be the
result of a sudden muscle contraction or a traumatic
event. Physical examination reveals point tenderness at
the medial epicondyle, which is often associated with a
10 to 15 flexion contracture of the elbow. Radiographic evaluation may show widening of the medial epicondyle apophysis compared with the contralateral
side. Patients are treated with a period of rest. Throwing
is avoided for 2 to 3 months and normal growth plate

American Academy of Orthopaedic Surgeons

Osteochondritis Dissecans
On the lateral side, the repetitive compression loads experienced during throwing result in microtrauma of the
articular cartilage that may lead to focal osteonecrosis
of the capitellum or the radial head. Osteochondritis
dissecans of the capitellum is seen in throwing athletes
13 to 15 years of age. The exact cause is still unknown,
but the pathology is attributed to trauma, vascular insult
resulting from the trauma, and family predisposition.
Osteochondritis dissecans may be associated with the
development of loose bodies, which can cause mechanical symptoms. Patients usually present with activityrelated pain that resolves with rest. If loose bodies are
present, patients may report clicking, catching, or locking. Radiographs will show irregularities of the capitellum, typically with a defect seen on the AP view, and
possibly loose body formation. The initial treatment of
osteochondritis dissecans is rest until symptoms resolve.
Surgical treatment is reserved for those patients with
loose bodies, which can be successfully removed using
an arthroscopic technique. Microfracture of the defect
to obtain a fibrocartilage scar may also be helpful. Fixation of osteochondral fragments can be attempted if
they are large enough.

Annotated Bibliography
Shoulder
Altchek DW, Hatch JD: Rotator cuff injuries in overhead athletes. Oper Tech Orthop 2001;11:2-16.
A review of the diagnosis and treatment of rotator cuff injuries in the overhead throwing athlete, with a focus on the
surgical technique for treating partial-thickness and fullthickness rotator cuff tears.

Andrews JR, Cain EL: Open treatment of anterior

shoulder instability in the overhead athlete. Oper Tech
Orthop 2001;11:9-16.
This article presents a review of the diagnosis and treatment of anterior shoulder instability in the overhead throwing
athlete, focusing on the technique of open surgical stabilization procedures.

Cohen B, Cole BJ, Romeo AA: Thermal capsulorrhaphy

of the shoulder. Oper Tech Orthop 2001;11:38-45.
The diagnosis of shoulder instability, including anterior,
posterior, and multidirectional instability, and the treatement

263

Shoulder and Elbow Injuries in the Throwing Athlete

of these different instabilities with thermal capsulorrhaphy, is
discussed.

Fealy S, Altchek DW: Athletic injuries and the throwing

athlete: Shoulder, in Norris TR (ed): Orthopaedic
Knowledge Update: Shoulder and Elbow 2. Rosemont,
IL, American Academy of Orthopaedic Surgeons, 2002,
pp 117-127.

Orthopaedic Knowledge Update 8

The anatomy of the ulnar nerve and the diagnosis and
treatment of ulnar neuropathy in the throwing athlete are reviewed.

Thompson WH, Jobe FW, Yocum LA, Pink MM: Ulnar

collateral ligament reconstruction in athletes: Musclesplitting approach without transposition of the ulnar
nerve. J Shoulder Elbow Surg 2001;10:152-157.

This chapter provides an extensive review of shoulder injuries in the overhead throwing athlete. Diagnosis, treatment,
and outcomes are presented.

In this study, reconstruction of the UCL using a musclesplitting approach resulted in a decreased rate of postoperative complications and improved outcomes compared with results of previous procedures.

Parten PM, Burkhart SS: The relationship of superior labral anteroposterior (SLAP) lesions and pseudolaxity to
shoulder injuries in the overhead athlete. Oper Tech
Sports Med 2002;10:10-17.

Williams RJ III, Urquhart ER, Altchek DW: Medial collateral ligament tears in the throwing athlete. Instr
Course Lect 2004;53:579-585.

This article reviews the three subtypes of a type II SLAP

lesion and how this condition contributes to the pathomechanic changes experienced by patients with internal impingement. Diagnosis and surgical treatment of patients with type II
SLAP lesions are also reviewed.

Pinto MC, Snyder SJ: SLAP lesions: Current operative

techniques and management. Oper Tech Orthop 2001;11:
30-37.
A review of SLAP lesions with a focus on diagnosis and
treatment of the four types of SLAP lesions, and a technique
review of the surgical repair of type II and IV lesions is presented.

Elbow
Bennett JB, Mehlhoff TL: Immature skeletal lesions of
the elbow. Oper Tech Sports Med 2001;9:234-240.
Pathologic entities in the young throwing athlete, including
how to make the diagnosis and how to treat the problem, are
discussed.

Conway JE, Lowe WR: Elbow medial ulnar collateral

ligament reconstruction. Oper Tech Sports Med 2001;9:
196-204.

Surgical treatment of MCL injuries, specifically using reconstruction of the MCL using a single ulnar tunnel and single
humeral tunnel performed through a muscle-splitting approach, is discussed.

Yadao MA, Field LD, Savoie FH III: Osteochondritis

dissecans of the elbow. Instr Course Lect 2004;53:599606.
A discussion of the natural history, clinical presentation,
classification, and treatment of osteochondritis dissecans of
the elbow is presented.

Classic Bibliography
Antoniadis G, Richter HP, Rath S, Braun V, Moese G:
Surprascapular nerve entrapment: Experience with 28
cases. J Neurosurg 1996;85:1020-1025.
Azar FM, Andrew JR, Wilk KE, Groh D: Operative
treatment of ulnar collateral ligament injuries of the elbow in athletes. Am J Sports Med 2000;28:16-23.
Conway JE, Jobe FW, Glousman RE, Pink MM: Medial
instability of the elbow in throwing athletes: Treatment
by repair or reconstruction of the ulnar collateral ligament. J Bone Joint Surg Am 1992;74:67-83.

This article discusses the anatomy of the MCL of the elbow, and the diagnosis of its injury, treatment, and outcomes.

DaSilva MF, Williams JS, Fadale PD, Hulstyn MJ, Ehrlich MG: Pediatric throwing injuries about the elbow.
Am J Orthop 1998;27:90-96.

Fealy S, Rohrbough JT, Allen AA, Altchek DW, Drakos

MC: Athletic injuries and the throwing athlete: Elbow,
in Norris TR (ed): Orthopaedic Knowledge Update:
Shoulder and Elbow 2. Rosemont, IL, American Academy of Orthopaedic Surgeons, 2002, pp 297-311.

Edelson G, Teitz C: Internal impingement in the shoulder. J Shoulder Elbow Surg 2000;9:308-315.

Elbow injuries in the throwing athlete including diagnosis,

treatment, and outcomes are discussed.

Gabel GT: Operative management of ulnar neuropathy

at the elbow in the throwing athlete. Oper Tech Sports
Med 2001;9:225-233.

264

Jobe CM: Posterior superior glenoid impingement: Expanded spectrum. Arthroscopy 1995;11:530-536.
Jobe CM, Pink MM, Jobe FW, Shaffer B: Anterior
shoulder instability, impingement and rotator cuff tears,
in Jobe FW, Pink MM, Glousman RE, Kvitne RS, Zemel
NP (eds): Operative Techniques in Upper Extremity
Sports Injuries. St Louis, MO, Mosby-Year Book, 1996,
pp 164-177.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Kuhn JE, Bey MJ, Huston LJ, Blasier RB, Soslowsky LJ:
Ligamentous restraints to external rotation of the humerus in the late-cocking phase of throwing: A cadaveric biomechanical investigation. Am J Sports Med 2000;
28:200-205.
Lester B, Jeong GK, Weiland AJ, Wickiewicz TL: Quadrilateral space syndrome: Diagnosis, pathology, and
treatment. Am J Orthop 1999;28:718-725.
Martin SD, Warren RF, Martin TL, Kennedy K, OBrien
SJ, Wickewicz TL: Suprascapular neuropathy: Results of

American Academy of Orthopaedic Surgeons

Chapter 23 Shoulder and Elbow Injuries in the Throwing Athlete

non-operative treatment. J Bone Joint Surg Am 1997;79:
1159-1165.
McFarland EG, Hsu CY, Neira C, ONeil O: Internal impingement of the shoulder: A clinical and arthroscopic
analysis. J Shoulder Elbow Surg 1999;8:458-460.
Schneider K, Kasparyan NG, Altcheck DW, Fantini GA,
Weiland AJ: An aneurysm involving the axillary artery
and its branch vessels in a major league baseball pitcher: A case report and review of the literature. Am J
Sports Med 1999;27:370-375.

265

Chapter

24

Shoulder and Arm Trauma: Bone

Robert P. Lyons, MD
Mark D. Lazarus, MD

Clavicle Fractures
The clavicle constitutes the only bony connection between the axial skeleton and the upper extremity. Displaced clavicle fractures risk injury to the cords of the
brachial plexus and subclavian artery and vein, which
pass between the medial curvature of the clavicle and
the first rib. Allmans classification system for clavicle
fractures is the most common and divides the clavicle
into thirds. Although somewhat arbitrary, this system
provides an efficient framework to stratify treatment
options and prognosis.

Midshaft Fractures
Acute Fractures
Fractures to the middle third segment are most common
and account for 81% of all clavicle fractures. Most middle third fractures heal with a sling or figure-of-8 dressing that is provided for symptomatic relief. An epidemiologic study of 535 isolated clavicle fractures found that
48% of middle third clavicle fractures were displaced
(using 3 mm as the displacement criterion), 19% were
comminuted, 68% of patients were men, and 61% of
fractures involved the left side.
A study of 1,430 clavicles from adult skeletons identified 73 clavicular fractures, of which 54 were
malunions. In middle third fractures, the lateral shaft
fragment was consistently displaced posteriorly to the
medial shaft fragment. In contrast, most medial third
fractures showed anterior displacement of the lateral
fragment, often forming a prominent anterior spike.
Twenty-four of 36 clavicles with significant shortening
had overriding of the bone fragments and angulation,
which tended to increase in severity the more lateral the
fracture. The maximal amount of angulation occurred at
the coracoclavicular junction, presumably because of
the deforming force of the upper extremitys weight
transmitted through the coracoclavicular ligamentous
complex. Although standing AP radiographs of clavicle
fractures display the inferior displacement of the lateral
fragment, this study suggests that the principal deformity in malunions is anterior angulation. Mild fixed

American Academy of Orthopaedic Surgeons

scapular winging was a common finding accompanying

clavicular malunions.
Indications for open reduction and internal fixation
(ORIF) of acute fractures include open fractures, neurovascular injury, or widely displaced fractures that tent,
and may compromise, the overlying skin. Some consider
wide displacement and comminution a relative indication for ORIF, citing impaired long-term function and
an increased chance of nonunion. However, the absolute
amount of shortening that can be accepted without
functional deficit has not been definitively established.
Techniques for fixation include plating or intramedullary pinning.
Recent studies documenting the patient-based outcomes following nonsurgical treatment of clavicular
fractures have shown that results are not as good as
once believed, with 31% of patients reporting shoulder
weakness and fatigability, paresthesias of the hand and
forearm, and an asymmetric or ptotic shoulder. Fractures with greater than 2 cm of shortening have been associated with a poor outcome. In addition, significant
comminution associated with displacement also may be
an indication for ORIF.
Another study compared 40 patients who received
nonsurgical treatment with a figure-of-8 bandage with
40 patients who underwent open reduction and intramedullary fixation with a 2.5-mm threaded pin. Each
group consisted of a similar profile of uncomplicated
midclavicular fractures. The group that underwent surgery experienced a high rate of complications including
eight superficial infections, three refractures, two delayed unions with pin breakage, and two nonunions.
These results suggest that nonsurgical treatment appears more advantageous than open intramedullary fixation for the treatment of most midclavicular fractures.
Malunions and Nonunions
The incidence of clavicular nonunions has been reported between 0.1% and 15%. Factors that predispose
patients to the development of a clavicular nonunion include open fractures, comminuted fractures, or initial
shortening greater than 2 cm. Anteroinferior plating of

267

Shoulder and Arm Trauma: Bone

Orthopaedic Knowledge Update 8

Figure 1 Radiograph (A) and intraoperative photograph (B) of a symptomatic clavicular malunion. C and D, Biplanar osteotomy with correction of deformity. Fixation was
achieved with a 3.5 mm dynamic compression locking plate. No intercalary graft was required.
midshaft clavicular nonunions with a 3.5-mm pelvic reconstruction plate, lag screw, and bone graft has been
described and resulted in a 100% union rate in 12 patients. The advantages of this technique include a longer
bicortical screw purchase because the anteroposterior
diameter of the clavicle is much greater than its superoinferior dimension. The plate in this position acts as a
buttress with theoretically less risk of screw pullout
from the lateral fragment and less chance of neurovascular injury during screw placement. Alternatively, a superiorly placed dynamic compression plate in compression mode can be used for more transverse fracture
patterns. The newer AO locking plates may be particularly advantageous for this indication. Care must be
taken to restore clavicular length and alignment during
nonunion correction. Prior reports have described the
use of a locked intramedullary device; however, the restoration of length with this device is more difficult. Finally, the use of a vascularized fibula graft to salvage a
recalcitrant clavicular nonunion with segmental bone
loss was described in three patients. Both pain and
shoulder function were improved and all patients
achieved union.

268

A recent patient-based outcome study of 15 patients

with clavicular midshaft malunion revealed a mean 2.9
cm of clavicular shortening. All patients had major functional deficits including chronic pain, shoulder weakness, and thoracic outlet symptoms for 1 year after injury. After corrective osteotomy, it was possible to
appose two fresh osseous surfaces in all patients without
the need for intercalary bone graft. The distal fragment
was typically rotated such that the flat superior surface
faced anteriorly, and this malrotation was corrected before fixation with a 3.5-mm limited contact dynamic
compression plate, using a minimum of six holes. This
plate was believed to be stronger than the reconstruction plate and easier to contour than the standard compression plate. Postoperatively the mean degree of
shortening was 0.4 cm and most patients were satisfied
with the result (Figure 1).
Distal Third Fractures
Neer classified distal third clavicle fractures into three
types. Type I fractures are the most common and remain
stable as the conoid and trapezoid ligaments remain intact. Type II fractures occur medial to the coracoclavicu-

American Academy of Orthopaedic Surgeons

Chapter 24 Shoulder and Arm Trauma: Bone

Orthopaedic Knowledge Update 8

Figure 2 A through C, fixation methods for type II distal clavicle fractures include coracoclavicular stabilization, using screws or tapes, with or without tension band fixation of
the distal clavicular segment. (Reproduced with permission from Lazarus MD: Fractures of the clavicle, in Bucholz RW, Heckman JD (eds): Rockwood and Greens Fractures in
Adults, ed 5. Philadelphia, PA, Lippincott Williams & Wilkins, 2001.)
lar ligaments, thus the lateral fragment is usually unstable and displaced inferiorly with a high incidence of
nonunion. Type III fractures extend into the acromioclavicular joint and usually do not involve ligamentous
injury.
Distal third fractures account for 15% of clavicular
fractures, and type II distal third fractures have a rate of
delayed union and nonunion between 30% and 45%.
Fixation techniques described in the literature include
transacromial Kirschner wires, coracoclavicular screws,
plates, Dacron tapes, and tension wires (Figure 2). Fixation between the clavicle and coracoid or clavicle and
acromion will often fail because it interferes with the
normal rotation of the clavicle that occurs with arm elevation. Thus, a second operation is required to remove
the fixation before full mobilization is commenced. A
modification of Neviasers technique has been described, in which two No. 1 polydioxanone sutures are
used as a superior figure-of-8 tension band between the
fracture fragments. If the distal fragment is too small,
the lateral drill hole can be passed through the acromion because the fixation is absorbable and the acromioclavicular joint is not violated. All fractures healed at 6
weeks with no complication.
A retrospective comparison of type II distal clavicle
fractures treated nonsurgically versus those that underwent open reduction and coracoclavicular stabilization
was reported. Nonsurgical treatment resulted in a 43%
nonunion rate; however, the nonunion had no significant effect on functional outcome or strength. This study
suggests that most type II distal clavicle fractures can be
treated nonsurgically.

Scapula Fractures
Epidemiology
Scapula fractures account for 3% to 5% of all fractures
involving the shoulder girdle. Injury to the scapula is
rare because it is enveloped by a well-developed muscular layer and lies flush with the thorax. However, other
injuries occur concomitantly in patients with scapula

American Academy of Orthopaedic Surgeons

fractures and these may delay the diagnosis. The associated mortality rate is 10% to 15% and this outcome
usually is secondary to pulmonary sepsis or head injury.
Half of all patients with scapula fractures have other insults to the ipsilateral extremity, including vascular and
brachial plexus injuries in 13%. Other associated injuries include hemopneumothorax, pulmonary contusion
with rib fractures, and spinal cord injury. Ninety percent
of scapula fractures are minimally displaced or nondisplaced, and thus can be treated nonsurgically.

Classification
The classification of glenoid fossa fractures was recently
modified to better estimate scapular body involvement
and provide more consistent guidance for choosing a
surgical approach. Type I injuries represent isolated involvement of the anteroinferior articular surface and
may be associated with a glenohumeral dislocation.
Treatment is based on instability criteria with fracture
fixation, either arthroscopically or through a deltopectoral approach, indicated for fractures that involve more
than 25% of the articular surface. Type II fractures involve the superior one third to one half of the articular
surface in continuity with the coracoid. Types I and II
can be treated surgically via a standard deltopectoral
approach. A Schanz screw used as a joystick or a dental
pic may aid reduction before fixation with 2.7-mm or
3.5-mm cortical screws.
Types III, IV, and V fracture patterns involve a variable portion of the lateral border of the scapula and
usually require the posterior approach described by Judet. Type V patterns, which involve a large separate
coracoid or superior articular surface fragment, may require a combined Judet and deltopectoral approach.
Surgical indications for glenoid fractures have been described as displacement greater than 5 mm or any displacement associated with subluxation of the humeral
head. For scapula neck fractures, surgery has been recommended if the glenoid is medially displaced greater
than 2 cm or if there is more than 40 of angular dis-

269

Shoulder and Arm Trauma: Bone

placement. Surgical indications for type V fractures include articular step-off of 5 mm or greater, severe diastasis of the articular surfaces, inferior displacement of
the glenoid fragment with inferior subluxation of the
humeral head, or severe disruption of the superior
shoulder suspensory complex.

Orthopaedic Knowledge Update 8

tion, a significant number of patients may still require
surgical intervention for postoperative stiffness. Unfortunately, associated injuries, especially those to the peripheral or central nervous system, will negatively affect
rehabilitation.

Superior Shoulder Suspensory Complex Injuries

Glenoid Fractures
Glenoid fractures are rare injuries that account for only
9% to 20% of all scapula fractures, and only 10% of glenoid fractures are substantially displaced. Common surgical indications for extra-articular glenoid fractures include angulation greater than 40, displacement of the
glenoid segment greater than 1 cm, or a fracture associated with ipsilateral clavicle fracture (such as floating
shoulder). Surgical indications for intra-articular glenoid
rim fractures include displacement greater than 10 mm,
recurrent glenohumeral instability, or for fractures involving more than 25% of the anterior rim or 33% of
the posterior rim. Complex intra-articular fractures are
treated with surgery when there is greater than 5 mm
displacement or persistent glenohumeral subluxation.
Nonsurgical treatment of minimally displaced fractures
has been successful. Thus, the decision for surgery must
be based not only on fracture character but also on patient age, hand dominance, and activity level.
Arthroscopic fixation of a displaced posterior glenoid rim fracture involving more that one third of the
glenoid cavity has been described. Ligamentotaxis on
the surgical extremity held in lateral decubitus traction
was believed to have assisted reduction.
Glenoid neck fractures are best visualized through a
posterior approach, which includes the more limited
muscle-sparing, or the extensile Judet approaches. A
muscle-sparing approach can be used for isolated posterior glenoid rim or isolated glenoid neck fractures.
The incision for the Judet approach begins at the
posterolateral corner of the acromion, extends along the
scapular spine, and then curves inferiorly along the medial scapular border. The deltoid is detached from the
scapular spine and reflected laterally. The infraspinatus
is then elevated from the scapular body on its neurovascular pedicle and reflected laterally, giving wide exposure to the posterior scapula. The fracture is identified
and directly reduced. Provisional fixation can be obtained with Kirschner wires by passing them through
the acromion into the glenoid fragment from a superior
direction. A 3.5-mm reconstruction plate is contoured
and applied to the lateral scapular body and posterior
glenoid. Intraoperative radiographs or fluoroscopy is
used to confirm anatomic reduction and prevent intraarticular screw penetration.
Postoperative stiffness is a concern. Prophylaxis
against heterotopic ossification should be considered in
all patients with closed head injury. Despite early mo-

270

The superior shoulder suspensory complex is a bony

soft-tissue ring made up of the glenoid, coracoid, acromion process, distal clavicle, acromioclavicular joint, and
coracoclavicular ligaments. The superior strut is the middle clavicle and the inferior strut is the lateral scapula.
Injury to any two of these elements is believed to result
in the potential for a large amount of scapula fracture
displacement with resultant disability to the patient. It is
generally believed that displacement of the scapular
neck will alter the relationship of the glenohumeral
joint with the acromion and create a functional imbalance. However, the exact amount of displacement or angulation required to create this functional imbalance has
not been quantified. Abduction weakness, decreased
range of motion, and nonunion are the frequently implicated complications of nonsurgical treatment.

Floating Shoulder Injuries

The floating shoulder constitutes a double disruption of
the superior shoulder suspensory complex and occurs in
only 0.1% of all fractures. In the past, this injury has
been considered unstable and thought to require surgical stabilization of one or both elements. Most glenoid
neck fractures exit the superior scapular border medial
to the base of the coracoid process, near the suprascapular notch. This results in a distal fragment consisting of
the glenoid and coracoid process, and a proximal fragment consisting of the acromion, scapular spine, and
scapular body. The distal fragment is still attached to the
proximal fragment by the coracoacromial ligament, and
through the coracoclavicular ligament and distal clavicular fragment by the acromioclavicular capsular ligaments.
In one study, 20 patients with a floating shoulder
were treated nonsurgically and functional scores were
equivalent to those from studies involving surgery. Although most of the scapula fractures in this study were
minimally displaced, the five patients who did have severe displacement of both scapula and clavicle fractures
had outcomes comparable to those with minimal displacement. Nonsurgical treatment of floating shoulder
injuries, especially those with less than 5 mm displacement, is recommended. The criteria for surgical intervention needs better delineation.
A recent study was the first to directly compare the
functional and clinical outcomes of surgical versus nonsurgical treatment of displaced ipsilateral fractures of
the clavicle and glenoid neck. The surgically treated pa-

American Academy of Orthopaedic Surgeons

Chapter 24 Shoulder and Arm Trauma: Bone

Orthopaedic Knowledge Update 8

tients, in whom clavicle and glenoid fractures were
fixed, had significantly better forward elevation but had
weaker external rotation (possibly related to the surgical approach). It was concluded that patient satisfaction
can be achieved with either surgical or nonsurgical care;
thus, the decision must be individualized for each patient. However, type of treatment was based on surgeon
preference, and the severity of fractures in each group
was not specified.
A masked retrospective radiographic review of 20
patients found that scapular neck fracture displacement,
angulation, and anatomic classification showed moderate interobserver reliability by plain films, which was
not enhanced by CT. CT may be helpful to assess possible intra-articular extension of a glenoid neck fracture
or occult injury to the superior shoulder suspensory
complex. However, whether such an injury would render the shoulder unstable and require surgical fixation
has yet to be clarified.
A recent biomechanical cadaver study testing medial
stability of ipsilateral fractures of the clavicular shaft and
scapular neck did not demonstrate instability (floating
shoulder) without additional disruption of the coracoacromial and acromioclavicular capsular ligaments. Although not included in the original description of the superior shoulder suspensory complex, the coracoacromial
ligament is believed to be an important stabilizer of scapular neck fractures because it is the only direct ligamentous connection between proximal and distal fragments.

Proximal Humeral Fractures

Epidemiology
Fractures of the proximal humerus account for 4% of
all fractures. The current incidence is 70 fractures per
100,000 people and is likely to increase because these
fractures are related to both patient age and osteoporosis. Eighty-five percent of these fractures are nondisplaced or minimally displaced and respond to treatment
with immobilization followed by early motion.
Risk factors for fracture of the proximal humerus
were evaluated with data collected from the European
Patient Information and Document Service study, which
prospectively followed 6,901 white women who lived independently at home. Overall, the incidence of proximal
humeral fracture was 6.6 per 1,000 person-years, which
occurred at a mean age of 82.2 years. The incidence of
proximal humeral fracture in women with osteoporosis and a low fall risk score (5.1 per 1,000 womanyears) was only slightly higher than in women who did
not have osteoporosis (4.6 per 1,000 woman-years), and
similar to the incidence in women without osteoporosis but a high fall risk score (5.5 per 1,000 womanyears).

American Academy of Orthopaedic Surgeons

Classification
Under Neers criteria a segment is nondisplaced when
radiographs reveal less than 1 cm of displacement or
less than 45 angulation of any one fragment with respect to the others. The number of fracture lines is important only if the displacement criteria are fulfilled. A
one-part fracture is nondisplaced and is the most common type. There are four types of two-part fractures: anatomic neck, surgical neck, greater tuberosity, and lesser
tuberosity. Three-part fractures involve either a greater
or lesser tuberosity fracture in conjunction with a fracture of the surgical neck. Four-part fractures have displacement of all four segments. Articular surface and
head-splitting fractures are also included in this category. Neer also characterized fracture-dislocations as either anterior or posterior dislocation of the articular
segment. Two-, three-, and four-part fractures can occur
as fracture-dislocations.
The AO/Orthopaedic Trauma Association classification system emphasizes the vascular supply to the articular segment.Type A fractures are the least severe and involve only one tuberosity with no isolation of the articular
segment.Type B fractures are extra-articular, involve both
tuberosities, and constitute a low risk of osteonecrosis.
Type C are intra-articular fractures involving the anatomic
neck and carry a high risk of osteonecrosis. Each type is
further classified into three subtypes. The complexity of
this classification system has limited its widespread appeal.
Much attention has been given to the poor interobserver reliability of Neers classification documented by
some studies. However, a recent study showed that kappa
values for interobserver variation improved substantially
among physicians who underwent two 45-minute training
sessions compared with a control physician group who did
not receive training in the Neer system. Another study
showed that 8 of 22 patients who underwent surgery for
three- and four-part fractures of the proximal humerus did
not correspond to any category of the Neer or AO classification system.Articular surface orientation on plain radiographs (medially oriented or not) appeared to be more
indicative of remaining soft-tissue attachments to the
head. Whether this factor has validity in predicting osteonecrosis or functional outcome has yet to be determined.

Two-Part Surgical Neck Fractures

In the two-part surgical neck fracture, the pectoralis major acts as a deforming force and displaces the shaft medially and anteriorly. Closed reduction is often possible.
If the fragments are impacted, axial traction is applied.
Then a gentle posteriorly and laterally directed force is
applied to the upper arm as the shaft is flexed and
brought underneath the head. The fragments are then
impacted and stability is evaluated. The surgeon should

271

Shoulder and Arm Trauma: Bone

Orthopaedic Knowledge Update 8

Figure 3 AP (A) and axillary (B) radiographs

of a three-part proximal humeral fracture
treated with percutaneous pinning.
alic vein, biceps tendon, and musculocutaeous nerve
from the anterior pin. The shoulder should be externally
rotated during placement of greater tuberosity pins and
these pins should engage the cortex of the humeral neck
20 mm from the inferior most aspect of the humeral
head. Finally, great care should be used to place
threaded pins into but not through the subchondral region of the humeral head (Figure 3).
Intramedullary rodding has been advocated by some
and has shown an 85% success rate in younger patients.
However, concerns over poor proximal locking-screw
purchase and subsequent proximal rod migration in
older patients with osteoporotic bone limit its appeal in
this patient population (Figure 4). In addition, antegrade intramedullary rods violate the rotator cuff and
articular cartilage. Locked intramedullary nailing of
proximal humeral fractures is best indicated for use in
patients with multiple trauma or pathologic fracture.
Heavy nonabsorbable sutures have been shown to be
an attractive choice for fixation of two- and three-part
fractures, especially in those patients with osteoporotic
bone. The use of sutures allows the incorporation of the
rotator cuff as a fixation point and precludes any chance
of hardware failure.

Two-Part Greater Tuberosity Fractures

Figure 4 Pullout of an antegrade intramedullary humeral nail in an elderly patient.

be aware of an oblique surgical neck fracture pattern
(low anteromedial to high posterolateral), which is usually irreducible by closed means. If the reduction cannot
be maintained, then either closed reduction with percutaneous pinning or ORIF is performed.
Percutaneous pinning also carries a risk to neurovascular structures. At risk are the main trunk of the axillary nerve and posterior humeral circumflex artery from
the greater tuberosity pin, the anterior branch of the axillary nerve from the proximal lateral pin, and the ceph-

272

Isolated fractures of the greater tuberosity are frequently missed. In one study, 58 of 99 fractures (58%)
were initially overlooked. There was a 64% rate of
missed diagnoses in one-part (nondisplaced greater tuberosity fracture) compared with a 27% rate of missed
diagnoses in Neer two-part fractures. The presence of
tenderness on the lateral wall of the greater tuberosity
(distal to the insertion of the rotator cuff) is an effective
clinical sign to confirm the correct diagnosis. Greater tuberosity fractures occur in 5% to 15% of anterior glenohumeral dislocations. After age 50 years, the incidence
of nerve injury with proximal humeral fracturedislocations as determined by somatosensory-evoked
potentials is 50%, and this includes axillary nerve as
well as infraclavicular brachial plexus injuries. Displaced

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

greater tuberosity fractures can be associated with rotator cuff tears, but in many surgical cases, the cuff remains intact and only a capsular rent is observed. If radiographs do not adequately show the fracture
displacement, a CT scan may be necessary. In addition,
ultrasound can detect occult nondisplaced greater tuberosity fractures.
Most authors advocate conservative treatment of fractures with less than 5 mm displacement and surgery for
displacement greater than 5 mm. Given the literature to
date, any greater tuberosity fracture displaced 5 mm or
more should be treated with reduction and fixation.
Isolated screw fixation of greater tuberosity fractures should be used only in younger patients because,
with aging, the bone of the tuberosity becomes fragile
and prone to comminution. The fracture can be fixed
percutaneously with arthroscopic assistance. The superior approach typically permits easier fragment reduction, whereas the deltopectoral approach allows easier
suture placement. The edge of the fragment can be
aligned with the edge of the fracture bed of the proximal humerus. The sutures are passed around the greater
tuberosity and through the attachment of the rotator
cuff tendons and then cross in a figure-of-8 fashion
across the fracture site to avoid overreduction of the
fragment. Because no hardware is used, there is no concern about hardware failure. Subacromial scarring, even
without displacement and capsular contracture, may
limit motion postoperatively. This complication is best
avoided by initiating passive range-of-motion exercises
early in the postoperative period.

Three- and Four-Part Fractures

In three-part fractures, one tuberosity is displaced and
there is a displaced unimpacted surgical neck fracture that
allows the head to be rotated by the attached tuberosity.
Four-part fractures involve displacement of all four segments (the greater and lesser tuberosities, the articular
segment, and the shaft), and the articular fragment is usually devoid of soft-tissue attachments. Three- and fourpart fractures represent 13% to 16% of proximal humeral
fractures. Surgical modalities have included plates, pins,
screws, staples, intramedullary nails, wires, sutures, or any
combination of these, and arthroplasty. Neer preferred
ORIF for two- and three-part fractures and arthroplasty
for four-part fractures. Others have approached all threeand four-part fractures that are amenable to stable internal fixation with minimal osteosynthesis.
Osteonecrosis of the humeral head after four-part
fracture of the proximal humerus has an incidence of
21% to 75%, which is believed to be the result of the injury pattern itself. It is important to preserve the ascending branch of the anterior humeral circumflex artery, which is the primary vessel supplying the humeral
head. The posteromedial vessels may preserve some

American Academy of Orthopaedic Surgeons

Chapter 24 Shoulder and Arm Trauma: Bone

blood flow to the humeral head when the anterior humeral circumflex artery has been interrupted. Plates
were applied lateral to the bicipital groove to avoid injury to the anterior humeral circumflex artery. No significant relationship was found between the method of fixation (cerclage wire versus plate) and the development
of osteonecrosis. Although most patients in this study
who had four-part fractures developed osteonecrosis,
the results were generally good.
Newer fixation devices have led to improved techniques for ORIF of two-, three-, and four-part fractures
of the proximal humerus. The use of a blade plate and
interfragmentary sutures has been shown to provide
firm fixation and a good functional result in the treatment of 34 three-part and 8 four-part fractures of the
proximal humerus in elderly patients. The blade plate
has primarily been used for two-part surgical neck fractures. The use of a blade plate and autogenous cancellous bone graft resulted in good or excellent results in
80% of patients with two-part nonunions (and in 90%
of low two-part nonunions). Low two-part nonunions
were easier to treat because the proximal fragment is
larger and thus easier to secure.
The advent of the locking compression plate, with a
plate specifically designed for the proximal humerus, is
an excellent new option for ORIF of proximal humeral
fractures. With this technique, it is crucial that the surgical approach preserve soft-tissue attachments as much
as possible. It is mandatory to use the drill guide that
threads directly into the screw hole in the plate because
this ensures that the screw threads will engage precisely
with the plate threads. Locking screws will not bring the
plate to the bone, which is not a critical feature of the
technique (as opposed to the standard dynamic compression plate) and may actually increase the blood supply to the bone and enhance the biology of healing.
Some plate-bone contact may be required to improve fracture alignment and diminish the prominence
of the implant. This goal can be accomplished by using a
few standard (nonlocking) screws. A mechanical study
in cadavers showed that the locking compression plate
for the proximal humerus was a more elastic implant
than the spiral blade, T-plate, and humeral nail, a property making this plate less likely to fail in osteoporotic
bone, compared with the other implants.

Valgus-Impacted Three- and Four-Part Fractures

The valgus-impacted pattern of a proximal humeral
fracture is characterized by preservation of the posteromedial capsular attachments to the humeral articular
surface. Radiographically, this pattern is defined by the
alignment between the medial shaft and head segments.
The results of treatment of the three-part valgusimpacted fracture variant are better than for the standard Neer three-part fracture. The intact posteromedial

273

Shoulder and Arm Trauma: Bone

soft tissue supplying residual perfusion to the humeral
head is presumably responsible for a better prognosis
with the valgus-impacted fracture. In a recent study,
80% of patients had a good or excellent result by Neers
outcome criteria with nonsurgical care. Patients will regain only 80% of abduction and flexion strength but do
not require full motion or strength to return to daily activities.
A new technique of minimal fixation for valgusimpacted fractures of the humeral head has been introduced. An open technique without any sharp deep softtissue dissection was used, and the split between the
tuberosities was opened to allow elevation of the impacted humeral head fragment. The tuberosities were
reduced to hold the humeral head fragment in place and
repaired with heavy absorbable suture (No. 2 Vicryl) between the rotator cuff insertions. No bone grafting, Kirschner wire, or additional fixation was used. No attempt
was made to secure the head to the shaft because the
fracture configuration was considered stable. For 11 patients with an average age of 55 years, the mean
follow-up was 69 months. The mean Constant-Morley
score as a percentage compared with the opposite side
was 86%. One patient (9%) developed osteonecrosis.
This rate of osteonecrosis was similar to other reported
rates using closed techniques. Most of the soft-tissue dissection occurred at the time of fracture, so if the principle of minimal further disruption of the soft tissues is
adhered to, an open technique such as this should not
increase the rate of osteonecrosis.

Role of Arthroscopy
Arthroscopy was used to preoperatively assess 80 shoulder fractures, including 52 proximal humerus fractures,
20 fracture-dislocations, and 8 glenoid and/or scapula
fractures. Overall, 20% of fractures were found to have
a full-thickness supraspinatus and/or infraspinatus cuff
tear and 30% had partial tears. Eighteen percent had
subscapularis tendon tears. Two-part fractures had a
31% incidence of complete labral tear, whereas only
10% of three- and four-part fractures had a complete labral tear, presumably because more energy is dissipated
at the fracture site in the more complex patterns. Proximal humeral fracture-dislocations showed a 56% incidence of complete labral tear.

Role of Arthroplasty
Incongruity of the humeral head or complete detachment of the articular blood supply are the main indications for humeral arthroplasty, which typically includes
some three-part fractures, most four-part fractures, humeral head-splitting injuries, and humeral head impression defects involving greater than 40% of the articular
surface. Other factors that may favor arthroplasty include excessive comminution, inadequate bone quality

274

Orthopaedic Knowledge Update 8

to allow for stable fixation, tuberosity fragments that include a substantial portion of the articular surface, and
inability to obtain an acceptable reduction. These patients are best treated with hemiarthroplasty or total
shoulder arthroplasty depending on the condition of the
glenoid surface and rotator cuff.
Hemiarthroplasty is primarily indicated for displaced
four-part fractures. Tuberosity osteolysis, malunion, and
nonunion are implicated as the main causes of failure.
Late arthroplasty after failed primary treatment of
proximal humeral fractures improves pain, range of motion, and function, but most authors have found that the
results of late arthroplasty are inferior compared with
proximal humerus fractures treated acutely with humeral head replacement.
Prognostic factors in prosthetic replacement for
acute proximal humeral fractures were studied in 32 patients retrospectively. Fracture type, gender, and type of
prosthesis were found to be irrelevant with respect to
outcome. Increased age and an increased preoperative
delay correlated with a poorer clinical outcome. Patients
who had surgery within 14 days had a better general
outcome. This study showed a strong correlation between the radiologic position of the tuberosities and the
quantitative as well as qualitative clinical outcome. Tuberosity complications were the most common complication and were seen in 50% of patients. Humeral offset
is defined as the distance between the geometric center
of the humeral head and the lateral edge of the greater
tuberosity. The average humeral offset of patients with a
good or excellent outcome (25.39 mm) was significantly
higher than the average offset in patients with unsatisfactory outcomes (20.20 mm). Patients with a humeral
offset of 23 mm or more had a significantly better outcome. Also, patients with a head height of 14 mm or less
had a better outcome. Lateralizing the tuberosities results in a better outcome and their distal transfer causes
a poorer outcome. Stable fixation and anatomic repositioning of the tuberosities must be achieved during surgery.
In a similar study reviewing outcome of hemiarthroplasty for fractures of the proximal humerus, it was determined that early surgical intervention within 2 weeks
of injury and accurate tuberosity reconstruction were
two factors that were found to have the greatest impact
on functional outcome. The height of the superior humeral articular surface with respect to the greater tuberosity is termed the head-to-tuberosity distance (HTD)
and is 8 3 mm as measured in cadaver specimens. The
tuberosity was considered malreduced if the fragment
was superior to the humeral head or if the HTD was
greater than 20 mm.
Proper height was established by placing the articular surface of the prosthetic head at the midpoint of the
glenoid, and with gentle traction, the prosthetic head remained within the glenoid fossa. The humeral compo-

American Academy of Orthopaedic Surgeons

Chapter 24 Shoulder and Arm Trauma: Bone

Orthopaedic Knowledge Update 8

nent was cemented in all patients and the tuberosities
were then reconstructed so that the greater tuberosity
was overlying the lateral fin or was repaired to the anterior fin in a three-fin design. Other studies indicate that
when a prosthesis is inserted for fracture, the lateral fin
should be a mean distance of 5.2 2.6 mm from the
posterior edge of the bicipital groove. In a recent biomechanical study, it was shown that a horizontal circumferential cerclage suture passed through the medial hole
of the prosthesis and around the tuberosity fragments
substantially improves fracture stability, and, in theory,
minimizes the potential for nonunion of the tuberosities.
Another study considered an HTD of 3 to 20 mm as
an anatomic reconstruction. Acceptable alignment and
position of the tuberosities was achieved in 79% of patients. The most common complication was malreduction of the greater tuberosity in the vertical plane. In patients with a nearly anatomic reduction of the greater
tuberosity, American Shoulder and Elbow Surgeons and
the Simple Shoulder Test scores, patient satisfaction, and
motion were all significantly better than in those patients in whom an anatomic reduction was not achieved.
Superior migration was associated with pain, impingement, and motion loss. These patients either had a
malunion or nonunion of the tuberosities. It is now recommended that the greater tuberosity be placed 10 mm
below the superior articular surface of the humeral
head, but not overreduced, as this places the supraspinatus under increased tension and jeopardizes tuberosity
fixation. Technical errors that can result in an increased
HTD include humeral lengthening by cementing the
prosthesis proud, using a head segment that is too thick,
and overreduction of the greater tuberosity distal to its
normal anatomic location.
Reasons for poor outcomes after hemiarthroplasty
for displaced fractures of the proximal humerus have
been analyzed. Initial tuberosity malposition was found
to be present postoperatively in 27% of patients and tuberosity detachment and migration were noted in 23%.
Final tuberosity malposition was present in 50% and
correlated with an unsatisfactory result, superior migration of the prosthesis, stiffness, weakness, and pain. The
worst association was a prosthesis that was too high and
too retroverted with a low greater tuberosity. This unhappy triad was associated with migration of the
greater tuberosity and proximal migration of the prosthesis under the acromion in all cases. The functional results after hemiarthroplasty for three- and four-part
proximal humerus fractures are directly associated with
tuberosity osteosynthesis. Lengthening of the prosthesis
greater than 10 mm because of a proud prosthesis significantly correlated with tuberosity detachment and
proximal migration of the prosthesis. The most common
mistake was excessive retroversion of the prosthesis. All
patients with retroversion exceeding 40 relative to the

American Academy of Orthopaedic Surgeons

transepicondylar axis had posterior migration of the

greater tuberosity and a poor functional result.
In a related study, shoulder arthroplasty for the
treatment of sequelae of fractures of the proximal humerus was evaluated. The most significant factor affecting functional outcome was greater tuberosity osteotomy. All patients who underwent greater tuberosity
osteotomy had either fair or poor results and did not regain active elevation above 90. Whenever possible, osteotomy of the greater tuberosity should not be done,
and the prosthesis should be implanted by attempting to
adapt the implant to the modified anatomy. This goal is
accomplished by changing the inclination or offset of a
modular prosthesis. All complications concerning the
greater tuberosity (nonunion, bone resorption, and tuberosity migration) occurred after osteotomy. Another
study showed that 10 of 24 patients who underwent osteotomy had a complication related to tuberosity nonunion, malunion, or resorption.
A promising surgical option for severe tuberosity
nonunions or long-standing surgical neck nonunions
may be a reverse prosthesis. In theory, by placing a glenosphere, the proximal humerus can gain a motion fulcrum without the necessity of a rotator cuff centering
force. There are long-term risks associated with this implant, including glenoid notching and fracture and loosening of the glenoid component. Also, implantation is
technically demanding. Long-term efficacy studies are
needed to determine if a reverse prosthesis is the answer for these difficult reconstructive cases.

Nerve Injury
A prospective study of 143 consecutive proximal humerus fractures found evidence of nerve injury by electromyography in 67% of patients. The axillary nerve was
involved in 58% and the suprascapular nerve in 48%. A
combination of nerve lesions was frequently seen. As
might be expected, nerve lesions were more common in
displaced fractures. Although the nerve lesions recovered in all patients, restoration of shoulder function was
less favorable. This finding has implications for patients
treated both surgically and nonsurgically.

Humeral Shaft Fractures

Acute Fractures
Since Sarmientos original description in 1977, functional bracing has been considered the gold standard
treatment for acute humeral shaft fractures. One recent
study revealed a 95% union rate in patients. The rate of
nonunion has been reported as low as 1.5% for closed
fractures and 5.8% for open fractures with the brace removed between 10 and 13 weeks. Varus angulation after
treatment with functional bracing is common. However,
the angulation in most patients is usually functionally
and cosmetically acceptable. The high rate of union and

275

Shoulder and Arm Trauma: Bone

no chance of postsurgical infection imply an earlier return to daily function and a lower cost of care.
Transverse fracture patterns are the most likely to
develop angular deformity. Axial distraction between
the fragments implies a high level of soft-tissue injury,
and these fractures are more likely to develop delayed
union or nonunion. In general, fractures at various levels heal at the same speed and with similar degrees of
angulation. It is not necessary for the brace to fully
cover every proximal or distal fragment. The brace
should begin approximately 1 inch (2.5 cm) distal to the
axilla and should terminate 1 inch proximal to the humeral condyles. Supra-acromial and supracondylar extensions are unnecessary. Active exercises may begin as
soon as symptoms allow. However, active abduction and
elevation of the shoulder must be avoided until after the
fracture is clinically stable because these movements
may produce angular deformities.
Indications for surgical intervention include open
fractures, segmental and pathologic fractures, bilateral
humeral fractures, fractures associated with vascular injuries, fractures where radial nerve palsy develops after
reduction, multiple trauma with substantial chest/head
injuries, floating elbow injuries (ipsilateral fractures of
both forearm and arm), obesity that prevents adequate
reduction, severe neurologic disorder, and inability to
maintain satisfactory alignment. For a patient in the recumbent position for an extended period, it is difficult
to control the fracture with closed treatment. Also, patients with concomitant chest injury are poor candidates
for sling and swathe treatment. Alignment is considered
acceptable with up to 20 anterior angulation, 30 of
varus angulation, and 3 cm of shortening.
A radial nerve palsy sustained during humeral fracture is not an indication for exploration of the nerve or
for internal fracture fixation. However, whenever the
nerve is explored and the fracture is not healed, internal
fixation is recommended. Patients with associated brachial plexus injury should undergo internal fixation because stabilization of the humeral fracture allows earlier
rehabilitation of the injured extremity. Fractures that occur in the proximal and distal quarters of the humerus
without an intervening diaphyseal fracture component
should be treated separately. These fractures are exposed through separate incisions and fixed with separate plates.
Plate osteosynthesis has been the most accepted
method for surgical treatment of both acute humeral
shaft fractures and nonunions. The rate of union after
compression plate fixation has been reported to be
greater than 95%. The disadvantages include considerable soft-tissue stripping, poor screw purchase in osteoporotic bone, difficulties in complex fractures, and
the risks associated with hardware removal.
Closed intramedullary nailing is considered for
pathologic fractures, segmental fractures, fractures of os-

276

Orthopaedic Knowledge Update 8

teopenic bone, and in patients with overlying burns and
in those with multiple traumas. The advantages of intramedullary nailing over plate fixation include preservation of fracture hematoma and local blood supply,
smaller surgical incision, a load-sharing as opposed to a
stress-shielding device, and the theoretical advantage of
earlier weight bearing. However, immediate weight
bearing in acutely plated fractures has also been shown
to have no detrimental effect.
Recent biomechanical studies show increased initial
fracture stability when nails are inserted antegrade for
proximal humeral shaft fractures and retrograde for distal humeral shaft fractures. Intramedullary nails can be
inserted for shaft fractures in the region 2 cm distal to
the surgical neck and 3 cm proximal to the olecranon
fossa. The nonunion rate for intramedullary nailing was
shown to be 5.6%, a rate comparable to that for plating.
However, a complication rate of 19% for humeral
locked nailing indicates that the procedure is not simply
benign and noninvasive. Antegrade nailing carries the
risk of shoulder pain and disability in 16% to 37% of
patients. Retrograde nailing has a significantly higher
risk of surgical comminution, and this comminution significantly increases the risk of nonunion. Great care
must be taken to avoid anterior cortical notching during
retrograde reaming. In addition, radial nerve exploration has been recommended before intramedullary rodding of distal third spiral fractures regardless of the
presence or absence of nerve palsy because the radial
nerve is in imminent danger when intramedullary rodding is attempted with this fracture pattern.
Several studies in recent years have compared plate
fixation of humeral shaft fractures with intramedullary
devices and have reached different conclusions. In reports in which plate and intramedullary rod fixation are
directly compared, the rate of complications associated
with intramedullary rodding appears higher than that
associated with plate fixation. The complications associated with intramedullary rodding are related to rates of
union (which appear to be somewhat lower than after
plate fixation) and increased functional symptoms such
as shoulder pain and weakness. Complications such as
infection, radial nerve palsy, delayed union, and failure
of fixation seem to occur at an equivalent rate with both
types of intervention.
A prospective multicenter study reported the first
clinical results of the Locking Compression Plate (Synthes, Paoli, PA) in 169 fractures in 144 patients, including 45 humeral fractures. In 86% of all fractures treated,
healing occurred within the expected time period and
without complication. The 1.5% infection rate was low
considering the heterogenous sample with numerous
open fractures and many revision surgeries. This report
provided several guidelines regarding when to use compression technique or a bridging technique with the
locking compression plate. Compression technique can

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

be used for joint fractures and simple metadiaphyseal
fracture patterns. The bridging technique is recommended for multifragmentary metadiaphyseal and diaphyseal fractures. A combined approach is warranted in
only two situations: (1) in joint fractures with multifragmentary metadiaphyseal components, the compression
technique should be used for the joint and bridging
technique used for the metadiaphyseal segment; (2) in
two-level fractures with different fracture patterns, the
compression technique is used for the simple fracture
part and the bridging technique for the multifragmentary segment.

Nonunion
Nonunion is defined as lack of union after 24 weeks.
Nonunion of the humeral shaft occurs in 2% to 10% of
nonsurgically treated fractures and up to 15% of fractures treated by primary ORIF. An increased incidence
of nonunion can be seen with open fractures, highenergy injuries, bone loss, soft-tissue interposition, unstable or segmental fracture patterns, impaired blood
supply, infection, and initial treatment with a hanging
arm cast. Preexisting elbow or shoulder stiffness can
cause increased motion at the fracture site and predispose to nonunion. Obesity, osteoporosis, alcoholism,
malnutrition, smoking, and noncompliance are all patient factors that may increase the risk of nonunion.
A 98% consolidation rate was reported after ORIF
of humeral nonunion using an anterolateral approach
and autogenous bone grafting. A wave plate was applied
in two instances when an intramedullary nail was in
place and its removal was believed to be too hazardous.
Only 2 of 51 patients had a transient sensory radial neuropathy. An anterolateral approach with routine identification and ample release of the radial nerve well beyond the nonunion ensures a very low rate of radial
nerve injury. This broad exposure allows a plate of sufficient length to be applied, which increases the probability of union.
A recent study reported on results of the use of
wave-plate fixation and autologous bone grafting in the
management of humeral nonunion with a retained intramedullary nail. None of the patients had a prominent
nail, which might represent a source of shoulder pain.
Healing occurred in all six patients at a mean of 16
weeks postoperatively. The 4.5-mm wave plates were
bent at two different locations so that the middle portion of the plate was standing 5 to 10 mm off the bone
at the level of the nonunion. At least three bicortical
screws were applied proximally and distally. Autologous
cancellous graft was packed under the elevated portion
of the plate at the site of nonunion. A high rate of nonunion has been shown after exchange nailing of humeral

American Academy of Orthopaedic Surgeons

Chapter 24 Shoulder and Arm Trauma: Bone

fractures, which is in contrast to the good results
achieved for exchange nailing of femoral and tibial nonunions.
Humeral nonunion defects of 6 cm or greater
present a reconstructive challenge. A vascularized bone
graft remains alive, does not resorb, maintains its structural characteristics, and can hypertrophy, thus increasing its structural integrity. The healing of a vascularized
bone graft is a process similar to acute fracture healing,
in contrast to the long-standing process of creeping substitution, which must be endured with a nonvascularized
allograft. The technique involves the creation of an intramedullary dowel of the vascularized fibula graft that
is placed in the medullary canal of the humerus. The
graft is impacted 1 to 2 cm at each end and the construct is stabilized with a 4.5-mm dynamic compression
plate (Synthes, Paoli, PA). The fibula is applied as an onlay graft if the construct is already stable, as with a retained intramedullary rod. The peroneal artery is usually
anastomosed end-to-side with the brachial artery. The
peroneal vein is anastomosed end-to-side with either
the venae comitantes of the brachial artery, or the basilic or cephalic vein. According to one recent study, 11
of 15 humeral nonunions united with this technique.
Three of these 11, however, sustained a fracture of the
fibula transplant at a mean of 8 months postoperatively
and required additional bone grafting and plating but
progressed to healing. Three other patients with early
fixation failure underwent plating and bone grafting and
eventually healed. One patient developed an infection
and required a second vascularized fibula graft. Three
patients had early revision of the venous anastomosis
because laser Doppler readings were below 40% of the
initial reading for 30 minutes and clinical observation of
the skin island showed questionable perfusion. Only 5
of 15 patients regained nearly normal function, illustrating the complexity of the humeral reconstruction.

Annotated Bibliography
Clavicle Fractures
Edelson JG: The bony anatomy of clavicular malunions.
J Shoulder Elbow Surg 2003;12:173-180.
Seventy-three fractures and 54 malunions were found in a
study of 1,430 clavicles from adult skeletons. A consistent pattern of clavicular shortening involving anterior-posterior angulation is described.

Grassi FA, Tajana MS, DAngelo F: Management of

midclavicular fractures: Comparison between nonoperative an open intramedullary fixation in 80 patients.
J Trauma 2001;50:1096-1100.
This study compared figure-of-8 bandaging to intramedullary fixation for midclavicular fractures and found that nonsurgical treatment appeared more advantageous for most patients
with midclavicular fractures.

277

Shoulder and Arm Trauma: Bone

Levy O: Simple minimally invasive surgical technique
for treatment of type 2 fractures of the distal clavicle.
J Shoulder Elbow Surg 2003;12:24-28.
A minimally invasive technique is described in 12 patients,
which uses absorbable suture fixation to treat Neer type II distal clavicle fractures. Union was achieved in all fractures with
no complications.

McKee MD, Wild LM, Schemitsch EH: Midshaft

malunions of the clavicle. J Bone Joint Surg Am 2003;85A:790-797.
Fifteen patients who had a malunion following nonsurgical
treatment of a clavicle fracture underwent corrective osteotomy through the original fracture line and internal fixation
without bone graft. The mean amount of clavicular shortening
was 2.9 cm preoperatively and 0.4 cm after surgery.

Postacchini F, Gumina S, DeSantis P, Albo F: Epidemiology of clavicle fractures. J Shoulder Elbow Surg 2002;11:
452-456.
An epidemiologic study of 535 isolated clavicle fractures
during an 11-year period was performed. Most patients (68%)
were men. The left side was involved in 61% and fractures of
the middle third were the most common (81%).

Rokito AS, Zuckerman JD, Shaari JM, Eisenberg DP,

Cuomo F, Gallagher MA: A comparison of nonoperative and operative treatment of type II distal clavicle
fractures. Bull Hosp Joint Dis 2002;61:32-39.
This retrospective study compared nonsurgical and surgical treatment of type II distal clavicular fractures. At an average follow-up of 53.5 months, 7 of 16 patients that were
treated nonsurgically had a nonunion. However, nonunion had
no significant impact on functional outcome or strength.

Scapula Fractures
Egol KA, Connor PM, Karunakar MA, Sims SH, Bosse
MJ, Kellam JF: The floating shoulder: Clinical and functional results. J Bone Joint Surg Am 2001;83-A:11881194.
Nineteen patients with a floating shoulder injury were
compared with respect to fracture healing, functional outcome,
patient satisfaction, and muscular strength. Treatment was
nonsurgical in 12 patients and surgical in 7 patients. There was
no significant difference between groups with regard to the
three functional outcome measures.

McAdams TR, Blevins FT, Martin P, DeCoster TA: The

role of plain films and computed tomography in the
evaluation of scapular neck fractures. J Orthop Trauma
2002;16:7-10.
Scapular neck fracture displacement, angulation, and anatomic classification showed moderate interobserver reliability
by plain films and was not enhanced when supplemented with
CT. CT may be useful to assess associated injuries to the superior shoulder suspensory complex, which could be missed
when using plain films alone.

278

Orthopaedic Knowledge Update 8

Schandelmaier P, Blauth M, Schneider C, Krettek C:
Fractures of the glenoid treated by operation: A 5 to 23
year follow-up of 22 cases. J Bone Joint Surg Br 2002;84:
173-177.
The results after ORIF of 22 displaced glenoid fractures
with a mean follow-up of 10 years is described. At follow-up,
the median Constant score was 94%. The score was less than
50% in four patients, including two patients who developed an
infection.

Williams GR Jr, Naranja J, Klimkiewicz J, Karduna A,

Iannotti JP, Ramsey M: The floating shoulder: A biomechanical basis for classification and management. J Bone
Joint Surg Am 2001;83-A:1182-1187.
After a standardized neck fracture was made in 12 cadaver
shoulders, the resistance to medial displacement was determined following sequential creation of an ipsilateral clavicle
fracture, coracoacromial ligament disruption, and acromioclavicular capsular disruption. In another group, resistance to medial displacement was determined following sequential release
of the coracoacromial and coracoclavicular ligaments. Ipsilateral scapular neck and clavicle fractures do not produce a
floating shoulder until the coracoacromial and acromioclavicular capsular ligaments are disrupted.

Proximal Humeral Fractures

Boileau P, Krishnan SG, Tinsi L, Walch G, Coste JS,
Mole D: Tuberosity malposition and migration: Reasons
for poor outcomes after hemiarthroplasty for displaced
fractures of the proximal humerus. J Shoulder Elbow
Surg 2002;11:401-412.
The results of hemiarthroplasty for displaced proximal humeral fractures is evaluated. Final tuberosity malposition occurred in 33 patients (50%) and correlated with superior migration of the prosthesis, stiffness, weakness, and persistent
pain.

Boileau P, Trojani C, Walch G, Krishnan SG, Romeo A,

Sinnerton R: Shoulder arthroplasty for the treatment of
sequelae of fractures of the proximal humerus. J Shoulder Elbow Surg 2001;10:299-308.
Seventy-one sequelae of proximal humeral fractures were
treated with shoulder arthroplasty with the same modular
prosthesis (Aequalis) in this multicenter study. The most significant factor affecting functional outcome was greater tuberosity osteotomy. Surgeons should accept the distorted anatomy of the proximal humerus and adapt the prosthesis to the
modified anatomy if possible, to avoid tuberosity osteotomy.

Brorson S, Bagger J, Sylvest A, Hobjartsson A: Improved interobserver variation after training of doctors
in the Neer system: A randomized trial. J Bone Joint
Surg Br 2002;84:950-954.
Fourteen doctors were randomly assigned to two training
sessions or to no training and asked to categorize 42 pairs of
plain radiographs of proximal humerus fractures according to
the Neer system. The kappa value for interobserver variation

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 24 Shoulder and Arm Trauma: Bone

improved (especially for specialists) from 0.30 to 0.79 with formal training in the Neer system.

Mighell MA, Kolm GP, Collinge CA, Frankle MA: Outcomes of hemiarthroplasty for fractures of the proximal
humerus. J Shoulder Elbow Surg 2003;12:569-577.

Court-Brown CM, Cattermole H, McQueen MM: Impacted valgus fractures of the proximal humerus. The results of non-operative treatment. J Bone Joint Surg Br
2002;84:504-508.

Eighty shoulders that had been treated with hemiarthroplasty for proximal humeral fractures were reviewed. Tuberosity complications occurred in 16 shoulders. Healing of the
greater tuberosity more than 2 cm below the humeral head
correlated with a worse functional result.

In this retrospective study of 125 patients with valgusimpacted fractures of the proximal humeral, all were treated
nonsurgically. At 1-year follow-up, 80.6% of patients had a
good or excellent result.

Demirhan M, Kilicoglu O, Altinel L, Eralp L, Akalin Y:

Prognostic factors in prosthetic replacement for acute
proximal humerus fractures. J Orthop Trauma 2003;17:
181-189.
This retrospective study investigated the effect of radiologic and other factors on the outcome of prosthetic replacement in 32 patients with acute proximal humeral fractures. The
humeral offset was directly correlated to the amount of forward elevation and Constant score. The head height was inversely correlated to the same parameters.

Frankle MA, Ondrovic LE, Markee BA, Harris ML, Lee

WE III: Stability of tuberosity reattachment in proximal
humeral hemiarthroplasty. J Shoulder Elbow Surg 2002;
11:413-420.
This biomechanical study compared the stability of different reconstructive techniques of tuberosity reattachment for
proximal humeral head replacement in four-part fractures using eight fresh-frozen cadaver shoulders. Mercury strain
gauges were used to measure bony fragment displacement
during cyclic loading. Repairs that used a medial circumferential cerclage had significantly lower displacements and strains.

Hockings M, Haines JF: Least possible fixation of fractures of the proximal humerus. Injury 2003;34:443-447.
An open technique with no deep dissection is described
for the fixation of valgus-impacted proximal humeral fractures. No bone grafting, Kirschner wires, or other fixation was
used. The mean Constant-Murley score as compared with the
opposite side was 86%.

Lee SH, Dargent-Molina P, Breart G: Risk factors for

fractures of the proximal humerus: Results for the
EPIDOS prospective study. J Bone Miner Res 2002;17:
817-825.
The EPIDOS study evaluated 6,901 white women age 75
years and older. The examination included measurements of
femoral neck bone mineral density, calcaneal ultrasound parameters, a functional clinical examination, and lifestyle questionnaire. The incidence of proximal humerus fractures was
greatly increased in women who had osteoporosis and a high
risk score for falling.

American Academy of Orthopaedic Surgeons

Ogawa K, Yoshida A, Ikegami H: Isolated fractures of

the greater tuberosity of the humerus: Solutions in recognizing a frequently overlooked fracture. J Trauma
2003;54:713-717.
Isolated greater tuberosity fractures were overlooked in 58
of 99 shoulders (59%) that had been initially examined at
other clinics. A smaller fragment correlated with a higher rate
of missed diagnosis. The presence of tenderness on the lateral
wall of the greater tuberosity is a clinically effective method to
avoid a missed diagnosis.

Park MC, Murthi AM, Roth NS, Blaine TA, Levine WN,
Bigliani LU: Two-part and three-part fractures of the
proximal humerus treated with suture fixation. J Orthop
Trauma 2003;17:319-325.
The radiographic and clinical outcomes of patients with
displaced two- and three-part proximal humeral fractures that
were treated with nonabsorbable cuff-incorporating sutures
were reviewed. Both groups had similar outcomes; some residual deformity did not preclude an excellent outcome.

Ring D, McKee MD, Perey BH, Jupiter JB: The use of a

blade plate and autogenous cancellous bone graft in the
treatment of ununited fractures of the proximal humerus. J Shoulder Elbow Surg 2001;10:501-507.
A blade-plate with autogenous cancellous graft was used
to treat proximal humeral nonunions in 25 patients with a
mean age of 61 years. Union was achieved in 23 of 25 patients
(92%). The results were classified as good or excellent in 20 of
25 patients.

Robinson CM, Page RS: Severely impacted valgus proximal humerus fractures. J Bone Joint Surg Am 2003;85A:1047-1055.
Twenty-five patients with severely impacted valgus proximal humeral fractures were treated with open reduction, fixation with screws or a buttress plate, and the fracture defect
was filled with Norian Skeletal Repair system bone substitute.
All fractures united within the first year and no patient had
signs of osteonecrosis at latest follow-up. The functional result
continued to be satisfactory for the 12 patients who were followed for 2 years.

Rowles DJ, McGrory JE: Percutaneous pinning of the

proximal part of the humerus: An anatomic study.
J Bone Joint Surg Am 2001;83-A:1695-1699.

279

Shoulder and Arm Trauma: Bone

Orthopaedic Knowledge Update 8

In 10 fresh-frozen cadaver shoulders the intact proximal

humerus was pinned using fluoroscopic guidance and a standard published technique. The specimens were dissected to determine the distance of each pin from vital neurovascular
structures. Some modifications of technique and a knowledge
of local anatomy are needed to avoid injury to the axillary
nerve, musculocutaneous nerve, and biceps tendon.

Fifteen patients with an average 9.3-cm segmental humeral

defect were treated with osteoseptocutaneous fibular transplant. This treatment can be successful but has a high rate of
complications.

Stoffel K, Dieter U, Stachowiak G, Gachter A, Kuster

MS: Biomechanical testing of the LCP: How can stability in locked internal fixators be controlled? Injury
2003;34(suppl 2):B11-B19.

Sixty-seven humeral shaft fractures were treated by

Sarmiento bracing over a 15-year period. At a mean of 10
weeks, 87% had healed clinically. Among nine cases of delayed
or nonunion leading to surgery, there were six cases with
transverse fractures. Reasons for failed nonsurgical treatment
included incorrect indication, significant axial deformity, and a
hyperextended position of the fracture fragments.

A biomechanical analysis including axial stiffness and torsional rigidity, fatigue testing, and finite element analysis were
undertaken to evaluate the locking compression plate. Optimal use of the locking compression plate as it relates to biomechanical principles is discussed. For comminuted fractures
of the humerus, it is recommended that innermost screws be
placed as close to the fracture as practicable. The distance between the plate and the bone should be kept small. Long
plates should be used to provide sufficient axial stiffness.

Visser CP, Coene LN, Brand R, Tavy DL: Nerve lesions

in proximal humeral fractures. J Shoulder Elbow Surg
2001;10:421-427.
Fourteen consecutive proximal humeral fractures were
evaluated with electromyograms in this prospective study. The
electromyogram showed denervation in 96 patients (67%),
with the axillary (58%) and the suprascapular nerves (48%)
most frequently involved.

Wijgman AJ, Roolker W, Patt TW, Raaymakers EL,

Marti RK: Open reduction and internal fixation of three
and four-part fractures of the proximal part of the humerus. J Bone Joint Surg Am 2002;84-A:1919-1925.
The long-term results were assessed for 60 patients with a
three- or four-part fracture of the proximal humerus who had
undergone ORIF with cerclage or a T-plate. Thirty-seven percent of patients had development of osteonecrosis and 77% of
these had a good or excellent Constant score.

Humeral Shaft Fractures

Gerber A, Marti R, Jupiter J: Surgical management of
diaphyseal humeral nonunion after intramedullary nailing: Wave-plate fixation and autologous bone grafting
without nail removal. J Shoulder Elbow Surg 2003;12:
309-313.
Sic patients with a nonunion of the humeral diaphysis after intramedullary nailing were treated with a wave-plate and
autologous bone grafting. Union can be achieved without removal of the intramedullary device.

Heitmann C, Erdmann D, Levin LS: Treatments of segmental defects of the humerus with an osteoseptocutaneous fibular transplant. J Bone Joint Surg Am 2002;84A:2216-2223.

280

Koch PP, Gross DF, Gerber C: The results of functional

(Sarmiento) bracing of humeral shaft fractures. J Shoulder Elbow Surg 2002;11:143-150.

Lin J, Shen PW, Hou SM: Complications of locked nailing in humeral shaft fractures. J Trauma 2003;54:943949.
Delayed unions, nonunions, and acute humeral shaft fractures in 159 patients were treated with humeral locked nails
and followed for an average 25.4 months. Surgical comminution was significantly higher in retrograde nailing and surgical
comminution and significantly increased the risk of nonunion.
Other complications included functional shoulder impairment,
angular malunion, and postnailing radial nerve palsy.

Marti RK, Verheyen CC, Besselaar PP: Humeral shaft

nonunion: Evaluation of uniform surgical repair in fiftyone patients. J Orthop Trauma 2002;16:108-115.
This article presents a review of the standard treatment of
aseptic humeral shaft nonunions involving an anterolateral approach, radial nerve identification, compression plating, and
autogenous bone grafting. This approach is reliable and
achieves union without a significant risk of complications.

Ring D, Jupiter JB: Internal Fixation of the humerus

with locking compression plates. Tech Shoulder Elbow
Surg 2003;4:169-174.
The drawbacks of previous methods of internal fixation
for humeral fractures and nonunions is discussed. The advantages of and technique for the application of Locking Compression Plates to humeral fractures and nonunions is presented. Preliminary results are also discussed.

Sommer C, Gautier E, Muller M, Helfet DL, Wagner M:

First clinical results of the Locking Compression Plate
(LCP). Injury 2003;34(suppl 2):B43-B54.
A prospective multicenter study of the combined European and American experience with the new locking compression plate system used to treat various fractures in 144 patients is presented. In this study, tibial fractures (n = 46) and
humeral fractures (n = 45) predominated. General principles
for the application of the locking compression plate to different fracture configurations as well as to humeral fractures,
are discussed. The low infection rate (1.5%) in this heteroge-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

nous sample with numerous open fractures and revision procedures attests to the good biology of the implant.

Chapter 24 Shoulder and Arm Trauma: Bone

Hintermann B, Trouillier HH, Schafer D: Rigid internal
fixation of fractures of the proximal humerus in older
patients. J Bone Joint Surg Br 2000;82:1107-1112.

Stannard JP, Harris HW, McGwin G Jr, Volgas DA,

Alonso JE: Intramedullary nailing of humeral shaft fractures with a locking flexible nail. J Bone Joint Surg Am
2003;85-A:2103-2110.

Mayo KA, Benirschke SK, Mast JW: Displaced fractures

of the glenoid fossa: Results of open reduction and internal fixation. Clin Orthop 1998;347:122-130.

A flexible humeral nail is described that allows both antegrade and retrograde insertion and static locking without violating the rotator cuff or humeral articular surface. Although
the nail functioned well in most patients, the use of the smaller
(7.5 mm) nail was associated with a higher complication rate.
This implant should be used with caution in any patient with a
medullary canal diameter of 8 mm or less.

McCormack RG, Brien D, Buckley RE, McKee MD,

Powell J, Schemitsch EH: Fixation of fractures of the
shaft of the humerus by dynamic compression plate or
intramedullary nail: A prospective, randomized trial.
J Bone Joint Surg Br 2000;82:336-339.

Classic Bibliography

Momberger NG, Smith J, Coleman DA: Vascularized

fibular grafts for salvage reconstruction of clavicle nonunion. J Shoulder Elbow Surg 2000;9:389-394.

Chapman JR, Henley MB, Agel J, Benca PJ: Randomized prospective study of humeral shaft fracture fixation: Iintramedullary nails versus plates. J Orthop
Trauma 2000;14:162-166.

Schai PA, Hintermann B, Koris MJ: Preoperative arthroscopic assessment of fractures about the shoulder. Arthroscopy 1999;15:827-835.

Edwards SG, Whittle AP, Wood GW II: Nonoperative

treatment of ipsilateral fractures of the scapula and
clavicle. J Bone Joint Surg Am 2000;82:774-780.

Tingstad EM, Wolinsky PR, Shyr Y, Johnson KD: Effect

of immediate weightbearing on plated fractures of the
humeral shaft. J Trauma 2000;49:278-280.

Galatz LM, Iannotti JP: Management of surgical neck

nonunions. Orthop Clin North Am 2000;31:51-61.

Webber MC, Haines JF: The treatment of lateral clavicle

fractures. Injury 2000;31:175-179.

Herscovici D Jr, Saunders DT, Johnson MP, Sanders R,

DiPasquale T: Percutaneous fixation of proximal humerus fractures. Clin Orthop 2000;375:97-104.

American Academy of Orthopaedic Surgeons

281

Chapter

25

Shoulder Instability
Andrew D. Pearle, MD
Frank A. Cordasco, MD

Natural History
The glenohumeral joint is the most mobile articulation
in the body and the most commonly dislocated diarthroidal joint, with peaks in the incidence of dislocation
occurring during the second and sixth decades of life. Instability of the glenohumeral joint ranges from subtle
increased laxity to recurrent frank dislocation. No single
disease or lesion is responsible for all types of shoulder
instability, and treatment has evolved to anatomically
address specific lesions.
Traumatic injury is the major cause of shoulder instability, accounting for approximately 95% of shoulder
dislocations. The sequela of traumatic anterior dislocation is associated with the age of the patient at the time
of initial dislocation and the degree of injury. Age at the
time of the initial dislocation is inversely related to the
recurrence rate. In patients younger than 20 years of
age, recurrent dislocation rates may be as high as 90%
in the athletic population. The rate of recurrences drops
to 50% to 75% in patients 20 to 25 years of age. In patients older than 40 years, anterior dislocation is associated with lower rates of redislocation, but high rates of
rotator cuff tears. Although the incidence of rotator cuff
tears in patients older than 40 years at the time of initial
dislocation is 15%, this incidence reaches 40% in patients older than 60 years. The degree of injury (presence and size of Bankart tear, presence and size of osseous lesions including Hill-Sachs defects and osseous
Bankart lesions, capsular tears such as a humeral avulsion of the glenohumeral ligament, and the presence of
associated rotator cuff pathology) is directly related to
the recurrence rate.

Pathophysiology
The normal humeral head translates only 1 mm from
the center of the glenoid during active motion. The glenohumeral joint is stabilized by both static and dynamic
stabilizers. The static restraints consist of the glenoid labrum, the articular anatomy, negative intra-articular
pressure, joint fluid adhesion, and the capsuloligamentous structures. Dynamic stabilizers of the joint include

American Academy of Orthopaedic Surgeons

the rotator cuff muscles, biceps, and periscapular muscles. In general, the capsular ligaments provide stability
at end range of motion; however, during midrange of
motion, the capsuloligamentous structures are lax and
the joint is stabilized by dynamic joint compression.
Glenohumeral instability occurs when there is a deficiency in the bony, soft-tissue, or dynamic muscular restraints to translation of the humeral head on the glenoid. Rehabilitation following instability episodes is
directed toward optimizing the dynamic stabilizers,
whereas surgical intervention restores the static stabilizers.
The oval glenoid is longest in its inferior-superior diameter and has a nearly flat articular surface. Although
the osseous shape of the glenoid does not contribute
greatly to stability, the peripheral chondral surface of
the glenoid is thickened, creating a concave articular
surface that augments glenohumeral stability. The labrum is a fibrous structure firmly bound to the glenoid
at its inferior margin and bound more loosely superiorly, where it is confluent with the origin of the tendon
of the long head of the biceps. The labral tissue contributes to glenohumeral stability by deepening the glenoid
by approximately 50%. The labrum is thought to act as
a chock block and has been shown to decrease resistance to humeral translation by 10% to 20%. In addition, the labrum serves as an attachment site connecting
the glenoid to the capsule, ligaments, and biceps tendon.
The capsule of the glenohumeral joint is lax under
normal circumstances, which allows for great range of
motion of the joint. The capsule attaches medially to the
glenoid and labrum and extends lateral to the surgical
neck of the humerus. In addition to providing structural
stability, the capsule maintains the negative joint pressure, which creates a negative vacuum that augments
joint stability.
The ligaments of the glenohumeral joint are discrete
bands that insert onto the glenoid labrum. The most important ligaments in the glenohumeral joint are the
superior glenohumeral ligament (SGHL), the middle
glenohumeral ligament (MGHL), and the inferior glenohumeral ligament complex (IGHLC) (Figure 1).

283

Shoulder Instability

Figure 1 Schematic drawing of the shoulder capsule showing the glenohumeral ligaments highlighting the IGHLC. A, anterior; P, posterior; B, biceps tendon; AB, anterior
band; AP, axillary pouch; PB, posterior band; and PC, posterior capsule. (Reproduced
with permission from OBrien SJ, Neves MC, Arnoczky SP, et al: The anatomy and
histology of the inferior glenohumeral ligament complex of the shoulder. Am J Sports
Med 1990;18:449-456.)

The SGHL originates on the superoanterior rim of

the glenoid, just anterior to the biceps origin, and
courses across the rotator interval to the lesser tuberosity adjacent to the bicipital groove. The SGHL is the primary static restraint to inferior translation of the adducted shoulder. The coracohumeral ligament (CHL) is
an extra-articular structure that arises from the base of
the coracoid and passes between the supraspinatus and
subscapularis, forming the roof of the rotator interval. It
is composed of an anterior and posterior band that inserts onto the lesser and greater tuberosity, respectively.
The CHL is a static restraint to anteroinferior translation in the adducted shoulder.
The MGHL has the most variable morphology of
the ligaments; however, no distinct variation in its structure has been correlated with increased rates of instability. The MGHL normally originates in the upper third of
the anterior glenoid rim and labrum and courses ob-

284

Orthopaedic Knowledge Update 8

liquely across the subscapularis tendon to its humeral
attachment on the lesser tuberosity. A Buford complex
is a normal anatomic variant defined by the absence of
the superoanterior aspect of the labrum and a broad,
cord-like MGHL. The MGHL prevents anterior translation when the shoulder is externally rotated and in the
middle range of abduction.
The IGHLC is the major static anterior stabilizer of
the glenohumeral joint, especially during abduction and
external rotation (Figure 2). The complex consists of an
anterior and posterior band as well as the axillary
pouch. It originates from the inferior labrum and inserts
on to the humeral neck. Detachment of the anterior labrum and anterior IGHLC is referred to as a Bankart
lesion and has classically been understood as the essential lesion in traumatic anterior instability (Figure 3).
However, in cadaveric models, minimal translation has
been noted with recreation of this lesion, suggesting that
capsular and ligamentous plastic deformation is an important pathologic component of instability. In a recent
study using MRI to quantitate the amount of capsular
elongation in patients with recurrent anterior dislocations, the authors showed that the anterior and inferior
portions of the shoulder capsule are elongated an average of 19% compared with the unaffected contralateral
shoulder.
A recent cadaveric study has demonstrated that the
posterior band of the IGHLC shows greatest strain in
flexion and internal rotation, the provocative positions
for posterior instability. This finding suggests that the
posterior band of the IGHLC is a static posterior stabilizer and should be firmly repaired in patients with posterior instability.
Dynamic stabilization of the glenohumeral joint occurs through joint compression by the rotator cuff, primarily the subscapularis and infraspinatus. Dynamic
joint compression from the rotator cuff musculature
limits superior-inferior translation and imparts stability
during abduction. The dynamic joint compression force
from the rotator cuff muscles is more important for stability than the static glenohumeral ligaments. The scapular stabilizers, including the trapezius, rhomboids, latissimus dorsi, serratus anterior, and levator scapulae
position the glenoid in an anteverted and superior position and provide dynamic coverage for the retroverted
humeral head; this increases the posterior and inferior
stability of the glenohumeral joint during motion.
The biceps tendon is thought to be a dynamic stabilizer of the joint, particularly at middle and lower elevation angles. The tendon provides stability in the anterior
and superior direction, imparting more posterior stability in external rotation of the humerus and anterior stability with internal rotation. In addition, because of the
intimate association of the bicipital anchor and the
SGHL and MGHL, lesions that involve the bicipital anchor, such as superior labral anterior and posterior

American Academy of Orthopaedic Surgeons

Chapter 25 Shoulder Instability

Orthopaedic Knowledge Update 8

Figure 2 The glenohumeral ligaments provide static restraint in different functional positions. A, With the shoulder in adduction and external rotation, the ligament SGHL and
MGHL are taut while the anterior band (AB) and posterior band (PB) of the IGHLC are lax. B, With the shoulder in abduction and external rotation, the AB of the IGHLC tightens
and the SGHL and MGHL become lax. (Reproduced with permission from Warner JP, Boardman ND III: Anatomy, biomechanics, and pathophysiology of glenohumeral instability, in
Warren RF, Craig EV, Altcheck DW (eds): The Unstable Shoulder. Philadelphia, PA, Lippincott-Raven, 1999, pp 51-76.)

(SLAP) tears, may increase anteroposterior as well as

superoinferior glenohumeral translation.

Patient Evaluation
Multiple static and dynamic structures contribute to
shoulder stability; therefore, proper classification of
shoulder instability is essential to identify injured structures and plan treatment. Shoulder instability is classified by the degree, frequency, etiology, and direction of
instability (Table 1). Patient history and physical examination should focus on accurately classifying the instability pattern.

History
A careful history is essential to begin to classify the patients instability. The etiology of the instability may be
readily apparent as the patient may describe a frank traumatic dislocation event, history of repetitive microtrauma
with overhead activity, or generalized laxity that is familial. Patients are usually able to clearly describe the frequency and chronicity of instability episodes. Pain or instability with particular movements or positions may
reveal the direction of instability. Patients with anterior instability report symptoms with the arm in an abducted and
externally rotated position. Posterior instability often occurs with the arm flexed, internally rotated, and adducted.
Patients may experience symptoms while pushing open a
door or heavy object. Patients with inferior instability often have pain while carrying heavy objects; they may also
experience traction paresthesias.

American Academy of Orthopaedic Surgeons

Figure 3 Schematic drawing of the anatomy of the Bankart lesion demonstrating

avulsion of the capsulolabral structures off the inferior rim of the glenoid. A = anterior;
P = posterior l(Reproduced with permission from Warner JP, Caborn DNM: Overview of
shoulder instability. Crit Rev Phys Rehabil Med 1992;4:145-198.)

Physical Examination
Examination of shoulder instability begins with a thorough evaluation of the cervical spine. Careful inspection

285

Orthopaedic Knowledge Update 8

Shoulder Instability

Table 1 | Classification of Shoulder Instability

Degree
Dislocation
Subluxation
Microinstability
Frequency
Acute (primary)
Chronic
Recurrent
Fixed
Etiology
Traumatic (macrotraumatic)
Atraumatic
Voluntary (muscular)
Involuntary
Aquired (microtrauma)
Direction
Unidirectional
Anterior
Posterior
Inferior
Bidirectional
Anterior-inferior
Posterior-inferior
Multidirectional

Imaging
Acute traumatic shoulder dislocations are evaluated
with a trauma series that includes an AP, transscapular
(Y) lateral, and an axillary view. The axillary view is especially important to confirm reduction. In more
chronic instability, additional views are useful to assess
bony anatomy and identify characteristic pathologic lesions. The West Point axillary view, which is taken with
the patient prone, the arm in 90 of abduction and neutral rotation, and the x-ray beam directed 25 posterior
to the horizontal plane and 25 medial to the vertical
plane, shows the anterior glenoid rim and may reveal
bony Bankart lesions. The Stryker notch view, taken
with the patient supine and the hand placed on top of
the head, shows the posterosuperior humeral head and
Hill-Sachs lesions. CT scans can be helpful in selected
patients with complex bony injuries and in evaluation of
glenoid and humeral head version.
MRI has become a helpful tool to evaluate patients
with acute and chronic shoulder instability. With MRI,
capsular and ligament detachments, labral lesions, rotator cuff tears, and bony trauma can be identified with
more accuracy than with radiographs or CT scan.

Examination Under Anesthesia and Arthroscopy

(Reproduced with permission from Backer M, Warren RF: Glenohumeral instabilities in
adults, in DeLee JC, Drez D Jr, Miller MD (eds): DeLee and Drezs Orthopaedic Sports
Medicine: Principles and Practice. Philadelphia, PA, Saunders, 2003, pp 1020-1034.)

of the shoulder girdle and upper extremity may reveal

obvious deformity such as an acute dislocation. Anterior
dislocations are characterized by prominence of the humeral head anterior, medial, and often inferior to the
shoulder joint, a hollow region beneath the lateral deltoid, and the arm held in an abducted, slightly externally
rotated position. Posterior dislocations may be less obvious but a posterior prominence is usually noted and the
arm is held in an internally rotated and adducted position. The coracoid is prominent as well. More subtle
findings on inspection of a reduced but unstable joint
include muscle atrophy caused by general neurologic
conditions, local nerve compression, and tendon tears.
The strength of the muscles surrounding the shoulder and the patients range of motion should be tested
and compared with the contralateral side. The function
of the axillary nerve should be evaluated by testing the
function of the deltoid and the sensation on the lateral
aspect of the shoulder.
Provocative tests for shoulder instability allow the
examiner to further classify the instability. Evaluation of
the contralateral side should be conducted for comparison. Tests that reproduce the symptoms of instability include the apprehension test and the posterior load and
shift test.

286

Although classification of the frequency, etiology, and

direction of instability may be evaluated with a thorough history and physical examination, the degree of instability is often best appreciated with examination under anesthesia or during arthroscopy. The axial load test
is used and the degree of instability in each direction is
graded.
Arthroscopic examination allows for complete assessment of injured structures in the glenohumeral joint.
The articular surfaces, labrum, glenohumeral ligaments,
integrity of the capsular tissue, biceps tendon, and rotator cuff should be carefully inspected and probed.

Treatment of Acute Dislocation

Acute dislocations should be reduced as quickly and
atraumatically as possible. After a trauma series is obtained, closed reduction is performed with the patient
relaxed with sedation or local anesthetics.
A recent MRI study has shown that the conventional position of immobilization in adduction and internal rotation results in significantly more separation and
displacement of a Bankart lesion than immobilization in
adduction and external rotation.
In the young, highly active athletic population with a
traumatic anterior dislocation, primary surgical intervention may be warranted. Several studies from the US
Military Academy at West Point have reported recurrent dislocation rates of 80% to 92% for cadets treated
with physical therapy after initial traumatic shoulder
dislocation. In this population, surgical intervention has

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 25 Shoulder Instability

been advocated. In a recent prospective randomized

clinical trial comparing arthroscopic stabilization with
nonsurgical treatment of first-time shoulder dislocations
in active duty soldiers, 75% of the nonsurgically treated
patients developed recurrent instability, versus 11% of
the arthroscopy treated patients. Average follow-up was
3 years. In another study from the US Military Academy
at West Point, the recurrence rate of instability after primary arthroscopic repair of initial anterior shoulder dislocations at 2- to 5-year follow-up was 12%.

Treatment of Recurrent Traumatic Instability

Whereas the goal of rehabilitation is the enhancement
of the dynamic stabilizers of the shoulder, the aim of
surgical intervention is the restoration of disrupted
static restraints. Primary stabilization should be considered for first-time traumatic shoulder dislocation with a
Bankart lesion confirmed by MRI in the athletic population younger than 25 years old. These patients have a
low rate of return to previous activity level if managed
with rehabilitation alone. In the older or less athletic
population, the primary indication for surgical stabilization is recurrent instability despite an appropriate rehabilitation program.
The goals for treatment of shoulder instability are
identical whether the procedure is arthroscopic or open.
Anatomic repair of the static stabilizers involves identification and reattachment of disrupted structures with
minimal disturbance of the length and attachments of
uninvolved tissues (Figure 4).
An open Bankart repair involves an anterior approach to the glenohumeral joint and repair of the detached capsulolabral structures to the glenoid through
drill holes or with suture anchors. With this timehonored procedure, the expected success rate is approximately 90% to 95%. Because the capsule is opened
during the approach and may be retensioned, imbricated, or shifted, open repair is traditionally favored in
situations of multiple recurrences with increased capsular laxity. In addition, open repair is generally considered a more durable procedure for patients who wish to
return to contact sports.
Arthroscopic stabilization techniques involve anatomic repairs without the takedown and reattachment
of the subscapularis tendon as required in the open procedure. The arthroscopic techniques have improved
greatly. Fixation devices for reapproximation of the capsulolabral lesions have evolved from staples to transglenoid sutures to cannulated bioabsorbable screws and suture anchors. In addition to shorter hospitalization times
and less postoperative pain, results of arthroscopic procedures show increased external rotation compared with
open techniques. The major disadvantage of arthroscopic techniques has been a higher recurrence rate in
comparison with open procedures. In most studies using

American Academy of Orthopaedic Surgeons

Figure 4 A, In the normal shoulder, the capsule and labrum serve to deep glenoid.
B, This effect is lost in the presence of a Bankart lesion. C, Anatomic repair restores
effective depth of glenoid concavity. D, Nonanatomic repair does not restore normal
chock-block effect of the capsulolabral structure. (Reproduced with permission from
Matsen FA III, Lippitt SB, Sidles JB. Harryman DT II: Stability, in Matsen FA III (ed):
Practical Evaluation and Management of the Shoulder. Philadelphia, PA, WB Saunders,
1994, p 59.)
older techniques for arthroscopic fixation, the recurrence rate has been between 10% to 20%, approximately twice as high as with open procedures. However,
in a recent review of the outcomes of 167 arthroscopic
Bankart repairs using suture anchor techniques, recurrence of instability was 4% at a mean postoperative
follow-up of 44 months with a mean loss of external rotation of 2.
Recently there have been two prospective studies
comparing open procedures with arthroscopic stabilization with biodegradable tacks for shoulder instability in
patients with Bankart lesions. In a prospective, randomized multicenter study, 30 patients were treated with arthroscopic stabilization and 26 patients were treated
with an open procedure. At 2-year follow-up, the recurrence rate in the arthroscopic group was 23% versus
12% in the open group. In another study, which was not
randomized, patients were evaluated at 3-year followup. The patients in open and arthroscopic groups had
similar demographics, although patients who underwent
open procedures had a greater number of dislocations
before surgery. In this study, the recurrence rate, including subluxations or dislocations, was 15% in the arthroscopic group compared with 10% in the open group.
The external rotation in abduction averaged 90 in the
arthroscopic group compared with 80 in the open
group.

287

Shoulder Instability
Traumatic glenohumeral defects have been associated with failed Bankart repairs, particularly when performed arthroscopically. Two bony lesions in particular,
the engaging Hill-Sachs lesion and the inverted-pear
glenoid, have been associated with high rates of recurrent instability after arthroscopic repair, particularly in
athletes who participate in contact sports. In the engaging Hill-Sachs lesion, the orientation of the Hill-Sachs
lesion is such that it engages the anterior glenoid in abduction and external rotation. In the inverted-pear glenoid, the normal pear-shaped configuration of the glenoid is altered because of significant anteroinferior
bone loss, which creates an inverted pear appearance.
With these osseous findings, open procedures with reconstruction of the osseous defects may be warranted.

Posterior Instability
Posterior instability occurs in 2% to 4% of patients with
shoulder instability. Causative factors for posterior instability include major trauma, repetitive microtrauma
as in overhead athletes, and generalized ligamentous
laxity. Acute posterior dislocations occur with load to a
flexed, adducted, and internally rotated upper arm and
are commonly missed at initial presentation. Patients
present with an adducted, internally rotated arm and
are unable to externally rotate the shoulder. The classic
radiographic findings on the AP view include a loss of
the humeral neck profile, a vacant glenoid sign, and an
anterior humeral head compression fracture (reverse
Hill-Sachs lesion). After recognition, gentle reduction
should be performed with lateral traction, external rotation, and abduction. Initial treatment after radiographic
confirmation of reduction is immobilization in slight extension and external rotation followed by physical therapy.
Most authors recommend a prolonged period of rehabilitation for symptoms of posterior instability. In patients who do not respond to nonsurgical treatment, a
variety of surgical and arthroscopic interventions have
been described. Open procedures include posterior capsulorrhaphy, bone block procedures, and glenoid or humeral osteotomy. One study reported results of open
posterior capsulorrhaphy for traumatic recurrent posterior subluxations in athletic patients; good or excellent
results were achieved in 13 of 14 patients. Arthroscopic
procedures include posterior capsular plication, thermal
shrinkage, and capsulolabral repair. In a recent retrospective review of 27 shoulders treated with arthroscopic repair using bioabsorbable tack fixation for a
posterior capsulolabral detachment (posterior Bankart
lesion), symptoms of pain and posterior instability were
eliminated in 92% of patients at a mean follow-up of 5.1
years. In another review of patients with traumatic unilateral posterior instability, 27 patients with posterior
Bankart lesions underwent arthroscopic repair using su-

288

Orthopaedic Knowledge Update 8

ture anchor fixation. At a mean follow-up of 39 months,
96% of these patients had a stable shoulder and were
able to return to their prior sports activity with few or
no limitations.

Multidirectional Instability
There is a common misconception that multidirectional
instability is limited to young sedentary patients with
generalized ligamentous laxity, an atraumatic history,
and bilateral symptoms. Although a subgroup of such
patients exists, shoulders with multidirectional instability are often seen in athletic patients, many of whom
have had an injury. Activities such as gymnastics or butterfly swimming may have resulted in repetitive microtrauma that selectively stretched out the shoulders,
and other joints may not be lax on examination. Additionally, Bankart lesions and humeral head impression
defects may be present in patients with multidirectional
instability, although less commonly than in patients with
unidirectional traumatic instability.
Symptoms typically include episodes of pain and instability that are positional or occur after minimal force.
The lesion in multidirectional instability was classically
thought to be a loose redundant inferior pouch with a
large rotator interval. However, the etiology of multidirectional instability appears to be multifactorial and
may include aberrant muscle firing patterns and abnormal connective tissue. In a recent MRI study, isometric
muscle activity leads to an off-centered humeral head
position in patients with multidirectional instability
compared with a recentered humeral head position during muscle firing in patients with traumatic instability.
Initial treatment focuses on rehabilitation, which has
a greater than 80% success rate if sustained for 6 to
12 months.
The goal of arthroscopic and open procedures is to
reduce the capsular volume of the joint while maintaining motion. Capsular shift, performed anteriorly or posteriorly, has an 85% to 90% success rate. Humeralbased as well as glenoid-based shift procedures have
been described and have yielded similar success rates.
Arthroscopic techniques include capsular plication,
rotator interval closure, and thermal capsular shrinkage.
Various arthroscopic devices have been developed to
deliver heat to capsular tissue. Investigators have noted
a 15% to 40% reduction in the length of collagenous tissue when it is heated to 65 to 72C. This reduction in
length has been shown to be caused by thermal denaturation of the collagen triple helix structure with subsequent reorganization into a random coil formation at a
shorter length. A recent retrospective review of the results of arthroscopic laser-assisted capsular shrinkage in
27 shoulders with multidirectional instability has shown
an overall success rate of 81.5% (with recurrent instability as a measure of failure) at an average follow-up of

American Academy of Orthopaedic Surgeons

Chapter 25 Shoulder Instability

Orthopaedic Knowledge Update 8

28 months. In another study using arthroscopic thermal
capsulorrhaphy to treat multidirectional instability in 30
shoulders, 76% of patients had a good or excellent result at an average follow-up of 3 years. The long-term
sequela of thermal shrinkage techniques is not known.
Recent studies of the behavior of lax ligaments treated
with electrothermal shrinkage in animal models have
found that although thermal shrinkage reduced laxity,
there was increased potential for creep and failure at
low physiologic stresses. A recent histologic analysis of
the capsule from seven patients after failed thermal capsulorrhaphy showed denudation of the synovial layer
and morphologic changes in the collagen layer that
included, in some cases, a hyalinization appearance.
Although no pathognomonic histologic changes were
identified, the study demonstrated that histologic abnormalities may be present for years after failed thermal
capsulorrhaphy.

Failed Instability Surgery

Recurrent instability after surgical treatment of instability may occur because of new trauma, incomplete treatment of pathologic lesions, poor rehabilitation, abnormal host healing response, or diagnostic error.
Treatment of recurrent stability after surgical stabilization is difficult. Prior to surgical intervention, careful
evaluation by history, physical examination, and diagnostic imaging should be performed to identify the
cause of the recurrence. Recurrent instability after
failed stabilization should be classified according to direction, degree, and frequency of instability.
Initial treatment almost always consists of extensive
rehabilitation. New traumatic lesions may heal, and
strengthening and improved control of the dynamic
muscular stabilizers may make the instability more tolerable.
Surgical intervention should be directed by the etiology of the recurrent instability. Common findings at revision surgery include excessive capsular redundancy,
uncorrected Bankart lesions, and bone loss from the glenoid or from the humeral head. Incomplete correction
of pathologic lesions, particularly Bankart lesions,
should be repaired anatomically. An inferior capsular
shift should be performed in patients with capsular laxity, particularly in those with a patulous axillary pouch.
One retrospective review investigated the outcome
of 50 patients who underwent revision anterior stabilization. Forty-nine of the procedures involved an open capsular shift, and 23 shoulders underwent concomitant repair of a Bankart lesion. At an average 4.7-year followup, good or excellent results were obtained in 78% of
patients. The authors identified factors associated with
poor results of the revision procedure that included
atraumatic causes of failure, voluntary dislocations, and
multiple prior stabilization attempts.

American Academy of Orthopaedic Surgeons

Irreparable capsular deficiency after failed stabilization procedures presents a particularly challenging situation. Capsular deficiency has been described after open
procedures because of subscapsularis tendon incompetence and after arthroscopic thermal capsulorrhaphy
caused by excessive thermal injury and tissue necrosis.
Deficiencies of the subscapularis tendon can be reconstructed using a transfer of the sternal head of the pectoralis major tendon. Capsular deficiencies can be reconstructed with a portion of the subscapularis tendon,
autograft tissue, or allograft tendon. A recent report in
which the iliotibial band was used to reconstruct the deficient capsular tissues in seven patients showed the
elimination of instability and the maintenance of a physiologic range of motion.
Osseous deficiency of the humerus or of the glenoid
is a rare etiology of recurrent instability after failed stabilization. Bone procedures to reconstruct osseous deficiencies include the Latarjet and the Bristow procedures
in which a portion of the coracoid is used to reconstruct
an inferior glenoid deficiency. Hill-Sachs lesions, an impaction fracture of the posterolateral margin of the humeral head, are found in more than 80% of patients
with traumatic anterior instability. These lesions are
thought to play a significant role in recurrent instability
if they comprise more than 30% of the proximal humerus articular surface or if they engage the glenoid.
Large Hill-Sachs lesions that engage the glenoid and
contribute to recurrent instability may be treated with
osteochondral allografts.
In patients with multiple failed stabilization procedures and recurrent, debilitating instability, arthrodesis
may be considered as a salvage procedure. A recent
study evaluated the efficacy of shoulder arthrodesis performed after an average of seven failed stabilization attempts. The results of eight patients were reviewed at a
mean follow-up of 35 months. All patients had achieved
bony union and all patients reported that they would repeat the surgery.

Chronic Dislocations
The diagnosis of chronic locked anterior or posterior
shoulder dislocations may be made after physical examination and proper radiographic analysis. Nonsurgical
treatment may be considered in elderly patients with
poor general mental status, especially when the condition is associated with limited pain. Surgical treatment is
necessary for reduction of the joint. Humeral head impaction fracture (Hill-Sachs lesions with chronic anterior dislocations and reverse Hill-Sachs lesions with
chronic posterior dislocations) may preclude stable reduction.

289

Shoulder Instability

Orthopaedic Knowledge Update 8

Figure 5 The Rockwood classification of ligamentous injuries to the acromioclavicular joint. Type I, no disruption of the acromioclavicular or coracoclavicular ligaments. Type II,
disruption of the acromioclavicular ligament (coracoclavicular ligaments remain intact.) Type III, disruption of the acromioclavicular and coracoclavicular ligaments. Type IV,
disruption of the acromioclavicular and coracoclavicular ligaments, and the distal end of the clavicle is displaced posteriorly into or through the trapezius muscle. Type V,
disruption of the acromioclavicular and coracoclavicular ligaments, along with disruption of the muscle attachments and creation of a major separation between the clavicle and
the acromion. Type VI, disruption of the acromioclavicular and coracoclavicular ligaments. Inferior dislocation of the distal clavicle in which the clavicle is inferior to the coracoid
process and posterior to the biceps and coracobrachialis tendons. (Reproduced with permission from Rockwood CA, Williams GR, Young DC: Shoulder instability, in Rockwood CA,
Green DP, Bucholz RN, Heckman JD (eds): Rockwood and Greens Fractures in Adults. Philadelphia, PA, Lippincott-Raven, 1996, p 1354.)

Acromioclavicular Separations
The acromioclavicular (AC) joint is a sturdy structure
that affixes the clavicle to the scapula. Injury to the AC
joint is common (comprising approximately 9% of
shoulder girdle injuries) and occurs most often in males
in their 20s. The AC joint is surrounded by and stabilized by a capsule and AC ligaments (superior, inferior,
anterior, and posterior). Additional joint stability is provided by the coracoclavicular ligaments (trapezoid and
conoid) as well as the coracoacromial ligament.
Injury to the AC joint is understood as a sequential
loss of AC stabilizers. The classification of AC joint instability reflects this anatomic progression of injury
(Figure 5) and is useful in directing treatment.
Radiographic evaluation of the AC joint may be performed with standard AP and lateral views of the shoulder. However, improved visualization of the AC joint is
achieved with a Zanca view, performed by tilting the
beam 10 to 15 cephalad. Additionally, only one third
to one half of the penetration strength for a standard
AP of the glenohumeral joint is used to more precisely
visualize the AC joint; standard shoulder radiographic
penetration will overpenetrate the less dense AC joint.
Stress views, using 5-lb weights placed on the wrist and
comparing both AC joints, have been advocated in the

290

past to help distinguish between type II and type III AC

joint separation; however, this technique is no longer favored because it usually does not influence the choice of
treatment.
Type I and II AC separations are considered incomplete lesions and may be treated conservatively with ice,
rest, and a brief period of immobilization followed by
therapy. Types IV, V, and VI are complete injuries that
usually require surgical reconstruction. Treatment of
type III injuries remains controversial.

Sternoclavicular Dislocations
The sternoclavicular joint is the only true articulation
between the clavicle and the axial skeleton. Sternoclavicular dislocations are usually the result of high-energy
force sustained during a motor vehicle accident or during contact sports activity.
Classification of sternoclavicular dislocations is
based on anatomic findings. Anterior dislocations are
most common and are present when the medial end of
the clavicle is displaced anterior or anterosuperior to
the sternum. Patients with anterior dislocations have a
palpable medial clavicular head mass that may be more
pronounced with abduction and elevation. Posterior dislocations are uncommon but more concerning. The me-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

dial end of the clavicle is displaced posteriorly or posterosuperiorly and may impinge on vital structures. The
patient presents with a hollow region just lateral to the
sternum, and may also present with dyspnea or dysphagia.
Radiographic evaluation involves standard chest radiographs as well as the serendipity view obtained by directing the x-ray beam 40 cephalad to visualize both
sternoclavicular joints. CT scan and/or MRI are currently the best techniques to evaluate the sternoclavicular joint and clearly distinguish the direction of the dislocation as well as any associated fractures. In patients
with posterior dislocation, CT and/or MRI allow for the
evaluation of mediastinal structures and visualization of
the medial clavicular head in relationship to these structures.
Acute reduction of anterior sternoclavicular dislocations may be attempted with traction of the extended
arm. Residual anterior instability is well tolerated but
may result in persistent deformity. Posterior dislocations
should be reduced in the operating room with care
taken to avoid damage to the mediastinal structures.

Annotated Bibliography
Pathophysiology
Urayama M, Itoi E, Hatakeyama Y, Pradhan RL, Sato
K: Function of the 3 portions of the inferior glenohumeral ligament: A cadaveric study. J Shoulder Elbow
Surg 2001;10:589-594.
The strain of the three portions or sections of the inferior
glenohumeral ligament in 17 fresh-frozen cadaveric shoulders
was studied during elevation and rotational maneuvers. The
anterior band and axillary pouch showed the greatest strain in
abduction and external rotation, confirming their role as anterior stabilizers. The posterior band showed the greatest strain
with flexion and internal rotation, suggesting a key role as a
posterior stabilizer.

Urayama M, Itoi E, Sashi R, Minagawa H, Sato K: Capsular elongation in shoulders with recurrent anterior
dislocation: Quantitative assessment with magnetic resonance arthrography. Am J Sports Med 2003;31:64-67.
Magnetic resonance arthrography was used to evaluate the
length of the anteroinferior, inferior, and posteroinferior capsule in 12 patients with unilateral recurrent anterior instability.
Unaffected shoulders were used as controls. The anteroinferior and inferior portions of the shoulder capsule were elongated an average of 19% in shoulders with recurrent anterior
dislocation compared with the unaffected shoulders.

Treatment of Acute Dislocation

Bottoni CR, Wilckens JH, DeBerardino TM, et al: A
prospective, randomized evaluation of arthroscopic stabilization versus nonoperative treatment in patients
with acute, traumatic, first-time shoulder dislocations.
Am J Sports Med 2002;30:576-580.

American Academy of Orthopaedic Surgeons

Chapter 25 Shoulder Instability

Twenty-one active duty military personnel between the
ages of 18 and 26 years with primary, traumatic shoulder dislocation were randomized to nonsurgical treatment or arthroscopic Bankart repair using a bioabsorbable tack. At an average of 3 years follow-up, 75% of patients treated nonsurgically
versus 11% of the patients treated with arthroscopic repair developed recurrent instability. Six of nine patients treated nonsurgically who developed recurrent instability required subsequent open Bankart repair.

Buss DD, Lynch GP, Meyer CP, Huber SM, Freehill MQ:
Nonoperative management for in-season athletes with
anterior shoulder instability. Am J Sports Med 2004;32:
1430-1433.
Thirty in-season athletes with either acute or recurrent anterior instability were treated with physical therapy and, if appropriate, bracing. Eighty-seven percent of athletes were able
to return to their sport in season and an average of 1.4 recurrent instability episodes per season per athlete occurred. No
further injuries were attributable to the shoulder instability.
Fifty-three percent of patients had subsequent surgical stabilization in the off-season.

DeBerardino TM, Arciero RA, Taylor DC, et al: Prospective evaluation of arthroscopic stabilization of
acute, initial anterior shoulder dislocations in young athletes: Two- to five-year follow-up. Am J Sports Med
2001;29:586-592.
Forty-eight cadets at the US Military Academy with 49 anterior dislocations were treated with primary arthroscopic repair using bioabsorbable tacks. At an average follow-up of 37
months, the average Rowe score was 92%, and 88% of shoulders remained stable. Factors associated with failure included
a history of bilateral shoulder instability, a 2+ sulcus sign, and
poor capsulolabral tissue at the time of repair. All patients
with stable shoulders returned to their preinjury levels of athletic activity. These results are favorable compared with nonsurgical treatment in young, active adults at the US Military
Academy.

Itoi E, Hatakeyama Y, Kido T, et al: A new method of

immobilization after traumatic anterior dislocation of
the shoulder: A preliminary study. J Shoulder Elbow
Surg 2003;12:413-415.
Forty patients were randomly assigned to immobilization
in external rotation versus immobilization in internal rotation
for 3 weeks after initial anterior dislocation. The average ages
in the internal and external groups were 38 years and 40 years,
respectively, with similar numbers of patients under the age of
29 years. At approximately 1-year follow-up, the recurrence
rate was 30% in the internal rotation group and 0% in the external rotation group. Anterior apprehension was positive in
14% of the internal rotation group without recurrence and 5%
of the external rotation group.

Miller SL, Cleeman E, Auerbach J, Flatow EL: Comparison of intra-articular lidocaine and intravenous seda-

291

Orthopaedic Knowledge Update 8

Shoulder Instability
tion for reduction of shoulder dislocations: A randomized, prospective study. J Bone Joint Surg Am 2002;84:
2135-2139.
In a prospective study, 30 patients with anterior glenohumeral dislocation were randomized to receive either intraarticular lidocaine or intravenous sedation before relocation
using the Stimson method. There was no significant difference
between the two groups with regard to pain, success of the
Stimson technique, or time required for reduction of the
shoulder. The lidocaine group spent significantly less time in
the emergency department and required less nursing resources.

Treatment of Recurrent Traumatic Instability

Karlsson J, Magnusson L, Ejerhed L, Huttenheim I,
Lundin O, Kartus J: Comparison of open and arthroscopic stabilization for recurrent shoulder dislocation in
patients with a Bankart lesion. Am J Sports Med 2001;
29:538-542.
One hundred eight shoulders with symptomatic, recurrent
anterior instability and a Bankart lesion underwent either
open stabilization or arthroscopic stabilization with bioabsorbable tack fixation. At a mean follow-up of 28 months, the recurrence rate, including both dislocations and subluxations,
was 9 of 60 (15%) in the arthroscopic group, compared with
5 of 48 (10%) in the open group. No significant differences
were found between the study groups for the Rowe or Constant scores at follow up. The only significant difference in
range of motion assessment was in external rotation in abduction, which was 90 (range, 50 to 135) in the arthroscopic
group and 80 (range, 25 to 115) in the open group. The
treatment groups in this study were not randomized.

Kim SH, Ha KI, Cho YB, Ryu BD, Oh I: Arthroscopic

anterior stabilization of the shoulder: Two to six-year
follow up. J Bone Joint Surg Am 2003;85:1511-1517.
One hundred sixty-seven patients with traumatic recurrent
anterior instability were treated with arthroscopic Bankart repair using suture anchors. At a mean follow-up of 44 months,
the rate of postoperative recurrence of instability was 4% with
a mean loss of external rotation of 2. Postoperative recurrence was related to an osseous defect of more than 30% of
the glenoid circumference. These results compare favorably
with historical results of open stabilization.

Sperber A, Hamberg P, Karlsson J, Sward L, Wredmark

T: Comparison of an arthroscopic and an open procedure for posttraumatic instability of the shoulder: A
prospective, randomized multicenter study. J Shoulder
Elbow Surg 2001;10:105-108.
Fifty-six patients with anterior instability and a Bankart lesion were randomized to either open stabilization or arthroscopic stabilization with the use of bioabsorbable tacks. Patients were evaluated at 2, 12, and 24 months postoperatively.
Thirty patients were surgically treated with the arthroscopic
technique and 26 patients with the open technique. In the arthroscopic group, there were recurrences in 7 of 30 patients

292

(23%) at a mean of 13 months (range, 5 to 21 months) after

surgery. In the open group, there were recurrences in 3 of 26
patients (12%) at a mean of 10 months (range, 2 to 23
months) after surgery (P = not significant). A tendency toward
more redislocations in the arthroscopic group was noted.

Posterior Instability
Kim SH, Ha KI, Park JH, et al: Arthroscopic posterior
labral and capsular shift for traumatic unidirectional recurrent posterior subluxation of the shoulder. J Bone
Joint Surg Am 2003;85:1479-1487.
Twenty-seven patients with unidirectional posterior instability were treated with arthroscopic labral repair and posterior capsular shift using suture anchors. At a mean postoperative follow-up of 39 months, all patients had a stable shoulder
by subjective and objective measurements except for one patient who had recurrent subluxation. Twenty-six patients returned to prior sports activity with few or no limitations. The
authors also describe arthroscopic findings in traumatic unidirectional recurrent posterior instability.

Williams RJ III, Strickland S, Cohen M, Altchek DW,

Warren RF: Arthroscopic repair for traumatic posterior
shoulder instability. Am J Sports Med 2003;31:203-209.
Twenty-seven shoulders in 26 patients with traumatic posterior shoulder instability were treated with arthroscopic repair using bioabsorbable tack fixation. At a mean follow-up of
5.1 years, no patients showed a range-of-motion deficit. There
was no instability greater than 1+ in the anterior, posterior, or
inferior directions. Symptoms of pain and instability were
eliminated in 24 patients (92%). Two patients (8%) required
additional surgery after arthroscopic repair of the posterior
Bankart lesion.

Multidirectional Instability
Favorito PJ, Langenderfer MA, Colosimo AJ, Heidt RS
Jr, Carolonas RL: Arthroscopic laser-assisted capsular
shift in the treatment of patients with multidirectional
shoulder instability. Am J Sports Med 2002;30:322-328.
Twenty-seven shoulders in 25 patients with multidirectional shoulder instability were treated with an arthroscopic
laser-assisted capsular shift procedure. At an average
follow-up of 28 months, 22 shoulders had no recurrent symptoms and required no further surgical intervention. In five
shoulders, treatment was considered a failure because of recurrent pain or instability and the need for an open capsular
shift procedure. With recurrent instability as a measure of failure, the overall success rate was 81.5%.

Fitzgerald BT, Watson BT, Lapoint JM: The use of thermal capsulorrhaphy in the treatment of multidirectional
instability. J Shoulder Elbow Surg 2002;11:108-113.
Thirty shoulders with multidirectional instability were
treated with arthroscopic thermal capsulorrhaphy. At a mean
follow-up of 36 months (range, 24 to 40 months), 3 excellent,
20 good, and 7 poor results were reported using the University

American Academy of Orthopaedic Surgeons

Chapter 25 Shoulder Instability

Orthopaedic Knowledge Update 8

of California, Los Angeles score. Twenty-three patients (76%)
returned to full activity.

McFarland EG, Kim TK, Banchasuek P, McCarthy EF:

Histologic evaluation of the shoulder capsule in normal
shoulders, unstable shoulders, and after failed thermal
capsulorrhaphy. Am J Sports Med 2002;30:636-642.
Shoulder capsules were evaluated histologically in 12 patients with traumatic anterior instability and in 7 patients who
experienced recurrent instability after a thermal capsulorrhaphy. The capsules of six fresh-frozen cadavers with no shoulder
lesions were used as controls. In patients who had a history of
traumatic instability, a denuded synovial layer was present in
58%, subsynovial edema in 58%, increased cellularity in 25%,
and increased vascularity in 83%. At the time of surgery, five
of seven shoulders in the failed thermal capsulorrhaphy group
(71%) were subjectively felt to be thin and attenuated. Denuded synovium was found in 100% of these patients, subsynovial edema in 43%, and changes in the collagen layer in
100%. Changes in the collagen layer in these patients included
the appearance of hyalinization in five patients (71%), increased collagen fibrosis in two patients (29%), and increased
cellularity in two patients (29%).

von Eisenhart-Rothe RM, Jager A, Englmeier KH, Vogl

TJ, Graichen H: Relevance of arm position and muscle
activity on three-dimensional glenohumeral translation
in patients with traumatic and atraumatic shoulder instability. Am J Sports Med 2002;30:514-522.
Open MRI and three-dimensional post processing technique was used to evaluate glenohumeral translation with different arm positions with and without muscle activation in 12
patients with traumatic and 10 patients with atraumatic instability. In patients with traumatic instability, increased translation was observed only in functionally important arm positions, whereas intact active stabilizers showede sufficient
recentering. In patients with atraumatic instability, a decentralized head position was recorded also during muscle activity,
suggesting alterations of the active stabilizers.

Wallace AL, Hollinshead RM: FrankCB: Electrothermal

shrinkage reduces laxity but alters creep behavior in a lapine ligament model. J Shoulder Elbow Surg 2001;10:1-6.
Using a lapine medial collateral ligament laxity model, the
acute effects of radiofrequency shrinkage were assessed. Thermal treatment resulted in restoration of laxity but a significant
increase in the cyclic and static creep strains compared with control sides. The authors concluded that radiofrequency electrothermal shrinkage is effective at reducing laxity but significantly
alters viscoelastic properties, posing a risk of recurrent
stretching-out at physiologic loads.

Failed Instability Surgery

Diaz JA, Cohen SB, Warren RF, Craig EV, Allen AA:
Arthrodesis as a salvage procedure for recurrent instability of the shoulder. J Shoulder Elbow Surg 2003;12:
237-241.

American Academy of Orthopaedic Surgeons

Eight patients with recurrent shoulder instability with an

average of seven prior stabilization attempts were treated with
glenohumeral arthrodesis. The average time to bony union after arthrodesis was 3.5 months (range, 2.5 to 5 months). At
mean follow-up of 35 months, the patients reported significant
overall subjective improvement as a group after fusion. None
of the patients reported instability postoperatively. All eight
patients stated that they would repeat the surgery again under
similar preoperative circumstances.

Iannotti JP, Antoniou J, William GR, Ramsey ML: Iliotibial band reconstruction for treatment of glenohumeral instability associated with irreparable capsular
deficiency. J Shoulder Elbow Surg 2002;11:618-623.
Seven patients with recurrent anterior instability after
failed surgery complicated by the loss of capsular tissue underwent reconstruction of the capsular ligaments using the iliotibial band. After iliotibial band reconstruction, the patients
showed significant improvement in their American Shoulder
and Elbow Surgeons score (P =0.0004), and no patient had
any persistent symptoms of instability. Physiologic range of
motion and function were maintained. The authors describe
their surgical technique for iliotibial band reconstruction for
capsular deficiency.

Acromioclavicular Separations
Schlegel TF, Burks RT, Marcus RL, Dunn HK: A prospective evaluation of untreated acute grade III acromioclavicular separations. Am J Sports Med 2001;29:699703.
Twenty patients with acute grade III AC separations were
treated nonsurgically with a sling for comfort through progressive early range of motion as tolerated and completed a 1-year
evaluation and strength-testing protocol. Subjectively, 4 of the
20 patients (20%) thought that their long-term outcome was
suboptimal, although 3 of them did not believe that the outcome should warrant surgery. Objective examination and
strength testing of the 20 patients revealed no limitation of
shoulder motion in the injured extremity and no difference
between sides in rotational shoulder muscle strength. The
bench press was the only strength test that showed a significant short-term difference, with the injured extremity being an
average of 17% weaker.

Classic Bibliography
Bankart ASB: The pathology and treatment of recurrent
dislocation of the shoulder-joint. Br J Surg 1938;26:2329.
Burkhart SS, DeBeer JF: Traumatic glenohumeral bone
defects and their relationship to failure of arthroscopic
Bankart repairs: Significance of the inverted-pear
glenoid and humeral engaging Hill-Sachs lesion.
Arthroscopy 2000;16:677-694.

293

Shoulder Instability
Curl LA, Warren RF: Glenohumeral joint stability: Selective cutting studies on the static capsular restraints.
Clin Orthop 1996;330:54-65.
Hovelius L, Augustini BG, Fredin H, Johansson O, Norlin R, Thorling J: Primary anterior dislocation of the
shoulder in young patients: A ten-year prospective
study. J Bone Joint Surg Am 1996;78:1677-1684.
Neer CS II, Foster CR: Inferior capsular shift for involuntary inferior and multidirectional instability of the
shoulder: A preliminary report. J Bone Joint Surg Am
1980;62:897-908.
OBrien SJ, Neves MC, Arnoczky SP, et al: The anatomy
and histology of the inferior glenohumeral ligament
complex of the shoulder. Am J Sports Med 1990;18:449456.
Rowe CR, Patel D, Southmayd WW: The Bankart procedure: A long-term end-result study. J Bone Joint Surg
Am 1978;60:1-16.

Orthopaedic Knowledge Update 8

Thomas SC, Matsen FA III: An approach to the repair
of avulsion of the glenohumeral ligaments in the management of traumatic anterior glenohumeral instability.
J Bone Joint Surg Am 1989;71:506-513.
Tossy JD, Mead NC, Sigmon HM: Acromioclavicular
separations: Useful and practical classification for treatment. Clin Orthop 1963;28:111-119.
Weaver JK, Dunn HK: treatment of acromioclavicular
injuries, especially complete acromioclavicular separation. J Bone Joint Surg Am 1972;54:1187-1194.
Wheeler JH, Ryan JB, Arciero RA, et al: Arthroscopic
versus nonsurgical treatment of acute shoulder dislocations in young athletes. Arthroscopy 1989;5:213-217.
Zabinski SJ, Calloway HG, Cohen S, Warren RF:
Revision shoulder stabilization 2 to 10 year results.
J Shoulder Elbow Surg 1999;8:58-65.

Speer KP, Deng X, Borrero S, Torzilli PA, Altchek DA,

Warren RF: Biomechanical evaluation of a simulated
Bankart lesion. J Bone Joint Surg Am 1994;76:18191826.

294

American Academy of Orthopaedic Surgeons

Chapter

26

Shoulder Reconstruction
Leesa M. Galatz, MD

Rotator Cuff Tear

Pathophysiology
The etiology of rotator cuff tears has not been clearly
elucidated. Focus has centered primarily on intrinsic and
extrinsic mechanisms of generating rotator cuff tears,
with reports in the early literature primarily focused on
an extrinsic mechanism. Three types of acromial morphology were defined (types I, II, and III), with increasing curvature of each type. Type III acromions, with
maximal curvature or hooking of the undersurface, were
associated with an increased incidence of rotator cuff
tears resulting from increased mechanical contact of the
cuff with the bone. Recent studies have provided data
that do not substantiate this early information and suggest a much more complex etiology for rotator cuff
tears.
Attention has focused on the role and contribution
of plain radiographic imaging for the diagnosis and detection of full-thickness rotator cuff tears. Acromial osteophytes or spurring, acromioclavicular degeneration,
and lateral clavicular and acromial sclerosis seen on
plain radiographs have been correlated with age rather
than rotator cuff disease. Rotator cuff disease, especially
involving full-thickness tears, has also been correlated
with age, complicating the issue. Although some cadaveric studies have demonstrated an association between
spurring of the anterior acromion and the acromioclavicular joint and rotator cuff tears, a direct causal relationship between degenerative radiographic changes
and full-thickness tears has not been definitively
proven. One recent study of radiographic changes in 40
patients with full-thickness cuff tears and a large group
of age-matched controls found no association between
acromial morphology and rotator cuff tears. Findings
that were indicative of cuff tears included sclerosis, osteophytes, subchondral cysts, and osteolysis of the
greater tuberosity. There was no relationship between
any of these findings in the greater tuberosity and tear
size.
Intrinsic mechanisms of rotator cuff degeneration
have been implicated in the etiology of rotator cuff

American Academy of Orthopaedic Surgeons

tears. Histologic changes in chronic rotator cuff tissue

tears have shown degenerative changes, but there is little evidence of acute inflammation. This finding reinforces the notion that changes in proteoglycan and
other extracellular matrix protein content in the aging
tendon may have a role in attritional changes leading to
tears. More research is necessary to make definitive
conclusions regarding the true etiology, and it is likely
that both intrinsic and extrinsic mechanisms as well as
patient-related factors contribute to the development of
rotator cuff disease.
After a rotator cuff is torn, several changes can
occur in the shoulder joint. Clinically these changes
manifest as pain and decreased strength and range of
motion. Radiographic changes can include proximal migration of the humeral head, narrowing of the acromialhumeral interval, sclerosis of the undersurface of the acromion and superior humeral head, acetabularization or
rounding of the undersurface of the acromion such that
it is congruent with the humeral head, and degenerative
changes of the glenohumeral joint. Most of these
changes occur only with massive, long-standing rotator
cuff tears. These structural changes to the joint with rotator cuff tears result from kinematic changes secondary
to loss of the normal muscle forces about the shoulder.
The purposes of the rotator cuff are to keep the humeral head centered on the glenoid (termed concavity
compression) by counteracting the superior vector of
the deltoid, and to add strength and dynamic stability to
glenohumeral motion.
Several recent basic science studies have been performed to investigate the changes in forces and motion
patterns associated with rotator cuff tears. One cadaveric study examined the effect of rotator cuff tears on
joint reaction forces at the glenohumeral joint. There
were no significant differences in joint reaction force
magnitude or direction with abduction motion between
the intact state, partial supraspinatus tear, or complete
supraspinatus tear. Extension of tears beyond the supraspinatus tendon resulted in a significant decrease in
the magnitude of the joint reaction force. Additionally,
the vector of the force changed direction toward the de-

295

Shoulder Reconstruction
ficient area. Clinically, this finding may explain why
some patients with fairly large tears may still maintain
function. The critical tear size at which the force couple
of the remaining musculature is overcome and pseudoparalysis of the shoulder results is yet to be determined, and clinically seems to differ between individuals.
Another cadaveric study was performed to compare
the effects of supraspinatus detachments, tendon defects, and muscle retraction. Simply detaching a portion
of the supraspinatus tendon had no effect on the force
transmitted by the rotator cuff. Detaching or creating a
defect of the entire supraspinatus tendon led to a moderate decrease in force transmission (11% and 17%, respectively). However, when the investigators simulated
muscle retraction of the tendon by incising medially to
detach the tendon from adjacent tendon tissue, there
were substantial reductions in force transmission (58%
with involvement of the entire supraspinatus). Side-toside repair of smaller defects restored force capability,
but a deficit remained even after side-to-side repair of
the entire supraspinatus tendon retraction simulation.
This study supports the cable concept of force transmission of the rotator cuff, and suggests that the amount of
retraction and not just transverse diameter may be an
important factor in functional deficit after a rotator cuff
tear.

Natural History
The severity of rotator cuff disease ranges from painful
cuffs without tears to partial-thickness tears to fullthickness tear. It is not known whether this is a continuous spectrum in which a patient will go from one stage
to the next as part of the natural history of their condition or whether some patients present at one stage,
never progressing to another. One longitudinal ultrasonographic study showed that over a time period of
5 years, approximately 28% of patients had an increase
in size of known full-thickness rotator cuff tears. Additionally, more recent data show a high prevalence of
full-thickness tears in the opposite shoulders of patients
with full-thickness tears, and this was also seen more often as patients get older. Research on the natural history of tears is ongoing and more information will likely
be available in the near future.

Evaluation
History and Examination
Patients with rotator cuff tears report a history of shoulder pain and/or weakness. Pain associated with rotator
cuff tears is usually located on the anterolateral aspect
of the shoulder and often radiates distally toward the
deltoid insertion. Pain that radiates below the elbow to
the hand should raise suspicion of cervical radiculopathy or peripheral nerve compression.

296

Orthopaedic Knowledge Update 8

Examination should begin with an inspection of both
shoulders to detect asymmetry or muscle atrophy. The
examiner should stand behind the patient and observe
the scapulae during overhead elevation for winging and
motion pattern. Range of motion should be recorded
for elevation in the scapular plane, external rotation at
the side, and external rotation in abduction to fully examine the anterior capsule. Internal rotation can be
tested by internal rotation of the shoulder with the arm
in 90 of abduction and by asking the patient to reach
behind the back. Side-to-side comparison is critical to
determine normal motion, as range of motion differs between individuals.
Supraspinatus strength is tested by external rotation
with the arm at the side with the shoulder in neutral or
slight internal rotation. Testing external rotation with
the shoulder in external rotation allows substitution by
the posterior deltoid. Placing the arm in 90 of abduction in the scapular plane with the thumbs down and
asking the patient to resist downward pressure is another way to isolate the supraspinatus. This test (Jobes)
is useful for both strength and pain. A lag sign (for testing the posterosuperior rotator cuff) is performed by
placing the arm in maximal external rotation. A patient
with a large or massive tear will not be able to maintain
the arm in this position and the hand will swing toward
neutral rotation. In a patient with a massive tear, including the teres minor, the hornblowers sign will be positive (the patient is unable to externally rotate the arm
to 90 with the arm in abduction). Positive results from
these two tests are usually associated with massive tears
and there is correlation with fatty degeneration of the
cuff musculature.
Subscapularis integrity is tested using the lift-off test
and the abdominal compression test. The lift-off test involves placing the arm behind the back, at the midline,
at waist level. The patient is asked to raise the hand off
the back, a motion requiring an intact subscapularis. An
abdominal compression test is performed by placing the
hand flat on the abdomen with a straight line between
the hand, wrist, and elbow. A patient with a torn subscapularis will either bend the wrist to keep the hand on
the abdomen or lose contact. Internal strength can be
tested in this position as well.
The Neer and Hawkins impingement signs are used
to detect pain related to rotator cuff disease by provoking impingement in the subacromial space by internally
rotating the shoulder, bringing the greater tuberosity
underneath the acromion, and minimizing space available for the cuff. A thorough examination should also
include evaluation of the acromioclavicular joint for
pain on direct palpation and pain with cross-body adduction. Biceps tendon pathology is often associated
with rotator cuff tears. The Speed and Yergason tests are
sensitive for biceps tendon pain.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Diagnostic Studies
Plain radiographs should be the initial radiographic
study of choice. A standard series should include an AP
radiograph of the shoulder with the arm in internal rotation, an AP radiograph of the scapula with the arm in
neutral or slight external rotation, an outlet view, and an
axillary view. The acromioclavicular joint is best visualized on the AP view of the shoulder. The glenohumeral
joint space is assessed on the scapular AP and the axillary views, in which degenerative changes are best detected. The outlet view enables evaluation of acromial
morphology. MRI is valuable for evaluating the soft tissues about the shoulder and is a sensitive and specific
test for rotator cuff pathology. The role of ultrasonography in the evaluation of the rotator cuff has gained popularity in recent years because it is highly accurate in
making the diagnosis of full-thickness and partialthickness tears. Because the reliability of ultrasonography is operator-dependent, significant experience should
be gained before this study is used independently. MRI
has largely replaced arthrography, which is rarely used;
however, arthrography remains a good alternative if an
MRI evaluation cannot be obtained and an ultrasound
is not available.

Treatment of Partial-Thickness Rotator Cuff Tears

Partial-thickness tears are often managed successfully
with nonsurgical treatment. An exercise program focuses on range of motion, stretching, and strengthening
of the rotator cuff, deltoid, and scapular stabilizers.
Partial-thickness tears that do not become asymptomatic with nonsurgical treatment can be effectively
treated surgically. Arthroscopic treatment entails dbridement of the tear after careful inspection of both the
articular and bursal sides. In general, if the tear involves
less than 50% of the thickness of the cuff, dbridement
should be adequate. If the tear involves more than 50%
of the thickness of the cuff or the remaining tissue is degenerative and easily dbrided, then the tear is expanded to a full-thickness tear and healthy tissue is repaired primarily. Recently, several arthroscopic methods
of imbricating partial-thickness tears and repairing laminated flaps have been described, and offer promising
options in the management of partial-thickness tears.

Treatment of Full-Thickness Rotator Cuff Tears

Nonsurgical treatment for full-thickness tears is the same
as for partial-thickness tears in terms of exercise and pain
control measures. Surgical treatment of symptomatic fullthickness rotator cuff tears is indicated in patients in
whom nonsurgical treatment has failed and who are candidates for surgery, those with acute tears, and patients
with tears associated with a sudden loss of strength and
function. Relative indications include younger people of
working age with a painful tear. Factors to consider in-

American Academy of Orthopaedic Surgeons

Chapter 26 Shoulder Reconstruction

clude the following: (1) tears are easier to repair when
they are smaller, and there is documented evidence that
tear size can increase with time; (2) there is a higher recurrence rate after repair of large and massive tears;
(3) long-standing chronic tears are associated with muscle
atrophy and fatty degeneration, which are irreversible
changes that can compromise functional outcome after repair; and (4) there may be attritional changes in the tendon edge of chronic tears that may compromise healing
potential.
Surgical Approaches
An open surgical approach involves an incision and removal of the anterior deltoid origin from the acromion
to maximize exposure. The major advantage of an open
repair is the extensive exposure and ability to mobilize
the cuff under direct observation. It also requires no arthroscopic skills. A mini-open repair is alternatively performed through a deltoid split rather than removing the
deltoid origin. To achieve a repair using a mini-open approach, arthroscopic assistance is used. Because exposure is more limited, inspection, cuff mobilization, and
the acromioplasty are performed with the use of the arthroscope, obviating the need for a more extensile approach. The major advantage of a mini-open repair is
that the deltoid is not removed from the acromion. Removal of the deltoid is one of the more painful aspects
of the open procedure, and failure of the deltoid repair
is a serious and painful complication for which there is
no reliable option for correction.
Arthroscopic repair of full-thickness tears has gained
recent popularity. As equipment and techniques continue to advance, this approach will likely continue to
become easier and more achievable with advanced arthroscopic skills. Potential advantages of an arthroscopic
approach include decreased scarring, easier rehabilitation, decreased pain, and improved cosmesis. Damage to
the deltoid is minimized with an arthroscopic approach.
One disadvantage to an arthroscopic repair may be a
higher recurrence rate after repair of large and massive
tears. The procedure requires advanced arthroscopic
skills, which may be considered by some to be a disadvantage of using the procedure; however, the use of
minimally invasive approaches is becoming mainstream.
Role of Acromioplasty
An acromioplasty has historically been a standard aspect of a rotator cuff repair based on Neers classic description of impingement syndrome and the theory that
contact with the rigid coracoacromial arch is the mechanical impetus for rotator cuff tears. An acromioplasty entails takedown of the coracoacromial ligament
from the anterior acromion and removal of bone from
the undersurface of the acromion to increase the space
available for the rotator cuff. Historically, removal of
large amounts of bone, including full-thickness removal

297

Orthopaedic Knowledge Update 8

Shoulder Reconstruction

Figure 1 A, Radiograph of a shoulder after failed rotator cuff repair and distal clavicle resection. The anchors have failed, the humeral head has migrated anteriorly and
superiorly, and degenerative changes have begun to involve the glenohumeral joint. B, MRI of the shoulder demonstrates the massive rotator cuff tear and proximal migration.
C, A sagittal oblique cut of the MRI demonstrates the cuff insufficiency as well as the deltoid attenuation.

series of tears involving the subscapularis, a lower Constant score correlated with duration of symptoms longer
than 6 months and the appearance of fatty degeneration
and atrophy of the subscapularis muscle as detected by
MRI. The authors recommended repair of the subscapularis before 6 months of symptoms to maximize functional outcome.

of bone from the anterior acromion in some instances,

was recommended. The more recent trend for acromioplasty is to remove enough bone to make the undersurface of the acromion flat to protect the origin of the deltoid. One anatomic study found that the average
thickness of the acromion was 7 mm. Therefore, removal
of large amounts of bone, as had been previously recommended, could potentially compromise the acromion
and deltoid origin.

Complications

Results of Treatment
Rotator cuff repair has historically been a reliable option for pain relief over time. This finding has been substantiated by two long-term follow-up studies. In a recent study, results of a large number of patients treated
with open cuff repair, V-Y plasty, tendon transposition,
and reinforcement with fascia lata were reported. Tear
size was the most important determinant of outcome
with regard to active motion, strength, rating of the result, patient satisfaction, and need for another operation. Another recent study reported over 90% good to
excellent results in a prospective series of patients after
open cuff repair on the shoulder. These patients were
studied longitudinally over a 10-year period. Outcome
at 10 years had not deteriorated from the 2-year results,
demonstrating the longevity of results after rotator cuff
repair.
One recent study showed 46 of 48 good to excellent
results after arthroscopic repair of medium to large fullthickness rotator cuff repairs. Forty-four of 45 patients
were satisfied with the results. Another study reported
on the results of arthroscopic rotator cuff repair of large
and massive chronic tears. Despite a high rate of satisfaction and significantly improved functional outcome,
ultrasonographic analysis revealed 17 of 18 patients had
recurrent tears. These results suggest that success in
terms of functional outcome and pain relief may not
correlate with anatomic healing of the rotator cuff. In a

The rate of complications after rotator cuff repair is low

in most series. Potential complications include stiffness,
infection, deltoid dehiscence after open repair, failure of
repair, and less commonly, neurovascular injury. Although the incidence of infection is low after rotator
cuff repair, recently one organism, Propionibacterium
acnes, has gained attention as a frequent cause of infection in the shoulder. This is a low virulence organism
that was found in the few infections in a large series of
mini-open rotator cuff repairs.
Another complication that has become increasingly
recognized is anterior-superior instability in the shoulder associated with massive, irreparable cuff tears. This
complication results from a combination of rotator cuff
insufficiency, disruption of the coracoacromial arch, and
anterior deltoid dehiscence. Anterior-superior instability
usually is an iatrogenic complication after open cuff repair and subacromial decompression where the deltoid
and rotator cuff repair have failed (Figure 1). Clinically,
it is recognized by proximal and anterior migration of
the humeral head to the subcutaneous position and obligates extension of the shoulder with attempted overhead elevation. This condition occurs because the intact
posterior and lateral deltoid exert a superior vector on
the humerus, with no rotator cuff or coracoacromial
arch to keep the head contained. Surgical attempts at
coracoacromial arch reconstruction and muscle transfers
for treatment of this problem have met with unpredictable results and limited success.

298

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 26 Shoulder Reconstruction

Treatment of the Massive Rotator Cuff Tear

Treatment of the massive rotator cuff tear is a controversial issue. It is important to consider that the massive
tear does not equal the irreparable tear. Many massive
tears are easily repairable, and some smaller tears and
many massive tears can be irreparable. Preoperative
evaluation of the massive tear should give some indication as to the likelihood of successful repair. MRI demonstrating significant atrophy and fatty degeneration of
the musculature (Figure 2) along with a history consistent with a chronic tear can indicate a difficult if not impossible repair. Most studies that have examined the anatomic results after rotator cuff repair have
demonstrated a higher rate of recurrence with larger
tears. Therefore, preoperative discussions with patients
should include a discussion of these results as well as
patient goals and expectations after surgery. Overall, repair of a massive tear can be potentially predictable for
pain relief, but it is much less reliable for restoring lost
strength.
Nonsurgical treatment should be attempted initially,
especially in the older patient with lower physical demands. Younger patients of working age in whom
strength is more important are exceptions; early intervention should be considered in these patients to avoid
future irreversible, degenerative changes that compromise results. Physical therapy helps to strengthen the remaining musculature about the shoulder and improve
kinematics to facilitate or regain range of motion. Nonsteroidal anti-inflammatory drugs and cortisone injections are options for providing pain relief.
Surgical treatment options include open or miniopen repair, dbridement, and arthroscopic repair. Most
studies of open and mini-open repair have found that an
increasing tear size correlates with inferior results. Simple dbridement and acromioplasty have been advocated, especially for irreparable tears, with a reported
83% satisfaction rate for this procedure at an average
6.5-year follow-up. In other studies, it was shown that
the results of treatment with dbridement were inferior
to results in a series in which the cuffs were repaired.
Results of dbridement deteriorated with time, and a
comparison of results of dbridement with those of
open repair found that patients who had open repair
had better strength, function, and outcome scores, although patient satisfaction was comparable.
Arthroscopic treatment is particularly advantageous
in the management of massive rotator cuff tears because
it allows assessment of whether the tear is repairable
before an open incision and splitting or detaching of the
deltoid is performed. An intact deltoid is critical for a
good result in a patient with a massive tear, whether or
not it is repaired. Limited, early results of arthroscopic
repair are promising. In a previously mentioned study of
arthroscopic repair of large and massive tears, there was

American Academy of Orthopaedic Surgeons

Figure 2 An MRI of a massive cuff tear. The cuff has retracted to the level of the
superior glenoid rim. The muscle belly has undergone severe atrophic changes and a
widened subtrapezial fat space (arrow) is evident. There is also proximal migration of
the humeral head relative to the glenoid.
a significant improvement in functional outcome and
pain relief, and a high degree of patient satisfaction despite a high recurrence rate. The margin convergence
technique has been used in patients undergoing arthroscopic repair of massive tears, with an improvement in
range of motion and high degree of patient satisfaction.
One recent study reported on the results of tuberoplasty, a procedure in which the bony excrescences of
the greater tuberosity are dbrided such that the humeral head conforms to the rounded undersurface of
the acromion. No subacromial decompression was performed in this series. Although the long-term results of
arthroscopic treatment in patients with massive tears
are not yet available, early results suggest that it can
play a role in the management of this challenging problem. A repair should be performed if possible. The role
of an acromioplasty alone in the treatment of a massive
rotator cuff tear remains debatable, but it is clear that
excessive bone removal along with compromise of the
anterior deltoid leads to significantly inferior results and
is therefore contraindicated. If an acromioplasty is performed, an arthroscopic approach and conservative
bone removal, if any, are recommended.
The final option for treatment of a massive rotator
cuff tear is a muscle transfer. A transfer of the pectoralis
major is performed for a chronic irreparable subscapularis tear, and a latissimus transfer is used for posterior
and superior cuff insufficiency. Other transfers have
been described, but results are not as favorable and they
are rarely used today. Indications for a latissimus transfer for an irreparable tear involving supraspinatus, infraspinatus, and teres minor tears are primarily pain and
loss of function. Relative indications are a young, active

299

Orthopaedic Knowledge Update 8

Shoulder Reconstruction
patient and an intact subscapularis. Inferior results are
reported for patients with concurrent subscapularis
tears. The deltoid must also be intact and functioning
well. A muscle transfer is a major reconstructive procedure with a long period of rehabilitation, and should
only be considered as a salvage operation for patients
willing and able to undergo the operation and comply
with the rehabilitative program.

Role of the Biceps Tendon

The role of the long head of the biceps tendon continues to generate interest and controversy. Its exact function, if any, in the shoulder is a subject of debate with
both cadaveric and clinical studies that support several
differing viewpoints. Nevertheless, the long head of the
biceps tendon is an increasingly recognized source of
pain in the shoulder in conjunction with rotator cuff pathology. The synovial lining of the glenohumeral joint is
continuous with the tenosynovium of the proximal biceps so any inflammatory process involving the joint can
affect the biceps tendon as well.
The biceps tendon should be carefully inspected during any rotator cuff repair to rule out any concurrent pathology. This evaluation is generally performed as part of
the arthroscopic assessment.The portion of the biceps tendon from the intertubercular groove should be pulled into
the joint for inspection as well. Lesions of the tendon can
include inflammation of the tendon, tearing, or instability
relative to the intertubercular groove. Instability is always
associated with subscapularis tears or tears of the rotator
interval. In some circumstances, the tendon may have already ruptured. This condition is more common in conjunction with larger tears.
Minor fraying involving only a minimal amount of
the tendon proper can be dbrided; however, a tear involving 50% or more of the tendon should be treated
with tenotomy or tenodesis. Any instability of the tendon should also be treated in this fashion. The issue of
whether to use tenotomy or tenodesis of the long head
of the biceps tendon is also a subject of debate. In general, tenodesis is indicated in younger patients (younger
than 55 years) because of concerns about muscle cramping and cosmesis. Potential disadvantages are increased
length of the procedure, possible pain at the tenodesis
site, and the possibility of an additional incision. Older
patients have done well with simple tenotomy, with little
to no noticeable deformity, but it is important to discuss
the possibility of either procedure with patients preoperatively.
Recently, biceps tenotomy has been performed as
part of the surgical procedure for massive rotator cuff
tears. In one series containing the largest number of patients treated with tenotomy for rotator cuff tears, tenotomy was found to be particularly effective for massive
tears.

300

Glenohumeral Joint Arthritis

Etiology
Glenohumeral joint arthritis results from multiple different etiologies, including primary osteoarthritis, the
synovial-based arthroses such as rheumatoid arthritis,
osteonecrosis, posttraumatic arthritis, malunion or nonunion of the proximal humerus, rotator cuff arthropathy,
and arthritis associated with instability or surgery for instability. The etiology of shoulder arthritis is important
because each of these entities is associated with certain
characteristics that impact the technique and/or outcome of the surgery.
Primary osteoarthritis is associated with osteophyte
formation and posterior glenoid wear. Posterior wear
can make glenoid implantation very challenging and at
times can even preclude glenoid insertion. One solution
to this problem is bone grafting the glenoid with internally fixed corticocancellous bone. A recent report on
this procedure in 21 shoulders with an average correction of 33 demonstrated that this is a technically demanding procedure with failure in eight patients. Rheumatoid arthritis, on the other hand, can be associated
with osteophyte formation but more typically presents
with central wear of the glenoid, osteopenia, bone erosions, and subchondral cyst formation.
Arthritis secondary to previous instability surgery is
thought to arise because of an imbalance in the soft tissues surrounding the shoulder. This condition most often
occurs after older instability procedures that involved
transfer or imbrication of the subscapularis have been
done, leaving the shoulder extremely deficient in external rotation. These shoulders also have large osteophyte
formation and preferential posterior glenoid wear. The
soft-tissue contractures and scarring about the shoulder
make the approach and exposure for shoulder arthroplasty extremely difficult. Static posterior subluxation of
the humeral head has recently been recognized as a possible etiology of early glenohumeral arthritis in young
adults. One recent article discussing a small group of
men (average age, 40 years) with early arthritis found an
increase in mean glenoid retroversion and marked static
posterior subluxation. The condition was unresponsive to
surgical treatment, and could indicate a risk for the development of glenohumeral arthritis in some patients.
Arthroplasty after nonunion or malunion of the
proximal humerus is another very difficult problem to
manage. The outcome is better if the reconstruction
does not involve osteotomy of the tuberosities. Therefore, every effort should be made to accommodate the
abnormal bony anatomy with a modular prosthesis (Figure 3). Osteonecrosis is characterized by sclerosis early,
bone collapse at the later stages, and degenerative
changes on both the glenoid and humeral sides at the
last stage. Stiffness is often a prominent characteristic of
arthritis associated with osteonecrosis.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 26 Shoulder Reconstruction

Figure 3 A, A healed proximal humerus fracture has left the surgical neck with a varus malunion. The glenohumeral joint shows evidence of posttraumatic arthritis.
B, A prosthesis has been placed without osteotomy of the surgical neck or greater tuberosity. Modularity of the prosthesis facilitates an attempt to recreate normal anatomy.

Surgical Treatment
Indications and Contraindications
The main indication for shoulder arthroplasty is endstage arthritis associated with pain unresponsive to nonsurgical measures. Contraindications include active infection, absence of both deltoid and rotator cuff
musculature, neuropathic arthropathy such as a Charcot
joint, and intractable instability. Relative contraindications to glenoid implantation include young patients, excessive bone loss from the glenoid, and rotator cuff arthropathy.
Prosthetic Arthroplasty: Hemiarthroplasty Versus Total
Shoulder Arthroplasty
Prosthetic replacement for primary osteoarthritis has
historically been associated with extremely favorable results. Whether to use hemiarthroplasty or total shoulder
arthroplasty remains a topic of debate. Advantages cited
for performing a hemiarthroplasty include less lateralization of the joint line, less time spent in the operating
room, less blood loss, easier procedure, and the fact that
conversion to total shoulder arthroplasty can be performed at a later date. Advantages of glenoid implantation include better pain relief and longer survival of the
arthroplasty. Various authors have reported good results
using both techniques. Most of the recent literature
however, supports the use of glenoid components because the results are better for pain relief in long-term
follow-up. One prospective, randomized study showed
superior results in the total shoulder group, with a 12%
revision rate in the hemiarthroplasty group at an average follow-up of 35 months. In this study, the only fac-

American Academy of Orthopaedic Surgeons

tors that made hemiarthroplasty advantageous over total shoulder replacement were shorter operating room
time and less blood loss. Another study of the long-term
follow-up of both procedures cited the 15-year survival
rate as 93% for the patients who had total shoulder arthroplasty and 75% for the patients who underwent
hemiarthroplasty.
Results of conversion of a hemiarthroplasty to a total shoulder arthroplasty at a later date has not proved
to be as satisfactory as originally thought. There is often
severe bone loss on the glenoid side after a hemiarthroplasty, which makes glenoid implantation more challenging (Figure 4). Pain relief in this group of patients is
not as predictable; additional surgery is often required.
In one series of 18 conversions, there was marked pain
relief and increased range of motion; however, results
were unsatisfactory in 7 because of limited motion and
additional surgery. Overall, although some controversy
still exists, total shoulder replacement including implantation of a glenoid component is emerging as the standard for the treatment of primary osteoarthritis.
It is well established in the literature that pain relief
is much better with total shoulder arthroplasty for rheumatoid arthritis. A recent study of 105 shoulder arthroplasties in patients with rheumatoid arthritis showed no
statistical difference in Constant score between humeral
head replacement and total shoulder arthroplasty. The
group that underwent total shoulder arthroplasty had a
high rate of glenoid lucencies (58%), but none required
revision, demonstrating excellent longevity of the glenoid in this population. Of significance, in the group
that had humeral head replacement, there was superior

301

Shoulder Reconstruction

Orthopaedic Knowledge Update 8

Figure 4 A, A hemiarthroplasty was placed for osteoarthritis in this patient, without relief of pain. The head is slightly proud and asymmetric glenoid wear has occurred such that
the center of the glenoid is worn to the base of the coracoid. B, The hemiarthroplasty was converted to a total shoulder arthroplasty. A pegged glenoid component was inserted to
restore joint space and help lateralize the joint line to a more anatomic location.
migration of the humeral component by more than
5 mm in 28% and medial migration by more than 2 mm
in 16%. This migration did adversely affect the outcome
in these patients. Overall, the 8-year survival rate was
92% in this population. The issue of bone wear in rheumatoid arthritis remains a critical complication in terms
of potential reconstruction. Rheumatoid shoulders tend
to have central wear of the glenoid, which can eventually preclude implantation of a prosthetic component.
Therefore, at this point, the standard of care is to perform a total shoulder arthroplasty in a rheumatoid
shoulder unless severe bone wear already makes this
impossible.
When osteonecrosis is diagnosed at an early stage
before the development of degenerative changes on the
glenoid, the results of hemiarthroplasty have been consistently favorable. In patients with long-standing disease with radiographic involvement of the glenoid or in
patients in whom cartilage wear is discovered at the
time of surgery, total shoulder arthroplasty is recommended.
Glenoid implantation is generally not recommended
in patients with rotator cuff arthropathy. These shoulders are characterized by severe degenerative changes
to the glenohumeral joint, with complete loss of the
joint space in combination with a massive rotator cuff
tear with subsequent loss of rotator cuff function, resulting in proximal migration of the humeral head such that
it articulates with the undersurface of the acromion.
There also is often a slight anterior superior subluxation. Glenoid implantation during arthroplasty for ro-

302

tator cuff arthropathy has historically been discouraged

because of concerns over early loosening. Earlier recommendations were to oversize the humeral head in
these shoulders. The technique of humeral head replacment has evolved, however, and anatomic head sizing is
now advised. Overstuffing the joint has been associated
with painful arthroplasty because of soft-tissue stretch.
This overstuffing also lateralizes the deltoid insertion,
compromising its function. Preservation of the coracoacromial arch is critical during humeral head replacement
for rotator cuff arthropathy to prevent anterior-superior
instability. Bipolar components have not demonstrated
improved results in these cases and are rarely used. Reverse ball and socket prostheses are still under investigation and may hold promise for future use in treating
rotator cuff arthropathy.
Total Shoulder Arthroplasty
Total shoulder arthroplasty is a reliable operation for
pain relief and good function can be anticipated. Recently, a large study reported the results of arthroplasty
for osteoarthritis using a third generation, anatomically
designed shoulder prosthesis. Patients were evaluated
using the age-adjusted Constant score an average of
30 months postoperatively. The score improved from an
average of 37 to 97 postoperatively. Good or excellent
results were obtained in 77% of patients, and 94% were
satisfied or very satisfied. Forward elevation was 145.
Another recent study investigated the effect of rotator cuff disease on the outcome of total shoulder arthroplasty. Rotator cuff tears are uncommon in patients with

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

osteoarthritis, occurring in less than 5%. In this series of
514 patients, 41 had a partial-thickness tear and 42 had a
full-thickness tear. Supraspinatus tears did not adversely
affect outcome, satisfaction, or mobility. Treatment of
the tears did not change outcome parameters. However,
shoulders with moderate or severe fatty degeneration of
the infraspinatus were associated with poorer results.
This study suggests that small tears limited to the supraspinatus do not negatively affect the outcome; however, more global degenerative changes in the rotator
cuff musculature will.
Another prospective study further substantiates the
fact that small tears of the supraspinatus do not adversely affect the outcome after total shoulder arthroplasty. This study was performed to determine the effect
of tears, preoperative glenoid bone erosion, and radiographic evidence of subluxation of the humeral head on
the outcome of shoulder arthroplasty. Tears of the supraspinatus tendon did not adversely affect the outcome. Radiographic posterior subluxation of the humeral head was associated with a lower outcome score,
and patients with moderate to severe glenoid erosion
did much better after total shoulder arthroplasty than
humeral head replacement. Based on these data, the use
of a glenoid component is recommended in shoulders
with preoperative glenoid erosion.
Glenohumeral arthritis can occur after surgery for
shoulder instability, and in general occurs more often after nonanatomic repairs where there is excessive shortening of the anterior capsule and subscapularis resulting
in severe limitation of external rotation. The reported
results for arthroplasty for arthritis of this etiology are
not as good as those for primary arthritis. One reason
may be that the patients are in general younger, and
have higher physical demands and expectations. Two recent studies showed definite improvements in range of
motion and pain relief. However, both studies had relatively high revision rates and some unsatisfactory results. The survivorship of the components at 10 years
was 61% and is significant because the average age of
patients undergoing arthroplasty was 46 to 47 years in
both studies.
Glenoid component loosening has been recognized
as one of the most common reasons for revision shoulder arthroplasty surgery. Although radiolucency around
the glenoid component occurs in a much higher number
of patients than the number that ultimately require revision for glenoid loosening, the glenoid component remains the most likely to fail of the shoulder arthroplasty
components. It has been shown in a series of 48 shoulders undergoing glenoid component revision surgery
that satisfactory pain relief was achieved in a much
higher percentage of patients that had a new glenoid
component inserted than in those who had glenoid component removal without reimplantation. Patients who
did not have reinsertion of a component underwent

American Academy of Orthopaedic Surgeons

Chapter 26 Shoulder Reconstruction

bone grafting of the defects; these patients may be candidates for glenoid component placement after graft
consolidation.
Nonprosthetic Treatment Options for Glenohumeral
Arthritis The role of arthroscopy in the treatment of
glenohumeral joint arthritis remains unknown and recent studies of results in the knee joint have made this a
controversial topic. Nevertheless, there are several recent reports of small studies of arthroscopic treatment
of arthritis in the shoulder suggesting favorable shortterm results. Indications for arthroscopic surgery for arthritis are ill defined, but generally include younger patients in whom it is desirable to prolong the need for
prosthetic arthroplasty, and those with concentric wear
of the glenoid, absence of severe joint contracture, loose
bodies, and minimal if any bone loss. Arthroscopic treatment is most useful in alleviating mechanical symptoms
from loose cartilage fragments or interposed soft tissue.
Contraindications to arthroscopy for the treatment of
arthritis include older patients with global arthritis who
are candidates for shoulder arthroplasty, marked posterior or otherwise nonconcentric wear of the glenoid, severe joint contracture, and bone loss on the humeral
side. Other arthroscopic procedures such as microfracture arthroplasty and glenoidplasty (burring nonconcentric wear) have been suggested, but their usefulness has
not yet been substantiated in the peer-reviewed literature.
Soft-Tissue Resurfacing With or Without Hemiarthroplasty Soft-tissue interposition arthroplasty has
received significant attention in recent years, given the
complications associated with early glenoid wear and
the frequency of radiolucencies associated with glenoid
implantation. The two most common materials used are
fascia lata, both allograft and autograft, and allograft
knee meniscus. One early report of fascia lata interposition is promising, with failure in only one patient in
whom the graft was found to have lost fixation at the
time of revision surgery. However, the numbers are
small and follow-up has been short term.
More recently, several centers have successfully used
the lateral knee meniscus as an interposition tissue. The
rationale for the use of the lateral meniscus is that it is
tissue designed to withstand significant load with weight
bearing under normal physiologic conditions. The lateral
meniscus is used because it is more discoid in shape and
gives better glenoid coverage. It is secured to the glenoid surface in place of a prosthetic component using
bone anchors in the glenoid in a circumferential fashion.
Early results show maintenance of the joint space radiographically and excellent functional results. Long-term
survival analysis of this construct is still not available.
The effect of severe posterior glenoid wear is also not
clearly delineated at this point, but it may be a relative

303

Orthopaedic Knowledge Update 8

Shoulder Reconstruction
contraindication to the use of soft-tissue interposition
because of the risk of continued bone loss that may
make future reconstructive options difficult or impossible.
The primary indication for soft-tissue interposition
arthroplasty is young patients with severe arthritis who
are candidates for arthroplasty, but have a high likelihood of needing revision surgery in the future because
of their young age. In general, this includes patients in
their 40s and active, higher-demand patients in their
early 50s. There is always a small risk of disease transmission with use of allograft material, and recipients
should be made aware of this risk. Although the longterm results of this procedure are still not available and
the indications are still developing, soft-tissue interposition in the shoulder holds significant promise for the
treatment of arthritis in the young patient.

Acromioclavicular Joint Arthritis

Symptomatic acromioclavicular joint arthritis is a clinical rather than a radiographic diagnosis. Examination
findings include tenderness with palpation at the acromioclavicular joint and pain with cross-body adduction.
Radiographic changes are extremely common and do
not correlate with symptomatic pathology that requires
treatment. Nonsurgical treatment of isolated acromioclavicular joint pain includes nonsteroidal antiinflammatory drugs, rest from inciting activity, and corticosteroid injections. If the condition is associated with
rotator cuff pathology, physical therapy may be of benefit. Nonsurgical treatment of at least 3 to 6 months is
generally recommended before surgical treatment
should be considered.
A persistently painful acromioclavicular joint that is
not responsive to nonsurgical treatment is surgically
treated with a distal clavicle resection that can be performed through an open or an arthroscopic approach. If
the procedure is performed through an open incision,
care should be taken to repair the acromioclavicular ligaments. The superior and posterior acromioclavicular
ligaments are the most important for maintaining stability. Disadvantages of the open approach are that the ligaments must be violated and subperiosteally removed
from the distal clavicle. A portion of the deltoid is also
removed from the anterior acromion and clavicle. This
removal is of little consequence if appropriately repaired, but failure to do so results in significantly compromised outcome.
An arthroscopic distal clavicle resection is performed with the arthroscope in the posterior or posterolateral portal into the subacromial space. The working
instruments are inserted through the rotator interval
portal redirected into the subacromial space or through
a separate portal to access the acromioclavicular joint.
This approach allows excision of the clavicle without

304

disturbing the ligaments and deltoid. Care must still be

taken not to violate the ligaments with the arthroscopic
instruments. The surgeon must adequately assess for
complete bone removal. The most commonly missed
bone is located superiorly and posteriorly, and symptoms may continue if resection is not thorough.

Annotated Bibliography
Rotator Cuff Tear
Cofield RH, Parvizi J, Hoffmeyer PJ, Lanzer WL, Ilstrup
DM, Rowland CM: Surgical repair of chronic rotator
cuff tears. J Bone Joint Surg Am 2001;83:71-77.
This is a prospective long-term study examining the results
of open surgical repair and acromioplasty of chronic rotator
cuff tears. Satisfactory pain relief was obtained in 96 of 105
shoulders. Tear size was the most important determinant of
outcome with regard to active motion, strength, rating of results, patient satisfaction, and need for revision.

Fenlin JM Jr, Chase JM, Rushton SA, Frieman BG: Tuberoplasty, creation of an acromiohumeral articulation:
A treatment option for massive, irreparable rotator cuff
repairs. J Shoulder Elbow Surg 2002;11:136-142.
The authors discuss results of tuberoplasty in 20 patients.
Overall results (improved pain relief and return to daily activities) were good.

Galatz LM, Griggs S, Cameron BD, Iannotti JP: Prospective longitudinal analysis of postoperative shoulder
function: A ten-year follow-up study of full-thickness rotator cuff tears. J Bone Joint Surg Am 2001;83:10521056.
The authors reported that early results of rotator cuff repairs do not deteriorate with time in this prospective longitudinal study.

Halder AM, ODriscoll SW, Heers G, et al: Biomechanical comparison of effects of supraspinatus tendon detachments, tendon defects, and muscle retractions.
J Bone Joint Surg Am 2002;84:780-785.
The effects of supraspinatus tendon detachments, tendon
defects, and muscle retraction on in vitro force transmission by
the rotator cuff to the humerus were compared.

Murray TF, Lajtai G, Mileski RM, Snyder SJ: Arthroscopic repair of medium to large full-thickness rotator
cuff tears: Outcome at 2- to 6-year follow-up. J Shoulder
Elbow Surg 2002;11:19-24.
Forty-eight arthroscopic repairs of medium to large rotator cuff tears were evaluated 2 to 6 years after surgery. There
were 35 excellent, 11 good, 2 fair, and no poor results. Only
one patient had clinical evidence of failed repair.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Parsons IM, Apreleva M, Fu FH, Woo SL: The effect of
rotator cuff tears on reaction forces at the glenohumeral
joint. J Orthop Res 2002;20:439-446.
Results from this study of nine cadaveric upper extremities indicate that the integrity of the rotator cuff has a significant effect on joint reaction forces.

Pearsall AW IV, Bonsell S, Heitman RJ, Helms CA, Osbahr D, Speer K: Radiographic findings associated with
symptomatic rotator cuff tears. J Shoulder Elbow Surg
2003;12:122-127.
Radiographs of 40 patients with a documented rotator cuff
tear were compared with those of asymptomatic age-matched
control patients. Results indicate that radiographs of patients
with rotator cuff tear have greater tuberosity radiographic abnormalities that are not seen in the asymptomatic patients.

Postacchini F, Gumina S: Results of surgery after failed

attempt at repair of irreparable rotator cuff tear. Clin
Orthop 2002;397:332-341.
This is a study of a small number of patients who had an
open attempt at rotator cuff repair including deltoid detachment and acromioplasty. At the time of surgery, the tears were
found to be irreparable. Shoulder function deteriorated in 11
patients. The authors recommended against attempted open
repair of irreparable cuff tears because the functional results
are generally poor.

Warner JJ, Higgins L, Parsons IM IV, Dowdy P: Diagnosis and treatment of anterosuperior rotator cuff tears.
J Shoulder Elbow Surg 2001;10:37-46.
According to results of this study, repair within 6 months
of subscapularis tear may produce a better functional outcome.

Yamaguchi K, Tetro AM, Blam O, Evanoff BA, Teefey

SA, Middleton WD: Natural history of asymptomatic rotator cuff tears: A longitudinal analysis of asymptomatic
tears detected sonographically. J Shoulder Elbow Surg
2001;10:199-203.
The natural history of asymptomatic rotator cuff tears was
studied over a 5-year period to determine risk of tear progression and development of symptoms.

Glenohumeral Joint Arthritis

Antuna SA, Sperling JW, Cofield RH, Rowland CM:
Glenoid revision surgery after total shoulder arthroplasty. J Shoulder Elbow Surg 2001;10:217-224.
Glenoid component revision surgery was performed on
48 shoulders. The indications for surgery were glenoid component loosening in 29 shoulders, implant failure in 14, and component malposition or wear leading to instability in 5 shoulders. Satisfactory pain relief was acute in 86% of patients who
had another glenoid component inserted and in 66% of patients who underwent glenoid component removal. Patients
who have bone grafting may be candidates for glenoid component placement after graft consolidation.

American Academy of Orthopaedic Surgeons

Chapter 26 Shoulder Reconstruction

Edwards TB, Boulahia A, Kempf J-F, Boileau P, Nemoz
C, Walch G: The influence of rotator cuff disease on the
results of shoulder arthroplasty for primary osteoarthritis: Results of a multicenter study. J Bone Joint Surg Am
2002;84:2240-2248.
Rotator cuff tears are uncommon in primary glenohumeral arthritis. This study examined the effect of fullthickness tears, partial-thickness tears, and fatty degeneration
of the rotator cuff on the outcome. Supraspinatus tears were
not found to influence the postoperative outcome. Additionally, treatment of these tears did not influence outcome parameters. However, shoulders with severe and moderate fatty
degeneration of the infraspinatus had poorer results.

Godeneche A, Boileau P, Favard L, et al: Prosthetic replacement in the treatment of osteoarthritis of the
shoulder: Early results of 268 cases. J Shoulder Elbow
Surg 2002;11:11-18.
This is a study reporting the results of 268 shoulder arthroplasties for primary osteoarthritis. Good to excellent results
were observed in 77% of patients and there was a 94% satisfaction rate. Mean active forward elevation was 145 postoperatively. In this study, glenoid radiolucent lines were present in
58% of cases and were associated with a less satisfactory result. Patients who underwent biceps tenodesis had better pain
relief. Complications occurred in 8.6% of cases.

Hill JM, Norris TR: Long-term results of total shoulder

arthroplasty following bone grafting of the glenoid.
J Bone Joint Surg Am 2001;83:877-883.
Bone grafting of the glenoid can restore bone stock in patients with structural defects.

Iannotti JP, Norris TR: Influence of preoperative factors

on outcome of shoulder arthroplasty for glenohumeral
osteoarthritis. J Bone Joint Surg Am 2003;85:251-258.
The purpose of this study was to evaluate the influence of
full-thickness rotator cuff tears, preoperative erosion of glenoid bone, radiographic evidence of subluxation of the humeral head, and preoperative range of motion on the outcome
shoulder arthroplasty. Repairable full-thickness tears of the
supraspinatus tendon did not affect outcome. Shoulders with
severe or moderate glenoid erosion had better results with total shoulder arthroplasty than hemiarthroplasty. Based on
these results, it is recommended that a glenoid component be
inserted in shoulders with glenoid erosion. A repairable tear
of the supraspinatus tendon is not a contraindication to the
use of a glenoid component.

Sperling JW, Antuna SA, Sanchez-Sotelo J, Schleck C,

Cofield RH: Shoulder arthroplasty for arthritis after instability surgery. J Bone Joint Surg Am 2002;84:17751781.
Arthroplasty was associated with significant pain relief.
There was not a significant difference between the hemiarthroplasty and total shoulder arthroplasty groups with regard
to external rotation, abduction, or pain. The survival of the

305

Shoulder Reconstruction
components at 10 years was 61%. The data from these studies
suggest that arthroplasty in this particular group of patients is
associated with good pain relief; however, there were high
rates of revision surgery and unsatisfactory results because of
component failure instability and glenoid arthritis in the hemiarthroplasty group.

Trail IA, Nuttall D: The results of shoulder arthroplasty

in patients with rheumatoid arthritis. J Bone Joint Surg
Br 2002;84:1121-1125.
A clinical and radiologic analysis of 105 shoulder arthroplasties in patients with rheumatoid arthritis was performed.
Constant scores and American Shoulder and Elbow Surgeons
scores were improved, and both scores were statistically significant.

Classic Bibliography
Bigliani LU, Cordasco FA, McIlveen SJ, Musso ES: Operative repair of massive rotator cuff tears: Long term
results. J Shoulder Elbow Surg 1992;1:120-130.
Brenner BC, Ferlic DC, Clayton ML, Dennis DA: Survivorship of unconstrained total shoulder arthroplasty.
J Bone Joint Surg Am 1989;71:1289-1296.
Burkhead WZ Jr, Hutton KS: Biologic resurfacing of
the glenoid with hemiarthroplasty of the shoulder.
J Shoulder Elbow Surg 1995;4:263-270.
Gartsman G, Khan M, Hammerman S: Arthroscopic repair of full-thickness tears of the rotator cuff. J Bone
Joint Surg Am 1998;80:832-840.
Gartsman GM, Roddey TS, Hammerman SM: Shoulder
arthroplasty with or without resurfacing of the glenoid
in patients who have osteoarthritis. J Bone Joint Surg
2000;82:26-34.

306

Orthopaedic Knowledge Update 8

Gerber C, Fuchs B, Hodler J: The results of repair of
massive tears of the rotator cuff. J Bone Joint Surg Am
2000;82:505-515.
Goutallier D, Postel JM, Bernageau J, Lavau L, Voisin
MC: Fatty muscle degeneration in cuff ruptures: Preand postoperative evaluation by CT scan. Clin Orthop
1994;304:78-83.
Hawkins RJ, Angelo RL: Glenohumeral osteoarthrosis:
A late complication of the Putti-Platt repair. J Bone
Joint Surg Am 1990;72:1193-1197.
Iannotti JP, Bernot MP, Kuhlman JR, Kelley MJ, Williams GR: Postoperative assessment of shoulder function: A prospective study of full-thickness rotator cuff
tears. J Shoulder Elbow Surg 1996;5:449-457.
Neer CS II: Replacement arthroplasty for glenohumeral
osteoarthritis. J Bone Joint Surg Am 1974;56:1-13.
Neer CS II, Watson KC, Stanton FJ: Recent experience
in total shoulder replacement. J Bone Joint Surg Am
1982;64:319-337.
Rodosky M, Bigliani L: Indications for glenoid resurfacing in shoulder arthroplasty. J Shoulder Elbow Surg
1996;5:231-248.
Rokito AS, Cuomo F, Gallagher MA, Zuckerman JD:
Long-term functional outcome of repair of large and
massive chronic tears of the rotator cuff. J Bone Joint
Surg Am 1991;81:991-997.
Walch G, Boileau P: Prosthetic adaptability: A new concept for shoulder arthroplasty. J Shoulder Elbow Surg
1999;8:443-451.

American Academy of Orthopaedic Surgeons

Chapter

27

Elbow and Forearm: Adult Trauma

David Ring, MD
Jesse B. Jupiter, MD

Distal Humeral Fractures

Fractures Involving the Metaphyseal Columns of the
Distal Humerus
Fractures of the distal humerus are usually bicolumnar,
and most will involve the articular surface. Although often depicted as Y- or T-shaped fractures with a simple
articular split, it is becoming evident that these relatively simple fracture patterns are not as common as the
more complex metaphyseal and articular fractures. Most
fractures involve one of the following factors that increase the challenge of surgical treatment: (1) low (distal) fracture of one or both columns (at the level of the
base of the olecranon fossa); (2) metaphyseal fragmentation of one or both columns; (3) complex fragmentation of the articular surface, including entirely or nearly
entirely articular fragments (especially those with a
coronal plane fracture line).
The surgical treatment of fractures of the distal humerus remains unsatisfactory in approximately 25% of
patients in most reported series. Some series have reported a greater complication rate in older patients,
probably because of poor bone quality, but perhaps also
resulting from more difficult injury patterns. On the
other hand, two recent series evaluated the results of
treatment using the Disabilities of the Arm, Shoulder,
and Hand outcomes instrument and found reasonably
good restoration of function from the patients perspective.
Olecranon osteotomy for exposure of fractures of
the distal humerus has waned in popularity primarily
because of problems related to the wires or screws used
to repair the osteotomy. Alternative exposures commonly in use include the recently introduced tricepsreflecting anconeus pedicle modification of the triceps
elevating exposure of Bryan and Morrey; the traditional
Campbell exposure (midline triceps split), with additional elevation of the triceps off of the proximal ulna;
and the reintroduced Allonso-Llamas exposure (in
which the triceps is elevated off of the back of the humerus, but its insertion onto the olecranon is preserved).
However, a recent cadaver study showed that olecranon

American Academy of Orthopaedic Surgeons

osteotomy provides the best exposure of the distal humerus. The authors of this study also report that with attention to detail in the creation and repair of the olecranon osteotomy the rate of complications and revisions
specifically to address prominent wires is acceptable.
The use of two plates in orthogonal planes has become well established in the treatment of fractures of
the distal humerus. However, there is biomechanical
support for the use of parallel plates placed on the direct medial and lateral aspects of the columns, and this
configuration is often better suited to the treatment of
complex fracture patterns. Newer plate designs include
precontoured shapes and smaller 2.7-mm screws placed
distally to facilitate internal fixation.
Total elbow arthroplasty is now an accepted treatment option for older, less active patients with fractures
of the distal humerus. Several studies, including a recent
comparative study of arthroplasty and plate and screw
fixation, document very good early results; however,
long-term follow-up is needed because elbow arthroplasties have a limited life span and eventually wear
out. In addition, patients who undergo total elbow arthroplasty require a 5-kg lifting restriction, and patients
who undergo this procedure are more prone to infection
and other complications than those who undergo total
hip or knee arthroplasty. The best candidate for total elbow arthroplasty is an older, infirm, and inactive patient
with a fracture of the distal humerus, or a patient with a
fracture of the distal humerus with an elbow joint that is
already compromised by rheumatoid arthritis.

Fractures Involving Primarily the Articular Surface

The discussion of fractures involving primarily the articular surface of the distal humerus is often limited to
fractures of the capitellum. It is now known that isolated fractures of the capitellum are uncommon. An apparent fracture of the capitellum nearly always extends
into the lateral trochlear lip (the so-called coronal shear
fracture) and may also involve a fracture of the lateral
epicondyle, impaction of the posterior aspect of the lateral column, fracture of the posterior aspect of the tro-

307

Elbow and Forearm: Adult Trauma

Orthopaedic Knowledge Update 8

Figure 1 Illustration of the lateral lip of the trochlea (1), lateral epicondyle (2), posterior aspect of the lateral column (3), posterior trochlea (4), and medial epicondyle (5).
Apparent fractures of the capitellum usually involve region 1, and injuries with greater
complexity can involve fractures in regions 2 through 5. ( Copyright D. Ring, MD &
J.B. Jupiter, MD.)

Figure 2 Three-dimensional reconstruction of a CT image; this is useful for understanding complex articular fractures of the distal humerus and planning surgical treatment. ( Copyright D. Ring, MD & J.B. Jupiter, MD.)

Figure 3 When the articular fragments do not fit, as shown in this photograph, there
is impaction of the posterior aspect of the distal humerus, which must be elevated to
allow accurate reduction of the articular fragments. ( Copyright D. Ring, MD & J.B.
Jupiter, MD.)
chlea, and fracture of the medial epicondyle (Figure 1).
Other variations such as fractures involving primarily
the trochlear side of the joint are occasionally encountered.
When a semicircular fracture fragment is identified
anterior to the distal humerus on the lateral radiograph
of a patient with an injured elbow, surgeons should look
for a second arc or semicircle on the fragment, which indicates that the fracture involves the trochlea. A trochlear defect is often apparent on the AP view. Because
these features and the more complex fractures of the
distal humeral articular surface can be difficult to detect
on standard radiographs, CT is especially helpful in diagnosing these injuries. In particular, three-dimensional

308

CT reconstructions with the ulna and the radius subtracted from the image are invaluable for understanding
the injury and planning surgical treatment (Figure 2).
Traction radiographs and fluoroscopy can help define
the fracture pattern, but they are not as useful for preoperative planning as three-dimensional CT reconstructions because they are usually obtained at the time of
surgery after the administration of anesthesia.
Surgical fixation can usually be achieved through a
lateral muscle interval with elevation of the extensor
carpi radialis brevis and part of the extensor carpi radialis longus off of the supracondylar ridge and anterior
humerus. If the lateral collateral ligament and epicondyle are intact, it is often possible to work through
an interval anterior to the lateral collateral ligament
(through the common extensor muscles) and rely on reduction of the metaphyseal fracture lines. For patients
with more complex fractures, there is nearly always a
fracture of the lateral epicondyle, which can be mobilized along with the origins of the lateral collateral ligament and common extensor muscles. The elbow joint
can then be subluxated, allowing a good view of the anterior articular surface of the distal humerus. If the posterior aspect of the trochlea or the medial epicondyle is
fractured, exposure through an olecranon osteotomy
may be preferable.
If the fracture fragments do not seem to fit back
onto the intact portions of the distal humerus, additional
impaction of the intact distal humerus should be suspected (Figure 3). In this situation, it will be necessary
to hinge open the posterior aspect of the lateral column
and the posterior aspect of the trochlea to properly re-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

align the anterior fracture fragments. The articular fragments are secured with countersunk screws (screws with
threads on the head, such as Herbert screws, can be
used), small threaded Kirschner wires, or screws entering from the nonarticular parts of the distal humerus,
such as the posterior aspect of the lateral column. The
lateral epicondyle is secured with a tension band wire to
incorporate the common extensor fascia; a screw or a
plate is also used if the fragment is large enough.
Ideally, active-assisted exercises and use of the arm
for light daily activities are started the day after surgical
fixation of a fracture of the distal humerus. However, if
the fracture is complex or the bone quality is poor and
the fixation is therefore not optimal, it may be preferable to immobilize the elbow for 4 weeks before initiating exercises. A healed stiff elbow will be easier to salvage than a nonhealed elbow, particularly in patients
with complex fractures.

Radial Head Fractures

Isolated radial head fractures usually involve part of the
radial head and heal well with nonsurgical treatment.
Sometimes they are displaced enough to block forearm
rotation and benefit from surgical treatment. The most
important complication of isolated partial radial head
fractures is elbow stiffness, and the most important aspect of treatment is mobilization of the elbow.
As documented in two recent case series, nonsurgically treated minimally displaced fractures of the radial
neck will occasionally fail to heal. The prevalence of
nonunion is obscured for several reasons: (1) nonunion
of the radial head is usually asymptomatic; (2) radiographs are rarely obtained until union in the treatment
of patients with this type of fracture; and (3) a fracture
of the radial head that is unhealed 1 year after the injury may still eventually heal 2 years or longer after the
injury.
Fractures of the radial headparticularly complex
fractures involving the entire head and neckare often
associated with one of the following five injury types:
(1) an Essex-Lopresti fracture-dislocation of the forearm or a variant thereof; (2) rupture of the medial collateral ligament complex; (3) dislocation of the elbow;
(4) dislocation of the elbow with fractures of the radial
head and coronoid processthe so-called terrible triad
of the elbow; and (5) an olecranon fracture-dislocationusually with an apex posterior fracture of the
proximal ulna or olecranon (the so-called posterior
Monteggia lesion). In each type, the associated fractures
and ligament injuries may compromise forearm or elbow instability, thereby increasing the importance of the
radial head.
Although the classification system of Mason is still
used, it is often modified. The Mason classification distinguished nondisplaced fractures (type 1), partial head

American Academy of Orthopaedic Surgeons

Chapter 27 Elbow and Forearm: Adult Trauma

fractures (type 2), and complete head fractures (type 3).
There are few data in the literature to support any subclassification based on the size or displacement of partial head fractures. Several important factors are not accounted for in this classification system, and they are
only partially accounted for in the more recent comprehensive classification of fractures, including: (1) the
number of fracture fragments; (2) lost or unrecoverable
articular fragments; (3) central impaction and deformation of the radial head; (4) the quality of the fracture
fragments (amount of subchondral bone, osteoporosis);
and (5) metaphyseal comminution. These factors all will
have a substantial impact on attempts at surgical repair
of a radial head fracture. In particular, a recent study
suggested that patients with fractures of the entire radial head that create more than three articular fragments have high rates of early failure, nonunion, and
poor forearm rotation after surgical fixation.
Excision of a complex fracture of the radial head
without prosthetic replacement is still a viable treatment
option. To do this safely, surgeons must make sure that
there is no forearm ligament injury by using the recently
described radius pull test and ensure that there is no
fracture of the coronoid (terrible triad). Although it is
usually preferable to treat an elbow fracture-dislocation
with either repair or replacement of the radial head,
reasonable results have been reported after radial head
resection as long as the coronoid is not also fractured.
Excision of the radial head without prosthetic replacement is considered primarily in older patients with limited demands and simple injury patterns.
Partial excision of the radial head has traditionally
been associated with less optimal results. Even a small
portion of the radial head can be important to stability
of the forearm or elbow. In the setting of an unstable
fracture-dislocation of the forearm or elbow, if a partial
radial head fracture has inadequate fragments for repair, it may be preferable to proceed with prosthetic replacement.
Surgical repair should be used only for patients with
simple fractures with three or fewer fragments with
good quality bone and little or no impaction. The implants must be placed directly laterally, with the arm in
neutral rotation to avoid impingement on the ulna with
forearm rotation. Another rough guide for placement is
provided by the arc defined by the radial styloid and
Listers tubercle. Implants placed on the articular surface of the radial head must be countersunk beneath the
articular surface.
Several metal radial head prostheses are in common
use. Two studies support the use of a loose-stemmed
metal prosthesis that acts as a stiff spacer, and another
study supports the use of a bipolar prosthesis that is cemented into the neck of the radius. Surgeons should be
careful not to place a prosthesis that is too long or wide
and always err toward using a slightly smaller prosthesis

309

Elbow and Forearm: Adult Trauma

Orthopaedic Knowledge Update 8

Figure 5 Three-dimensional reconstruction of a CT image showing that even very

small coronoid fractures can be associated with recurrent elbow instability, particularly
when they involve the anteromedial facet of the coronoid. ( Copyright D. Ring, MD &
J.B. Jupiter, MD.)

Figure 4 Radiograph showing wear of the capitellum that can result when a metal
radial head prosthesis is too long; the medial side of the ulnohumeral joint will be
hinged open in such instances. Therefore, it is important not to place too large an
implant. ( Copyright D. Ring, MD & J.B. Jupiter, MD.)

because the major problem is overstuffing of the joint,

which can lead to problems with wear of the capitellar
cartilage and potential instability or malalignment of the
elbow (Figure 4). Some of the older prosthetic designs
had stems that were too big for the average radial neck,
which could result in inadequate placement with the radial head not well seated into the shaft. Biomechanical
studies have confirmed the ability of metal radial head
prostheses to restore near-normal stability to the elbow
after radial head excision.

The lateral collateral ligament characteristically fails

by avulsion of its lateral epicondylar origin, with midsubstance tears and avulsion from the ulna occurring
less commonly. This facilitates repair with either suture
anchors or drill holes through the bone of the lateral
epicondyle. Instability after dislocation of the elbow is
usually associated with fracture of the radial head and
coronoid process. Elbow dislocations without major
associated fractures are also occasionally unstable.
Whether they occur in older or younger patients after a
high-energy injury, there are usually extensive avulsions
of the common extensor and flexor muscles from the
distal humerus. Older patients may have primarily
lateral-sided injuries, with a tendency for the elbow to
rotate on the relatively preserved medial soft tissues.
Some patients with unstable elbow dislocations have a
small fracture of the anteromedial facet of the coronoid
processa variant of posteromedial varus rotational instability pattern injuries (Figure 5). Unstable elbow dislocations are treated with either extensive soft-tissue reattachment to the distal humerus, cross pinning of the
elbow joint, or hinged external fixation.

Elbow Dislocations
It is now well recognized that elbow dislocations are associated with complete or near-complete disruption of the
capsuloligamentous stabilizers of the elbow and that the
progression of injury is typically from lateral to medial. It
is possible to dislocate the elbow with the anterior band
of the medial collateral ligament still intact. The focus of
treatment has shifted from the medial to the lateral collateral ligament complex. The medial collateral ligament
is repaired only if treatment of associated fractures and
the lateral collateral ligament fails to restore stability.

310

Coronoid Fractures
Although a recently published biomechanical cadaver
study suggested a limited effect of small coronoid fractures on elbow instability, clinical data document that
even very small fractures can lead to troublesome elbow
instability. For instance, although the coronoid fractures
associated with terrible triad pattern fracturedislocations of the elbow are nearly always transverse
fractures that are less than 30% of the coronoid height,
terrible triad elbows can dislocate in spite of cast immo-

American Academy of Orthopaedic Surgeons

Chapter 27 Elbow and Forearm: Adult Trauma

Orthopaedic Knowledge Update 8

bilization, are prone to dislocation after surgical treatment (particularly if radiocapitellar contact is not restored), and can have a high prevalence of unsatisfactory
treatment results. Because of the poor results associated
with the treatment of this type of fracture, many surgeons
are routinely repairing the coronoid fracture, usually
with suture placed through drill holes in the ulna.
Fractures of the anteromedial facet of the coronoid
process are usually associated with injury to the lateral
collateral ligament complex and can be difficult to manage. These have been labeled posteromedial varus rotational instability pattern injuries. This type of fracture is
repaired with a plate and screws through a medial exposure between the two heads of the flexor carpi ulnaris
(where the ulnar nerve usually sits), more anteriorly
through a split in the flexor pronator mass, or both.
Large coronoid fractures are nearly always associated with an olecranon fracture-dislocation. These fractures typically involve multiple fragments, particularly
when part of a posterior olecranon fracture-dislocation.
They can be accessed and manipulated through the olecranon fracture itself and fixed with a dorsal plate and
screws with or without an additional medial plate. The
results of treatment of anterior transolecranon fracturedislocations have been quite good, whereas the treatment of posterior olecranon fracture-dislocations has
been associated with a greater number of complications
and revisions and diminished results. Patients with large
fractures with complex comminution or poor bone
quality may occasionally benefit from protection with
hinged external fixation.

simple olecranon fractures, and good results have been

reported; however, 20% of patients had a second surgery for plate removal. Furthermore, it may not be wise
to rely on screws to secure a small osteopenic proximal
fragment. The long-term results of successful surgical repair of olecranon fractures are good.

Diaphyseal Forearm Fractures

Mechanics
Several recent studies have examined the impact of rotational malalignment of the radius and ulna on forearm rotation. These studies consistently show that rotational
malalignment of the ulna has less impact on forearm rotation than radius malrotation. In both the radius and the
ulna, forearm motion was not impacted unless there was
substantial malalignment (at least 30 malrotation)an
amount of malalignment that is unlikely to occur in the
management of patients. Prior studies have shown a more
substantial impact with even minor angular malalignments of the radius and ulna, particularly loss of the normal radial bow.
A biomechanical study of dynamic compression plates
applied to transverse diaphyseal osteotomies of the radius
showed that even relatively small plates could achieve a
bending stiffness comparable to that of the intact bone
and that longer plates with bicortical screws at the ends of
the plates had better torsional stiffness. This is consistent
with the findings of previous studies and emphasizes the
advantages of longer plates and bicortical screws in a setting with substantial torsional stresses such as the forearm.
These advantages may be even more pronounced in patients with comminuted fractures.

Olecranon Fractures
The Mayo classification of olecranon fractures characterizes fractures based on the three most important factors in treatment considerations: comminution, displacement, and fracture-dislocation.
There is wide variation regarding the treatment of
olecranon fractures. A biomechanical study found that a
large screw was best for internal fixation of olecranon
fractures, but reported that the stability was insufficient
to allow active elbow exercises. Conversely, internal fixation of olecranon osteotomies and fractures with small
caliber wires with immediate active mobilization has
been shown to be very successful in practice, at least
when specific techniques are used.
The major disadvantage of tension band wiring continues to be a high rate of subsequent surgical procedures for removal of symptomatic prominent hardware,
which has been reported in one long-term study as ranging between 43% and 81% (depending on technique);
however, in another study, it was reported to be only
13% when specific techniques intended to limit the
prominence of the wires were used. Plates are becoming
a more popular treatment option for fixation of even

American Academy of Orthopaedic Surgeons

Ulnar Fractures
Two recently published meta-analyses of the treatment
of isolated fractures of the ulnar diaphysis (the so-called
nightstick fracture) identified only retrospective case series and therefore could not make definitive recommendations regarding treatment. Fractures with greater than
50% of displacement or 10 of angulation can impact
forearm rotation. Because surgical treatment of these
fractures is straightforward, it is recommended that surgeons have a low threshold for surgical treatment for
displaced fractures. For less displaced fractures, nonsurgical treatmenteven simple symptomatic treatment
will usually result in union with good function, but the
time to complete healing can be quite prolonged. Therefore, surgical treatment is also a reasonable option for
motivated patients with less displaced fractures.
Diaphyseal ulnar fractures with anterior or lateral
dislocation of the proximal radioulnar joint (anterior or
lateral Monteggia lesions) are uncommon in adults.
Plate and screw fixation of the ulna in anatomic alignment usually restores good function. Open reduction of
the proximal radioulnar and radiocapitellar joints is

311

Elbow and Forearm: Adult Trauma

Orthopaedic Knowledge Update 8

rarely necessary. Residual malalignment of these joints

usually reflects residual malalignment of the ulna.
Posterior Monteggia fractures occur more commonly
in adult patients, particularly when posterior olecranon
fracture-dislocations are included. Few of these injuries
occur at the diaphyseal level. Most occur at the level of
the metaphysis (just distal to the coronoid process) or
through the olecranon (olecranon fracture-dislocation).
These injuries are often associated with fractures of the
radial head and coronoid and injury to the lateral collateral ligament complex. Posterior Monteggia fractures are
more challenging to treat, and the results are not as predictable because (1) it can be difficult to obtain solid fixation of the proximal ulna fragment, particularly in patients with osteoporosis; (2) the fracture of the coronoid
can be comminuted and difficult to repair; (3) the fracture
of the radial head increases the risk of diminished forearm rotation and proximal radioulnar synostosis; and (4)
the ulnohumeral joint can be unstable, and this is often not
well appreciated. Complications and secondary procedures can be limited by using a plate contoured to wrap
around the dorsal surface of the olecranon and proximal
ulna to provide more secure fixation of the proximal fragment; by obtaining good exposure and secure fixation of
the coronoid; by obtaining stable fixation or replacement
of the radial head; and by repairing the lateral collateral
ligament when injured.

Radial Fractures
Although textbooks describe isolated fractures of the
ulnar diaphysis without proximal or distal radioulnar
joint injury (nightstick fractures), isolated radial fractures are often omitted, with classification systems skipping directly to Galeazzi fractures (fracture of radial diaphysis and dislocation of the distal radioulnar joint). A
recent article suggests that isolated fractures of the diaphyseal radius are unlikely to be associated with major
injury to the distal radioulnar joint (triangular fibrocartilage complex) when the ulnar fracture is greater than
7.5 cm from the radiocarpal joint. It may be unwise to
be complacent about the distal radioulnar joint based
on the location of the fracture, but what this study supports is the idea that many isolated fractures of the radius occur without major radioulnar ligament injury
(Figure 6). After stable anatomic fixation of the radius,
the distal radioulnar joint should be evaluated and compared with a preoperative examination of the opposite,
uninjured side. In the absence of substantial instability,
immediate active mobilization may be safe and worthwhile. If the distal radioulnar joint is unstable there are
several treatment options: (1) repair of a large ulnar styloid fracture if present; (2) repair of the triangular fibrocartilage complex; (3) immobilization of the forearm in
midsupination; and (4) cross-pinning of the radius and
the ulna in midsupination.

312

Figure 6 Preoperative radiograph of a patient with a radial diaphysis fracture. Many

such fractures do not have associated injury to the distal radioulnar joint and can be
managed with early mobilization after plate fixation. Preoperative radiographic alignment and location of the radius fracture can be helpful, but surgeons should always
base treatment decisions on a careful evaluation of the distal radioulnar joint after
plate fixation. ( Copyright D. Ring, MD & J.B. Jupiter, MD.)

Annotated Bibliography
Distal Humeral Fractures
Frankle MA, Herscovici D Jr, DiPasquale TG, Vasey
MB, Sanders RW: A comparison of open reduction and
internal fixation and primary total elbow arthroplasty in
the treatment of intraarticular distal humerus fractures
in women older than age 65. J Orthop Trauma 2003;17:
473-480.
The authors of this study retrospectively reviewed 24 patients older than 65 years with bicondylar (C2 or C3) fractures
of the distal humerus that were treated with either total elbow
arthroplasty (12 patients) or plate and screw fixation (12 patients). At a short-term follow-up of between 2 and 6 years
(average just under 4 years), the functional results were somewhat better in the total elbow arthroplasty than the internal
fixation group, with an average arc of flexion of 113 versus
100 and an average Mayo Elbow Performance Index of 95
versus 88 points, respectively.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Pajarinen J, Bjorkenheim JM: Operative treatment of
type C intercondylar fractures of the distal humerus:
Results after a mean follow-up of 2 years in a series of
18 patients. J Shoulder Elbow Surg 2002;11:48-52.
In this study, 8 of 18 patients (all of whom were older than
50 years) had an unsatisfactory result after plate and screw fixation of a bicolumnar fracture of the distal humerus. Poor
bone quality and open fracture were also associated with inferior results.

Ring D, Jupiter JB, Gulotta L: Articular fractures of the

distal part of the humerus. J Bone Joint Surg Am 2003;
85:232-238.
In this study, 21 patients with apparent fractures of the
capitellum had more complex fractures of the articular surface
of the distal humerus. The authors report that fixation with
buried screws and wires achieved healing with slight settling of
the fracture fragments in two patients and no major osteonecrosis. Ten patients required a second surgical procedure, and
five patients had an unsatisfactory functional result related to
persistent elbow stiffness.

Ring D, Gulotta L, Chin K, Jupiter JB: Olecranon osteotomy for exposure of fractures and nonunions of the
distal humerus. J Orthop Trauma 2004;18:446-449.
In this study, 45 patients had an apex distal chevronshaped olecranon osteotomy repaired with Kirschner wires directed out the anterior ulnar cortex distal to the coronoid process, and bent 180 and impacted into the olecranon
proximally with two 22-gauge figure-of-8 stainless steel tension wires. The only failure occurred in a patient who returned
to athletic activities too soon. Only six patients (13%) had a
subsequent surgical procedure performed specifically to remove the wires.

Schildhauer TA, Nork SE, Mills WJ, Henley MB: Extensor mechanism-sparing paratricipital posterior approach
to the distal humerus. J Orthop Trauma 2003;17:374-378.
Extra-articular fractures (Opthopaedic Trauma Association [OTA] type A) and simple articular distal humeral fractures with simple or multifragmentary metaphyseal involvement (OTA type C1 and C2) were treated by elevating the
triceps off of the back of the humerus and working through
the medial and lateral paratricipital windows (an exposure described previously by Allonso-Llamas).

Wilkinson JM, Stanley D: Posterior surgical approaches

to the elbow: A comparative anatomic study. J Shoulder
Elbow Surg 2001;10:380-382.
In this study, the median-exposed articular surface for triceps splitting, triceps reflecting, and olecranon osteotomy approaches was 35%, 46%, and 57%, respectively.

Radial Head Fractures

Liow RY, Cregan A, Nanda R, Montgomery RJ: Early
mobilisation for minimally displaced radial head frac-

American Academy of Orthopaedic Surgeons

Chapter 27 Elbow and Forearm: Adult Trauma

tures is desirable: A prospective randomised study of
two protocols. Injury 2002;33:801-806.
In this study, 60 patients with minimally displaced fractures of the radial head were prospectively randomized to undergo either immediate active mobilization or 5 days of splint
immobilization. All patients achieved excellent results. Patients treated with immediate mobilization had less pain and
better elbow function at 1 week. There were no adverse consequences of immediate mobilization.

Moro JK, Werier J, MacDermid JC, Patterson SD, King

GJ: Arthroplasty with a metal radial head for unreconstructible fractures of the radial head. J Bone Joint Surg
Am 2001;83:1201-1211.
In this study, 25 patients with metal radial head prostheses
were evaluated an average of 39 months after injury. The average flexion arc was 132, the average forearm arc was 146, the
average disabilities of the arm, shoulder, and hand score was
17, and the average Mayo Elbow Performance Index score
was 80. There were eight unsatisfactory results related to associated injuries, psychiatric problems, and secondary gain.
Asymptomatic lucencies around the stem were the rule because the implant functions as a spacer and is not intended to
have a tight fit in the radial neck.

Pomianowski S, Morrey BF, Neale PG, Park MJ,

ODriscoll SW, An KN: Contribution of monoblock and
bipolar radial head prostheses to valgus stability of the
elbow. J Bone Joint Surg Am 2001;83-A:1829-1834.
In this study, three metal radial head prostheses were
shown to restore valgus stability in the medial collateral ligament deficient elbow close to the status with an intact radial
head.

Ring D, Psychoyios VN, Chin KR, Jupiter JB: Nonunion

of nonoperatively treated fractures of the radial head.
Clin Orthop 2002;398:235-238.
The authors of this study describe five patients with a radial neck nonunion after a minimally displaced radial head
fracture. These fractures are usually asymptomatic.

Ring D, Quintero J, Jupiter JB: Open reduction and internal fixation of fractures of the radial head. J Bone
Joint Surg Am 2002;84:1811-1815.
In this study, 56 patients in whom an intra-articular fracture of the radial head had been treated with open reduction
and internal fixation were evaluated at an average of 48
months after injury. The authors report that good results were
obtained in all of the patients with isolated partial radial head
fractures, in 4 of 15 of those with partial radial head fractures
that were part of a complex injury to the elbow or forearm,
and in 11 of 12 of those with fracture of the entire head into
two or three large fragments. Among patients with fracture of
the entire head into greater than three fragments, 13 of 14 had
an unsatisfactory result, with three early failures and six nonunions.

313

Elbow and Forearm: Adult Trauma

Orthopaedic Knowledge Update 8

Smith AM, Urbanosky LR, Castle JA, Rushing JT, Ruch

DS: Radius pull test: Predictor of longitudinal forearm
instability. J Bone Joint Surg Am 2002;84:1970-1976.

the arm (a mainstay of effective treatment and not associated

with major problems in prior clinical series).

In this sequential cutting study in cadavers, the authors

showed that after radial head resection 3 mm of proximal radial migration with longitudinal traction indicated disruption
of the interosseous membrane and migration of 6 mm or
greater indicated gross longitudinal instability with disruption
of all ligamentous structures of the forearm.

Karlsson MK, Hasserius R, Besjakov J, Karlsson C, Josefsson PO: Comparison of tension-band and figure-ofeight wiring techniques for treatment of olecranon fractures. J Shoulder Elbow Surg 2002;11:377-382.

Elbow Dislocations
McKee MD, Schemitsch EH, Sala MJ, ODriscoll SW:
The pathoanatomy of lateral ligamentous disruption in
complex elbow instability. J Shoulder Elbow Surg 2003;
12:391-396.
Six patterns of injury to the lateral collateral ligament injury were observed in 62 patients with a surgically treated dislocation or fracture-dislocation of the elbow (proximal avulsions in 32, bony avulsions of the lateral epicondyle in 5,
midsubstance ruptures in 18, ulnar detachments of the lateral
collateral ligament in 3, ulnar bony avulsions in 1, and combined patterns in 3). The common extensor origin was also
ruptured in 41 patients (66%).

Coronoid Fractures
Ring D, Jupiter JB, Zilberfarb J: Posterior dislocation of
the elbow with fractures of the radial head and coronoid. J Bone Joint Surg Am 2002;84-A:547-551.
The authors of this article describe 7 of 11 terrible triad elbows that redislocated in a splint after manipulative reduction.
Five patients, including four who were treated with resection
of the radial head, experienced redislocation after surgical
treatment. Only four patients reported satisfactory results, all
of whom had retained the radial head and two of whom had
lateral collateral ligament repair. The subsequent letters to the
editor adds additional perspective to the current concepts regarding the treatment of these injuries.

Olecranon Fractures
Bailey CS, MacDermid J, Patterson SD, King GJ: Outcome of plate fixation of olecranon fractures. J Orthop
Trauma 2001;15:542-548.
Near normal motion, strength, and disabilities of the arm,
shoulder, and hand scores were observed after plate fixation
of 25 displaced olecranon fractures. The authors report that
20% of patients requested plate removal.

Hutchinson DT, Horwitz DS, Ha G, Thomas CW, Bachus KN: Cyclic loading of olecranon fracture fixation
constructs. J Bone Joint Surg Am 2003;85:831-837.
As might be expected based on the size of the implants
alone, a large screw limits displacement of an olecranon osteotomy better than Kirschner wires with a tension band wire
in biomechanical tests in cadavers. Unfortunately, the authors
of this study interpreted their data as disproving the tension
band concept (a basic engineering principle that cannot be disproved) and as discouraging immediate active mobilization of

314

In a long-term study of olecranon fractures, one subset of

patients was treated with tension band wiring and the other
was treated with a simple figure-of-8 wire. The authors report
that the results were comparable for both groups, with a high
rate of hardware removal occurring in both groups after tension band wiring (81% of patients) and after figure-of-8 wiring
(43%).

Karlsson MK, Hasserius R, Karlsson C, Besjakov J, Josefsson PO: Fractures of the olecranon: A 15- to 25-year
followup of 73 patients. Clin Orthop 2002;403:205-212.
In a long-term follow-up of 70 patients who were treated
for olecranon fractures, the authors report excellent or good
results in 96%, slight loss of elbow flexion and extension and
mild or moderate degenerative changes in over 50%. The authors conclude that adequately treated fractures of the olecranon have a favorable long-term outcome.

Diaphyseal Forearm Fractures

Dumont CE, Thalmann R, Macy JC: The effect of rotational malunion of the radius and the ulna on supination
and pronation. J Bone Joint Surg Br 2002;84:1070-1074.
In this cadaver study, the authors report that substantial
rotational malalignment of the radius and/or ulna was necessary before forearm rotation was affected.

Handoll HH, Pearce PK: Interventions for isolated

diaphyseal fractures of the ulna in adults. Cochrane
Database Syst Rev 2004;2:CD000523.
A review of trials was performed to determine the effects
of different treatment methods (in adults) for isolated fractures of the ulnar shaft. The authors concluded that there was
insufficient evidence from randomized trials to determine the
most appropriate treatment method.

Hertel R, Eijer H, Meisser A, Hauke C, Perren SM: Biomechanical and biological considerations relating to the
clinical use of the Point Contact-Fixator: Evaluation of
the device handling test in the treatment of diaphyseal
fractures of the radius and/or ulna. Injury 2001;32(suppl
2):B10-B14.
In this study of 83 diaphyseal forearm fractures in 52 patients that were repaired using a Point Contact-Fixator (Synthes, Paoli, PA), 76 bones healed with callus without further
intervention. Stripping of the hexagonal slot was reported to
be a problem at removal of the implant.

Kasten P, Krefft M, Hesselbach J, Weinberg AM: How

does torsional deformity of the radial shaft influence the

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

rotation of the forearm? A biomechanical study. J Orthop Trauma 2003;17:57-60.
The authors of this study report that significant loss of
forearm rotation was not observed until a minimum of 30 of
rotational malunion.

Rettig ME, Raskin KB: Galeazzi fracture-dislocation: A

new treatment-oriented classification. J Hand Surg Am
2001;26:228-235.

Chapter 27 Elbow and Forearm: Adult Trauma

Chapman MW, Gordon JE, Zissimos AG: Compression
plate fixation of acute fractures of the diaphysis of the
radius and ulna. J Bone Joint Surg Am 1989;71:159-169.
Josefsson PO, Gentz CF, Johnell O, Wendeberg: Dislocations of the elbow and intraarticular fractures. Clin Orthop 1989;246:126-130.
Mason ML: Some observations on fractures of the head
of the radius with a review of one hundred cases. Br J
Surg 1959;42:123-132.

In this review of 40 patients with Galeazzi fracturedislocations, the authors suggest that more proximal fractures
are less likely to have distal radioulnar joint instability. Among
22 fractures in the distal third of the radius (within 7.5 cm of
the radiocarpal joint), 12 had intraoperative distal radioulnar
joint instability. Among 18 more proximal fractures, only one
had intraoperative distal radioulnar joint instability after plating of the radius.

McKee MD, Wilson TL, Winston L, Schemitsch EH, Richards RR: Functional outcome following surgical treatment of intra-articular distal humeral fractures through
a posterior approach. J Bone Joint Surg Am 2000;82:
1701-1707.

Classic Bibliography

Regan W, Morrey BF: Fractures of the coronoid process

of the ulna. J Bone Joint Surg Am 1989;71:1348-1354.

Anderson LD, Sisk D, Tooms RE, Park WI III:

Compression-plate fixation in acute diaphyseal fractures
of the radius and ulna. J Bone Joint Surg Am 1975;57:
287-297.

Ring D, Jupiter JB, Sanders RW, Mast J, Simpson NS:

Transolecranon fracture-dislocation of the elbow. J Orthop Trauma 1997;11:545-550.

Beredjiklian PK, Nalbantoglu U, Potter HG, Hotchkiss

RN: Prosthetic radial head components and proximal
radial morphology: A mismatch. J Shoulder Elbow Surg
1999;8:471-475.
Broberg MA, Morrey BF: Results of treatment of
fracture-dislocations of the elbow. Clin Orthop 1987;216:
109-119.

American Academy of Orthopaedic Surgeons

Ring D, Jupiter JB, Simpson NS: Monteggia fractures in

adults. J Bone Joint Surg Am 1998;80:1733-1744.
Schemitsch EH, Richards RR: The effect of malunion
on functional outcome after plate fixation of fractures
of both bones of the forearm in adults. J Bone Joint Surg
Am 1992;74:1068-1078.

315

Chapter

28

Elbow Reconstruction
Scott P. Steinmann, MD
William B. Geissler, MD

Diagnostic Studies
For evaluation of most elbow disorders, plain radiographs are usually sufficient. Standard views should include AP, lateral, and oblique. Radiographs are often
the only imaging study needed in the clinic or emergency department if an arthritic process or a minor fracture is detected. If a more detailed examination is desired, CT is helpful to define an arthritic process, detect
loose bodies, or pinpoint the location of heterotopic
bone. Occasionally, a suspected radial head fracture will
not be seen on plain radiographs but will be confirmed
by CT. CT can also be quite helpful for defining the extent of a fracture to help determine a surgical repair
strategy.
Fluoroscopy plays a very important role in examination of the elbow. Elbow instability is often subtle and
sometimes it can be difficult to determine the pattern of
subluxation based solely on the clinical examination.
Fluoroscopic examination under anesthesia should be
considered a standard first step before many elbow surgical procedures. In the trauma patient obvious fractures may be seen on radiographs but fluoroscopic
evaluation under anesthesia may demonstrate any associated ligamentous injuries. Similarly, an occult instability pattern may be detected with fluoroscopic examination under anesthesia in a patient undergoing an
elective elbow procedure.
MRI can be helpful in examining a patient with suspected medial or lateral instability. Although MRI cannot be a substitute for a thorough clinical examination,
it can detect frank disruption of the collateral ligament
and partial tears. Unusual causes of elbow pain such as
a glomus tumor or an osteoid osteoma may also be detected by MRI. Occasionally, distal biceps tendinopathy
can be seen on MRI, which may correlate with a clinical
diagnosis of partial biceps tendon tear.

Arthroscopy
Elbow arthroscopy has become a more common procedure over the past decade. Elbow arthroscopy is technically demanding to perform and requires surgeon expe-

American Academy of Orthopaedic Surgeons

rience in advanced arthroscopic techniques. Advantages

of arthroscopic treatment of elbow disorders include improved articular visualization and potentially decreased
postoperative pain. Patients may also benefit from a decreased morbidity and a faster postoperative recovery.
Presently, elbow arthroscopy can be performed for resection of symptomatic plica, removal of loose bodies,
synovectomy for inflammatory arthritis, release of capsule in patients with contractures, removal of osteophytes, treatment of osteochondritis dissecans, dbridement of lateral epicondylitis, and treatment of elbow
fractures. Elbow arthroscopy is difficult because of the
smaller joint working space and the unique articular
anatomy of the elbow. The risk of injury to neurovascular structures is not insignificant. Contraindications to
elbow arthroscopy include significant prior trauma or
surgical scarring. Prior surgery is not a contraindication
but loss of joint space as a result of prior trauma makes
joint visualization very difficult. Previous trauma may
also distort the local anatomy, making accurate identification of neurovascular structures difficult. A subluxating ulnar nerve is not a contraindication to arthroscopy,
but should be identified before beginning the surgical
procedure. Before ulnar nerve transposition can be performed, the nerves location must be known before a
medial portal can be established; an open incision on
the medial side or intraoperative ultrasound can be used
to identify the nerve.

Arthroscopic Technique
Patients undergoing elbow arthroscopy are typically
placed under general anesthesia, which allows for muscle relaxation and permits placing the patient in either a
prone position or the lateral decubitus position, which
might otherwise not be tolerated by an awake patient
(Figures 1 and 2). Regional nerve blocks can be used as
anesthesia for elbow surgery; however, any position
other than the supine position may be difficult for the
patient to maintain. Additionally, if a nerve block has
been administered, it is impossible to assess the patient
immediately postoperatively for potential nerve injury.

317

Elbow Reconstruction

Orthopaedic Knowledge Update 8

Figure 1 Lateral decubitus position for

elbow arthroscopy. A dedicated arm holder is useful for positioning the elbow at
the ideal height.

Figure 2 View of elbow arthroscopy after

final positioning and draping.
Elbow arthroscopy can be performed in either the supine, prone, or lateral decubitus position.
The lateral decubitus position is fairly easy to maintain and is probably the most popularly used position.
At the time of surgery, the ulnar nerve is palpated to be
sure of its location and to check that it does not subluxate from the cubital fossa. It is best to mark all portal
sites before surgery when the elbow is not distended
and palpation of bony landmarks can be done more pre-

318

cisely. Distending the elbow with saline before making

the initial starting portal and inflating the tourniquet is
extremely helpful. With the elbow joint distended, the
major neurovascular structures are positioned further
from the starting portal and entry into the joint is also
easier.
Once the arthroscope has been successfully placed
into the joint, visualization can be maintained by either
pressure distention of the capsule or by retraction. Pres-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Figure 3 Portals marked with a surgical pen prior to beginning arthroscopy. A long

Chapter 28 Elbow Reconstruction

Figure 4 Arthroscopic view of coronoid as viewed from the anterolateral portal.

dotted line represents the area of the ulnar nerve. The retractor is placed between
anteromedial portals.

sure distention works quite well but will eventually lead

to significant fluid extravasation during a long arthroscopic procedure.The use of retractors is an alternative to
pressure distention. Retractors for elbow arthroscopy are
simple lever-type retractors such as a Howarth or a large
blunt Steinmann pin. Retractors are placed into the elbow
joint via accessory portals, which are typically 2 to 3 cm
proximal to the standard arthroscopic viewing portals.
By holding the capsule away from the joint, retractors allow adequate visualization with a low-pressure inflow system.
Portal Placement
Safe initial portal entry can be made either from the medial or lateral side of the elbow depending on the preference of the surgeon (Figure 3). The anterolateral portal is
often the initial portal placed and is usually placed just anterior to the sulcus between the capitellum and radial
head. This is an easily located anatomic area to find in
most patients.The radial nerve is in close proximity to this
portal. Once the arthroscope has been placed into the
joint from the anterolateral portal, the surgeon should be
able to visualize the trochlea, coronoid process, coronoid
fossa, and the medial aspect of the radial head (Figure 4).
After it has been confirmed that the arthroscope is in the
elbow joint, then an anteromedial working portal can be
established. This maneuver is easiest to perform from an
inside-out approach.The arthroscope is removed from the
sheath and a blunt trocar placed back in and then gently
pushed straight across the joint. With gentle pushing and
rotation, the tip of the trocar will begin tenting the skin on
the medial side of the elbow. The skin is then incised over
the tip of the trocar and the sheath pushed out of the skin.
A working cannula can then be placed over the tip of the
sheath and pushed back into the joint.

American Academy of Orthopaedic Surgeons

An anteromedial portal may be placed before the

anterolateral portal based on the surgeons preference.
The anteromedial portal is placed 2 cm distal and 2 cm
anterior to the medial epicondyle. This will penetrate
the common flexor origin and the brachialis. The medial
antebrachiocutaneous nerve is at risk with its portal, often only several millimeters from the starting position.
The posterolateral portal is an excellent initial viewing portal for the posterior aspect of the elbow. Unlike
the anterior portals, where palpation of the median or
radial nerve is not possible, establishment of posterior
portals should be a safe undertaking once the ulnar
nerve is palpated and identified. The posterolateral portal is made lateral and level with the tip of the olecranon with the elbow flexed 90. The trocar should be
aimed at the center of the olecranon fossa. Viewing is
often difficult initially posteriorly, because the fat pad
normally occupies a large portion of the potential joint
space. The direct posterior portal is a good initial working portal. Through this portal loose bodies and osteophytes in the entire olecranon fossa can be removed
while viewing from a posterolateral portal. This portal is
established 2 cm proximal to the tip of the olecranon at
the proximal margin of the olecranon fossa. After creating this portal, a shaver can be placed into the joint and
dbridement performed. The arthroscope and shaver
can be switched back and forth between the direct and
the posterolateral portal to enable efficient completion
of posterior dbridement.
Arthroscopic Treatment of Degenerative Arthritis
Degenerative arthritis is accompanied by osteophyte formation and capsular contraction. In addition to removal
of loose bodies, the surgeon should be prepared to remove
all impinging osteophytes and to release a potentially
tight, thickened capsule. It is the rare patient who has an

319

Elbow Reconstruction
isolated loose body, perhaps after an osteochondral injury.
After initial joint inspection and removal of all obvious
loose bodies, bony work should commence. A shaver or
burr can be used to remove osteophytes from the radial
and coronoid fossae of the humerus and the tip of the
coronoid can be excised. The medial coronoid should also
be examined for osteophytes, which may be missed if not
positively identified. It is helpful to perform most of the
bony work using a burr before completion of the capsulectomy because neurovascular structures are better protected and visualization is improved before muscle or soft
tissue impedes the arthroscopic view. Once the osteophytes have been excised, the anterior capsule can be removed. It is often helpful to take the capsule off the humerus as a first step. Care should be taken when excising
capsule just anterior to the radial head. The radial nerve
is at great risk of injury at this location. Often a small fat
pad can be visualized in this area, the radial nerve being
just anterior.

Complications
Complications that can arise from elbow arthroscopy include compartment syndrome, septic arthritis, and nerve
injury. In a report of 473 elbow arthroscopies, there
were four types of minor complications in 50 procedures, including infection, nerve injury, prolonged drainage, and contracture. The most common complication
was persistent portal drainage. Neurologic complications
were limited to transient nerve palsy. The rate of permanent neurologic injury appears to be higher in the elbow
than in the knee or shoulder. The risk of nerve injury is
higher in patients with rheumatoid arthritis or in those
undergoing a capsular release. Significant nerve injury
during elbow arthroscopy has been reported involving
the radial, median, and ulnar nerves. The use of retractors is probably the most important factor in preventing
nerve injury. In some instances, arthroscopic identification of nerves will allow for safer capsulectomy.

Instability
Pathophysiology
Interest in elbow instability has increased over the past
several years as more has been learned about the pathoanatomy of elbow dislocation. There are essentially
three patterns of acute elbow instability: (1) posterolateral rotatory, (2) valgus, and (3) varus posteromedial rotatory instability. Posterolateral rotatory instability is the
most common mechanism of acute instability and can
progress from a simple dislocation to a complex
fracture-dislocation (terrible triad). Valgus instability occurs most commonly as a chronic overload problem,
particularly in throwers. Varus posteromedial rotatory
instability has recently been recognized as a significant
pattern of instability, usually associated with a medial
coronoid fracture.

320

Orthopaedic Knowledge Update 8

Knowledge of the static and dynamic constraints of
the elbow is helpful in understanding the progression of
acute elbow instability. The three primary static constraints are the ulnohumeral articulation, anterior band
of the medial collateral ligament (MCL), and the lateral
ulnocollateral ligament (LUCL). The secondary constraints include the radiocapitellar articulation, common
flexor and extensor origins, and the capsule. The flexor
and extensor muscle groups across the elbow play a major role in creating dynamic instability and in producing
compressive forces at the elbow joint. Particularly, the
anconeus, because of its origin in the lateral epicondyle
and broad insertion on the ulna, is designed to function
as a dynamic stabilizer against posterolateral rotatory
instability. An elbow will remain stable if the three primary constraints are intact. If the MCL is disrupted, the
radial head and coronoid become more critical for stability. If the coronoid is fractured, the radial head becomes the prime stabilizer and must be reconstructed or
replaced. Likewise, if the radial head is fractured, stability will be difficult to achieve if the coronoid fracture remains unrepaired.

Posterolateral Rotatory Dislocation

A simple elbow dislocation is usually the result of posterolateral rotatory instability. This injury pattern occurs
during a valgus moment to the elbow with simultaneous
forearm supination. Progressively the radial head and
coronoid rotate under the capitellum and dislocation occurs. The disruption of the soft tissues begins laterally
and progresses to the medial side of the elbow (Figure
5). This progressive disruption occurs in a circle pattern
simultaneously anteriorly and posteriorly and has been
referred to as the Horii circle and is analogous to the
Mayfield circle of progressive carpal instability.

Varus Posteromedial Rotatory Dislocation

Similar to posterolateral rotatory instability, varus posteromedial rotary dislocation also occurs during axial
loading of the elbow during flexion. A varus moment
occurs with internal rotation of the forearm resulting in
a fracture of the anteromedial coronoid with disruption
of the LUCL. Unlike posterolateral rotatory instability,
where dislocation can occur without a fracture, the key
to varus posteromedial rotatory dislocation is an anteromedial coronoid fracture (Figure 6).

Recurrent Valgus Instability

Valgus instability can occur as an acute event or after
chronic valgus overload. This condition involves a distinctly different pattern of instability as compared with
posterolateral rotatory instability. As an acute event it
results in rupture of the MCL from a pure valgus load
to the elbow. As the injury occurs, the force is transmitted from medial to lateral, potentially resulting in radial

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 28 Elbow Reconstruction

Figure 5 Fluoroscopic examination under anesthesia demonstrating disruption of the lateral collateral ligament and lateral soft-tissue
attachments. A, Resting position. B, Gross lateral instability is seen after varus stress is applied.

Figure 6 A, CT scan showing varus posteromedial rotatory instability. The comminuted fracture of the coronoid and collapse pattern is best seen on the three-dimensional
reconstruction (B).

head fracture. If in this situation the radial head is excised, the elbow is at significant risk of remaining permanently unstable in valgus. This injury pattern does not
typically result in an acute dislocation. Acute rupture of
the MCL may occur as a catastrophic terminal event in
high-demand overhead throwing athletes but is usually
the result of a chronic valgus overload.

American Academy of Orthopaedic Surgeons

Posterolateral Rotatory Instability

Recurrent instability following simple dislocation of the
elbow involves posterolateral rotatory instability in almost every case. The essential lesion is detachment or
attenuation of the LUCL. Although both the MCL and
LUCL are usually disrupted, the LUCL is at risk of not
healing because of repetitive varus stresses applied to

321

Elbow Reconstruction

Orthopaedic Knowledge Update 8

Figure 7 Demonstration of lateral pivot shift

test for posterolateral rotatory instability. The
test is begun with the elbow extended. The elbow is then placed in supination with a moderate force. As the elbow is flexed to 40 or
more a valgus force torque is increasingly applied. Reduction of the ulna and radius together on the humerus occurs suddenly with a
palpable visible clunk. The reduction is what is
sensed and palpated, not the dislocation.

Figure 8 Comminuted radial head fracture. Pieces of the radial head have been removed (A). The shaft of the radius is then reamed and the end of the fracture is planed to a
level position. The radial head implant is then ready for placement (B).

the elbow, caused by the force of gravity when the

shoulder is abducted. The typical patient has a history of
recurrent painful snapping or locking of the elbow. A
careful history would demonstrate that symptoms occur
during extension arc of motion with the forearm in supination. Examination for posterolateral rotatory instability is best performed with the patient supine and the affected extremity overhead (Figure 7).

Radial Head Replacement

Radial head replacement is indicated for comminuted
fractures of the radial head associated with elbow instability (Figure 8). Routine replacement of comminuted
radial head fractures without associated MCL disruption or interosseous membrane disruption has not tradi-

322

tionally been performed. This viewpoint has begun to

change as newer prostheses have been developed and as
greater awareness occurs of the accelerated degenerative process that can occur at the ulnohumeral joint after radial head excision even with an initially stable
MCL. In long-term studies in which patients were examined after radial head excision, a high percentage of patients developed significant radiographic ulnohumeral
arthrosis. Additionally, radial head excision may cause
decreased grip strength, wrist pain, and progressive valgus instability.

Elbow Contractures
Most elbow contractures are mild and usually well tolerated. The functional range of motion has been defined

American Academy of Orthopaedic Surgeons

Chapter 28 Elbow Reconstruction

Orthopaedic Knowledge Update 8

as between 30 full extension to 130 flexion. Mild loss
of terminal extension is usually well tolerated, but there
are certain exceptions. Basketball players with loss of
extension typically report that the contracture affects
their jump shot; loss of extension affects gymnasts as
well. Primary loss of flexion is usually not well tolerated
because it often affects the ability to perform common
activities of daily life.
The pathogenesis of elbow contractures is poorly understood. Patients with an injured elbow typically will
hold the extremity in a flexed position, which minimizes
the intra-articular pressure from a joint effusion. A fibrous tissue response occurs as a result of periarticular
muscular and capsular tearing. The capsule hypertrophies to a thick contracted structure (normal capsules
are transparent and thin). There are severe intrinsic and
extrinsic factors involved in an elbow contracture. Intrinsic factors include joint incongruity, loose bodies,
synovitis, articular surface ankylosis, and capsular contracture, all of which may be amenable to treatment
with arthroscopic elbow contracture release. Extrinsic
factors include heterotopic bone formation, myositis ossification, collateral ligament contracture, and muscle
contractures, all of which are more amenable to treatment with open elbow contracture release. In many patients, intrinsic and extrinsic factors coexist.
Heterotopic ossification has been shown to occur in
3% of patients with elbow dislocations. It has been documented that the incidence of heterotopic ossification
increases to 20% in patients with an elbow dislocation
associated with a fracture. Heterotopic ossification occurs in 5% of patients with isolated neuraxis trauma and
increases dramatically to 75% to 86% when brain injury
is associated with elbow trauma. An increased incidence
of heterotopic ossification has also been shown to occur
in patients with thermal injuries.
Patients with less than 40 of extension or 105 of
flexion are candidates for elbow contracture release.
The ulnar nerve is particularly vulnerable to injury and
scarring. In patients with preoperative ulnar nerve
symptoms and loss of marked flexion of the elbow, neurolysis and transposition of the ulnar nerve should be
considered.
The traditional timing of release was to wait at least
1 year after the initial injury. Other previous recommendations included waiting until bone scans became cold
and nonreactive and to follow serial alkaline phosphatase levels until a normal range was reached. However, in reality, patients rarely gain motion after
6 months, with the exception of pediatric patients. During this time, the soft-tissue contractures continue to
mature. New recommendations have been made that
suggest elbow contracture release be considered after
6 months when patients have no significant improvement from physical therapy.

American Academy of Orthopaedic Surgeons

The medial over-the-top release allows decompression to the ulnar nerve in patients with preoperative ulnar nerve symptoms. The lateral approach has been
popularized in several studies; this type of release is particularly advantageous in the treatment of patients without preoperative ulnar nerve symptoms. In this approach, the origin of the extensor digitorum communis
and extensor carpi radialis brevis is elevated to expose
the joint capsule. The joint capsule is then excised. The
posterior compartment is slightly more difficult to approach from the lateral side. It is important to approach
the posterior compartment by subperiosteally elevating
the triceps and working proximal to distal. The key is to
approach the olecranon fossa proximally and avoid dissection and involvement of the lateral ulnohumeral
complex. The olecranon fossa is dbrided, and the tip is
excised.
The risk of recurrent heterotopic ossification is low,
but some patients may be candidates for postoperative
radiation. Although there is significant evidence in the
literature to support the use of postoperative radiation
in patients with acetabular fractures who have undergone hip replacement, its use in patients with heterotopic ossification of the elbow remains undefined.
One study recently reported the results of 20 patients with complete ankylosis of the elbow caused by
heterotopic bone formation who underwent surgical release without severe injury to the central nervous system. In this series, the average arc of ulnar humeral motion was 81 in patients with burns and 94 in patients
with ankylosis secondary to trauma. Six of 11 limbs in
the burn group and 5 of 9 patients in the trauma group
had good results. The authors concluded that surgeons
should be aware of the small risk of recurrent heterotopic ossification, mild pain, and recurrent contracture
after surgical release. However, they believed the procedure was effective and safe.

Techniques of Elbow Reconstruction

Rheumatoid arthritis, posttraumatic arthritis, and osteoarthritis are the three major types of arthritis that affect the elbow. Pain is the primary complaint of patients
with these disorders, although some may also experience stiffness, weakness, and potentially instability. Several treatment options are available, depending on the
type and stage of arthritic changes.

Synovectomy
Rheumatoid arthritis is the most frequent type of arthritis that affects the elbow joint. Four stages of rheumatic
disease of the elbow have been described based on radiographic and pathologic criteria. In stage I rheumatic
disease, synovitis is present, but a normal joint surface
has been maintained. In stage II, mild to moderate synovitis is present and evidence of joint space narrowing is

323

Orthopaedic Knowledge Update 8

Elbow Reconstruction
seen on radiographs, but the joint contour is still maintained. In stage III, mild to moderate synovitis and mild
to moderate alteration of the joint surface are typically
present, and there is loss of joint space. In stage IV, mechanical instability with bone on bone articulation and
complete joint space destruction is evident on radiographs.
When conservative measures fail to relieve symptoms, arthroscopic or open synovectomy is frequently
recommended. The role of radial head excision with
synovectomy, however, remains controversial. Good
pain relief and preservation of motion have been reported with arthroscopic and open synovectomy.
The commonly described approaches for open synovectomy include extended lateral approach, mediolateral approach, and transolecranon approach. A total
synovectomy may be performed using the extended lateral approach with radial head excision. The proximal
ulna is displaced medially for adequate medial compartment synovectomy when this approach is selected.
Arthroscopy allows a complete synovectomy to be
done without the need for large capsular incisions,
which are required for any open approach. The arthroscope with a blunt trocar is introduced into the anterior
compartment after it is inflated with approximately
20 mL of sterile lactated Ringers solution. A proximal
anterolateral portal is made using either the inside-out
or outside-in technique. The view offered by the more
proximal portals allows excellent visualization of both
the medial and lateral anterior compartments of the elbow. The goal is to excise the excessive synovitis, taking
care not to aggressively cut through the anterior capsule. The arthroscope and shaver will need to be
switched between the portals to gain good access for the
synovectomy. Once the synovium has been resected, the
radiocapitellar articulation is evaluated. If there is significant involvement of the radial head, then radial head
excision may be required. A synovectomy of the posterior compartment is then performed.
Open synovectomy of the elbow is an accepted procedure with good results. The published results on this
remain consistent. Pain relief is regularly seen in 80% to
90% of patients at 3- to 5-year follow-up. Although recurrence of synovitis is common at 10- to 20-year
follow-up, approximately two thirds of patients were
satisfied with the results of the procedure. Range of motion will improve in approximately 40% of patients, decrease somewhat in approximately 15%, and the remainder will be unchanged. Arthroscopic synovectomy
is a highly demanding procedure. One study of 14 arthroscopic synovectomies revealed an early success rate
of 95%. However, at 3-year follow-up, approximately
60% of patients reported satisfaction with the results.
One study reported on 46 elbows that underwent arthroscopic synovectomy for the treatment of inflammatory arthritis. All 46 elbows had improved motion and

324

decreased pain. One patient developed a synovial fissula, two patients required repeat synovectomy, and one
patient required a total elbow arthroplasty.

Outerbridge-Kashiwagi (Ulnohumeral) Arthroplasty

In patients with primary degenerative arthritis or posttraumatic degenerative arthritis who have pain on terminal flexion and extension, the traditional procedure
has been the Outerbridge-Kashiwagi (ulnohumeral) arthroplasty. This procedure may be performed open or
arthroscopically. In the open procedure, an incision is
made 3 cm proximal to the tip of the olecranon with
splitting of the triceps. The tip of the olecranon is excised, and a foramenectomy in the distal humerus can
then be made using a 4.5-mm drill bit and then enlarged
with a burr. Care must be taken not to involve the medial and lateral columns of the elbow.
This procedure is done arthroscopically through a trocar, and a 4.5-mm drill bit is used to create a foramina in
the center of the olecranon fossa. Once the initial foramina has been made with the drill, it is then enlarged by an
arthroscopic burr inserted in the posterior central portal.
In one report of 12 patients, all had relief of pain and locking at follow-up of 3 to 30 months. A similar experience
was reported in 21 patients who underwent dbridement
and removal of loose bodies in both anterior and posterior compartments. In this group, the average follow-up
was 3 years, and 84% noted satisfactory outcomes.

Interposition Arthroplasty
Patients with posttraumatic arthritis may present with
stiffness, pain, or a combination of both. Traditionally,
total joint arthroplasty has not provided favorable experience in the management of patients with posttraumatic arthritis. In younger patients with posttraumatic
arthritis, an interposition arthroplasty is the treatment
of choice. If this procedure is performed, the elbow is
often protected with an external fixator, which decompresses the joint to decentralize motion while the tissues
are healing.

Arthrodesis
The functional use of the hand depends on the elbow. It
has been show that approximately 50% loss of elbow
motion results in an 80% loss of function to the upper
extremity. Additionally, elbow flexion is usually not well
tolerated. Arthrodesis of the elbow is not compatible
with satisfactory function because range of motion of
the elbow is essential for use of the hand. There is no
single ideal position for arthrodesis. Indications for arthrodesis include the presence of intractable sepsis and
in patients for whom there is no possibility of total elbow reconstruction. Because arthrodesis is primarily a
salvage procedure, its use in young men who perform
heavy labor is controversial.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 28 Elbow Reconstruction

Figure 9 AP (A) and lateral (B) radiographs

of a patient with an unlinked total elbow implant. The patient had good bone stock and
competent ligaments. If bone stock or ligaments are found to be of poor quality at the
time of surgery, the prosthesis can be converted to a linked implant by exchanging the
intra-articular components.

Total Elbow Arthroplasty

Although the results with early elbow prosthesis were
initially disappointing, improvements in surgical techniques and implant designs have made total elbow arthroplasty a reliable procedure for many patients with
rheumatoid arthritis or osteoarthritis, selected older patients with posttraumatic arthritis, and selected patients
with comminuted acute fractures and nonunions.
There are two types of prosthetic joint designs in
use: unlinked (previously referred to as nonconstrained)
and linked (semiconstrained). Recently, implants have
been designed that can be converted intraoperatively
from unlinked to linked. Total elbow arthroplasty is contraindicated in patients who are candidates for alternative procedures, such as synovectomy, dbridement, and
open reduction and internal fixation. Previous infection
is a relatively strong contraindication; however, recent
studies suggest that total arthroplasty can be successfully performed in selected patients using staged dbridements with negative cultures.
The decision to use unlinked or linked arthroplasty
depends on the amount of bone destruction, the status
of the capsule ligamentous tissues of the elbow joint,
and the surgeons experience. Unlinked designs seek to
replicate the axis of motion to the elbow to optimize ligamentous balance and maintain joint stability. Major
forces about the elbow are absorbed by the soft tissues
and theoretically protect the bone-cement interface. To
have sufficient stability, an unlinked arthrodesis requires
adequate bone stock and competent ligaments (Figure
9). The loss of continuity of the soft tissues that can be
seen after trauma or long-standing rheumatoid arthritis
would be a contraindication for an unlinked prosthesis.

American Academy of Orthopaedic Surgeons

Several linked semiconstrained, softly hinged prostheses have been designed for a degree of laxity that
permits the soft tissue to absorb some of the stresses
that would normally be applied to the prosthesis-bone
interface. Static loading conditions of the elbow can result in forces equal to three times body weight, and dynamic loading can equal up to six times body weight.
These tremendous forces can ultimately lead to aseptic
implant loosening and failure. A linked prosthesis may
be used in a patient with bone loss (such as a patient
with humeral nonunion or chronic instability) resulting
from previous trauma or erosion from advanced rheumatoid arthritis (Figure 10). The stability provided by a
linked arthroplasty allows for a complete release of soft
tissues, which may lead to more predictable gains in motion.
Various surgical approaches have been described for
total elbow arthroplasty. A distal-based triceps flap has
been used, particularly for unlinked implants. A standard posterior approach to the elbow is made; then a
triangular or rectangular distal-based triceps flap is
made. Care is taken to retain the triceps tendon attachment to the olecranon. Dissection is then continued
along the lateral side of the olecranon, and the joint is
exposed and hinged on the intact medial collateral ligament. The attachment of the medial collateral ligament
is preserved. Dissection can then continue proximally
along the lateral collateral ligament insertion on the humerus, and release of the anterior capsule can be done
to gain more exposure. The tip of the base of the olecranon and radial head may then be excised to gain further
exposure for joint arthroplasty. Management of the ulnar nerve is controversial. Most surgeons recommend

325

Elbow Reconstruction

Orthopaedic Knowledge Update 8

Figure 10 AP (A) and lateral (B) radiographs of a 68-year-old woman with rheumatoid arthritis and elbow trauma with bone deformity and ligament instability. C, AP radiograph
after a linked total elbow implant was inserted. The decision to link the elbow was made because of the bone deformity and ligament instability.

identification and protection of the nerve without formal transposition.

A Bryan-Morrey approach is typically used for a
linked arthroplasty. The triceps is elevated from medial
to lateral off the olecranon in continuity with the anconeus muscle. Before this approach is attempted, the ulnar nerve should be well identified and protected. A
long periosteal strip of the triceps off of the olecranon
and proximal ulna is made, creating a continuous extensor mechanism sling. The triceps tendon with the long
periosteal sleeve is then reattached with sutures through
multiple drill holes in the olecranon.
There are numerous published reports of unlinked
elbow arthroplasties. Pain relief and restoration by functional arc of motion have been reported in more than
90% of patients. The designers of this implant report excellent results in patients with rheumatoid arthritis, with
only a 1.5% incidence of postoperative dislocation and
a low incidence of aseptic loosening.
The success of linked total elbow arthroplasty has
also been well documented. A series of 58 modified
Coonrad semiconstrained total elbow arthroplasties followed for 3.8 years postoperatively yielded a 91% excellent or good result, with 84% of patients having no pain.
Similar results were reported with the use of a semiconstrained prosthesis (GSB III Elbow System, Zimmer,

326

Warsaw, IN), with 91% excellent or good results reported at 4-year follow-up.
In another study, 26 patients underwent either a
linked or unlinked total elbow arthroplasty; all procedures were performed by a single surgeon. The authors
reported that no significant differences were found in
functional performance or progressive radiolucent loosening and concluded that, when properly performed, total elbow arthroplasty with either type of prosthesis
yielded satisfactory results.
Recently, linked elbow arthroplasty has been recommended for the treatment of elderly patients with severely comminuted intra-articular fractures of the distal
humerus. Additionally, elbow arthroplasty has been recommended for distal humeral nonunions (Figure 11).
Under certain conditions, it is standard practice to excise humeral condyles during insertion of the linked total elbow prosthesis. One study recently reviewed the
results of condylar resection on forearm, wrist, and hand
strength in 32 patients who underwent total elbow arthroplasty. The normal contralateral limb served as the
control, and strength values were given as a percentage
of the normal side. The humeral condyles were intact in
16 patients and had been resected in the other 16 patients. The authors found no significant difference between the two groups with regard to strength of pronation and supination, wrist extension, or grip strength.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 28 Elbow Reconstruction

Figure 11 A, AP radiograph of a 75-year-old

woman with a 2-year history of a distal humeral nonunion. B, AP radiograph after a linked
total elbow implant was inserted. For this type
of arthroplasty, the distal supracondylar bone
fragments are removed, and the patient is
started on immediate range-of-motion exercises.

Figure 12 A, AP radiograph of a 72-year-old

woman 8 months after undergoing total elbow
arthroplasty. The patient fell and thereby sustained the humeral periprosthetic fracture.
B, AP radiograph showing a custom-made,
long-stem humeral component with interlocking screws. Two screws were placed proximal
and distal to the fracture .

American Academy of Orthopaedic Surgeons

327

Orthopaedic Knowledge Update 8

Elbow Reconstruction
There was no significant difference between the two
groups with regard to Mayo Elbow Performance Scores
(79 in the group with intact condyles versus 77 in the
group with resection of the condyles).
The most devastating complication of total elbow arthroplasty is infection. Infection rates between 2% to
5% have been reported. Late infections with persistent
sepsis despite dbridements should be managed by removal of the implant and all cement. Ulnar nerve neuropathy is common after total elbow arthroplasty. Usually, it most often results in transient numbness, which
resolves itself. However, residual sensory and motor
symptoms are occasionally reported. The nerve may be
compressed, and the elbow may be subluxated to replace the components and damaged by retraction or
thermal injury from extravasated bone cement. It is important to take care to protect the ulnar nerve during
the procedure. Intraoperative fractures of one or both
humeral columns may also occur, usually at the MCL
column as a result of stress on the MCL during the procedure. Open reduction and internal fixation of the fracture column may be performed either with intramedullary screws or potentially a plate. This also usually
involves conversion of an unlinked to a linked component.
Periprosthetic fractures around the elbow have been
reported. If minimally displaced, the fracture may be
managed nonsurgically with bracing. Displaced or unstable fractures may require surgical intervention. Previous options included plate fixation around the prosthesis or further stabilization with allografts. Recently, the
use of custom-made long-stem implants that act as intramedullary rods to cross over the fracture site with interlocking screws has been reported (Figure 12.)

Emerging Concepts
The outcome of total elbow arthroplasty has markedly
improved over the past two decades as a result of improved implant designs and surgical techniques. Recent
studies have demonstrated that total elbow arthroplasty
in patients with an inflammatory arthritis is quite successful. Patients who are difficult to treat, such as those
with osteoarthritis and posttraumatic arthritis, remain
problematic because implant survival rate has been
lower. Opportunities for improvement include advances
in implant design and implantation techniques. Titanium, cobalt-chromium, and polyethylene are the most
common materials currently used in elbow arthroplasty.
Whether ceramic or other materials will improve the
longevity of elbow arthroplasty will require further investigations. The optimal stem shape and length for total
elbow arthroplasty has yet to be determined. Newer implants that allow the surgeon to decide intraoperatively
whether linked or unlinked components may be best for
the patient have recently been developed.

328

Annotated Bibliography
Arthroscopy
Kelly EW, Morrey BF, ODriscoll SW: Complications of
elbow arthroscopy. J Bone Joint Surg Am 2001;83:25-34.
A large study of 473 elbow arthroscopies was conducted in
449 patients. A serious complication (joint space infection) occurred in four (0.8%). Minor complications, such as superficial
infection, minor contracture, and transcient nerve palsy, occurred after 50 (11%) of the procedures. Although there were
no permanent nerve injuries, the risk to neurovascular structures is emphasized, particularly in patients with rheumatoid
arthritis.

Steinmann SP: Elbow arthroscopy. JASSH 2003;3:199207.

Review of technique for elbow arthroscopy. This article
describes the operating room setup and indications for elbow
arthroscopy. Discussion of technique for treatment of different
conditions is included.

Instability
Ball CM, Galatz LM, Yamaguchi K: Elbow Instability:
Treatment strategies and emerging concepts. Instr
Course Lect 2002;51:53-61.
A description of biomechanics of elbow instability and relevant anatomy is presented. Treatment of acute and chronic
instability is discussed.

Thompson WH, Jobe FW, Yocum LA, Pink MM: Ulnar

collateral ligament reconstruction in athletes: Musclesplitting approach without transposition of the ulnar
nerve. J Shoulder Elbow Surg 2001;10:152-157.
The technique for MCL reconstruction is reviewed. The
authors describe in a large series the step-by-step procedure
for ligament reconstruction.

Radial Head Replacement

Moro JK, Werier J, MacDermid JC, Patterson SD, King
GJ: Arthroplasty with a metal radial head for unreconstructible fractures of the radial head. J Bone Joint Surg
Am 2001;83:1201-1211.
Patients treated with a metal radial head arthroplasty had
mild to moderate impairment of elbow and wrist function. Use
of a metal radial head, however, was found to be safe and effective treatment.

Elbow Contractures
Ring D, Jupiter JB: Operative release of complete ankylosis of the elbow due to heterotopic bone in patients
without severe injury of the central nervous system.
J Bone Joint Surg Am 2003;85:849-857.
The authors found that attempts to regain motion in this
class of patients are both worthwhile and safe in their experience with 11 elbows in seven patients.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 28 Elbow Reconstruction

King GJ: New frontiers in elbow reconstruction: Total

elbow arthroplasty. Instr Course Lect 2002;51:43-51.

Jones GS, Savoie FH: Arthroscopic capsular release of

flexion contractures (arthrofibrosis) of the elbow.
Arthroscopy 1993;9:277-283.

This article reviews the literature comparing linked and

unlinked total elbow arthroplasty and the relative indications
for each type of prosthesis.

King GJ, Morrey BF, An K-N: Stabilizers of the elbow.

J Shoulder Elbow Surg 1993;2:165-174.

Techniques of Elbow Reconstruction

McKee MD, Pugh DM, Richards RR, Pedersen E, Jones

C, Schmitsch EH: Effect of humeral condylar resection
on strength and functional outcome after semiconstrained total elbow arthroplasty. J Bone Joint Surg Am
2003;85:802-807.
The authors found condylar resection had a minimal, clinically irrelevant effect on forearm, wrist, and hand strength and
no significant effect on the Mayo Elbow Performance score
following total elbow arthroplasty in 16 patients.

Classic Bibliography
Evans EB: Orthopaedic measures in the treatment of
sever burns. J Bone Joint Surg Am 1966;48:643-669.
Ewald FC, Simmons ED Jr, Sullivan JA, et al: Capitellocondylar total elbow replacement in rheumatoid arthritis. J Bone Joint Surg Am 1993;75:498-507.
Froimson AI: Interposition arthroplasty of the elbow, in
Morrey BF (ed): The Elbow. New York, NY, Raven
Press, 1994, pp 329-342.
Garland DE, Hannscom DA, Keenan MA, et al: Resection of heterotopic ossification in the adult with head
trauma. J Bone Joint Surg Am 1985;67:1261-1269.
Hastings H II, Cohen MS: Posttraumatic contracture of
the elbow: Operative release using a new approach.
Trans ASES 1996, 13-32.
Hastings H II, Graham TJ: The classification and treatment of heterotopic ossification about the elbow and
forearm. Hand Clin 1994;10:417-437.
Hedley AK, Mead LP, Hendren DH: The prevention of
heterotopic bone formation following total hip arthroplasty using 600 rad in a single dose. J Arthroplasty
1989;4:319-325.
Herold N, Schroder HA: Synovectomy and radial head
excision in rheumatoid arthritis: Eleven patients followed for 14 years. Acta Orthop Scand 1995;66:252-254.
Hotchkiss RN, An Kn, Weiland AJ, et al: Treatment of
severe elbow contractures using the concepts of Ilizarov.
61st Annual Meeting Proceedings. Rosemont, IL, American Academy of Orthopaedic Surgeons, 1994.
Ikeda M, Oka Y: Function after early radial head resection for fracture: A retrospective evaluation of 15 patients followed for 3-18 years. Acta Orthop Scand 2000;
71:191-194.

American Academy of Orthopaedic Surgeons

King GJ, Zarzour ZD, Rath DA, Dunning CE, Patterson

SD, Johnson JA: Metallic radial head arthroplasty improves valgus stability of the elbow. Clin Orthop 1999;
368:114-125.
Kraay MJ, Figgie MP, Inglis AE, et al: Primary semiconstrained total elbow arthroplasty: Survival analysis in
113 consecutive cases. J Bone Joint Surg Br 1994;76:636640.
Lee BP, Morrey BF: Arthroscopic synovectomy of the
elbow in rheumatoid arthritis: A prospective study.
J Bone Joint Surg Br 1997;79:770-772.
Ljung P, Jonsson K, Larsson K, et al: Interposition arthroplasty of the elbow in rheumatoid arthritis.
J Shoulder Elbow Surg 1996;5:81-85.
Modabber MR, Jupiter JB: Reconstruction for posttraumatic conditions of the elbow joint. J Bone Joint
Surg Am 1995;77:1431-1446.
Morrey BF: Primary degenerative arthritis of the elbow:
Treatment by ulnohumeral arthroplasty. J Bone Joint
Surg Br 1992;74:409-413.
Morrey BF, Adams RA: Semiconstrained arthroplasty
for the treatment of rheumatoid arthritis of the elbow.
J Bone Joint Surg Am 1992;72:479-490.
Morrey BF, Adams RA, Bryan RS: Total elbow arthroplasty for post-traumatic arthritis of the elbow. J Bone
Joint Surg Br 1991;73:607-612.
Nestor BJ, ODriscoll SW, Morrey BF: Ligamentous reconstruction for posterolateral rotatory instability of the
elbow. J Bone Joint Surg Am 1992;74:1235-1241.
Nowicki KD, Shall LM: Arthroscopic release of a posttraumatic flexion contracture in the elbow: A case report and review of the literature. Arthroscopy 1992;8:
544-547.
ODriscoll SW: Arthroscopic treatment for osteoarthritis of the elbow. Orthop Clin North Am 1995;26:691-706.
ODriscoll SW, Morrey BF: Surgical reconstruction of
the lateral collateral ligament, in Morrey BF (ed): Master Techniques in Orthopedic Surgery: The Elbow. New
York, NY, Raven Press, 1994, pp 169-182.
Ogilvie-Harris DJ, Gordon R, MacKay M: Arthroscopic
treatment for posterior impingement in degenerative arthritis of the elbow. Arthroscopy 1995;11:437-443.

329

Elbow Reconstruction
Ring D, Jupiter JB: Fracture-dislocation of the elbow.
J Bone Joint Surg Am 1998;80:566-580.
Ross G, McDevitt ER, Chronister R, Ove PN: Treatment of simple elbow dislocation using an immediate
motion protocol. Am J Sports Med 1999;27:308-311.
Rymaszewski L, Glass K, Parikh R: Post-traumatic elbow contracture treated by arthrolysis and continued
passive motion under brachial plexus anesthesia. J Bone
Joint Surg Br 1996;76:S30.
Rymaszewski LA, Mackay I, Amis AA, Miller JH:
Long-term effects of excision of the radial head in rheumatoid arthritis. J Bone Joint Surg Br 1984;66:109-113.
Savoie FH III, Nunley PD, Field LD: Arthroscopic management of the arthritic elbow: Indications, technique,
and results. J Shoulder Elbow Surg 1999;8:214-219.

330

Orthopaedic Knowledge Update 8

Thal R: Arthritis, in Savoie FH, Field LD (eds): Arthroscopy of the Elbow. New York, NY, Churchill Livingstone
1996, pp 103-116.
Tulp NJ, Winia WP: Synovectomy of the elbow in rheumatoid arthritis: Long term results. J Bone Joint Surg Br
1989;71:664-666.
Vasen AP, Lacey SH, Keith MW, et al: Functional range
of motion of the elbow. J Hand Surg Am 1995;20:288291.
Wright TW, Wong AM, Jaffe R: Functional outcome
comparison of semiconstrained and unconstrained total
elbow arthroplasties. J Shoulder Elbow Surg
2000;9:524-531.

American Academy of Orthopaedic Surgeons

Chapter

29

Wrist and Hand: Trauma

Emily Anne Hattwick, MD, MPH
Edward Diao, MD

Distal Radius Fractures

Distal radius fractures are the most common orthopaedic fractures. These fractures are associated with osteoporosis in the elderly and high-energy trauma in
younger patients. Successful treatment outcomes are
correlated with accuracy of articular reduction, restoration of normal anatomic relationships, and early efforts
at regaining motion in the fingers and wrist.
The mechanism of injury is most commonly an axial
load through the hand and wrist. Useful imaging studies
include AP, lateral, and oblique radiographs, arthrogram, CT, and MRI. Reconstructed CT studies are helpful in planning for articular reduction of die-punch lesions, and MRI studies identify associated soft-tissue
injuries in complex fractures such as to the triangular fibrocartilage complex (TFCC) and scapholunate, and the
lunotriquetral ligaments.
Several classification systems describe fracture patterns. Restoration of radial height (average, 13 mm),
volar tilt (average, 11), radial inclination (average, 23),
and articular surface are associated with good outcomes
(Figure 1). In extra-articular fractures, radial shortening
is associated with the greatest loss of wrist function and
incidence of wrist degenerative changes, including
changes to the TFCC. Intra-articular fractures with
greater than 1 mm of articular incongruity are also associated with early radiocarpal degenerative changes.

Extra-articular Distal Radius Fractures

Extra-articular fractures with minimal displacement can
be treated nonsurgically with closed reduction and
splinting or casting with the wrist in a neutral position.
Studies have shown increased carpal tunnel pressures
associated with increasing wrist flexion and extension
positions, and extreme flexion or extension should be
avoided when immobilizing wrist fractures. Closed reduction and percutaneous pinning relies on either intrafocal manipulation and pinning or manual traction,
reduction, and pinning to hold the fracture in appropriate anatomic alignment. A percutaneous or limited incision technique is involved. Anatomic studies confirm

American Academy of Orthopaedic Surgeons

the risk of injury to the sensory branch of the radial

nerve and the radial artery with percutaneous pinning
through the snuffbox. Limited incisions or protective
devices such as angiocatheters can be used to protect
the distal radial sensory nerve branches from injury. External fixation should be used in unstable fractures to
provide distraction at the fracture site and prevent collapse and loss of reduction over time. External fixation
relies on ligamentotaxis to maintain a reduction, but excessive traction across the wrist ligaments leads to stiffness and can precipitate complex regional pain syndrome. Clinical range of motion of the fingers
intraoperatively along with radiographic evaluation of
joint distraction can help avoid overdistraction. The addition of two Kirschner wires to facilitate external fixation has been shown to increase stability of the fracture
with digital and forearm motion, in some patterns approaching a 3.5-mm AO plating technique. Bicolumnar
plating uses orthogonal plates, which provide greater
stability with smaller lower profile plates. This paperclip system of plating uses several small incisions to approach the radial, dorsal, and ulnar columns of the distal
radius and place plates at 90 to each other. Multiple incisions and increased technical difficulty are disadvantages to this system. Dorsal plating has historically been
associated with extensor tendon irritation, plate prominence, and tendon rupture. Newer systems have been
developed using lower profile plates with recessed screw
heads but there is still a low incidence of tendon irritation and synovitis. Recently, volar plates that are similar
in concept to a blade plate have been introduced. These
more rigid plates with locking screw or peg constructs
have been designed to support the subchondral bone at
the articular surface, and even the dorsal articular surface if necessary.

Intra-articular Distal Radius Fractures

Indications for surgical treatment of intra-articular fractures are based on patient age and activity level and include articular step-off and gap greater than 2 mm, oblique volar fractures, die-punch fractures, significant

331

Orthopaedic Knowledge Update 8

Wrist and Hand: Trauma

Figure 1 A, Normal anatomy of the distal radius on lateral radiograph. Volar or palmar tilt is 11 to 12. B, Normal anatomy of the distal radius on AP radiograph. Radial
inclination is 22 to 23 and radial height between the tip of the radial styloid and the ulnar border of the radius is 11 to 13. (Reproduced with permission from Jafarnia K,
Jupiter J. Distal radius fracture: Anatomy, biomechanics and classification, in Trumble T (ed): Hand Surgery Update 3. Rosemont, IL, American Society for Surgery of the Hand,
2003, p 84.)

dorsal comminution involving more than one third the

anteroposterior diameter of the radius, and fractures
that lose reduction in the first weeks after injury. Treatment options are similar to those for extra-articular
fractures. Closed reduction and percutaneous fixation,
external fixation, plating using orthogonal plating, and
dorsal or volar plating have been described to restore
articular and cortical alignment and relationships. Arthroscopy has been used to assess and improve these
described treatment options. A study of 34 intraarticular fractures revealed improved clinical outcomes
with arthroscopically-assisted reduction compared with
classic open reduction and internal fixation of these
fractures. Complex fractures may benefit from bone
grafting at the time of surgery. Autograft, allograft, and
synthetic grafts all are available, as well as bone cement,
which can provide temporary structural support before
being absorbed.
At the time of definitive treatment, assessment of
distal radioulnar joint (DRUJ) function is critical to ascertain adequate pronation and supination without undue instability. Complications associated with distal radius fractures include malunion, nonunion, extensor
pollicis longus rupture, and early degenerative changes.
Reconstructive options will be discussed in detail in
chapter 30.

332

Distal Ulna Injuries

When distal radius fractures are displaced more than 1
cm, ulnar-sided injuries such as TFCC tears or ulnar styloid fractures are the rule. Concomitant assessment of
the stability of the DRUJ is key to deciding when to repair distal ulna fractures or ligamentous injuries. The ulnar styloid is devoid of ligamentous attachments, but the
fovea at the base of the styloid is where the ulnar-sided
ligaments of the TFCC attach. Nondisplaced fractures
proximal to the ulnar styloid can be treated with cast
immobilization. Displaced fractures or bony avulsions of
the ligamentous structures in the face of a grossly unstable DRUJ should be repaired. An unstable DRUJ can
be defined as an ulnar head that subluxates its full width
out of the sigmoid notch with the forearm in a neutral
position.
In patients with complex elbow injuries, wrist radiographs should be obtained and a careful clinical examination of the wrist should be performed. In an EssexLopresti injury pattern, radial head pathology is coupled
with an interosseous membrane injury extending to the
DRUJ, creating an unstable relationship between the radius and ulna. Proximal migration of the radius results
in secondary DRUJ pathology and ulnocarpal abutment. Following appropriate treatment of the bony pathology (radial head, shaft fracture), pinning the DRUJ

American Academy of Orthopaedic Surgeons

Chapter 29 Wrist and Hand: Trauma

Orthopaedic Knowledge Update 8

Figure 2 Trumble classification of distal ulna

fractures/DRUJ injuries. Type 1 is a disruption
or fracture of the fibrocartilage disk of the
TFCC. Type 2A is an avulsion of the tip of the
ulnar styloid leaving the TFCC intact and stable. Type 2B disrupts the stability of the TFCC
with a fracture through the base of the styloid
and loss of integrity of the ulnar attachments
of the TFCC. Type 3 is a fracture through the
metaphyseal/diaphyseal ulna with minimal
comminution. Type 4 is a comminuted fracture
involving the metaphyseal/diaphyseal region
of the ulna. In both types 3 and 4 the articular
surfaces of the distal ulna are uninvolved. In
type 5, the fracture is near the epiphysis and
enters the sigmoid notch disrupting the DRUJ
articulation. Type 6 fractures disrupt the TFCC,
the DRUJ, and the articular surface of the
ulna. (Courtesy of Thomas Trumble, MD.)

for 6 weeks in neutral position may facilitate ligamentous healing. Essex-Lopresti injuries that have been neglected or failed prior treatment may require radial
head implant reconstruction.
Indications for fixation of distal ulna fractures includes displaced fractures of the base of the styloid, sigmoid notch fractures, and Galeazzi fracture patterns
(Figure 2). Methods of fixation include using a headless
screw such as a Herbert screw, wire or suture tensionbanding, and excision of the fragment with soft-tissue
repair (Figure 3). Repairable TFCC tears can be approached through an open incision or increasingly with
arthroscopically-assisted outside-in or the inside-out approaches. More sutures can be placed using the
outside-in technique but the dorsal sensory branch of
the ulnar nerve is at risk and the carpal ligament repair
is more difficult (Figures 4 and 5).

Carpal Fractures
The scaphoid is the most commonly fractured carpal
bone. More than half of the bone is covered by articular
cartilage. The dorsal surface of the scaphoid has a nonarticular ridge where the dorsal carpal branch of the radial artery enters and provides blood supply to the
proximal pole and 80% of the scaphoid. The superficial
volar branch of the radial artery supplies the remainder
of the blood supply to the scaphoid. The usual mechanism of injury is an axial load across a hyperextended,
ulnarly deviated wrist. Pain with resisted pronation,
snuffbox tenderness, and scaphoid tuberosity tenderness
should raise the suspicion of a scaphoid fracture. Radiographs should include an AP, lateral, and PA view of the
scaphoid with the hand in ulnar deviation, and an ob-

American Academy of Orthopaedic Surgeons

Figure 3 Methods of fixation for unstable distal ulna fractures. (Courtesy of Thomas
Trumble, MD.)

lique 45 view with the hand in pronation. Even with

normal radiographs, a high level of clinical suspicion
should lead to immobilization in a thumb spica splint or
cast and reevaluation in 14 to 21 days. Repeat films after 14 to 21 days can reveal an occult scaphoid fracture.
Bone scans have an 85% to 93% positive predictive
value at 72 hours after injury and have been used when

333

Wrist and Hand: Trauma

Figure 4 Illustration of the ligaments of the ulnar side of the wrist. PRU = palmar
radioulnar; UL= ulnolunate; UC = ulnocapitate; UT = ulnotriquetral; LT = lunotriquetral
(palmar region); TC = triquetrocapitate; TH = triquetrohamate; IOM = interosseous
membrane; R = radius, U = ulna; L = lunate; T = triquetrum; P = pisiform. (Courtesy of
Thomas Trumble, MD.)
radiographs are not revealing and clinical suspicion remains high. Sagittal and coronal CT scans have been
used, and MRI allows immediate identification of fractures and later evaluation for osteonecrosis.
Once a scaphoid fracture has been identified, the location and degree of displacement influences treatment
choices. Several classification schemes exist but the most
helpful in planning treatment is derived from Herbert
and Fisher defining a fracture as stable (occult, incomplete, or nondisplaced) or unstable (displaced, comminuted, dislocated, and combined). A delay in treatment
of less than 28 days was associated with a 5% nonunion
rate in one study, whereas a delay of more than 28 days
led to a rate of 45%.
Nonsurgical treatment of stable fractures is with a
below-elbow thumb spica cast for 2 to 5 months depending on the location of the fracture: distal waist, 2 to
3 months; midwaist, 3 to 4 months; proximal third, 4 to 5
months. Some authors still recommend 6 weeks in a
long arm cast followed by 6 weeks in a short arm cast.
Fracture healing can be determined by clinical assessment of the degree of tenderness and with plain radiographs, CT, or MRI. Athletes should not return to play
until studies show a healed fracture, although high-level
competitors have returned to play in custom casts. In a
recent prospective study, the incidence of union of stable fractures treated nonsurgically was 35 of 45 patients.
Displaced fractures more than 1 mm, those with a
radiolunate angle greater than 15, those with an intrascaphoid angle greater than 35, scaphoid fractures associated with perilunate dislocations, and proximal pole
fractures require surgical treatment (Figure 6.) Optimal

334

Orthopaedic Knowledge Update 8

treatment is with screw fixation. Percutaneous or miniopen fixation of minimally displaced scaphoid fractures
allows minimal dissection and preservation of extrinsic
ligaments. During a dorsal approach to the scaphoid, the
physician should be careful to preserve the blood supply
when entering the dorsal ridge by limiting the exposure
to the proximal half of the scaphoid. Volar approaches
can be useful to expose the entire scaphoid, and are particularly useful to reduce significant flexion deformity of
the scaphoid. The rates of union for open reduction and
internal fixation, both standard and percutaneous fixation, have been shown to be 89% to 100%; some studies
found faster times to union and earlier return to work
when compared with nonsurgical treatment.
Lunate dislocations are more common than lunate
fractures and are associated with a spectrum of ligamentous injuries in the wrist as described by Mayfield (Figure 7). Avulsion fractures of the lunate are treated with
4 weeks of immobilization and gradual return to full
mobility. Larger fragments that are displaced can be
treated with screw fixation, but osteonecrosis may result
from the injury.
Hamate fractures usually occur from a direct blow
such as from a baseball bat or a golf club. The most
common presentation is a hook of hamate fracture and
is best visualized with a carpal tunnel radiograph.
Hamate body and proximal pole fractures are less common. Because these fractures are often difficult to see
on plain radiographs, CT scans are helpful in symptomatic patients. Excision of the hook of hamate fragment is
the preferred treatment to avoid the sequelae of fracture nonunion and persistent pain. At the time of surgery, releasing Guyons canal and identification of the
ulnar artery and nerve is recommended.
Triquetral avulsion fractures frequently result from a
fall on an outstretched hand and can be treated with a
brief period of immobilization (as with a wrist sprain)
without resultant instability.

Carpal Instability
Scapholunate Instability
Scapholunate injuries are not always recognized acutely
and present as a spectrum of injuries involving the
scapholunate interval and extrinsic ligaments such as
the radioscaphocapitate and scaphotrapezial ligaments.
These injuries can be classified as occult (partial
scapholunate ligament injury, normal radiographs), dynamic (incompetent or torn scapholunate ligament, bxabnormal stress radiographs), static or complete
scapholunate dissociation (complete tear of scapholunate ligaments and torn extrinsic ligaments, abnormal
static radiographs), and dorsal intercalated segment instability (DISI), and scapholunate advanced collapse
(SLAC) wrists (chronic patterns frequently with degenerative changes). DISI is a progressive pattern of inter-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 29 Wrist and Hand: Trauma

Figure 5 Palmar classification of TFCC injuries. Class 1A injuries involve the horizontal or central portion of the TFCC disk and are treated nonsurgically or with dbridement.
Class 1B tears represent an avulsion of the peripheral portion of the TFCC from the insertion on the distal ulna. Class 1C tears involve disruption of the ulnocarpal ligaments
creating an avulsion of the TFCC from its carpal attachment. Class 1D tears are an avulsion of the TFCC from its radial attachment. (Courtesy of Thomas Trumble, MD.)

Figure 6 A and B, Transscaphoid perilunate carpal dislocation. Note the widened scapholunate interval suggesting a scapholunate ligament tear. (Courtesy of the University of
California, San Francisco, CA.)

carpal instability with a dorsiflexed lunate and a volarflexed scaphoid, frequently caused by a disruption of
the scapholunate ligament (Figure 8). Volar intercalated
segment instability (VISI) describes the opposite deformity with a volar-flexed lunate, frequently caused by
disruption of the lunotriquetral ligament. The abnormal
distribution of forces across the midcarpal and radiocarpal joints leads to pain, weakness, and early degenerative changes. SLAC describes a scenario in which the
chronic dissociation between the scaphoid and lunate
results in wear, sclerosis, and a degenerative arthritis initially involving the radioscaphoid and capitolunate
joints (Table 1).
Careful clinical and radiographic examination is important. Common symptoms include difficulty bearing
loads across the wrist, symptomatic dysfunction, and abnormal kinematics through the full range of motion.

American Academy of Orthopaedic Surgeons

Tenderness directly over the scapholunate ligament can

be expected and a positive scaphoid shift test (Watson
test) is helpful but must be compared with the contralateral side to confirm true ligamentous laxity (Figure 9).
Radiographic diastasis of the scapholunate joint
greater than 2 mm is associated with a complete tear, although the reliability of this measurement on plain radiographs has been questioned and contralateral comparison views are essential. Neutral PA, clenched fist
AP, true lateral, ulnar deviation PA, and fully flexed lateral radiographs provide helpful information. Important
radiographic findings include a scapholunate angle
greater than 70, dorsal flexion of the lunate greater
than 15 on a neutral lateral radiograph, and the
scaphoid ring sign in which the flexed distal pole of the
scaphoid projects a ring pattern at the distal scaphoid
on the AP projection. MRI studies show poor sensitivity

335

Wrist and Hand: Trauma

Orthopaedic Knowledge Update 8

Figure 8 The normal scapholunate angle (A) averages 47 with a range of 30 to

60. An angle greater than 80 (B) indicates complete scapholunate dissociation or a
displaced scaphoid fracture. (Reproduced with permission from Garcia-Elias M: Carpal
instabilities and dislocations, in Green DP, Hotchkiss RN, Pederson WC (eds): Greens
Operative Hand Surgery. Philadelphia, PA, Churchill Livingstone, 1993, p 878.)

Figure 7 Mayfields staging of the progression of perilunate injury and instability.

Stage I is a scapholunate ligament disruption. Stage II is a disruption of the capitolunate articulation. The lunotriquetral ligament is disrupted in stage III and complete
dislocation of the lunate with loss of the short radiolunate ligament occurs in stage IV.
(Reproduced with permission from the Mayo Foundation, Rochester, MN.)

zial trapezoid and scaphocapitate fusions have better outcomes and successful fusion rates than scapholunate fusions (one seventh of the scapholunate fusion rate in one
study), but may require later revisions if progressive arthritic changes occur in the remaining mobile joints.

Lunotriquetral Instability
with significant normal variation, but magnetic resonance arthrograms have shown 93% sensitivity for perforations. The most effective test is arthroscopy.
The treatment of scapholunate instability depends on
the stage of instability. Occult scapholunate instability
without kinematic abnormalities can be treated with arthroscopic or open dbridement, and a period of postoperative immobilization. Dynamic instability should be assessed arthroscopically. Incomplete tears are dbrided and
pinned followed by dorsal capsulodesis. Complete tears
require repair or reconstruction of the scapholunate ligament and dorsal capsulodesis. Some authors advocate
dbridement of the scaphoid and lunate and pinning to
promote healing.The more accepted treatment is open repair with sutures through drill holes or suture anchors.
Dorsal capsulodesis is recommended. The Blatt capsulodesis describes taking a proximally based capsuloligamentous flap and inserting it into a notch on the dorsal
scaphoid distal to the axis of rotation.A distally based flap
has been described, and a strip of dorsal intercarpal ligament has been used to control rotatory subluxation of the
scaphoid. More recently, ligamentous reconstruction of
the scapholunate interval with bone-ligament-bone constructs has been described with limited follow-up
(capitate-hamate and cuneonavicular). Some physicians
advocate the use of a Herbert bone screw to stabilize the
scapholunate joint, but still allow some rotation between
these bones. Intercarpal arthrodeses such as scaphotrape-

336

A sudden twisting injury or forced dorsiflexion of the

wrist can result in rupture of the lunotriquetral ligament. This injury can cause lunotriquetral instability,
which can manifest as a VISI deformity when associated
with damage to extrinsic wrist ligaments. This ligamentous injury can be imaged with PA and radial/ulnar deviation radiographs and magnetic resonance arthrography. Treatment options include acute primary repair,
reconstruction of the ligamentous structures with tendon or bone-ligament-bone construct, and dorsal capsulodesis for dynamic instability. Arthrodesis has been advocated, but nonunion can occur. Bone suture anchors
now provide excellent alternatives to creating bone tunnels for tendon-based ligamentous reconstruction.

Metacarpal and Phalangeal Fractures

Most metacarpal and phalangeal fractures can be
treated nonsurgically with closed reduction, short-term
splinting, and early protected range of motion. Indications for surgical treatment include displaced intraarticular fractures, unstable diaphyseal fractures, rotational deformity, open fractures, fractures with
associated tendon injuries, and multiple fractures (Figures 10 and 11).

Metacarpal Fractures
In general, metacarpal fractures have apex-dorsal angulation caused by the pull of the intrinsic muscles. Acceptable angulation in metacarpal diaphyseal fractures

American Academy of Orthopaedic Surgeons

Chapter 29 Wrist and Hand: Trauma

Orthopaedic Knowledge Update 8

Table 1 | Stages of Scapholunate Instability

Stage

Occult

Dynamic

Injured ligaments

Partial SLIL

Radiographs

Normal

Incompetent or torn
SLIL; partial palmar
extrinsics
Usually normal

Stress radiographs

Normal;
Abnormal fluoroscopy

Abnormal

Scapholunate
Dissociation
Complete SLIL, volar or
dorsal extrinsics
SL gap > 3 mm
SL angle > 70

Grossly abnormal

DISI

SLAC

Complete SLIL, volar

extrinsic; secondary
s: RL, ST
SL angle > 70
SL gap > 3 mm
RL angle > 16
CL angle < 15
Unnecessary

As in stage IV

I. Styloid DJD
II. RS DJD
III. CL DJD
IV. Pancarpal
Unnecessary

DISI, dorsal intercalated segment instability; SLAC, scapholunate advanced collapse; SL, scapholunate; RL, radiolunate; SLIL, scapholunate interosseous ligament; DJD, degenerative joint disease;
CL, capitolunate; RS, radioscaphoid; ST, scaphotrapezium
(Courtesy of S Wolfe, MD.)

increases from 15 to 20 in the index and middle

metacarpals to 40 to 50 in the ring and small metacarpals. This increase is a result of the increased mobility at the carpometacarpal joint of the fourth and fifth
metacarpals at the hamate compared with the more
fixed unit of the second and third metacarpals. At these
levels, rotational deformity will lead to scissoring of the
fingers. Long oblique and spiral diaphyseal fractures of
the index and small fingers can be unstable and may
need surgical treatment. Kirschner wire fixation minimizes soft-tissue damage and can be used for transverse,
longitudinal, or a combination of fixation patterns. Two
wires may be required to control rotation. Plate fixation
can be used dorsally with early motion to prevent adhesions. Short periods of splinting with early protected
motion will help maximize return to full function.

Phalangeal Fractures
Phalangeal fractures may have significant angulation resulting from the opposing pull of intrinsic and extrinsic
tendons across fracture sites. Intrinsic structures insert
relatively proximally and function as flexors. Extrinsic
structures insert relatively distally and function as extensors. Phalangeal fractures consequently tend toward
apex-volar angulation. Closed reduction is best accomplished by flexion of the distal fragment to match the
volar angulation of the proximal fragment. Stable closed
reduction may be limited by these forces, and internal
fixation with Kirschner wires or plating may be more reliable in ensuring acceptable bony alignment. Distal
phalanx tuft fractures frequently involve the nail bed,
and suture repair of the nail bed may be sufficient to reduce the bony fragments and allow a bony or fibrous
union. Alternatively, longitudinal pin fixation may be
used. Fibrous union of comminuted distal tuft fractures

American Academy of Orthopaedic Surgeons

Figure 9 Watsons scaphoid shift test. The thumb is placed on the volar scaphoid
tuberosity and pressure applied while moving the hand from ulnar to radial deviation.
The scaphoid will move from extension in ulnar deviation to flexion in radial deviation.
Palmar pressure on the scaphoid tubercle prevents scaphoid flexion and will produce
dorsal subluxation of the scaphoid in a patient with a torn or lax scapholunate ligament producing a painful snap. (Reproduced with permission from Watson HK, Weinzweig J: Intercarpal arthrodesis, in Green DP, Hotchkiss RN, Pederson WC (eds):
Greens Operative Hand Surgery. Philadelphia, PA, Churchill Livingstone, 1993, p 115.)

usually results in a stable, pain-free digit over time.

Flexor digitorum profundus-avulsion fractures can be
treated with open reduction and repaired over a button
or with suture anchors. Volar fractures of the base of the

337

Wrist and Hand: Trauma

Orthopaedic Knowledge Update 8

Figure 10 A, Intra-articular condylar fracture of the middle phalanx with 2 mm of displacement. B, Limited open reduction and screw fixation of intra-articular condyle fracture.
pezium. Screw fixation can be used if the fragment is
large enough and early motion can be started. Rolandos fracture, a Y-shaped intra-articular fracture of the
base of the thumb, can be treated with closed reduction
and Kirschner wires or open reduction and plate and/or
screw fixation using small implants. Accurate reduction
of the articular surface is important to prevent carpometacarpal arthritis.

Intra-articular Fractures

Figure 11 Scissoring of the fingers after index and middle metacarpal fractures prevents the patient from making a composite fist.
distal phalanx should be pinned or they will displace
with the pull of the flexor on the fragment.

Thumb Fractures
Fractures of the base of the thumb metacarpal require
anatomic reduction and fixation as needed to maintain
the reduction. Nondisplaced intra-articular fractures can
be treated in a thumb spica cast for 4 weeks. Bennetts
fracture-dislocations of the metacarpal from a bony
fragment attached to the volar beak ligament are usually unstable from the pull of the abductor pollicis longus on the metacarpal base (Figure 12). The reduction
maneuver is traction, pronation, and ulnar pressure over
the base of the metacarpal. Kirschner wires can be
placed across the fracture, into the adjacent index
metacarpal, or from the thumb metacarpal into the tra-

338

Proximal interphalangeal (PIP) joint fracturedislocations result from axial loading of the PIP joint in
hyperextension and are difficult to treat. Volar lip fractures involving less than 30% of the articular surface are
treated in extension blocking splints to maintain a congruent joint reduction while gradually increasing motion. The articular contour is less important than preserved range of motion in these fractures. Fractures
involving more than 30% of the joint surface frequently
are unstable and difficult to control. Fixation options include volar plate arthroplasty, open reduction, external
fixation, or dynamic traction (Figure 13). Use of osteochondral grafts from sources such as the distal dorsal
hamate bone has been described.

Ligament Injuries of the Hand

PIP joint injuries represent the most common ligamentous injuries in the hand. These range from jamming
the finger to irreducible dislocations and fracturedislocations. Dorsal dislocations are most common and
varying degrees of volar plate, collateral, and accessory
collateral ligaments are torn with the injury. The volar
plate, soft tissues, and fracture fragments can be interposed between joint surfaces, preventing a concentric
reduction. Stable motion is the goal; dynamic traction,
extension blocking splints or pins, open reduction, and

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 29 Wrist and Hand: Trauma

Figure 12 Methods of percutaneous fixation

of Bennetts fracture. Kirschner wire fixation
into the second metacarpal (A) or across the
fracture into the trapezium and base of the
second metacarpal after closed reduction (B).
(Reproduced with permission from Stern P:
Fractures of the metacarpals and phalanges,
in Green DP, Hotchkiss RN, Pederson WC
(eds): Greens Operative Hand Surgery.
Philadelphia, PA, Churchill Livingstone, 1993,
p 781.).
volar plate arthroplasty all are options allowing early
motion. Volar dislocations usually result from a rotatory
subluxation of the joint on one collateral ligament disrupting the extensor mechanism at the lateral band. The
condyle can be entrapped by the stout fibers of the lateral band preventing reduction of the joint. A true volar
dislocation can rupture the central slip, changing the
treatment to immobilization in extension with gradual
increases in flexion.
Metacarpophalangeal (MCP) joint collateral ligament injuries usually involve the radial collateral ligament and are more common in the index and small digits. Examining the finger in full flexion stretches the
collateral ligaments. If there is no subluxation, malrotation, or persistent lateral deviation, the injury can be
treated conservatively. The joint should be immobilized
in 30 of flexion for 3 weeks and then buddy taped to
the radial digit for an additional 2 to 3 weeks. Unstable
injuries require repair of the ligament with pull-out sutures or suture anchors. Persistent pain and instability
may require reconstruction of the radial collateral ligament.

Gamekeepers Thumb
Ulnar collateral ligament (UCL) injury in the thumb
MCP joint is known as gamekeepers thumb or skiers
thumb. The mechanism of acute injury is a radially directed force to the flexed MCP joint of the thumb. Testing the thumb in 30 of flexion isolates the proper collateral ligament, whereas positioning the thumb in
extension tests the volar plate and accessory collateral
ligament. The lack of a defined end point or 30 of laxity
greater than the contralateral thumb in both extension
and flexion suggests a complete tear of the ligament.

American Academy of Orthopaedic Surgeons

Figure 13 Diagram of Eatons method of volar plate arthroplasty using suture or wire
through the volar plate and a trough in the distal insertion site tied over a button. VP =
volar plate. (Reproduced with permission from Glickel SZ, Barron OA, Eaton RG: Dislocations and ligament injuries in the digits, in Green DP, Hotchkiss RN, Pederson WC
(eds): Greens Operative Hand Surgery. Philadelphia, PA, Churchill Livingstone, 1993, p
778.)

339

Orthopaedic Knowledge Update 8

Wrist and Hand: Trauma

ments and without cell migration from exogenous tissues. Tendon repair is clearly influenced by the
biomechanical environment. The application of mechanical forces can change the time course of improving
strength characteristics as well as change the load displacement curves over time. It is also clear that there is
no repair system that can match the strength and stiffness of native uninjured tendons. Advances in tendon
repair strive to improve both the strength and the stiffness characteristics at the repair site through innovations of multistrand methods, different configurations of
the strands, and recent innovations of nonsuture implants.

Anatomy
Figure 14 Radiographic views of a physeal gamekeepers avulsion fracture. Failure of
the physis occurs before failure of the collateral ligament.

UCL avulsions can be associated with a small bony fragment, which usually involves minimal articular surface.
Distal tears are five times more common than proximal
tears. Skeletally immature patients present with bony
avulsions more commonly than ligamentous avulsions
(Figure 14). The Stener lesion involves complete rupture
of the UCL distally with interposition of the adductor
aponeurosis. The aponeurosis prevents reapposition of
the UCL to the proximal phalanx. Thus, complete ruptures should be surgically repaired primarily, with pullout sutures or with suture anchors. Partial ruptures,
however, can be treated with 4 weeks of immobilization
followed by 2 weeks of gradual motion and a total of 3
months of protected activity. Differentiating between
partial and complete tears is important in choosing a
treatment plan. Ultrasound has been used with moderate reliability, whereas MRI provides the most accurate
diagnosis with up to 94% specificity for Stener lesions.
Thumb carpometacarpal dislocations involve rupture
of the volar beak ligament and radial collateral ligament. A PA stress radiograph of both thumbs pressed
together along their radial borders moves the proximal
metacarpal laterally and will show a shift of the metacarpal on the trapezium. The reduction must maintain
the metacarpal-trapezial relationship and casting or pinning can be used to hold the reduction. If the metacarpal is unstable, reconstruction of the ligament with
flexor carpi radialis or abductor pollicis brevis should be
considered.

Flexor Tendon Injuries

Rapid and reliable restoration of function is the goal of
flexor tendon surgery. Healing of injured or lacerated
tendons has been demonstrated on a biologic basis without significant input from the circulating blood and ele-

340

In the forearm, the flexor digitorum profundus muscle

and musculotendinous structures are dorsal to the flexor
digitorum superficialis. This relationship persists in the
hand where the flexor digitorum superficialis tendon
crosses the palm volar to the flexor digitorum profundus. At the level of the A1 pulley, flexor digitorum superficialis divides and moves dorsal to the flexor digitorum profundus and inserts on the proximal end of the
middle phalanx. The flexor digitorum profundus tendon
then continues to insert on the proximal end of the distal phalanx.
The course of the flexor tendon system in the hand
has been divided into zones used to describe the level of
injury and prognosis for repair (Figure 15). Zone I is
distal to the insertion of the flexor digitorum superficialis. In the thumb, zone TI is overlying and distal to the
interphalangeal joint. Zone II (no-mans land) starts at
the distal palmar crease and ends at the flexor digitorum superficialis insertion. Zone TII in the thumb is
between the MCP joint and the interphalangeal joint.
Zone III is the zone of lumbrical origin and extends
from the distal palmar crease to the distal edge of the
transverse carpal ligament. Zone TIII refers to the
flexor tendons to the thumb as they pass through the
thenar region of the hand. Zone IV is the carpal tunnel,
and zone V is the distal forearm. The pulley system in
the fingers and hand provides a biomechanical advantage for maximal tendon excursion and range of motion.
The vascular supply to the flexor tendons dorsally is via
the vincula, which contain vessels coursing from intraosseous vessels. The volar flexor tendon receives its
blood supply from small vessels in the peritenon.

Flexor Tendon Repair

Increasing the caliber of the suture and increasing the
number of strands in the core suture improves overall
tendon repair performance in terms of load to failure,
gap formation, and cyclic loading strength. At the same
time, increasing the number of strands of core suture
may increase the work of flexion. This suture method

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 29 Wrist and Hand: Trauma

Figure 15 Zones of injury on the flexor surface of the hand (A) and the extensor surface
of the hand (B). (Reproduced with permission
from Strickland JW: Flexor tendon: acute injuries, in Green DP, Hotchkiss RN, Pederson WC
(eds): Greens Operative Hand Surgery. Philadelphia, PA, Churchill Livingstone, 1993,
p 1856.)

may translate to a potential increase in adhesion formation in the biologic/clinical setting if efforts are not
made for earlier and more aggressive immobilization.
Recent studies comparing different suture methods and
number of suture strands conclude that a four-strand
modified Kessler technique provides a repair that is
strong enough to reliably allow immediate controlled
active flexion postoperatively (Figure 16).
The strength of a conventional core suture repair
can be doubled with the use of simple continuous running suture of 6-0 prolene, when the sutures are placed
within the substance of the tendon rather than just the
epitenon. Some physicians advocate the placement of
peripheral suture first, followed by core suture. This
method has been shown to decrease bunching and overlap of the tendon repair site.
Nonsuture techniques have been explored for flexor
tendon repair applications. Internal splint repairs have
been associated with good mechanical strength, but also
excessive work of flexion and tendon necrosis. An internal tendon implant consisting of a corkscrew-like device
called Teno-Fix (Ortheon Medical, Winter Park, FL) has
recently been approved for use in the United States.
This device is introduced into either end of the cut tendon and connected and tensioned by a stainless steel
woven cable and crimped beads. Early clinical trials, canine studies, and biomechanical testing suggest that this
Teno-Fix device may facilitate early active motion after
tendon repair.

Pulley Repair and Reconstruction

If critical pulleys are deficient, loss of flexor tendon
function results. The A2 and A4 pulleys are the most significant pulleys and should be preserved at all costs.

American Academy of Orthopaedic Surgeons

Figure 16 Modified Kessler suture technique with 4-0 monofilament nylon and 6-0
prolene deep epitendinous suture. (Reproduced with permission from Strickland JW:
Acute injuries, in Green DP, Hotchkiss RN, Pederson WC (eds: Greens Operative Hand
Surgery. Philadelphia, PA, Churchill Livingstone, 1993, p 1861.)

Step-cuts in the pulley make direct repair easier when

the pulley needs to be divided. An irreparable pulley
can be reconstructed with tendon grafts, retinacular tissues, or processed tissue grafts. A study examining flexor
digitorum profundus tendon repair compared work of
flexion when A2 pulleys were kept intact, incised, or enlarged. Gliding excursions were reduced in the intact
pulley group in comparison with the incised or enlarged
pulley groups, and work of flexion increased. This study
supports the notion of partial or complete pulley release
after tendon repair in improving overall gliding excursion. Preservation of A2 and A4 pulleys, however, pre-

341

Orthopaedic Knowledge Update 8

Wrist and Hand: Trauma

vents flexor bowstringing and maximizes end range of
motion.
An antiadhesion barrier gel has been used for prevention of adhesion in Europe for flexor tendon injuries. In one clinical study of zone II flexor tendon repairs, the active PIP joint motion was improved (89.5
compared with 65 in controls). Polyvinyl alcohol shields
have been used to decrease tendon adhesions but have
also shown an increase in repair site rupture.
5-fluorouracil is also being investigated to reduce peritendinous adhesions; adverse tendon healing effects
have not been shown.

Rehabilitation of Flexor Tendon Injuries

Many different protocols have been followed to protect
flexor tendon repair but prevent adhesions that ultimately affect motion. Immobilization is recommended
for children and patients with cognitive deficits. Passive
mobilization, a variant of early passive mobilization, and
an active place and hold model of mobilization have
been described. The goal, regardless of the protocol chosen, is early motion while protecting the repair sufficiently to allow healing. Current studies are aimed at
providing repairs that led to earlier active motion. A recent study used flexor tendon force transducers to show
that tenodesis motion requires less force than place and
hold motion. A tendon healing study found that breaking strength and cellular activity increased with both
early motion and tension. Therefore, current recommendations of four- to eight-strand repairs with a deep epitendinous running suture are directed at earlier active
motion.

Extensor Tendon Injuries

The extensor mechanism of the hand is a complex structure that when damaged requires recognition of injury,
appropriate repair, and early motion to preserve maximum function. More than 60% of injuries to the extensor mechanism are also associated with bony, joint, or
ligamentous injuries. Suture repair of extensor tendon
laceration should be with a modified Kessler or modified Bunnell repair using 4-0 synthetic nonabsorbable
suture. The extensor tendon excursion is less than the
flexor tendon excursion; therefore, preservation of
length and prevention of gapping is important to prevent loss of motion. The extensor mechanism, like the
flexor system, is divided into zones for discussion of injury and treatment (Figure 15). The even zones are overlying the bones (phalanges, metacarpals, and forearm
bones) and the odd zones are overlying the joints (distal
interphalangeal [DIP], PIP, and MCP joints, and wrist).

Mallet Finger
Disruption of the extensor tendon at the level of the
DIP joint as a tendinous or bony avulsion results in a

342

mallet finger deformity. This condition presents clinically as lack of active extension at the DIP joint. Closed
mallet fingers are best treated by continuous splinting in
extension for 8 weeks. This treatment is effective in patients with acute injury, and in patients in whom diagnosis and treatment are delayed for weeks or even months.
Active PIP joint motion is important and should be
maintained during the splinting period. Weaning from
the splint begins after 8 weeks, and night splints are recommended for an additional 4 to 8 weeks. The skin
should be monitored for maceration and necrosis; splinting changes may be necessary. Splints should hold the
DIP joint in slight hyperextension. A slight residual extensor lag (< 10) may result from closed treatment. Significant persistent extensor lags, open mallet injuries,
and large bony avulsions should be treated with open
repair; however, complications from surgical treatment
of mallet injuries approach 50%.

Boutonniere Deformity
Disruption of the central slip results in loss of extension
at the PIP joint and hyperextension of the DIP joint or
a boutonniere deformity. The primary injury is to the
central slip, the jammed finger. The clinical picture is a
result of subluxation of the lateral bands from an extensor position to a flexor position at the PIP joint. Treatment with splinting in extension at the PIP joint allows
for DIP joint motion and allows maintenance of normal
lateral band length and position.
Surgical indications for an acute boutonniere deformity include a displaced avulsion fracture at the base of
the middle phalanx, axial and lateral instability of the
PIP joint associated with loss of active or passive extension of the finger, or failed nonsurgical treatment. Passing a suture through the central tendon and securing it
to the middle phalanx (with or without the bony fragment) accomplishes primary repair of an avulsion fracture. Dorsal fixation of the lateral bands addresses the
soft-tissue boutonniere deformity, and transarticular pin
fixation of the PIP joint is often used to secure full extension. If primary repair is not possible, portions of the
lateral bands can be sutured together in the dorsal midline to reconstruct the central slip.

Zone V Injuries
The most common injury to the extensor mechanism in
zone V is from a fight bite sustained through a laceration from a human tooth when striking a blow to the
face. This injury occurs with the MCP joint in flexion.
Deeper injury to the extensor tendon will be proximal
to the skin wound when the fingers are extended. These
partial tendon injuries can be treated conservatively.
The skin wound, however, should be treated aggressively with thorough wound assessment, cultures, surgical dbridement, broad-spectrum intravenous antibiot-

American Academy of Orthopaedic Surgeons

Chapter 29 Wrist and Hand: Trauma

Orthopaedic Knowledge Update 8

ics, and initial splinting with gradual transition to
range-of-motion exercises. Devastating complications
including articular damage, septic arthritis, and osteomyelitis can arise from a human bite.
Rupture of a sagittal band can result from a blunt injury to the MCP joint. Most commonly, the radial sagittal band is damaged and the extensor tendon subluxates
into the ulnar gutter. This injury presents as lack of full
extension at the MCP joint. The tendon can be seen on
the ulnar aspect of the MCP joint and passively reduced
into a central position. Active flexion produces a painful
snap as the tendon dislodges into the ulnar gutter. Nonsurgical treatment of an acute injury can be attempted
with the finger splinted in extension for 6 weeks. If this
treatment fails or if the diagnosis is delayed, surgical repair is indicated. Primary repair reinforced with an
ulnar-based flap of extensor tendon is strong enough to
allow early active motion.
Closed rupture of the extensor pollicis longus following a distal radius fracture has become a well-recognized
entity. Attritional rupture secondary to irritation from
fracture fragments, ischemia, or hemorrhage within the
tendon sheath are commonly proposed theories of injury.
The patient will present several weeks after fracture with
loss of interphalangeal joint extension and weak adduction of the thumb. Nonsurgical treatment is not appropriate and direct surgical repair is often not possible. Free
tendon grafting with an intercalated graft or extensor indicis proprius-to-extensor pollicis longus transfer are two
common techniques for reconstruction and both are associated with good results.

Rehabilitation of Extensor Tendon Injuries

Static splinting has been the standard with good longterm results following extensor tendon repairs. Recently,
early active motion using dynamic splinting has shown
better functional results. Immobilization of tendons is
associated with adherence of the tendon to surrounding
soft tissues, producing loss of both extension and flexion. The immobilized tendon loses strength over time,
and aggressive rehabilitation following prolonged immobilization can further attenuate the repair. Controlled motion has been shown to increase tensile
strength of tendons, improve gliding properties, increase
repair-site DNA, and accelerate changes in the surrounding vascularity.
Dynamic extension splinting allows several millimeters of extensor gliding without placing stress across the
repair site. A splint holding the wrist in extension and
the MCP joints in 10 to 15 of flexion allows limited
motion at the interphalangeal joints both in active flexion and passive extension. Improved outcomes have
been shown using dynamic extension splinting for proximal injuries.

American Academy of Orthopaedic Surgeons

Nerve Injury and Repair

Injury
Peripheral nerve injuries can be classified by injury severity. Neurapraxia describes temporary failure of conduction in a nerve usually caused by contusion or
stretch in injury. Neurapraxic injuries heal without significant delay or sequelae. Axonotmesis describes a disruption of the nerve fibrils but an intact endoneurium
and epineurium. The intact neural tube functions as a
conduit to allow nerve healing and regeneration. Nerve
regeneration can be followed by an advancing Tinels
sign. A fibroblastic reaction within the nerve sheath can
impede nerve growth leading to a neuroma in continuity. These injuries can result from crush or severe stretch
mechanisms. A completely lacerated nerve is an example of neurotmesis in which the axons and endoneurial
tubes are disrupted. The proximal nerve ending undergoes retrograde degeneration to the next node of Ranvier where a disorganized tangle of regenerating axons
forms a neuroma. The distal nerve ending undergoes
wallerian degeneration, resulting in an end bulb glioma.
These injuries are best treated with surgical repair.

Repair
Nerve regeneration following repair is dependent on the
type of injury, location, degree of contamination, residual gap, tension, and many other factors currently being
investigated. Studies in animals have identified growth
factors such as fibroblast growth factor-1 and insulinlike growth factor that promote nerve regeneration and
improve outcomes. Different constructs such as reversed
vein entubulation, synthetic tubes, and fibrin glue are
being studied to replace nerve grafts. One animal study
in rat sciatic nerve found that muscle-enriched vein
graft potentiates Schwann cell proliferation and is a
promising alternative conduit for nerve grafting. Studies
using synthetic conduits have shown favorable functional outcomes for digital nerve repair with a gap of 4
mm or less.
Nerve repair techniques include epineurial suturing,
group fascicular repair, and individual fascicular repair.
Epineural repair and group fascicular repair are most
commonly used and the decision of which method to
use is dependent on the level of nerve injury, and the
appearance of and the ability to line up the fascicles. No
study has shown better results with one repair method
over the other, which may result from the difficulty in
appropriately matching the fascicles. Sensorimotor mapping can be used intraoperatively to maximize pairing of
motor and sensory fascicles. This method requires significant patient cooperation and meticulous repair by the
surgeon.
The external epineurium is a layer of connective tissue that can hold a suture better than internal epineurium and perineurium. Tension in the repair, however,

343

Orthopaedic Knowledge Update 8

Wrist and Hand: Trauma

must be avoided and nerve grafting should be considered if there is too much stretch to allow approximation
of the nerve endings. Early nerve repair of transected
nerves should be done within 10 to 14 days. Crushed
and contused nerves can be followed over time, and
early electromyograms (obtained 2 to 3 weeks after injury) can be compared with later studies (obtained after
6 weeks to 3 months) to identify areas of healing. If
there are no changes at 3 months, nerve exploration is
indicated. Waiting longer than approximately 18 months
leads to intraneural fibrosis and muscle fibrosis that
may not be reversible.

Rehabilitation and Results

Postoperative care involves splinting to protect the repair from stress for 2 to 3 weeks followed by gradual
range of motion across the repair site. Regeneration occurs at a rate of approximately 1 mm per day. Results of
nerve repair are better in the younger patient population, possibly because of more effective cortical reeducation. In patients younger than age 20 years, 75%
achieved two-point discrimination of less than 6 mm after digital nerve repair. In adults, only 25% to 50%
achieve that level of recovery. Median and ulnar nerve
recovery can improve for up to 5 years after repair. The
results of median and ulnar nerve repair are better for
lesions below the elbow than above. Studies have shown
that 29% to 37% of patients regain two-point discrimination less than 12 mm, and 14% to 27% regain functional motor activity. Sensory reeducation is now recognized as an important component in nerve recovery and
involves desensitization, motion, and tactile discrimination training. Sensory reeducation has been shown to be
more difficult when both the flexor tendon and nerve
are repaired. This situation suggests that fine motor control assists sensory interpretation in the fingers.

Fingertip Injuries
These common injuries require careful assessment and
treatment to maintain proper coverage and functionality
of the finger following injury. Appropriate padding and
sensibility must be preserved in the fingertip. Nail bed
injuries also need careful treatment to avoid painful and
cosmetically unappealing results.

Replantation and Local Flaps

The viability of replanted digits and hands has been
shown over many microsurgical centers to be approximately 80%. Revascularized parts have higher success
rates because some venous outflow is usually preserved.
Replantation should be considered when the amputated
part is a thumb, multiple digits, partial hand (through
the palm), wrist, forearm, elbow, sharp amputation
above the elbow, and individual digits distal to the
flexor digitorum superficialis insertion, and almost any

344

part in a child. Other factors impacting the decision to

replant a part are ischemic time, patient age, crush versus sharp injury, and segmental injuries. Warm ischemia
time should be less than 6 hours; in digits where there is
no muscle, warm ischemia time can be up to 12 hours.
The amputated part should either be immersed in saline
in a bag or sterile container, or wrapped in salinesoaked gauze and then placed on ice. Replantation with
bony stabilization should proceed first, followed by tendon repair, arterial anastomosis, venous anastomosis,
nerve repair, and finally careful skin approximation or
closure, avoiding tension across neurovascular structures.
Soft-tissue coverage of defects associated with amputations depends on the size of the defect, the required
sensibility of that area, and other functional needs. Cutaneous flaps such as a groin flap are useful for covering
moderate soft-tissue defects of the hand. The radial
forearm flap is a fasciocutaneous flap that can be used
either as a pedicled flap or a free flap. Myocutaneous
flaps using the rectus abdominis, latissimus dorsi, and
gracilis muscles can be taken to provide larger softtissue coverage options. In these flaps, the associated
nerve can be taken and the flap used for reinnervation
and replacement of lost motor function in the extremity.

Infections
Infections in the hand are challenging to treat because
of the anatomic spaces in the hand, which can allow organisms to spread quickly from distal points of injury to
proximal locations. The areas of the hand that facilitate
the spread of infection are the dorsal cutaneous space,
thenar and midpalmar spaces, Paronas space, the interdigital web spaces, the tendon sheaths, and articular
spaces.

Bacterial Infections
The most common bacterial pathogen in hand infections
is Staphylococcus aureus. Vancomycin-resistant S aureus
is increasingly prevalent in community-acquired infections. Many infections begin as cellulitis, which, if recognized early, can be treated with antibiotics, elevation,
and immobilization. The fundamental principles of
treatment are appropriate antibiotic therapy, adequate
dbridement and drainage, a period of immobilization
and elevation, and early remobilization. Tetanus immunization needs to be up to date. Untreated cellulitis may
develop into an abscess, which must be drained before
antibiotics will be effective.
A paronychia is an infection involving the nail bed
that evolves into an abscess that is easily drained by lifting the paronychial skin off the nail plate. Collar button
abscesses are located in the webspace and palmar abscesses usually involve the thenar or hypothenar bursas.
Infected wounds overlying the dorsal MCP joint are of-

American Academy of Orthopaedic Surgeons

Chapter 29 Wrist and Hand: Trauma

Orthopaedic Knowledge Update 8

ten a result of contact with teeth or an open mouth.
These wounds may involve the extensor mechanism
proximal to the wound. Evaluation of the extensor with
the hand in a fully flexed fist position may reveal partial
tendon laceration and allows appropriate dbridement
of the involved extensor mechanism. A felon is a fingertip pulp infection that requires surgical drainage to prevent a compartment syndrome and potential necrosis of
the entire pulp of the finger.
Purulent flexor tenosynovitis warrants early attention because it is a potentially devastating infection with
significant long-term morbidity. This infection involves a
penetrating injury or hematogenous seeding inside the
flexor sheath. Because the sheath extends across the
palm, the infection can easily track into the forearm. Patients with this type of infection will present with Kanavels signs of a digit resting in flexion, pain with passive
extension, pain along the flexor sheath, and fusiform
swelling in the digit. The most sensitive sign is pain with
passive extension. Flexor tenosynovitis can be treated
with irrigation of the flexor sheath once it has been
opened. A preferred method of irrigation is to place a
16-gauge catheter in the proximal sheath and a penrose
drain in the distal sheath in the finger. Every 2 hours, 50
mL of sterile saline is irrigated through the sheath for
48 hours. Motion should begin within the week to prevent excessive stiffness.
Antibiotic coverage of hand infections should be tailored to the cultured organism. As mentioned, the most
common pathogen is Staphylococcus but human bite
wound infections often contain Eikenella corrodens and
virulent cat bite infections are caused by Pasteurella
multocida. Penicillin is the drug of choice for these bite
infections.
Necrotizing fasciitis is a severe, rapidly spreading infection that carries mortality rates of 8% to 73%. There
are two groups of organisms responsible for this infection. Group one includes anaerobic bacteria such as Enterobacter and non-group A streptococci. Group two
includes group A streptococci with S aureus or Staphylococcus epidermidis. Patients may present with wounds
from minor trauma that progress rapidly from cellulitis
and low-grade fever to multisystem organ failure, coma,
and hemodynamic compromise. Patients in the intensive
care unit are at increased risk of infection by more unusual pathogens such as Pseudomonas and Acinetobacter. These patients require multiple antimicrobial
agent therapy because of multiagent resistant organisms.

Fungal, Mycobacterial, and Viral Infections

Fungal infections in the hand occur in the paronychium,
nail, and skin. Chronic paronychial infections are usually
caused by Candida albicans. Onychomycosis, or fungal
nail bed infection, is commonly caused by dermato-

American Academy of Orthopaedic Surgeons

phytes such as Trichophyton rubrum. Superficial skin infections are caused by dermatophytes and Candida.
These relatively superficial nail and skin infections are
treated with topical antifungal agents, with conversion
to oral agents when treatment appears ineffective. Granulomatous or mycobacterial infections show a predilection for synovium. They may present in a manner
similar to rheumatoid arthritis and frequently involve
the wrist. Cultures must be directed toward detecting
these slow-growing pathogens. Mycobacterium tuberculosis, M avium-intracellulare, and M marinum have been
described in the hand and may be resistant to multiple
chemotherapeutic agents. Herpetic whitlow presents as
a clear vesicular lesion on the tip of the finger caused by
herpes simplex virus type 1 or type 2. Treatment is nonsurgical and the infection is self-limiting and resolved
over 3 to 4 weeks.

Burns
Burns over a large surface area are devastating injuries,
resulting in months of hospitalization, multiple procedures, and prolonged wound care. Over 2 million burn injuries are reported to require medical care each year, and
approximately 6% result in death, usually from smoke inhalation. Most burn injuries are relatively minor and patients are discharged following outpatient treatment. Of
those patients who require hospitalization, approximately
20,000 are admitted directly or by referral to hospitals
with special capabilities in the treatment of burn injuries.
The average hospital stay is 2 months. Criteria for referral
to a burn unit are shown in Table 2.
Burns are classified according to the depth of thermal injury, which in turn guides prognosis and treatment
plans. The trend is toward classifying burns as partial
thickness, which heal on their own, and full thickness,
which require skin grafting.
Partial-thickness burns include first- and seconddegree burns. First-degree burns only involve the epidermis and require removal of the source of injury and
analgesic care. The skin heals within 1 week and there is
no permanent damage.
Second-degree burns destroy the epidermis and involve varying amounts of the dermis. The more superficial second-degree burns heal within 2 weeks and leave
little permanent scarring. This injury is usually accompanied by blisters that progress over time. Deep seconddegree burns produce more scarring, which increases
the time to healing. Healing depends on residual epithelial cells in deep dermal sweat glands and hair follicles.
Hypertrophic scarring can lead to prolonged healing
and infections so that excision and skin grafting may be
considered.
Third-degree or full-thickness burns involve all of
the dermis and varying amounts of underlying fat, muscle, and bone. The skin may appear white and waxy and

345

Wrist and Hand: Trauma

Table 2 | Burn Unit Referral Criteria: American Burn

Association 2000
Burn Injuries That Should Be Referred to a Burn Unit:
Partial-thickness burns greater than 10% total body surface area
Burns that involve the face, hands, feet, genitalia, perineum, or major
joints
Third-degree burns in any age group
Electrical burns, including lightning injury
Chemical burns
Inhalation injury
Burn injury in patients with preexisting medical disorders that could
complicate management, prolong recovery, or affect mortality
Any patients with burns and concomitant trauma (such as fractures) in
which the burn injury poses the greatest risk of morbidity or mortality.
In such cases, if the trauma poses the greater immediate risk, the
patient may be initially stabilized in a trauma center before being
transferred to a burn unit. Physician judgment will be necessary in such
situations and should be in concert with the regional medical control
plan and triage protocols.
Burned children in hospitals without qualified personnel or equipment
for the care of children
Burn injury in patients who will require special social, emotional, or
long-term rehabilitative intervention.

the severity of the burn can be underestimated. The skin

is without sensation and capillary refill and has a leathery appearance.
The pathology of burns has been divided into zones
of injury. The zone of coagulation describes the most
damaged tissue with coagulated vessels and no blood
supply. The zone of stasis has sluggish blood flow but is
impacted by early removal of the thermal or chemical
source and appropriate intervention. The zone of hyperemia produces an inflammatory response to the injury
via bradykinins and histamine producing capillary leaking, edema, and third-spacing. This can further decrease perfusion to the zone of stasis.
Rapid loss of intravascular fluid and protein occurs
during the first 6 hours after a burn. By 36 to 48 hours,
capillary integrity is restored but tissue edema continues
as a result of protein loss into the burned area and systemic hypoproteinemia. Fluid management, respiratory
care, and electrolyte monitoring are essential initial
steps in caring for severe burn injuries. Wound care can
then be addressed and many options exist for managing
the different types of burn wounds.
Antibiotic coverage is important because the skins
immunoprotective function is lost when burned. Silver
sulfadiazide is the most common topical agent for burn
wounds but can produce a transient leukopenia when
used for large burns. Bacitracin, mafenide, betadine, and
gentamicin ointments are also used. Skin substitutes are
also available, such as split-thickness porcine xenografts
and synthetic grafts. Dbridement and autogenous split-

346

Orthopaedic Knowledge Update 8

thickness grafting is the most desirable treatment for
wounds needing coverage, but large volume burns may
not leave enough autograft skin.

Injection Injuries
High-pressure injection of material into the hand can
produce significant tissue necrosis, edema, and even
compartment syndrome. Injection forces have been reported to be between 3,000 and 12,000 psi. The quantity
of material injected is difficult to ascertain, but radiographs are sometimes helpful in identifying how far the
material has spread. Injected fluids follow bursal planes
and tendon sheaths and create both toxic and inflammatory damage to the tissues around them. Negative prognostic factors are presentation for treatment more than
10 hours after injury, injection pressures greater than
7,000 psi, and injection with oil paint.
The initial presentation may be less impressive than
anticipated, but over several hours, the swelling, pain,
and loss of function can significantly change, requiring
prompt action. Amputation rates have decreased from
48% to 16% because of early irrigation and wide dbridement. Antibiotic coverage of Gram positive, Gram
negative, and anaerobic organisms is important, and a
tetanus shot must be administered if the patients immunizations are not up to date. Close monitoring and repeat dbridements may be needed.

Vascular Injuries
Acute vascular injuries to the hand can occur from
blunt or penetrating trauma. Patients can present with
gross ischemia, progressive hematoma, acute or delayed
thrombosis, compartment syndrome, aneurysm development, and distal embolization. The degree to which collateral supply can compensate for the injury is dependent on surrounding tissue damage, vasomotor control,
and systemic disease. Acute arterial injuries that present
with gross hypoperfusion are treated as emergencies
and interventions to return blood flow to the affected
part are urgently undertaken. Arterial reconstruction
should be considered for critical arterial injury with impending cell death and for noncritical arterial injuries
when there is concern for adequate collateral supply, associated nerve injury, and extensive soft-tissue injury.
When the artery has sustained extensive damage, vein
grafts can be used to replace the damaged area and do
not compromise the ultimate reperfusion results. After
perfusion is restored, a compartment syndrome may develop and close observation for signs of increasing compartment pressures is essential.

Annotated Bibliography
Distal Radius Fractures
Boyer MI, Galatz LM, Borrelli J Jr, Axelrod TS, Ricci
WM: Intra-articular fractures of the upper extremity:

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

New concepts in surgical treatment. Instr Course Lect
2003;52:591-605.
New techniques of internal fixation, postoperative rehabilitation, and emphasis on functional as well as radiographic
outcome have refined the surgical treatment of complex fractures of the glenoid, humeral head, supracondylar and intracondylar humerus, olecranon, radial head, distal radius, and
distal radioulnar joint over the past decade. Early stabilization
and rehabilitation of these injuries leads to soft-tissue stabilization and facilitates the patients ability to place the hand in
three-dimensional space.

Hanel DP, Jones MD, Trumble TE: Wrist fractures.

Orthop Clin North Am 2002;33:35-57.

Chapter 29 Wrist and Hand: Trauma

central placement of the screw demonstrated 43% greater
stiffness, 113% greater load at 2 mm of displacement, and 39%
greater load at failure. Central placement of the screw in the
proximal fragment of the scaphoid offers a biomechanical advantage in the internal fixation of an osteotomy of the
scaphoid waist. Clinical efforts and techniques that facilitate
central placement of the screw in the fixation of fractures of
the scaphoid waist should be encouraged.

Polsky MB, Kozin SH, Porter ST, Thoder JJ: Scaphoid

fractures: Dorsal versus volar approach. Orthopedics
2002;25:817-819.

The benefits of a well-reduced and well-healed wrist fracture are predictable. After either closed or open reduction, the
integrity of the volar ulnar corner of the radius, articular stepoff, metaphyseal comminution, and DRUJ stability should be
assessed. Reconstruction of the subluxated or dislocated
DRUJ starts with the reduction of the radius, frequently obviating the need to address fractures involving the ulnar head
and styloid. Most importantly, the results of treatment reflect
surgical decision over the fixation method.

Twenty-six patients with scaphoid fractures were treated

with internal fixation using a cannulated differential pitch
compression screw. Sixteen patients underwent a dorsal approach (group 1) and 10 patients a volar approach (group 2).
Average time from injury to surgery was 6.6 months (range,
0.3 to 19 months) for group 1 and 8.3 months (range, 0.3 to 24
months) for group 2. The rate of union, determined by radiographs and clinical examination, was 81% in group 1 and 80%
in group 2. No significant differences were noted between the
groups for dorsiflexion/palmar flexion, radial deviation, grip
strength, and pain level.

Herzberg G, Forissier D: Acute dorsal trans-scaphoid

perilunate fracture-dislocations: Medium-term results.
J Hand Surg Br 2002;27:498-502.

Skoff HD: Postfracture extensor pollicis longus tenosynovitis and tendon rupture: a scientific study and personal series. Am J Orthop 2003;32:245-247.

The purpose of this study was to investigate the mediumterm results (mean follow-up, 8 years) of a series of 14 transscaphoid dorsal perilunate fracture-dislocations treated surgically at an average of 6 days following injury. Eleven patients
underwent open reduction and internal fixation through a dorsal approach. Combined palmar and dorsal approaches were
used in three fractures, open reduction and internal fixation in
two, and proximal row carpectomy in one. The Mayo Wrist
Score revealed five excellent, three good, five fair, and one
poor result. The average score was 79% (range, 55% to 95%).
All internally fixed scaphoids healed and no lunate or
scaphoid fragment osteonecrosis with collapse was observed.
Carpal alignment was satisfactory in most patients. Posttraumatic radiologic midcarpal and/or radiocarpal arthritis were
almost always observed at follow-up, but this did not correlate
with the Mayo Wrist Score.

McCallister WV, Knight J, Kaliappan R, Trumble TE:

Central placement of the screw in simulated fractures of
the scaphoid waist: A biomechanical study. J Bone Joint
Surg Am 2003;85:72-77.
Recent reports on internal fixation of acute fractures of
the scaphoid waist have demonstrated higher rates of central
placement of the screw when cannulated screws were used
than when noncannulated screws were used. This cadaveric
study was designed to determine whether central placement in
the proximal fragment of the scaphoid offers a biomechanical
advantage. Central placement of the screw in the proximal
fragment of the scaphoid had superior results compared with
those using eccentric positioning of the screw. Fixation with

American Academy of Orthopaedic Surgeons

A review of treatment of 200 consecutive patients with

distal radius fractures found that the incidence of rupture of
the extensor pollicis longus tendon is 3%. Diagnosis is based
on persistent dorsal wrist pain and a positive retroflexion sign.
In the prerupture setting, recommended treatments include a
third dorsal compartment release with or without an extensor
retinacular patch graft. If after an acute rupture primary repair
is not possible, a palmaris longus graft or a transfer from the
extensor indicis proprius to the extensor pollicis longus tendon
can be used in the subacute or chronic setting. Results of all
treatments seem to be clinically satisfactory.

Slade JF III, Gutow AP, Geissler WB: Percutaneous internal fixation of scaphoid fractures via an arthroscopically assisted dorsal approach. J Bone Joint Surg Am
2002;84(suppl 2):21-36.
In a consecutive series of 27 fractures (17 waist fractures
and 10 proximal pole fractures) treated with arthroscopically
assisted dorsal percutaneous fixation, CT confirmed 100%
union at an average of 12 weeks. Eighteen fractures were
treated within 1 month after the injury, and nine were treated
more than 1 month after the injury. In this series, the fractures
that were treated early (less than 1 month after the injury)
healed more quickly than those treated later.

Wigderowitz CA, Cunningham T, Rowley DI, Mole PA,

Paterson CR: Peripheral bone mineral density in patients with distal radial fractures. J Bone Joint Surg Br
2003;85:423-425.

347

Wrist and Hand: Trauma

Fractures of the distal forearm are widely regarded as the
result of fragility. This study examines the extent to which
patients with Colles fractures have osteopenia. Bone mineral
density was measured in the contralateral radius of 235
women presenting with Colles fractures over a period of 2
years. Although women of all ages had low values for ultradistal bone mineral density, the age-matched values were particularly low among premenopausal women age 45 years or
younger. This finding did not occur because of the presence of
women with early menopause. This large survey confirms and
extends the findings from earlier small studies. The authors
conclude that it is particularly important to investigate young
patients with fractures of the distal forearm to identify those
with osteoporosis, to seek an underlying cause, and to consider
treatment.

Flexor Tendon Injuries

Angeles JG, Heminger H, Mass DP: Comparative biomechanical performances of 4-strand core suture repairs
for zone II flexor tendon lacerations. J Hand Surg [Am]
2002;27:508-517.
This study compared four-strand core suture repairs, the
modified Becker, modified double Tsuge, Lee, locked cruciate,
Robertson, and Strickland suture repairs. Work of flexion and
ultimate tensile strength were compared, as well as cyclic loading. The greatest interference for gliding was in the modified
Becker repair, and the least in the modified double Tsuge repair. None of these repairs had mean gaps after 1,000 cycles to
a 3.9 N pulp pinch load approach the clinically important limit
of 3 mm of gap. Ultimate tensile strength was highest in the
modified Becker (69.4 8.2N) compared to the modified double Tsuge (60.3 15.3N) and locked cruciate (64.1 16.2N),
but this was not statistically significant. These authors stated
that ease of performance for the surgeon and less interference
with tendon gliding favored the locked cruciate and modified
double Tsuge repairs compared with the modified Becker repair.

Boyer MI, Strickland JW, Engles D, Sachar K, Leversedge FJ: Flexor tendon repair and rehabilitation:
State of the art in 2002. Instr Course Lect 2003;52:137161.
The application of modern multistrand suture repair techniques as well as postoperative rehabilitation protocols emphasizing the application of intrasynovial repair site excursion
has led to a protocol for treatment of intrasynovial flexor tendon lacerations emphasizing a strong initial repair followed by
the application of postoperative passive motion rehabilitation.
Protocols for the reconstruction of failed initial treatment
have likewise undergone modification given new findings on
the biologic and clinical behavior of flexor tendon grafts. Currently accepted treatment protocols following flexor tendon
repair and reconstruction are based on current clinical and scientific data.

Orthopaedic Knowledge Update 8

digitorum superficialis after repair of both flexor tendons in zone II: A biomechanical study. J Bone Joint
Surg Am 2002;84:2039-2045.
The effects of two strategies to improve postoperative
gliding in a human cadaveric hand were studied. Complete lacerations and repairs were made to the profundus and superficialis tendons at a location where both repair sites would pass
beneath the A2 pulley with the proximal interphalangeal joint
in 45 of flexion. Pulley plasty and resection of one slip of the
flexor digitorum superficialis tendon both significantly decreased gliding resistance compared with repair of both slips
(P < 0.001). There was no difference in the mean gliding resistance between the pulley plasty and one-slip resection groups.
The flexor digitorum superficialis slip was stronger after repair
with a Becker suture than after repair with a modified Kessler
or a zigzag suture. Both pulley plasty and resection of one slip
of the flexor digitorum superficialis reduce gliding resistance
after tendon repair in zone II of the hand.

Zhao C, Amadio PC, Zobitz ME, Momose T, Couvreur

P, An KN: Effect of synergistic motion on flexor digitorum profundus tendon excursion. Clin Orthop 2002;
396:223-230.
The dog model was used to evaluate synergistic motion on
the flexor digitorum profundus. Eighty percent of lacerations
were created and repaired using 4-0 Ticron modified Kessler
core suture with circumferential epitenon running 6-0 nylon
suture repairs. Metal markers were placed and then divided
into two experimental groups. One group had passive flexion
and extension of the digits with the wrist fixed in 45 of flexion; in the second group, dogs received synergistic wrist motion with passive digit flexion combined with wrist extension,
and passive digit extension with wrist flexion. Time points
were 1, 3, and 6 weeks. At 1, 3, and 6 weeks after surgery, the
synergistic motion group was superior in total excursion, extension excursion, flexion excursio, and overall percentage of
motion compared with unoperated controls.

Extensor Tendon Injuries

Chester DL, Beale S, Beveridge L, Nancarrow JD, Titley
OG: A prospective, controlled, randomized trial comparing early active extension with passive extension using a
dynamic splint in the rehabilitation of repaired extensor
tendons. J Hand Surg Br 2002;27:283-288.
Two methods of rehabilitation were compared after extensor tendon repairs in zones IV through VIII. Nineteen patients
followed an early active mobilization program and 17 patients
followed a dynamic splinting regimen. Data were collected at
4 weeks and at final follow up (median = 3 months). Extension
lag, flexion deficit, and total active motion were measured. At
4 weeks, the patients on a dynamic splinting program had better total active motion; however, there were no significant differences in the two groups at final follow-up.

Nerve Injury and Repair

Paillard PJ, Amadio PC, Zhao C, Zobitz ME, An KN:
Pulley plasty versus resection of one slip of the flexor

348

Bell Krotoski JA: Flexor tendon and peripheral nerve

repair. Hand Surg 2002;7:83-109.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

When peripheral nerve injury is combined with flexor tendon injury, sensibility is directly impaired. There is a loss in the
sense of finger or thumb position, pain, temperature, and
touch/pressure recognition, in addition to the tendon injury.

Day CS, Buranapanitkit B, Riano FA, et al: Insulin

growth factor-1 decreases muscle atrophy following denervation. Microsurgery 2002;22:144-151.
The purpose of this study was to evaluate the histologic,
immunohistochemical, and electrophysiologic differences between normal, denervated, and insulin-like growth factor-1
denervated muscle over an 8-week period. Denervated mice
gastrocnemius muscles demonstrated a decrease in muscle
weight, a decrease in myofiber diameter, an absence of muscle
regeneration, an early increase in the number of neuromuscular junctions, and a decrease in fast-twitch and maximum tetanic strength compared with normal muscle up to 8 weeks following denervation. Insulin-like growth factor-1 denervated
muscle, on the other hand, sustained muscle diameter and
muscle weight, maintained a smaller number of neuromuscular junctions, and relatively sustained fast-twitch and maximum tetanic strength compared with normal muscle over 8
weeks. These data suggest that insulin-like growth factor-1
may help prevent muscle atrophy and secondary functional
compromise after denervation.

Classic Bibliography
Allen CH: Functional results of primary nerve repair.
Hand Clin 2000;16:67-72.
Diao E, Hariharan JS, Soejima O, Lotz J: Effect of peripheral suture depth on strength of tendon repairs.
J Hand Surg Am 1996;21:234-239.
Dunning CE, Lindsay CS, Bicknell RT, Patterson SD,
Johnson JA, King GJ: Supplemental pinning improves
the stability of external fixation in distal radius fractures
during simulated finger and forearm motion. J Hand
Surg Am 1999;24:992-1000.

American Academy of Orthopaedic Surgeons

Chapter 29 Wrist and Hand: Trauma

Garberman SF, Diao E, Peimer CA: Mallet finger: Results of early versus delayed closed treatment. J Hand
Surg Am 1994;19:850-852.
Geissler WB, Freeland AE, et al: Intracarpal soft-tissue
lesions associated with an intra-articular fracture of the
distal end of the radius. J Bone Joint Surg Am 1996;78:
357-365.
Gelberman RH, Boyer MI, Brodt MD, Winters SC, Silva
MJ: The effect of gap formation at the repair site on the
strength and excursion of intrasynovial flexor tendons:
An experimental study on the early stages of tendonhealing in dogs. J Bone Joint Surg Am 1999;81:975-982.
Lotz JC, Hariharan JS, Diao E: Analytic model to predict the strength of tendon repairs. J Orthop Res 1998;
16:399-405.
Okafor B, Mbubaegbu C: Mallet deformity of the finger:
Five-year follow-up of conservative treatment. J Bone
Joint Surg Br 1997;79:544-547.
Rockwell WB, Butler PN, et al: Extensor tendon: Anatomy, injury, and reconstruction. Plast Reconstr Surg
2000;106:1592-1603.
Soejima O, Diao E, Lotz JC, Hariharan JS: Comparative
mechanical analysis of dorsal versus palmar placement
of core suture for flexor tendon repairs. J Hand Surg
Am 1995;20:801-807.
Strickland JW: Development of flexor tendon surgery:
Twenty-five years of progress. J Hand Surg Am 2000;25:
214-235.
Toby EB, Butler TE, et al: A comparison of fixation
screws for the scaphoid during application of cyclical
bending loads. J Bone Joint Surg Am 1997;79:1190-1197.
Wolfe SW, Austin G, et al: A biomechanical comparison
of different wrist external fixators with and without
K-wire augmentation. J Hand Surg Am 1999;24:516-524.

349

Chapter

30

Wrist and Hand Reconstruction

Debra M. Parisi, MD
Thomas E. Trumble, MD

Wrist Imaging
The wrist is a relatively small joint in which multiple
bony articulations and soft-tissue structures act synergistically to provide motion while maintaining necessary
stability. Several different imaging modalities are available to assist the physician in the diagnosis of wrist pain
and/or dysfunction. The interpretation of diagnostic images is often dependent on the technique or skill of the
clinician evaluating the results.
Plain radiographs supplemented with stress radiographs should be the first imaging examinations ordered
when attempting to determine wrist pathology. Supplemental studies (such as CT, MRI, bone scan, and arthrography) may be used for further assessment of wrist
pathology (Table 1).
Prior to the widespread use of diagnostic wrist arthroscopy, triple injection wrist arthrography was the
gold standard for diagnosing carpal instability. The radiocarpal, midcarpal, and carpometacarpal (CMC) joints
are injected sequentially with radiopaque dye. After
each injection, the wrist is brought through a range of
motion and then a static or fluoroscopic image is obtained to ensure that the dye has not extravasated into
another compartment. If dye has traversed from one
compartment to another, the integrity of one or more of
the intraosseous wrist ligaments has been compromised.
A magnetic resonance arthrogram is frequently used in
conjunction with fluoroscopic arthrography to improve
the diagnostic accuracy of ligament injuries.

Wrist Arthroscopy
Eleven access portals are currently used to access the
entire wrist joint. Radiocarpal portals include the 3-4,
4-5, 6R, 6U, and 1-2 portals. Midcarpal portals include
the midcarpal radial, midcarpal ulnar, triquetral hamate,
and triscaphe portals. Distal radioulnar portals include
the proximal and distal radioulnar joint (DRUJ) portals.
Arthroscopy of the wrist is the most accurate and
specific method of diagnosing mechanical wrist pathology. Direct visualization of ligament disruption, abnormal motion, and articular pathology can be directly ob-

American Academy of Orthopaedic Surgeons

served during a diagnostic arthroscopic procedure.

Arthroscopy also is useful for diagnosing partial and
complete tears of the triangular fibrocartilage complex
(TFCC) and chondral lesions.
Arthroscopy also has important therapeutic roles.
Arthroscopic dbridement of the TFCC or excision of
the distal ulna can be combined with diagnostic arthroscopy. The radial styloid also may be resected arthroscopically instead of with an open procedure. Under arthroscopic visualization, the styloid may be resected until
the radial attachment of the radioscapholunate ligament
is reached. TFCC pathology may also be addressed arthroscopically.
Treatment of a symptomatic wrist ganglion may be
achieved by arthroscopic exploration of the wrist before
arthroscopic or open resection is performed. The wrist
scope is placed in the 4-5 or 6R portal and the camera is
then directed toward the presumed ganglion. Alternatively, the scope can be placed in the 3-4 portal through
the ganglion. If the stalk of the ganglion is visualized,
the surgeon can use a second portal to remove it along
with a window of capsule using the arthroscopic shaver,
or it may be marked and treated with an open approach. The recurrence rate using the arthroscopic technique may be less than with open techniques.
Wrist arthroscopy also is being used to facilitate the
reduction of intra-articular distal radius fractures. In a
1999 study, fractures treated with open reduction and
plate and screw fixation or external fixation and percutaneous Kirschner wires were compared with fractures
treated with arthroscopically guided reduction, percutaneous Kirschner wires, and external fixation. The
strength, range of motion, and radiographic appearance
were significantly better in patients in the arthroscopically assisted group after a mean follow-up of 31
months. Arthroscopic evaluation of intra-articular distal
radius fractures also has the advantage of providing visualization of the surrounding soft-tissue structures
(such as the intracapsular ligaments and the TFCC complex), which facilitates the diagnosis of injuries to these
structures and helps in directing additional treatment if
needed.

351

Orthopaedic Knowledge Update 8

Wrist and Hand Reconstruction

Wrist Arthritis Secondary to Trauma

Table 1 | Wrist Imaging

Study

Demonstrates

CT
MRI

Bony anatomy
Soft-tissue integrity,
bone viability/
vascularity

Bone scan

Bone turnover activity

Wrist arthrography

Ligament and cartilage

integrity

Recommended to
Evaluate for:
Fracture, tumor
Osteonecrosis, TFCC
tears, ulnar collateral
ligament injuries,
occult fractures
(scaphoid), tumor
Osteonecrosis,
infection
Ligament injury, TFCC
tears

Malunion of Distal Radius Fractures

Malunion following a displaced fracture of the distal
metaphysis of the radius impairs the normal function of
the radiocarpal joint and the DRUJ. The deformity of a
malunited distal radius fracture may result in decreased
grip strength and range of motion (including flexionextension, ulnar-radial deviation, and pronationsupination) and an unacceptable cosmetic deformity.
Shortening of the radius relative to the distal end of the
ulna, in combination with a rotational deformity of the
distal fragment, often results in incongruence and/or instability of the DRUJ.
Patients with a malunited distal radius fracture often
report pain and loss of function. Younger patients tend
to be more symptomatic with smaller degrees of deformity than older, less active patients. The pain and dysfunction often are related to the prominent distal ulna.
A distal ulna resection (the Darrach procedure) has
been advocated in the past to treat the deformity and
abutment. Currently, the preferred treatment of a symptomatic distal radius malunion is corrective osteotomy
to restore the anatomy of the distal radius. If the DRUJ
cannot be salvaged, a distal ulna resection with or without a closing wedge osteotomy may be indicated.
Criteria for treating distal radius nonunions with a
corrective osteotomy with tricortical bone graft include:
(1) a loss of radial height of greater than 4 to 5 mm;
(2) a loss of 10 or more or radial inclination; and/or
(3) a reversal of the palmar tilt to 15 or more of dorsal
angulation or an increase in palmar tilt to 25 or more.
A contraindication to corrective osteotomies includes
the development of significant traumatic arthritis. The
goal of surgery is to correct as much of the deformity as
possible in all three planes (Figure 1). Careful preoperative planning using templates made from high-quality
radiographs is essential.

352

Scapholunate Advanced Collapse

Scapholunate advanced collapse (SLAC) is the most
common form of wrist arthritis. It is the predictable outcome of an untreated injury to the scapholunate interosseous and palmar radioscaphoid ligaments. When
these ligaments are disrupted, there is a shift in the
pressure distribution among the carpal bones during
motion and loading. The resulting degenerative changes
advance in a predictable fashion and can be divided into
four stages (Figure 2). Initially, in stage I, increased contact pressure results in joint space narrowing between
the articular surface of the radial styloid and the distal
scaphoid. As the disease progresses, the entire articular
surface between the scaphoid fossa of the distal radius
and the radial curvature degenerates. In stage III, the
capitolunate joint becomes involved with joint space
narrowing, sclerosis, and cyst formation. Pancarpal arthrosis occurs during stage IV. The radiolunate joint is
consistently spared.
If conservative treatment measures fail, surgical
treatment to correct the abnormal loads is attempted.
With disruption of the scapholunate and palmar radioscaphoid ligaments, the scaphoid rotates into excessive
flexion. This situation creates the excessive loading of
the radioscaphoid, scapholunate, and capitolunate joints.
Once there is evidence of stage I SLAC, ligament reconstruction cannot consistently correct the excessive flexion deformity of the scaphoid. Thus, a radial styloidectomy is the procedure of choice to treat stage I disease.
For wrists with more advanced degenerative arthritis
(stage II and stage III), a motion-preserving reconstructive procedure is recommended. Either a capitatelunate-hamate-triquetrum (four-corner) arthrodesis with
scaphoid excision or a proximal row carpectomy may be
performed with reliable results. Attempts to limit the fusion to the capitolunate joint to achieve higher postoperative carpal motion have been unsuccessful because of
high rates of nonunion.
With proximal row carpectomy, a high percentage of
wrist motion and grip strength are maintained. However, pain may persist, and many patients fail to return
to work. Conversion to wrist arthrodesis has been required in up to 15% of patients. A carefully performed
four-corner fusion with excision of the scaphoid results
in a comparative level of patient satisfaction and grip
strength relative to proximal row carpectomy. However,
patients with four-corner fusions have approximately
15 to 20 less motion than patients with proximal row
carpectomy. A four-corner fusion may be preferable after the midcarpal joint has become involved (stage III).
After the disease has progressed to stage IV (pancarpal
arthritis), total wrist arthrodesis is the treatment of
choice.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 30 Wrist and Hand Reconstruction

Figure 1 Corrective osteotomy with tricortical bone graft for a distal radius malunion. The osteotomy and graft shape should be designed to correct the radial height (A), radial
inclination (B), and palmar tilt (C). (Reproduced with permission from Trumble TE: Fractures and malunions of the distal radius, in Trumble TE (ed): Principles of Hand Surgery and
Therapy. Philadelphia, PA, 2000, p 172.)

Scaphoid Nonunion Advanced Collapse

Untreated scaphoid nonunions also lead to a predictable progressive pattern of wrist arthrosis, termed
scaphoid nonunion advanced collapse (SNAC) (Figure
3). Similar to a wrist with SLAC, degenerative changes
first develop in the radial styloid (stage I). In stage II
SNAC, there is joint space narrowing of the entire radioscaphoid articulation. In stage III SNAC, the capitate
migrates proximally, resulting in a loss of carpal height.
As the disease progresses to its final stage (IV), the
midcarpal and radiocarpal articulations develop degenerative changes. The radiolunate joint and the articular
surface under the proximal fragment of the scaphoid are
usually preserved.
As with the treatment of SLAC, various techniques
have been described for the treatment of SNAC. The
appropriate procedure depends on multiple factors, including the stage of the progression and individual patient factors. In patients with a scaphoid nonunion who
have developed degenerative changes of the radial styloid (stage I SNAC), radial styloidectomy in addition to
open reduction and internal fixation with bone grafting

American Academy of Orthopaedic Surgeons

of the scaphoid is usually successful in halting progression of the arthrosis. If there is evidence of more advanced carpal arthrosis (stage II or greater SNAC),
bone grafting will not reliably relieve the patients
symptoms and a salvage procedure is more appropriate
(proximal row carpectomy or four-corner fusion).
An option for patients without capitolunate arthrosis
is excision of the distal pole of the scaphoid. A recent
study suggests that this procedure improves range of
motion and increases grip strength in patients with stage
II SNAC. Patients who do not have radiocarpal disease
may also benefit from either a four-corner fusion and
scaphoidectomy (Figure 4) or a proximal row carpectomy.
A proximal row carpectomy is a good procedure to
eliminate pain and preserve wrist motion. However,
once the progression of arthrosis involves the capitolunate articulation, a proximal row carpectomy is no
longer indicated. Therefore, stage III SNAC is most appropriately treated with a four-corner fusion and
scaphoidectomy, and stage IV SNAC is best treated with
a total wrist arthrodesis.

353

Wrist and Hand Reconstruction

Figure 2 The stages of SLAC. In stage I, there is beaking of the radial styloid. In stage
II, there is narrowing and arthrosis of the radioscaphoid joint. In stage III, there is
arthrosis between the capitate and the scaphoid and/or lunate as the capitate displaces between the scaphoid and lunate with the carpal collapse. In stage IV, all of
the above changes occur along with degeneration of the radiolunate joint. (Reproduced with permission from Trumble TE, Gardner, GC: Arthritis, in Trumble TE (ed):
Principles of Hand Surgery and Therapy. Philadelphia, PA, 2000, p 406.)

Figure 4 Radiographs showing scaphoidectomy and four-corner fusion using an innovative plate and screw system.

Osteonecrosis of the Carpus

Kienbcks Disease
The etiology of idiopathic lunate osteonecrosis (Kienbcks disease) is not fully understood. Approximately
20% of lunates have only a single vessel as its nutrient

354

Orthopaedic Knowledge Update 8

Figure 3 The stages of SNAC. Similar to SLAC, stage I is hallmarked with radial styloid
arthritis. In stage II, there is progression of the arthritis to the scaphoid fossa. In stage
III, capitolunate arthritis is observed. In stage IV, there is diffuse carpal arthritis with
sparing of the lunate fossa. (Reproduced with permission from Knoll VD, Trumble TE:
Scaphoid fractures and nonunions, in Trumble TE (ed): Principles of Hand Surgery and
Therapy. Philadelphia, PA, 2000, p 167.)
supply. An injury to this vessel cannot be compensated
for by collateral flow, and the lunate with a single nutrient vessel therefore may be more susceptible to osteonecrosis. The role of ulnar length in the development
of Kienbcks disease is still uncertain. When the distal
ulna articular surface sits more proximally than the articular surface of the distal radius (ulna negative variance), abnormally increased shear forces across the radiolunate joint may compromise a marginally perfused
lunate.
Patients with Kienbcks disease present with an insidious onset of wrist pain that is localized over the middorsum of the wrist. Examination shows only mild
tenderness with palpation over the lunate; however, decreased carpal range of motion, particularly in extension, is frequently noted. Plain radiographs may show
obvious density changes. MRI has replaced bone scan as
the best test to diagnose early stages of Kienbcks disease. It is important to differentiate avascular changes
involving the entire lunate consistent with Kienbcks
disease and ulnar-sided lunate changes related to impaction syndrome. Intraosseous ganglia also can be mistaken for Kienbcks disease.
Kienbcks disease progresses in a predictable fashion through four radiographically defined stages, which
are helpful in guiding treatment (Figure 5) (Table 2). Pa-

American Academy of Orthopaedic Surgeons

Chapter 30 Wrist and Hand Reconstruction

Orthopaedic Knowledge Update 8

Table 2 | Stages of Kienbck Disease

Stages

Radiographic Findings

Recommended Treatment

No findings on plain
radiographs, but + bone
scan and MRI
Lunate sclerosis without
collapse

Activity modification

3A
3B

Fragmentation and collapse

of lunate
DISI deformity

Degenerative changes of
radiocarpal or midcarpal
joint on plain radiographs

Negative variance: Radial

shortening vascularized
graft to lunate
Neutral or positive variance:
capitate shortening,
capitohamate arthrodesis,
vascularized graft to
lunate
Same as stage 2
Intercarpal arthrodesis
(scaphotrapeziotrapezoid
or scaphocapitate)
Proximal row carpectomy or
total wrist arthrodesis

DISI=Dorsal intercalated segmental instability

Figure 5 The radiographic stages of Kienbcks disease. (Reproduced with permission from Allan CH, Trumble TE: Kinebocks disease, in Trumble TE (ed): Principles of
Hand Surgery and Therapy. Philadelphia, PA, 2000, pp 441-443.)
tients with stage I disease may improve with activity
modification and immobilization, and this should be the
first line of treatment. If there is no clinical improvement or if the patient advances to stage II disease, surgery may be indicated. Arthroscopic inspection and synovectomy and/or dbridement may have a role in the
treatment of early Kienbcks disease, particularly because the natural history of untreated Kienbcks disease remains elusive. In patients without a fixed collapse
of the lunate and scaphoid rotation (stages I, II, and
IIIA), a lunate-salvaging procedure may allow for revascularization of the lunate, maintaining carpal kinematics. In patients with ulnar-negative variance and stage I,
II, or IIIA disease, an unloading or joint leveling proce-

American Academy of Orthopaedic Surgeons

dure may be considered. Radial shortening with or without vascularized bone graft to the lunate is the most
successful procedure. If the patient with stage I, II, or
IIIA disease has ulnar-positive or neutral variance, radial shortening will not decrease the load on the lunate.
In this situation, capitate shortening with capitohamate
fusion has been shown to successfully decrease the load
across the radiolunate articulation.
Stage IIIB Kienbcks disease is characterized by
carpal instability with either scaphoid hyperflexion or
widening of the scapholunate interval and subsequent
migration of the capitate. After the disease has progressed to this stage, simply addressing the load on the
lunate does not correct the instability. The scaphoid
must be stabilized. This goal can be accomplished by an
intercarpal arthrodesis that bridges the midcarpal joint,
such as a scaphotrapeziotrapezoid arthrodesis or a
scaphocapitate arthrodesis. The goal is to stabilize the
scaphoid in a nonrotated position to prevent abnormal
kinematics and subsequent degenerative changes in the
wrist, while simultaneously transferring some of the radiocarpal load away from the lunate.

Preisers Disease
Idiopathic osteonecrosis of the scaphoid (Preisers disease) occurs less frequently than Kienbcks disease. Patients generally report pain at the radial aspect of the
wrist. Radiographs at the time of presentation show
sclerosis of the involved areas and fragmentation of the

355

Wrist and Hand Reconstruction

Figure 6 The vascular supply of the scaphoid. MC 1 = first metacarpal, Tz = trapezium, S = scaphoid, R= radius. (Reproduced with permission from Trumble TE: Fractures and dislocations of the carpus, in Trumble TE (ed): Principles of Hand Surgery
and Therapy. Philadelphia, PA, 2000, p 94.)
proximal articular surface. This disease must not be confused with posttraumatic proximal osteonecrosis. Vascularized bone grafting can be successfully used if fragmentation of the proximal pole has not occurred. A
salvage procedure may be required. The most common
procedures used to treat Preisers disease are proximal
row carpectomy or scaphoidectomy and four-corner fusion. Recently, arthroscopic inspection and dbridement
has been described for its treatment.

Capitate Osteonecrosis
Capitate osteonecrosis is a very rare condition. It is usually associated with high-dose steroid use, chemotherapy, or trauma (such as transperilunate transcapitate
fracture-dislocations). Surgical treatment options include vascularized grafting, fragment excision with tendon interposition arthroplasty, and four-corner arthrodesis. Most recently, scaphocapitolunate arthrodesis has
been used to treat osteonecrosis of the capitate. Osteonecrosis of the other carpal bones is rarely seen.

Orthopaedic Knowledge Update 8

used to assess location of the fracture, size of the fragments, extent of collapse, and progression of union (Figure 7). Preoperatively, MRI is a good tool to evaluate
the vascularity of the proximal pole (Figure 8).
Many patients with scaphoid fractures have subtle
symptoms leading to a delay in diagnosis. It may be
months or years before radiographs are taken that reveal the scaphoid nonunion causing the patients pain.
By this time, the scaphoid has usually collapsed into a
humpback type of deformity. The proximal pole is extended through the pull of the scapholunate ligament
and the influence of the triquetrohamate articulation,
while the scaphotrapeziotrapezoid articulation exerts a
flexion moment on the distal scaphoid.
The duration of the nonunion influences the probability of surgical success. A scaphoid nonunion of less
than 5 years duration is more likely to heal with appropriate surgical treatment than a nonunion that has been
untreated for longer than 5 years. Proper resection of all
fibrous material, thorough dbridement of the sclerotic
margins, and adequate bone grafting are necessary to
achieve healing of a nonunion. Precise surgical technique and positioning of the implant is critical for obtaining union. The use of a variable-pitch screw enables
compression between the two fragments and enhances
union results. The placement of a compression screw
centrally within the scaphoid fragments also increases
the stability of fixation and the likelihood of union.
Vascularized bone grafts are a good treatment option for proximal pole nonunions when osteonecrosis is
confirmed by MRI. The vascularized graft is harvested
from the dorsal aspect of the distal radius and rotated
on a regional pedicle. The radial artery has dorsal
branches defined in reference to the extensor compartments. The artery of Zaidemberg (1,2-intercompartmental supraretinacular artery), located between the first and second compartments, is most
commonly used (Figure 9).

Scaphoid Nonunion and Osteonecrosis

Triangular Fibrocartilage Complex and the

Distal Radioulnar Joint

Osteonecrosis of the proximal pole of the scaphoid is

frequently posttraumatic and is associated with nonunion of a proximal pole or scaphoid waist fracture. This
condition is unique and should be approached differently than Preisers disease. Most of the scaphoid is vascularized via a dorsal branch of the radial artery, which
perforates the distal third of the dorsal cortex of the
scaphoid. Intraosseous blood flow is retrograde, leaving
the proximal pole with a tenuous blood supply (Figure
6). The more proximal a scaphoid fracture, the more
likely is delayed healing or nonunion and osteonecrosis.
Fine-cut (1-mm) sagittal and coronal CT images provide
excellent detail of the fracture fragments and can be

The TFCC is composed of the triangular fibrocartilage

and its supporting ligaments. The triangular fibrocartilage is a meniscus-like structure with a high proportion
of type II collagen in addition to type I collagen. Surrounding the periphery of this structure is the meniscal
homolog, a fibrocartilaginous rim that stabilizes the triangular fibrocartilage to the ulnar styloid and the ulnar
collateral ligament. This structure has a distinct blood
supply that enters from the periphery of the TFCC. The
ulnar periphery of the triangular fibrocartilage has the
richest blood supply and the best potential for healing
(Figure 10).

356

Anatomy and Biomechanics

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 30 Wrist and Hand Reconstruction

Figure 7 Fine-cut coronal (A) and sagittal (B) CT images provide excellent detail of fracture fragments (arrowheads) of the scaphoid nonunion. (Reproduced with permission
from Trumble TE, Fractures and dislocations of the carpus, in Trumble TE (ed): Principles of Hand Surgery and Therapy. Philadelphia, PA, 2000, p 102.)
The TFCC arises from the articular cartilage at the
corner of the sigmoid notch of the radius and inserts onto
the base of the ulnar styloid and volarly into the ulnocarpal ligament complex formed by the ulnar triquetral ligament and the ulnar lunate ligament. The ligaments and
the TFCC, which support the DRUJ and ulnar portion of
the carpus, form a three-walled pyramidal structure. The
triangular fibrocartilage is the floor of this pyramid. The
ligaments forming the walls of this pyramid stabilize the
carpus to the triangular fibrocartilage and the ulnar styloid so that they maintain their relationship to the TFCC
while rotating around the ulna. The ulnar triquetral ligament and the ulnar lunate ligament form the volar wall of
this box. The undersurface of the extensor carpi ulnaris
tendon forms the ulnar wall of the box, and the dorsal radial triquetral ligament forms the dorsal wall of this compartment. These structures maintain the relationship between the carpus and the ulna.
In addition to the DRUJ and ulnocarpal ligaments,
the TFCC is a key stabilizer of the DRUJ and a stabilizer of the ulnar carpus. The amount of load transferred
to the distal ulna from the carpus is directly proportional to the ulnar variance. In neutral ulnar variance,
approximately 20% of the load is transmitted via the
TFCC. With positive ulnar variance, the load across the
TFCC is increased, with a resultant thinning of its central disk. In pronation, the radius moves proximally relative to the ulna, while in supination the radius moves
distally. This movement results in a relative positive
variance in pronation and a relative negative variance in
supination. Thus, a greater load is transferred to the distal ulna from the carpus via the TFCC in pronation than

American Academy of Orthopaedic Surgeons

in supination. In pronation, the radius also moves volarly relative to the ulna, tightening the dorsal ligament
fibers of the TFCC. In supination, the radius moves dorsally such that the volar ligament fibers of the TFCC
tighten. As these respective ligaments tighten, they help
to support the carpus on the distal forearm.

Triangular Fibrocartilage Complex Injuries

TFCC injuries can be classified into two basic categories: type I, traumatic injuries, and type II, degenerative
lesions. Traumatic lesions are classified according to the
location of the tear within the TFCC; degenerative lesions are associated with positive ulnar variance and often have associated pathology (for example, lunotriquetral ligament attenuation or carpal arthrosis) (Table 3).
Type IA lesions are usually treated with arthroscopic
dbridement of the unstable portion of the tear that has
occurred in the avascular zone. If more than two thirds
of the central disk is dbrided, the DRUJ becomes unstable. The peripheral 2 mm of the triangular fibrocartilage must be maintained to avoid DRUJ instability.
Chronic type IB lesions can be difficult to diagnose.
Examination shows ulnar-sided wrist pain and mild
DRUJ instability. Radiographs may demonstrate an
ulna styloid fracture or may be normal. Diagnostic arthroscopy reveals that the TFCC has lost its normal tension. Isolated peripheral tears are repaired arthroscopically. If the tear is associated with an ulna styloid
fracture and instability, an open reduction with internal
fixation of the ulnar styloid fragment or styloidectomy is
performed before reattaching the TFCC to the remaining distal ulna. Treatment of this injury consists of ar-

357

Wrist and Hand Reconstruction

Orthopaedic Knowledge Update 8

Figure 8 MRI of scaphoid nonunion showing proximal pole avascularity. Each image shows hypoechoic signal in the proximal pole suggestive of avascularity.
throscopic or open repair of the TFCC tear. Three or
four sutures are placed through the most volar aspect of
the tear, which is then tied over the capsule (Figure 11).
The diagnosis of a type IC lesion is made arthroscopically after noting a loss of tension in the ulnar extrinsic ligaments, as well as easy and direct visualization of the pisotriquetral joint. This lesion may be repaired arthroscopically
or openly depending on the size of the defect. Type ID lesions are frequently associated with distal radial fractures.
This corner of the TFCC has poor vascularity. However, if
the articular cartilage of the sigmoid notch is disrupted by
fracture, or intraoperatively by the surgeon, healing to vascularized bone readily occurs (Figure 12).
The outcome of TFCC injuries also depends on the
chronicity of the tear. Patients with acute tears, which are
repaired within 3 months after injury, recover 80% of the
grip strength and range of motion as is present on the contralateral side. Subacute injuries (3 months to 1 year) are
still amenable to direct repair, but regain less strength and
range of motion. Arthroscopic repairs result in greater
range of motion, grip strength, and patient satisfaction
compared with open repairs. Chronic injuries to the TFCC
frequently benefit from ulnar shortening to decrease the
load distributed to the distal ulna via the TFCC, with or
without dbridement of the TFCC.

Degenerative TFCC Lesions (Ulnocarpal Impaction

Syndrome)
Figure 9 Vascularized bone grafts for the scaphoid are based on the artery of Zaidemberg (1,2-intercompartmental supraretinacular artery). A, The 1,2 intermetacarpal
artery is shown (arrow). B, The vascularized pedicle has been harvested. A burr is used
to create a recipient channel for the graft. C, Two Kirschner wires are used to stabilize
the bone graft. Note the vascularized pedicle.(Reproduced with permission from Thiru
RG, Ferlic DC, Clayton ML, McClure DC: Arterial anatomy of the triangular fibrocartilage of the wrist and its surgical significance. J Hand Surgery [Am] 1986;11:258263.)

358

Type II or degenerative lesions of the TFCC are related

to chronic overload of the ulnar side of the wrist. Once
symptomatic, chronic ulnar abutment or impaction is
progressive and deterioration of the TFCC as well as
ulnar-sided articulations occur over time. Abutment may
be idiopathic; patients with positive ulnar variance are
more susceptible to overload with chronic perforations

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

of their TFCC than patients with neutral or negative ulnar variance (Figure 13). Ulnar impaction also may be
secondary to a change in wrist anatomy after a traumatic insult. Malunion of the distal radius with excessive
shortening and dorsal tilt, growth arrest after a distal radius physeal fracture, or proximal migration of the radius following a radial head excision may all result in ulnar impaction syndrome.
The progressive degenerative changes of type II
TFCC lesions associated with ulnar abutment are classified into five stages: A through E (Table 3). The primary
goal of treatment of ulnar impaction is to unload or decompress the ulnar carpus and ulnar head. This decompression can be accomplished by several different techniques: ulnar shortening osteotomy, partial ulnar head
resection (wafer procedure), or an ulnar salvage procedure. Type IIA and IIB lesions are early stages in the
degenerative process. There is no perforation of the triangular fibrocartilage, and thus, no dbridement is necessary, and an open ulnar shortening osteotomy should
sufficiently unload the ulnocarpal joint. Type IIC lesions
are preferentially treated with an arthroscopic dbridement of the perforation. An arthroscopic wafer resection may then be performed through the dbrided area
of the triangular fibrocartilage. Conversely, an ulnar
shortening procedure can be done in addition to the arthroscopic dbridement (Figure 14). The major disadvantage of an ulnar shortening procedure is that the
ulna may be slow to unite or may even fail to unite. The
wafer resection procedure does not have the complications associated with an osteotomy; however, this procedure may accelerate the onset of DRUJ arthritis. Type
IID and IIE lesions represent end stages of ulnar impaction syndrome, with arthritic changes and instability. If
there is no notable lunotriquetral instability or DRUJ
arthrosis, the treatment of choice is ulnar shortening followed by arthroscopic dbridement. If there is demonstrable lunotriquetral instability after dbriding the
TFCC and the frayed lunotriquetral ligament, an ulnar
shortening osteotomy is performed. Lunotriquetral stability often improves after the osteotomy because of
tightening of the ulnar extrinsic ligaments. If, however,
significant lunotriquetral instability persists, the lunotriquetral joint should be percutaneously pinned with two
0.045-inch Kirschner wires. If significant arthritic
changes are noted at the DRUJ, an ulnar salvage procedure (such as a modified Darrach, an ulna hemiresection with tendon interposition, or a distal ulna replacement arthroplasty) is a better surgical option.

Instability and Arthritis of the DRUJ

Patients with DRUJ pathology can be divided into three
categories: those with instability only, those with arthritis only, and those with both arthritis and instability of
the DRUJ. In patients with DRUJ instability, the radius

American Academy of Orthopaedic Surgeons

Chapter 30 Wrist and Hand Reconstruction

Figure 10 Vascular supply to the TFCC enters via the periphery. Br = branch, Ant =
anterior, Int = interosseous. (Reproduced with permission from Thiru RG, Ferlic DC,
Clayton ML, McClure DC: Arterial anatomy of the triangular fibrocartilage of the wrist
and its surgical significance. J Hand Surgery [Am] 1986;11:258-263.)
impinges on the ulna, causing a painful click during attempts to pronate and supinate the wrist. Over time, this
impingement can result in degenerative changes within
the joint. As degenerative changes develop in the
DRUJ, the instability between the ulna and radius may
lessen.
DRUJ instability is the result of bone deformity, ligamentous injury, or a combination of the two. In patients with chronic instability without bony deformity,
the radioulnar ligaments are usually irreparable. Treatment consists of some type of soft-tissue reconstruction.
Reconstruction of the distal radioulnar ligaments can
potentially restore stability without substantial loss of
motion or strength. Several techniques of reconstruction
have been described. A careful evaluation of the patient
is necessary, however, because significant joint incongruity and frank arthritis of the DRUJ are contraindications to reconstruction.
A recent study assessed DRUJ ligament reconstruction using a palmaris tendon graft passed through bone
tunnels to restore both the volar and dorsal ligaments.
Stability was restored in 12 of 14 of patients who underwent this DRUJ ligament reconstruction. These patients
returned to full activities and recovered 85% of motion
and strength. This procedure was considered effective
for restoring DRUJ stability; however, it requires a competent sigmoid notch and did not fully correct associated ulnocarpal instability.
Recently, attempts to manage early DRUJ arthrosis
have focused on retaining the ulnar head and altering
the contact surface by performing either an ulna shaft
shortening osteotomy or dbriding osteophytes from the
proximal margin of the joint. More traditionally, patients with DRUJ arthritis have been treated by either
ulna head resection or distal radioulnar fusions. Patients
with stable arthritic joints have classically been treated
with excision and careful repair of the surrounding joint
capsule (Darrach or modified Darrach procedure).

359

Orthopaedic Knowledge Update 8

Wrist and Hand Reconstruction

Table 3 | TFCC Lesions

Traumatic
Type
IA
IB

IC

ID

Degenerative
Location
Central tears without
instability
Peripheral tear at the base
of the ulnar styloid

Treatment
Arthroscopic dbridement of
unstable portion
Isolated TFCC tear: Arthroscopic
repair
Associated with ulna styloid fixation:
ORIF of ulna styloid and repairing
TFCC
Avulsion from ulnar extrinsic Arthroscopic versus open repair
ligaments

Type
IIA

Detachment from sigmoid

notch; often associated
with distal radius fixation

IID

Arthroscopic versus open repair

IIB

IIC

IIE

Character
Treatment
Wearing without perforation Ulnar shortening
or chondromalacia
Wearing with
Ulnar shortening
chondromalacia of either
lunate or ulna

Perforation of triangular
fibrocartilage with lunate
chondromalacia
Perforation of triangular
fibrocartilage, lunate,
and/or ulna
chondromalacia, LT
disruption without
instability
Generalized arthritic
changes, LT disruption
with volar intercalcated
segmental instability

Arthroscopic dbridement
of TFCC; ulnar shortening
or wafer resection
Arthroscopic dbridement
of TFCC and LT ligaments;
ulnar shortening
If DRUJ arthrosis, modified
Darrach or ulna
hemiresection
Arthroscopic dbridement
of TFCC and LT ligaments;
ulnar shortening
If instability persists after
shortening, pin LT joint
If DRUJ arthrosis, modified
Darrach, hemiresection or
ulna replacement
arthroplasty

ORIF = open reduction and internal fixation; LT = lunotriquetral

Patients with instability associated with arthritis are

at increased risk for persistent instability and are best
treated with a procedure designed to stabilize the ulna
and buffer the distal radioulnar articulation, such as an
ulnar hemiresection with tendon interposition. In
younger patients, with posttraumatic DRUJ arthritis and
an intact TFCC, a Sauv-Kapandji procedure is an alternative approach described to maintain grip strength. In
the Sauv-Kapandji procedure, the DRUJ is fused and a
pseudarthrosis is created proximal to the DRUJ to allow for rotation of the radius around the ulna. The
Sauv-Kapandji procedure is often subject to complications related to an unstable proximal stump.
Distal ulna implant arthroplasty is an attractive alternative for the treatment of DRUJ arthritis and instability (Figure 15). Normal mechanical function and stability of the DRUJ requires an intact ulnar head. A
functional ulnar head implant would alleviate impingement and restore near-normal load transmission. Thus,
ulnar head replacement may be an alternative to the
modified Darrach procedure. In the past, patients with
recurrent instability of the ulna caused by excessive
proximal excision of the radial head, excessive resection

360

of the distal ulna, or combined radial head excision and

distal ulna excision were treated with arthrodesis of the
radius and ulna (creation of a one-bone forearm) performed as a salvage procedure. A total DRUJ prosthesis
is now available and has many advantages over the onebone forearm. Currently, there are several designs available for ulnar head implants and total DRUJ prostheses. However, although these devices show promise,
there are no long-term clinical studies that clearly delineate the indications for their use.

Degenerative Arthrosis of the Thumb and

Fingers
Thumb Carpometacarpal Joint
The CMC joint of the thumb has a biconcave saddle
shape, which imparts little intrinsic bony stability. In a
recent study, 16 ligaments surrounding the trapeziometacarpal joint were identified. The deep volar (anterior oblique) ligament, which passes from the trapezium
to the volar beak of the thumb metacarpal, and the dorsoradial ligament play the most significant role in maintaining CMC stability. The deep volar ligament resists

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 30 Wrist and Hand Reconstruction

Figure 11 Arthroscopically assisted TFCC repair using the outside-in technique. A, Shows the 3-4 and 6R portals used to access the TFCC. A needle is used to pierce the TFCC.
(B) A suture is threaded through the needle and grasped. The suture is tied to the capsule. (Reproduced with permission from Trumble TE: Distal radioulnar joint triangular
fibrocartilage complex, in Trumble TE (ed): Principles of Hand Surgery and Therapy. Philadelphia, PA, 2000, pp 136-137.)
the dorsal subluxation force created by pinch. Attenuation of the volar (anterior oblique) ligament allows for
dorsoradial subluxation of the metacarpal base and initiates articular cartilage degeneration. Degenerative
changes begin volarly and progress dorsally.
Patients with thumb basal joint arthritis can be divided
into those with arthritis isolated to the CMC joint (type
A), those with arthritis isolated to the scaphotrapeziotrapezoid joint (type B), and those with pantrapezial arthritis
(type C). Patients with involvement of the CMC joint can
further be divided into those with instability (dorsal subluxation of the metacarpal on the trapezium with or without compensatory metacarpophalangeal joint hyperextension) and those without instability. In patients with
metacarpophalangeal joint hyperextension coexisting
with dorsal CMC joint subluxation (swan neck thumb), it
is critical to correct the metacarpophalangeal joint hyperextension via capsulodesis or arthrodesis to eliminate the
long moment arm that continues to exert a dorsally directed force on the CMC joint during pinch activities.
CMC joint arthritis is most commonly treated with excision of the distal half or the entire trapezium with tendon interposition and some form of ligament reconstruction. The flexor carpi radialis can be used to stabilize the
joint and serve as the anchovy interpositioned between

American Academy of Orthopaedic Surgeons

Figure 12 Arthroscopic repair of a TFCC type ID lesion. (Reproduced with permission

from Trumble TE, Gilbert M, Vedder N: Isolated tears of the triangular fibrocartilage:
Management by early arthroscopic repair. J Hand Surg [Am] 1997;22:57-65.)

the scaphoid and the metacarpal (ligament reconstruction

tendon interposition); or a slip of the abductor pollicis lon-

361

Wrist and Hand Reconstruction

Orthopaedic Knowledge Update 8

Figure 14 An ulna shortening osteotomy should be performed in patients who have

type II TFCC lesions and positive ulnar variance. (Reproduced with permission from
Trumble TE: Distal radioulnar joint and triangular fibrocartilage complex, in Trumble TE
(ed): Principles of Hand Surgery and Therapy. Philadelphia, PA, 2000, p 139.)

Figure 13 Ulnar impaction syndrome. Radiographic changes are localized to the ulnar aspect of wrist and are associated with ulnar positive variance. (Reproduced with
permission from Conduit DP: Carpal avascular necrosis, Trumble TE (ed): Hand Surgery Update 3: Hand, Elbow, and Shoulder. Rosemont, IL, American Society for Surgery
of the Hand, 2003, p 218.)

gus can also be used as a sling to suspend the metacarpal

and recreate a volar beak ligament, after excision of the
trapezium (abductor pollicis longus suspensionplasty).
Studies have shown that basal joint arthroplasties with a
firm interposition graft provide a greater degree of stability than grafts composed of tendon. Thus, an alternative
technique is interposition with a costochondral allograft
spacer (lifesaver technique). One half of the flexor carpi
radialis is harvested and used to stabilize the spacer (Figure 16).
Other options include thumb metacarpal extension
osteotomy for patients with early degenerative changes
isolated to the volar aspect of the joint and CMC joint
arthrodesis in younger patients who require strength
and stability. Scaphotrapeziotrapezoid arthritis is a contraindication for CMC joint arthrodesis. Arthroscopic
dbridement of the CMC joint may also play a role in
the treatment of CMC joint arthritis. For patients with
an early disease stage, arthroscopic synovectomy and
electrothermal shrinkage of the trapeziometacarpal capsule to provide symptomatic relief has been described.
For patients with more advanced disease, hemitrapeziectomy and complete trapeziectomy with electrothermal
shrinkage of the anterior oblique ligament may be performed arthroscopically.

Metacarpophalangeal Joint Arthrosis

Figure 15 A, PA radiograph of failed Darrach resection caused by radioulnar impingement. B, A distal ulna prosthesis (Avanta Orthopaedics, San Diego, CA) restored alignment and stability to the DRUJ. (Reproduced with permission from Adams BD: Distal
radioulnar joint, in Trumble TE (ed): Hand Surgery Update 3: Hand, Elbow, and Shoulder. Rosemont, IL, American Society for Surgery of the Hand, 2003, p 154.)

362

When contemplating treatment of arthrosis of the

metacarpophalangeal joint of the thumb, the entire
thumb axis must be considered. If the interphalangeal
and CMC joints are not diseased, arthrodesis is the procedure of choice. The metacarpophalangeal joint of the
thumb is often fused because of instability, pain, and deformity, which impairs pinch.
In patients with rheumatoid arthritis and osteoarthritis, arthroplasty of the thumb metacarpophalangeal
joint with newer articulating implants that preserve col-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 30 Wrist and Hand Reconstruction

Figure 16 Lifesaver technique of CMC joint arthroplasty. MC I = first metacarpal,

FCR = flexor carpi radialis. (Reproduced with permission from Trumble TE, Garnder GC:
Arthritis, in Trumble TE (ed): Principles of Hand Surgery and Therapy. Philadelphia, PA,
2000, p 429.)

lateral ligaments, may be performed in conjunction with

fusion of a diseased interphalangeal joint. Arthroplasty
of the metacarpophalangeal joint may be preferred to
arthrodesis in this clinical scenario because it preserves
thumb length, improves position of thumb tip for opposition, and decreases the stress transferred to the CMC
joint. However, after arthroplasty, a deficient ulnar collateral ligament and the decreased ability to generate
pinch force may ultimately result in failure of the arthroplasty.
Whereas arthrodesis is commonly indicated in the
proximal interphalangeal (PIP) and distal interphalangeal joints of the fingers, arthrodesis is very functionally limiting in the metacarpophalangeal joint and is
rarely performed. Soft-tissue deformity in the metacarpophalangeal joints caused by rheumatoid arthritis is often severe, which increases the difficulty of obtaining a
good result from arthroplasty. If the metacarpophalangeal collateral ligaments are intact, an articulating
prosthesis can be used (Figure 17). If the collateral ligaments are deficient, a Silastic spacer is a better implant
choice. The status of the PIP joint and the wrist must
also be considered when contemplating metacarpophalangeal arthroplasty.

Proximal and Distal Interphalangeal Joints

The PIP joint is the soul of the hand and thus there is
no good position for arthrodesis. PIP joint arthrodesis
(in 40 of flexion for the index finger, 45 of flexion for
the middle finger, 50 flexion for the ring finger, and 55
of flexion for the small finger) is a poor secondary option. However, arthrodesis does provide a stable, painfree digit and is indicated for high-demand patients with
PIP joint arthrosis. A significant amount of lateral stress
is placed on the PIP joint of the index finger during
pinch; therefore, PIP arthrodesis should be considered
in all patients with index PIP disease.

American Academy of Orthopaedic Surgeons

Figure 17 Preoperative (A) and postoperative (B) AP radiographs of a patient with

posttraumatic metacarpophalangeal joint arthritis, treated with a newer, anatomically
designed metacarpophalangeal joint arthroplasty C, Photograph of implant components. (Ascencion Orthopaedics, Austin, TX.)
Loss of motion of the PIP joint results in the inability to make a fist. Therefore, arthroplasty is a viable alternative when motion of the joint needs to be preserved. Classically, Silastic joint spacers have been used
to preserve motion, but these implants are not very durable. Recently, several two-piece prostheses have been
anatomically designed specifically for PIP joint arthroplasty. Early results suggest that these prostheses may
function better than an arthrodesis. Similar to the effects of rheumatoid arthritis on the metacarpophalangeal joints, rheumatoid disease often causes softtissue imbalance of the interphalangeal joints, which
must be considered at the time of surgery.
The distal interphalangeal joint is most commonly
involved in degenerative osteoarthritic conditions. For
patients who do not respond to nonsurgical treatment,
surgical options include resection arthroplasty, implant
arthroplasty, and arthrodesis. Arthrodesis in 10 to 20
of flexion is the preferred treatment.

363

Wrist and Hand Reconstruction

Orthopaedic Knowledge Update 8

Figure 19 Illustration of endoscopic carpal tunnel release. (Reproduced with permission from Trumble TE: Compressive neuropathies, in Trumble TE (ed): Principles of
Hand Surgery and Therapy. Philadelphia, PA, 2000, p 334.)

Figure 18 Thumb opposition is a complex motion requiring trapeziometacarpal abduction, flexion, and pronation. (Reproduced with permission from Trumble TE: Tendon
transfers, in Trumble TE (ed): Principles of Hand Surgery and Therapy. Philadelphia, PA,
2000, p 351.)

Nerve Compression Syndromes

Carpal Tunnel Syndrome
Carpal tunnel syndrome (CTS) is a condition of compression of the median nerve as it traverses the wrist
and causes a constellation of related signs and symptoms. Patients often report intermittent or constant
numbness or tingling in the median nerve distribution,
which may be associated with pain. Night pain that
awakens the patient from sleep is also common. In severe cases, there may be denervation of the thenar muscles with resultant weakness and atrophy.
CTS is the most common entrapment neuropathy,
occurring in 0.1% to 10% of the general population.
Risk factors include obesity, hypothyroidism, diabetes,
pregnancy, renal disease, inflammatory arthritis, acromegaly, mucopolysaccharidosis, genetic predisposition, advancing age, smoking, and repetitive or extreme wrist
flexion at work. However, in more than 95% of patients,
CTS is idiopathic.
CTS is a clinical diagnosis based on history and physical examination and is confirmed by electrodiagnostic
studies. Other pathologies (such as cervical radiculopathy, brachial plexopathy, thoracic outlet syndrome, apical
lung tumor, pronator syndrome, cubital and ulnar tunnel
syndromes, and peripheral neuropathy) may cause hand
numbness; these disease processes must be ruled out before making the diagnosis of CTS. A combination of
findings in the history and physical examination is more
accurate than the same findings in isolation. CTS is accurately diagnosed 86% of the time when night pain, a positive Semmes-Weinstein monofilament test, a positive
carpal tunnel compression test (Durkhans sign), and a

364

consistent Brigham hand diagram are all present. The

probability of having CTS when all of these tests are
negative is 0.68%. Electrodiagnostic studies do not add
to the diagnostic reliability of this combination of tests.
Electrodiagnostic studies (nerve conduction velocities and electromyography) are used to confirm the clinical diagnosis. A pathologic nerve conduction velocity
study includes decreased action potential amplitude, increased distal latency, and a decreased velocity. A distal
motor latency of more than 4.5 ms and a sensory latency
of more than 3.5 ms is abnormal. Abnormal electromyographic findings include increased insertional activity, fibrillations at rest, positive sharp waves and complex
repetitive discharges, and decreased motor unit recruitment.
Nonsurgical treatment of CTS includes activity modification, night splinting, steroid injection into the carpal
canal, and oral medication (such as nonsteroidal antiinflammatory drugs and vitamin B6). Neither nonsteroidal anti-inflammatory drugs nor vitamin B6 has been
definitively shown to be effective in isolation. Steroid
injection into the carpal tunnel combined with nocturnal
splinting has a short-term success rate of 80% in relieving symptoms. However, after 12 to 18 months, only
22% of patients treated with this regimen remain symptom free.
Surgical treatment is indicated for patients who have
not responded to nonsurgical treatment and for patients
with thenar weakness, atrophy, or electrodiagnostic evidence of denervation (Figure 18). Several wellcontrolled studies indicate that there is no appreciable
benefit of internal neurolysis or epineurotomy. Also,
there does not seem to be a role for tenosynovectomy
with carpal tunnel release in patients with idiopathic
CTS.
Regardless of the technique used, the many anatomic variations in the region of the carpal tunnel demands that care be taken during the procedure. There

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 30 Wrist and Hand Reconstruction

have been major complications reported in association

with all techniques of carpal tunnel release including
open, endoscopic, and limited open techniques (with or
without specialized instrumentation). Several prospective randomized studies have shown that endoscopic release results in better early patient satisfaction scores
and earlier return to work (Figure 19).
The incidence of persistent symptoms after carpal
tunnel release range from 1% to 25%. Incomplete
transverse carpal ligament release is the most common
cause of recurrent symptoms. Transverse and miniincisions have a higher incidence of incomplete release.
Endoscopic carpal tunnel release does not have an increased risk of incomplete release when compared with
the traditional open technique. Other causes of recurrent CTS include incorrect diagnosis, double crush phenomena, concomitant peripheral neuropathy, a persistent carpal tunnel space occupying lesion, and iatrogenic
median nerve injury.

Pronator Syndrome and Anterior Interosseous Nerve

Compression Syndrome
At the level of the elbow, the median nerve has two
branches: one branch innervates the pronator teres,
flexor carpi radialis, and palmaris longus; the other
branch forms the anterior interosseous nerve. In most
cases, the median nerve then dives to pass between the
superficial and deep heads of the pronator teres. Up to
15% of patients have a Martin-Gruber anastomosis
where the intrinsic motor fibers that traveled within the
median nerve from the level of the brachial plexus cross
over at this level to return to the ulna nerve. At the
level of the elbow and proximal forearm, the potential
sites of median nerve compression are the supracondylar process, the ligament of Struthers, the bicipital aponeurosis, the lacertus fibrosus, fascia between the two
heads of the pronator (most common cause), and under
the fascial arch of the flexor digitorum superficialis.
The incidence of both pronator syndrome and anterior interosseous nerve compression syndrome is 100
times less common than CTS. Pronator syndrome is
more common in women. Patients generally present
with reports of pain in the anterior proximal forearm.
Hand numbness, nocturnal paresthesias, and objective
neurologic findings are rare. Silent motor weakness is
common. Frequently, there is a decrease of nerve conduction across the elbow. Additionally, pronator syndrome can be differentiated from CTS by sensory disturbance in the distribution of the palmar cutaneous
branch of the median nerve and a positive proximal
Tinels sign. Specific provocative maneuvers help isolate
the site of entrapment. Exacerbation of symptoms with
(1) resisted elbow flexion with forearm supination suggests the compression is under the bicipital aponeurosis;
(2) resisted forearm pronation with elbow extension

American Academy of Orthopaedic Surgeons

Figure 20 Decompression of the proximal median nerve involves release of the lacertus fibrosus and lengthening the humeral head of the pronator teres (A and B), and
release of the vascular leash proximal to the flexor digitorum sublimis (FDS) and release of the flexor digitorum sublimis fascia (C and D). FDP = flexor digitorum profundus. (Reproduced with permission from Trumble TE: Compressive neuropathies, in
Trumble TE (ed): Principles of Hand Surgery and Therapy. Philadelphia, PA, 2000, p
339.)

suggests entrapment between the two heads of the pronator; and (3) resisted long finger PIP joint flexion suggests the nerve is trapped under the origin of the flexor
digitorum sublimis. During surgical decompression, all
of these sites should be sufficiently released (Figure 20).
Pronator syndrome is often associated with medial epicondylitis. Conservative treatment of the epicondylitis
often relieves the pronator syndrome as well.
The anterior interosseous nerve is the largest branch
of the median nerve, arising 5 to 8 cm distal to the level
of the lateral epicondyle. The anterior interosseous
nerve is a pure motor branch innervating the flexor pollicis longus, flexor digitorum profundus to the index finger and occasionally the middle finger, and the pronator
quadratus. Compression of the anterior interosseous
nerve results in weakness or paralysis of one or more of
these muscles. If only one muscle is involved, the com-

365

Wrist and Hand Reconstruction

Orthopaedic Knowledge Update 8

Figure 21 Submuscular transposition of the ulnar nerve. FCU = flexor carpi ulnaris.
(Reproduced with permission from Trumble TE: Compressive neuropathies, in Trumble
TE (ed): Principles of Hand Surgery and Therapy. Philadelphia, PA, 2000, p 328.)
pression may be misdiagnosed as a tendon rupture. The
typical symptom, however, is the inability to form an
O with the thumb and index finger. There also will be
changes on the electromyogram recording of the flexor
pollicis longus. Parsonage-Turner syndrome (brachial
neuritis), which is generally associated with intense pain
during its onset, must be differentiated from a mechanical compression. The incidence of true entrapment neuropathy is very low; therefore, surgical decompression
should be reserved for patients who have no recovery
after 3 months.

Cubital Tunnel Syndrome

Although the incidence of cubital tunnel syndrome is
second only to CTS, it is a relatively uncommon disorder. The anatomic structures that can potentially cause
constriction of the nerve include the arcade of Struthers
(the medial intermuscular septum), the medial head of
the triceps, the ligament of the cubital tunnel or Osbornes ligament (the most common cause), the anconeus epitrochlearis (an accessory muscle), and the fascia
of the flexor carpi ulnaris. The nerve may subluxate over
the medial epicondyle causing chronic irritation. Cubital
tunnel syndrome may be associated with prior elbow
trauma. A patient may seek treatment many years after
an elbow (for example, lateral condyle) fracture with a
tardy ulnar nerve palsy. Cubital tunnel syndrome also
may be associated with medial epicondylitis. Patients
with cubital tunnel syndrome generally present with
numbness along the ring and small fingers without involvement of the medial forearm. Elbow flexion exacerbates the symptoms. Nocturnal symptoms are very common because people generally sleep with their elbows
flexed.
On examination, patients have a positive Tinels sign
with gentle percussion at the cubital tunnel. Monofilament and two-point testing often detect decreased sensation in the ulnar nerve distribution, including the ul-

366

Figure 22 The ulnar tunnel can be divided into three sections. Lesions in zone I (1 in
the figure) tend to produce a combined sensory and motor deficit. Lesions in zone II
(2) generally produce pure motor deficit. Lesions in zone III (3) usually produce pure
sensory deficits. FCU = flexor carpi ulnaris. (Reproduced with permission from Trumble
TE: Compressive neuropathies, in Trumble TE (ed): Principles of Hand Surgery and
Therapy. Philadelphia, PA, 2000, p 330.)
nar aspect of the dorsum of the hand. Wartenbergs sign
(an abducted small finger) may occur early and indicates intrinsic muscle weakness. In more severe cases, atrophy of the first dorsal interosseous and adductor pollicis with a concomitant Froments paper sign may be
seen. Severe cases may also be associated with clawing
of the ulnar digits. Weakness may be masked by a
Martin-Gruber anastomosis distal to the level of compression. Nerve conduction velocity studies are very
helpful for confirming the diagnosis; when they are negative, nonsurgical management (nocturnal soft splints to
prevent hyperflexion of the elbow) should be the mainstay of treatment.
Surgical decompression of all potential sites of entrapment (and possible transposition) is indicated for
symptomatic patients with positive nerve conduction velocity studies. Intramuscular transposition may result in
recurrent entrapment from perineural scarring. One
prospective, randomized study comparing in situ release,
subcutaneous transposition, and submuscular transposition did not find any statistically significant difference in
outcome. However, the results were slightly better in the
patients who underwent a subcutaneous or submuscular
transposition (Figure 21). In situ decompression is only
indicated for patients with mild symptoms and a nonsubluxating nerve. Submuscular transposition may be favored for patients with a thin layer of subcutaneous fat.
Muscle wasting is a poor prognostic sign for surgery because it suggests the presence of irreversible nerve damage. Ideally, surgical intervention should be performed

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 30 Wrist and Hand Reconstruction

Figure 23 A, The anterior approach (Henrys approach) for decompression of the distal portion of the radial nerve and the posterior interosseous nerve. B, The posterior
approach (Thompsons approach) for radial nerve decompression. ECRL = extensor carpi radialis longus, ECRB = extensor carpi radialis brevis, EDC = extensor digiti communis,
APL = abductor pollicis longus. (Reproduced with permission from Trumble TE: Compressive neuropathies, in Trumble TE (ed): Principles of Hand Surgery and Therapy. Philadelphia, PA, 2000, p 328.)

before there is notable atrophy. Patients with milder

compression are best treated with night splints.

with resection or grafting, depending on the competency

of the palmar arch.

Ulnar Tunnel Syndrome

Radial Nerve Entrapment

Ulnar tunnel syndrome is a rare condition that usually

results from the effect of some mass in the ulnar tunnel
(Guyons canal) such as a ganglion from the wrist joint
or an aneurysm of the ulnar artery. The ulnar tunnel has
been divided into three zones: zone I is proximal to the
bifurcation of the ulnar nerve, zone II surrounds the
deep motor branch and is dorsal and ulnar, and zone III
is palmar and radial, surrounding the superficial palmar
branch of the ulnar nerve (Figure 22). The most common cause of compression is different in each zone.
Nerve compression in zone I and II is most commonly
associated with either ganglia or fractures of the hook
of the hamate and ulnar nerve compression in zone III
is most frequently the result of a thrombosis or aneurysm of the ulnar artery.
Numbness in the ring and small fingers without involvement of the dorsum of the hand is the most common presenting symptom. Patients also may report pain
with flexion or extension and/or weakness of the hand.
Atrophy of the ulnar hand intrinsic muscles can occur,
especially in the adductor pollicis and the first dorsal interosseous muscle. The Allen test should be performed
to rule out ulnar artery pathology. Treatment depends
on the underlying pathology. In patients with concomitant CTS, release of the carpal tunnel is often sufficient
for decompressing the ulnar nerve. If a ganglion cyst is
suspected, surgical exploration and excision is indicated.
Fractures of the hook of the hamate are treated with excision. Ulnar artery aneurysm and thrombosis is treated

Compressive lesions of the radial nerve are relatively

rare. The five sites for compression of the radial nerve
are the fibrous bands near the radiocapitellar joint, a
vascular leash proximal to the supinator muscle (leash
of Henry), the leading edge of the extensor carpi radialis brevis, the proximal edge of the supinator muscle fascia (arcade of Frhse), and the distal edge of the supinator fascia. The most common site of compression is at
the proximal edge of the supinator muscle. Patients may
present with painless weakness of finger or thumb extension (posterior interosseous nerve compression syndrome) or may present with pain mimicking lateral epicondylitis without neurologic deficit (radial tunnel
syndrome).
Patients with posterior interosseous nerve compression syndrome generally present with insidious, painless
finger extension weakness and radial wrist deviation
(the extensor carpi radialis brevis, extensor carpi radialis
longus, and supinator are innervated above the arcade
of Frhse and therefore generally not involved). Electrodiagnostic studies will confirm the diagnosis. If there
is no improvement after 1 to 3 months of nonsurgical
treatment, surgical release of all the potential sites of
compression is indicated.
The pain of radial tunnel syndrome generally occurs
during lifting, especially with the elbow and wrist in extension. Pain associated with radial tunnel syndrome is
more distal than lateral epicondylitis (approximately
4 cm distal to the lateral epicondyle) and may be pro-

American Academy of Orthopaedic Surgeons

367

Wrist and Hand Reconstruction

Orthopaedic Knowledge Update 8

Table 4 | Best Combination of Tendon Transfers for Radial

Nerve Palsy

Table 5 | Tendon Transfer for the Treatment of Low Median

Nerve Palsy

Flexor Carpi Radialis Transfer (Starr, Brand, Tsuge)

Wrist extension: PT to extensor carpi radialis brevis
Thumb extension: PL to rerouted extensor pollicis longus
Finger extension: flexor carpi radialis to EDC

Four Reliable Opponensplasties (to abductor pollicis brevis)

Flexor digitorum sublimis of ring
Extensor index proprius
Abductor digiti mini
Used in reconstruction of the congenital hypoplastic thumb
Palmaris longus
Used in patients with long standing CTS
Provides greater abduction than opposition

Superficialis Transfer (Modified Boyes)

Wrist extension: PT to extensor carpi radialis brevis
Thumb extension: PL to extensor pollicis longus
Finger extension: flexor digitorum sublimis IV to EDC
Flexor Carpi Ulnaris Transfer (Jones)
Wrist extension: PT to extensor carpi radialis brevis
Thumb extension: PL to rerouted extensor pollicis longus
Finger extension: flexor carpi ulnaris to EDC
PT = pronator teres; PL = palmaris longus; EDC = extensor digiti communis

voked with resisted wrist extension or supination when

the elbow is in extension, or resisted middle finger extension. Electrodiagnostic tests are generally normal but
may show increased latencies in the area of compression
during provocative forearm positioning. Electrodiagnostic testing can be very useful in delineating radial tunnel
syndrome from the more common lateral epicondylitis
for which it is often confused. Injection of an anesthetic
into the radial tunnel should produce a posterior interosseous nerve palsy and alleviate pain, confirming
the diagnosis. Approaches directed anteriorly (Henrys)
or posteriorly (Thompsons) to the mobile wad provide
a wider exposure to identify the possible locale of radial
nerve compression (Figure 23).

Wartenbergs Disease
Compression of the superficial branch of the radial
nerve is known as Wartenbergs disease. Injuries to the
superficial branch of the radial sensory nerve have been
reported after tight handcuff placement and excessively
tight wrist taping. The superficial branch of the radial
sensory nerve pierces the deep fascia between the dorsal border of the brachioradialis and the extensor carpi
radialis longus muscles and then continues to travel distally in the subcutaneous plane. Direct compression or
shear stress can injure this nerve resulting in an annoying neuropathy that is difficult to treat.

Reconstruction for Nerve Paralysis

Radial Nerve Palsy
Depending on the level of nerve damage, patients with
radial nerve palsy will have functional limb deficits resulting from the loss of extension of the wrist joint and
metacarpophalangeal joints of the fingers, and from loss
of extension and radial abduction of the thumb. The inability to actively extend the wrist and to stabilize it for

368

all hand activities is probably the most significant functional deficit for patients with a radial nerve palsy. Patients with posterior interosseous nerve palsy have radial deviation with wrist extension caused by paralysis
of the extensor carpi ulnaris, and unopposed force of
the extensor carpi radialis longus and extensor radialis
brevis (innervated by the radial nerve proper). Patients
with a complete palsy of the radial nerve have a wrist
drop in addition to the loss of wrist and finger extension. Insufficient recovery of function after observation
for 6 to 12 months is an indication for tendon transfers.
Serial examinations and electromyograms are indicated
to assess recovery. If there is evidence of recovery of
function, continued observation is warranted. Available
donor muscles include all of the extrinsic muscles innervated by the median and ulnar nerves. All tendon transfers for radial nerve palsy include transferring the
pronator teres to the extensor radialis brevis for wrist
extension. Options for restoration of hand function in
radial nerve palsy include the flexor carpi radialis transfer, the superficialis transfer, and the flexor carpi ulnaris
transfer for finger extension (Table 4).

Median Nerve Palsy

Tendon transfers also may be effective treatment of low
median nerve palsy. The major deficit associated with
low median nerve palsy is the loss of thumb opposition.
Thumb opposition is a complex motion requiring trapeziometacarpal abduction, flexion, and pronation. Lack
of pronation is primarily caused by the denervation of
the abductor pollicis brevis. The goal of these tendon
transfers is to reconstruct the course of the fibers of the
abductor pollicis brevis. The vector of the abductor pollicis brevis muscle intersects the pisiform. Tendon transfers that are distal to the pisiform provide more thumb
flexion than abduction, and tendon transfers that are
proximal to the pisiform provide more thumb abduction
than flexion. Table 5 lists transfers commonly used to
treat low median nerve palsy.
Additional deficits associated with proximal (high)
medial nerve palsies include loss of flexion of the index
and long fingers as well as the interphalangeal joint of
the thumb. When the ulnar nerve is intact, side-to-side

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

transfers of the index and long finger flexor digitorum
profundus to the flexor digitorum profundus of the ring
and small fingers, with sufficient tension to restore digital cascade, is a simple and effective procedure. Restoration of thumb interphalangeal joint flexion is achieved
by transferring the brachioradialis to the flexor pollicis
longus. Occasionally, loss of active forearm rotation
from paralysis of the pronator teres and quadratus will
require biceps tendon rerouting. Reattaching the distal
biceps tendon laterally on the radius can improve active
pronation.

Ulnar Nerve Paralysis

Paralysis of the ulnar nerve distal to the innervation of
the flexor digitorum profundus results in significant
hand deformity and loss of function. With low ulnar
nerve palsy, the index and long fingers do not show
clawing because the lumbricals to these two fingers are
innervated by the median nerve, thus preventing dynamic deformity. Clawing does not occur in high ulnar
nerve palsy because the flexor digitorum profundus (as
well as the lumbricals) to the small and ring fingers is
paralyzed in this injury. Thus, there is no longer unopposed flexion of the interphalangeal joints.
Power pinch restoration can be achieved by an extensor carpi radialis brevis transfer to the adductor pollicis via a palmaris longus tendon graft. One option for
restoration of power pinch is transferring the flexor digitorum sublimis of the ring finger to the adductor pollicis. The use of the extensor carpi radialis brevis is an excellent option for restoration of power pinch because
wrist extension and power pinch are synergistic.
Transfers to correct an ulnar palsy claw deformity
may be static or dynamic. Static transfers act as tenodeses and do not require an innervated muscle. A static
distal tenodesis is accomplished by passing a free tendon graft from the ulnar lateral band of the extensor
mechanism of the ring finger around the deep transverse metacarpal ligament onto the radial lateral band
of the extensor mechanism of the small finger. Dynamic
tenodeses consist of tendon transfers (such as flexor digitorum sublimis to intrinsic muscles) that pass palmar to
the deep transverse metacarpal ligament, through the
lumbrical canals to the radial bands of the extensor
mechanism. As the wrist flexes, tension is generated, creating active metacarpophalangeal flexion with interphalangeal extension. The intact ring finger flexor digitorum
sublimis, extensor carpi radialis longus, brachioradialis,
extensor index proprius, or extensor digiti minimi may
be used.
The additional functional loss associated with high
ulnar nerve palsy when compared with a more distal lesion is the loss of ring and small finger flexion. Performing a side-to-side tendon transfer between the index and
long flexor digitorum profundus to the flexor digitorum

American Academy of Orthopaedic Surgeons

Chapter 30 Wrist and Hand Reconstruction

profundus of the ring and small fingers can restore this
functional deficit.
In treating combined palsies, the surgeon must define what muscle-tendon units are functional, what
spare units are available for transfer, and what function
is needed. The preferred transfers then can be systematically chosen.

Brachial Plexus Injuries

The etiology of injuries of the brachial plexus can be
classified as open penetrating injuries, closed or traction
injuries, injuries caused by radiation, and obstetric palsies. Traction injuries are defined by the anatomic location, either supraclavicular or infraclavicular. Supraclavicular injuries are further divided into preganglionic and
postganglionic injuries depending on where in the nervous system the avulsion has occurred. Preganglionic lesions occur proximal to the dorsal root ganglion and
have a poor prognosis because they are lesions of the
central nervous system, which lacks the capacity to regenerate. Postganglionic and infraclavicular lesions have
a better prognosis because these are lesions of the peripheral nervous system. Most traction injuries are supraclavicular.
The diagnosis depends on an accurate history and
physical examination. MRI can identify root avulsions
and the site of injury in postganglionic lesions. Electromyograms and nerve conduction velocity studies may
be used to localize lesions of the brachial plexus preoperatively, and to follow their progress postoperatively.
Within 3 weeks after an injury, changes will be noted on
an electromyogram. Nerve conduction velocity studies
with somatosensory evoked potentials can help differentiate preganglionic from postganglionic lesions.
The decision to proceed with surgical intervention is
not always straightforward and depends on the nature
of the injury. Timing of surgical intervention also is critical; repairs attempted more than 6 months after injury
have a poor prognosis. Partial nerve injuries have the
best prognosis. Nerve conductions studies should be
used to evaluate the patient at 1 month and then at
3 months after the injury. Patients who have persistent
deficits (such as lack of elbow flexion) at 3 months without evidence of improvement should be considered for
surgical exploration. When the entire plexus is involved,
the prognosis is poor, and immediate exploration may
be indicated. Nerve grafts and transfers can be performed in an attempt to circumvent nerve root avulsions and restore neurologic function. A common pattern is injury to the upper and middle trunks. This injury
pattern can be treated with nerve grafts to the suprascapular, musculocutaneous, axillary, and radial nerves
(Figure 24).
The options for surgical reconstruction include neurolysis, primary nerve repair, nerve repair using nerve

369

Orthopaedic Knowledge Update 8

Wrist and Hand Reconstruction

of the brachial plexus. Good results have also been obtained with ulnar nerve fascicle transfers to the biceps
and brachialis branches of the musculocutaneous nerve
for improved elbow flexion strength.

Annotated Bibliography
Wrist Imaging
Steinborn M, Schurmann M, Staebler A, et al: MR imaging of ulnocarpal impaction after fracture of the distal
radius. AJR Am J Roentgenol 2003;181:195-198.

Figure 24 Cable grafts are used to bypass the large zone of scar tissue in the region
of the scalene muscle just distal to the brachial plexus roots. Cable grafts are in place
in the brachial plexus (A). Close-up view of the cable nerve graft (B). (Reproduced with
permission from Trumble TE: Brachial plexus injuries, in Trumble TE (ed): Principles of
Hand Surgery and Therapy. Philadelphia, PA, 2000, p 305.)
graft, nerve transfers, tendon transfers, free muscle
transfers, and arthrodesis and/or tenodesis to stabilize
joints. Depending on the injury and the patient, one or
more of the techniques may be used in an attempt to
improve function. The first priority of reconstruction is
to reestablish elbow flexion. Without the ability to position the hand in space, hand function is severely compromised. The second goal is to stabilize the shoulder.
Because it is difficult to position the extremity in a patient with a fused shoulder, if a shoulder arthrodesis is
indicated, this procedure should be the last step in the
series of reconstructive procedures. The next set of objectives includes obtaining and maintaining wrist and
digit motion.
The most proximal muscles are more successfully
reinnervated after nerve reconstruction, because they
require less axonal input and are a shorter distance
from the site of injury. After 12 to 18 months without
neurologic input, the motor end plates in the muscle
completely degenerate and the muscle loses its ability to
be successfully reinnervated. The success rate for reinnervation of the muscles innervated by the axillary, suprascapular, and musculocutaneous nerves approaches
70% to 80%. Unfortunately, more distal muscles in the
forearm and hand have a much poorer prognosis for
reinnervation. Younger patients and patients who receive nerve grafts within the first 3 months after injury
have the best prognosis.
A recent study demonstrated reliable deltoid reconstruction for upper arm brachial plexus injury by nerve
transfer to the deltoid using the nerve to the long head
of the triceps in conjunction with spinal accessory nerve
transfer to the suprascapular nerve. The ipsilateral C7
nerve root also may be transferred with or without simultaneous transfer of the spinal accessory nerve to the
suprascapular nerve to treat C5 and C6 root avulsions

370

Ulnocarpal impaction is a common finding after distal radius fracture. MRI can detect characteristic bone marrow
changes of the lunate early after the trauma. A significant correlation exists between MRI findings and the extent of posttraumatic ulnar variance and pain levels.

Wrist Arthroscopy
Shih JT, Lee HM, Hou YT, Tan CM: Arthroscopicallyassisted reduction of intra-articular fractures and soft
tissue management of distal radius. Hand Surg 2001;6:
127-135.
Arthroscopy was used to help reduce intra-articular fracture of the distal radius and treat soft-tissue injuries in 33
acute patients. The fractures were treated by reduction under
arthroscopic control and percutaneous fixation with or without external fixation. The TFCC was torn in 18 of 33 patients
(54%). All tears were peripheral and were repaired with arthroscopic procedures. Scapholunate ligament injuries with instability of the scapholunate joint were noted in 6 patients
(18%). This injury was treated with scapholunate dbridement
and stabilization of the joint with Kirschner wires. Four patients (12%) had lunotriquetral ligament injuries; three of
these patients were treated with Kirschner wire transfixion.
Six patients (18%) had chondral fractures. All 33 patients
healed without measurable incongruity of the joint surface
and excellent or good results according to the modified Mayo
wrist score.

Wrist Arthritis Secondary to Trauma

Cohen MS, Kozin SH: Degenerative arthritis of the
wrist: Proximal row carpectomy versus scaphoid excision and four-corner arthrodesis. J Hand Surg [Am]
2001;26:94-104.
Two cohort populations of 19 patients from separate institutions who underwent either a scaphoid excision and fourcorner arthrodesis or proximal row carpectomy for scapholunate advanced collapse were compared. The results of the
study suggest that both proximal row carpectomy and
scaphoid excision and the four-corner arthrodesis are motionpreserving options for the treatment of scapholunate advanced collapse. There were minimal subjective or objective
differences in short-term follow-up evaluations.

Soejima O, Iida H, Hanamura T, Naito M: Resection of

the distal pole of the scaphoid for scaphoid nonunion

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

with radioscaphoid and intercarpal arthritis. J Hand
Surg [Am] 2003;28:591-596.
Nine patients with recalcitrant scaphoid nonunion and associated degenerative arthritis were treated by resection of the
distal scaphoid fragment. At an average follow-up of 28.6
months, these patients had less pain and improved motion and
grip strength compared with that found at their preoperative
evaluation. Radiographically, neither additional degeneration
nor progress of degenerative changes was noted after surgery
in eight of nine patients.

Osteonecrosis of the Carpus

Shin AY, Bishop AT: Pedicled vascularized bone grafts
for disorders of the carpus: Scaphoid nonunion and
Kienbcks disease. J Am Acad Orthop Surg 2002;10:
210-216.
The use of reverse-flow pedicled vascularized bone grafts
from the dorsal distal radius makes it possible to transfer bone
with a preserved circulation and viable osteoclasts and osteoblasts. The resultant primary bone healing without creeping
substitution within the dead bone is an alternative to conventional bone grafting. Recent advances in understanding the
anatomy and physiology of vascularized pedicled bone grafts
have increased their use in treating a variety of carpal maladies such as scaphoid nonunions and Kienbcks disease.

Scaphoid Nonunion and Osteonecrosis

McCallister WV, Knight J, Kaliappan R, Trumble TE:
Central placement of the screw in simulated fractures of
the scaphoid waist: A biomechanical study. J Bone Joint
Surg Am 2003;85:72-77.
This cadaveric study was designed to determine whether
central placement in the proximal fragment of the scaphoid
offers a biomechanical advantage. Eleven matched pairs of
scaphoids were removed from fresh cadaveric wrists, osteotomized and fixed with either eccentric or central placement of a
Herbert-Whipple cannulated screw. Central placement of the
screw in the proximal fragment of the scaphoid had superior
results compared with those using eccentric positioning of the
screw. Clinical efforts and techniques that facilitate central
placement of the screw in the fixation of fractures of the
scaphoid waist should be encouraged.

Steinmann SP, Bishop AT, Berger RA: Use of the 1,2 intercompartmental supraretinacular artery as a vascularized pedicle bone graft for difficult scaphoid nonunion.
J Hand Surg [Am] 2002;27:391-401.
Fourteen patients with established scaphoid nonunion
were treated with vascularized pedicle bone grafting. All nonunions healed at a mean of 11.1 weeks. Wrist motion was minimally affected by surgery. Intercarpal and scaphoid angles
were improved after surgery, particularly in patients with preoperative humpback deformity who had undergone previous
interposition grafting. Vascularized bone grafts are indicated in
proximal pole fracture nonunions, in the presence of osteonecrosis, and after conventional grafts. Radiocarpal arthritis, if
present before surgery, is a poor prognostic sign.

American Academy of Orthopaedic Surgeons

Chapter 30 Wrist and Hand Reconstruction

Trumble TE, Kearny R: Scaphoid Nonunions. Orthopaedic Knowledge Online Website. American Academy
of Orthopaedic Surgeons. Available at: http://
www5.aaos.org/oko/hand_wrist/scaphoid_nonunion/
pathophysiology/pathophysiology.cfm. Accessed October 29, 2004.
The pathophysiology, etiology, diagnosis, management options, and surgical procedures commonly used to treat
scaphoid nonunions are discussed.

Triangular Fibrocartilage Complex and the Distal

Radioulnar Joint
Adams BD, Berger RA: An anatomic reconstruction for
the distal radioulnar ligaments fro posttraumatic distal
radioulnar joint instability. J Hand Surg [Am] 2002;27:
243-251.
Fourteen patients with posttraumatic DRUJ instability
were treated with a DRUJ ligament reconstruction using a
technique that is anatomic, reproducible, and requires less dissection than other described techniques. All of the patients
had joint instability and an irreparable TFCC. Stability was restored and symptoms relieved in 12 of 14 patients. All patients
attained nearly full pronation and supination. This procedure
is effective for an unstable DRUJ when the articular surfaces
are intact and other wrist ligaments are functional. It can be
used in conjunction with a distal radius osteotomy.

Cober SR, Trumble TE: Arthroscopic repair of triangular fibrocartilage complex injuries. Orthop Clin North
Am 2001;32:279-294.
The TFCC is a functionally and anatomically intricate
group of structures located at the ulnar aspect of the wrist. Injury to this structure affects the biomechanics of the wrist and
makes functional restoration difficult. This article reviews the
anatomy, biomechanics, diagnosis, and arthroscopic treatment
of TFCC injuries.

Ditano OA, Trumble TE, Tencer AF: Biomechanical

function of the distal radioulnar and ulnocarpal wrist
ligaments. J Hand Surg [Am] 2003;28:622-627.
This study was designed to provide quantitative information about the functions of the ligaments that stabilize the
DRUJ. This joint permits the radius to rotate around a nearly
fixed ulna allowing supination and pronation of the hand. Understanding their function is important in developing procedures for reconstruction. Using a ligament tension transducer,
the tension of six ligaments of the DRUJ was determined in
nine cadaver arms in pronation and supination of the hand. In
supination of the forearm all ligaments except for the dorsal
distal radioulnar ligament had equivalent tensions, indicating
their role in stabilizing the joint to this motion. In pronation,
ligament tensions generally were lower but were distributed
over all six ligaments tested. The dorsal ulnocarpal ligament
tension was equivalent in both supination and pronation, unlike the other ligaments that had greater tensions in supination.

371

Orthopaedic Knowledge Update 8

Wrist and Hand Reconstruction

Scheker LR, Babb BA, Killion PE: Distal ulnar prosthetic replacement. Orthop Clin North Am 2001;32:365376.
The advantages and disadvantages of four prostheses are
discussed and early clinical results of a new design are presented.

Tomaino MM, Weiser RW: Combined arthroscopic

TFCC debridement and wafer resection of the distal
ulna in wrists with triangular fibrocartilage complex
tears and positive ulnar variance. J Hand Surg [Am]
2001;26:1047-1052.
TFCC dbridement with arthroscopic wafer resection was
performed in 12 patients with TFCC tears and positive ulnar
variance. Eight patients had complete pain relief; four patients
had only minimal residual symptoms. Grip strength improved.

Degenerative Arthrosis of the Thumb and Fingers

Fulton DB, Stern PJ: Trapeziometacarpal joint arthrodesis in primary osteoarthritis: a minimum two-year
follow-up. J Hand Surg [Am] 2001;26:109-114.
The authors report their retrospective review of 59 trapeziometacarpal arthrodeses with follow-up from 2 to 20 years.
Only four nonunions (7%) occurred, and only one of these
was symptomatic and required reoperation. Peritrapezial arthrosis was seen in seven patients at the last follow-up.

Nerve Compression Syndromes

Trumble TE, Diao E, Abrams RA, Gilbert-Anderson
MM: Single-portal endoscopic carpal tunnel release
compared with open release: A prospective, randomized
trial. J Bone Joint Surg Am 2002;84:1107-1115.
This prospective, randomized, multicenter study compares
open carpal tunnel release with single-portal endoscopic carpal tunnel release. The open method was performed in 95
hands in 72 patients, and the endoscopic method was performed in 97 hands in 75 patients. Follow-up evaluations with
use of validated outcome instruments and quantitative measurements of grip strength, pinch strength, and hand dexterity
were performed at 2, 4, 8, 12, 26, and 52 weeks after the surgery. Complications were identified. The cost of the procedures and the time until return to work were recorded and
compared between the groups. During the first 3 months after
surgery, the patients treated with the endoscopic method had
better validated outcomes, better subjective satisfaction scores,
as well as significantly greater grip strength, pinch strength,
and hand dexterity. The open technique resulted in greater
scar tenderness during the first 3 months after surgery as well
as a longer return to work interval. No technical problems
with respect to nerve, tendon, or artery injuries were noted in
either group. There was no significant difference in the rate of
complications or the cost of surgery between the two groups.

Reconstruction of Nerve Paralysis

Fridn J, Lieber RL: Mechanical considerations in the
design of surgical reconstructive procedures. J Biomech
2002;35:1039-1045.

372

The principles of muscle-tendon units as they relate to tendon transfers are reiterated. The importance of considering
muscle architecture and length-tension relationships when
choosing an appropriate donor is discussed. The limitations of
excursion relative to connective tissue factors as well as consequences of overstretching musculotendinous units are also delineated. The role of synergism and joint moment arm changes
are outlined.

General
Trumble TE (ed): Hand Surgery Update 3: Hand, Elbow,
& Shoulder. Rosemont, IL, American Society for Surgery of the Hand, 2003.
A comprehensive review of the cutting-edge advances as
well as core knowledge in upper extremity surgery is presented.

Trumble TE (ed): Comprehensive Review for Hand Surgery [book on CD-ROM]. Rosemont, IL, American Society for Surgery of the Hand, 2003.
An advanced review of the core concepts of hand anatomy, biomechanics, and pathology as well as diagnostic and
treatment methods relating to hand surgery are presented.

Classic Bibliography
Almquist EE: Capitate shortening in the treatment of
Kienbcks disease. Hand Clin 1993;9:505-512.
Berger RA: The ligaments of the wrist. Hand Clin 1997;
13:63-82.
Bettinger P, Linsheid R, Berger R, Cooney W, An K: An
anatomic study of the stabilizing ligaments of the trapezium and trapeziometacarpal joint. J Hand Surg [Am]
1999;24:786-798.
Cooney WP, Linscheid RL, Dobyns JH: Triangular fibrocartilage tears. J Hand Surg [Am] 1994;19:143-154.
Doi K, Hattori Y, Otsuka K, Abe Y, Yamamoto H: Intraarticular fractures of the distal aspect of the radius: Arthroscopically assisted reduction compared with open
reduction and internal fixation. J Bone Joint Surg Am
1999;81:1093-1110.
Feldon P, Terrono AL, Belsky MR: Wafer distal ulna resection for triangular fibrocartilage tears and/or ulna
impaction syndrome. J Hand Surg [Am] 1992;17:731737.
Fernandez DL: Malunion of the distal radius: Current
approach to management. Instr Course Lect 1993;42:99113.
Gilula LA: Carpal injuries: Analytic approach and case
exercises. AJR Am J Roentgenol 1979;133:503-517.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Green DP, Hotchkiss RN, Pederson WC (eds): Greens
Operative Hand Surgery, ed 4. Philadelphia, PA,
Churchill Livingston, 1999.
Kirschenbaum D, Schneider LH, Adams DC, Cody RP:
Arthroplasty of the metacarpophalangeal joints with use
of silicone-rubber implants in patients who have rheumatoid arthritis: Long-term results. J Bone Joint Surg
Am 1993;75:3-12.
Levinsohn EM, Rosen ID, Palmer AK: Wrist arthrography: Value of the three-compartment injection method.
Radiology 1991;179:231-239.
Malerich MM, Clifford J, Eaton B, Eaton R, Littler JW:
Distal scaphoid resection arthroplasty for the treatment
of degenerative arthritis secondary to scaphoid nonunion. J Hand Surg [Am] 1999;24:1196-1205.
Szabo RM, Slater RR Jr, Farver TB, Stanton DB, Sharman WK: The value of diagnostic testing in carpal tunnel syndrome. J Hand Surg [Am] 1999;24:704-714.
Trumble TE, Gilbert M, Murray LW, Smith J, Rafijah G,
McCallister WV: Displaced scaphoid fractures treated

American Academy of Orthopaedic Surgeons

Chapter 30 Wrist and Hand Reconstruction

with open reduction and internal fixation with a cannulated screw. J Bone Joint Surg Am 2000;82:633-641.
Trumble TE, Gilbert M, Vedder N: Ulnar shortening
combined with arthroscopic repairs in the delayed management of triangular fibrocartilage complex tears.
J Hand Surg [Am] 1997;22:807-813.
Trumble TE, Nyland W: Scaphoid Nonunions: Pitfalls
and Pearls. Hand Clin 2001;17:611-624.
Trumble TE, Rafijah G, Gilbert M, Allan CH, North E,
McCallister WV: Thumb trapeziometacarpal joint arthritis: partial trapeziectomy with ligament reconstruction
and interposition costochondral allograft. J Hand Surg
[Am] 2000;25:61-76.
Trumble TE, Shon FG: The physiology of nerve transplantation. Hand Clin 2000;16:105-122.
Weiss AP, Weiland AJ, Moore JR, Wilgis EF: Radial
shortening for Kienbcks disease. J Bone Joint Surg Am
1991;73:384-391.

373

Chapter

31

Biomechanics of Gait
Alberto Esquenazi, MD

Introduction
Gait analysis is a useful clinical tool and a recognized
medical procedure for evaluating and treating patients
with ambulatory impairments. It is challenging for many
physicians to achieve a clear understanding of gait analysis data and to meaningfully interpret the clinical applicability of the data to a patients impairment, disability,
or handicap. Familiarity with the complex physiologic
interactions of normal gait and movement biomechanics, functional anatomy, normal and abnormal patterns
of motor control, and with the technology used for its
assessment will contribute to better care for patients
with ambulatory difficulties.
Gait can be described as an interplay between the
two lower limbs, one in touch with the ground, producing sequential restraint and propulsion, while the other
swings freely and carries with it the forward momentum
of the body. By the age of 4 to 8 years, most healthy individuals have established a similar manner of walking
because of a common, basic anatomic and physiologic
makeup. However, because of inherent differences in
body proportions, level of coordination, motivation, and
other factors, each persons gait pattern is unique. Despite these complexities, gait patterns are highly repeatable both within a subject and between subjects; however, each person has a unique walking style.
The modern quantitative study of human locomotion dates back to the early part of the 19th century
with Muybridges sequential photographs. Inman and
associates, from the University of California, refined the
simultaneous recording of multiple muscle group activity during normal ambulation and published their findings in a textbook, which has become a classic reference
for the field.

Normal Locomotion
Based on the timing of reciprocal floor contacts, the gait
cycle can be defined as a single sequence of functions by
one limb. Each gait cycle has two basic components
the stance phase, which designates the duration of foot
contact with the ground; and the swing phase, which oc-

American Academy of Orthopaedic Surgeons

curs when the foot is in the air for limb advancement.

The stance phase can be subdivided into five subphases:
initial contact, loading response, midstance, terminal
stance, and preswing. The swing phase can be divided
into three functional subphases: initial swing, midswing,
and terminal swing (Figure 1) (Table 1).
The stance phase can alternately be subdivided into
three periods according to floor contact patterns. Both
the beginning and the end of the stance phase are the
double support period during which both feet are in
contact with the floor. Single limb support begins when
the opposite foot is lifted for the swing phase. The broad
normal distribution of the periods of floor contact during the gait cycle are 40% for the swing phase and 60%
for the stance phase (each double support time period
accounts for approximately 10% of the stance phase).
These ratios apply to individuals with normal gait patterns who are walking at a self-selected, comfortable
speed. These proportions will vary greatly with changes
in walking velocity. For example, walking more slowly
will reduce single limb support time and will increase
double limb support time. With increasing cadence
(steps per minute), the double limb support period
steadily decreases, and disappears during running.
The step period is the time measured from initial
contact in one foot to the subsequent occurrence of the
same event in the other foot. There are two steps in
each stride or gait cycle. The stride period can be defined as the time from initial ground contact of one foot
until the next ground contact for the same foot; the
stride period is normally equal for left and right strides.
Stride length is the distance covered during one stride.
The stride period is often normalized to the full gait cycle for the purpose of averaging gait parameters over
several strides both within and between subjects. The
step period is useful for identifying and measuring
asymmetry between the two sides of the body, especially
in pathologic conditions. Step length is the distance covered during one step. Side-to-side symmetry of step
length and step time is characteristic in individuals with
normal gait.

377

Biomechanics of Gait

Orthopaedic Knowledge Update 8

Figure 1 Time-elapsed pictorial depiction of the normal gait cycle with numbercoded phases as described in Table 1.
Phases of the gait cycle are shown from
initial contact to ipsilateral initial contact.

Table 1 | Phases of the Gait Cycle and Reference Points

Stance Phase

Reference Point

Initial contact
Loading response

1. The instant the foot contacts the ground

2. From flatfoot position until the opposite foot is
on the ground for swing
3. From the time the opposite foot is lifted until
the ipsilateral tibia is vertical
4. From heel rise until the opposite foot contacts
the ground (contralateral initial contact)
5. From initial contact of the opposite foot and
ends with ipsilateral toe-off
Reference Point

Midstance
Terminal stance
Preswing
Swing Phase
Initial swing

Midswing
Terminal swing

6. Begins with lift-off of the foot from the floor and

ends when the foot is aligned with the opposite
foot
7. Begins when the foot is aligned with the opposite foot and ends when the tibia is vertical
8. Begins when the tibia is vertical and ends when
the foot contacts the ground (initial contact)

Under normal conditions, a comfortable walking

speed corresponds to the speed at which the energy cost
per unit distance is optimal. Energy efficiency is dependent on unrestricted joint mobility and the precise timing and intensity of muscle action. The result of abnormal biomechanics is increased energy cost, usually with
a compensatory decrease in walking speed.

378

The kinematics (geometry of motion without regard

to the forces that cause it) of the gait cycle are organized to minimize the movement of the bodys center of
gravity in both the vertical and horizontal planes, which
results in energy-efficient movement. The average total
displacement of the center of gravity in a stride is less
than 5 cm in individuals with normal gait. Impairments
of the gait mechanism can result in decreased energy efficiency caused in part by increased excursions of the
center of gravity.
The clearance mechanisms for the swinging limb require specific coordinated events to achieve swing phase
limb-length reduction. These include knee flexion in
preswing, coordinated hip and knee flexion in the early
to midswing phases, and ankle dorsiflexion in the midswing phase. Stabilization of the pelvis is also critical
and is achieved by the hip abductors, which control the
amount of pelvic drop (tilt) in swing phase. To further
ensure limb clearance while optimizing energy efficiency, the stance limb has several mechanisms to provide sufficient clearance for the swinging limb. These include midstance ankle dorsiflexion control (to prevent
excessive forward rotation of the tibia through the
forces exerted by the ankle plantar flexors), midstance
to terminal stance knee extension, and terminal stance
heel rise (ankle plantar flexion).
To achieve the most efficient gait while maximizing
step length, controlled, coordinated leg movements must
occur. For step length to be greatest, the swinging limb
must have hip flexion that will occur in coordination
with terminal swing knee extension. At the same time

American Academy of Orthopaedic Surgeons

Chapter 31 Biomechanics of Gait

Orthopaedic Knowledge Update 8

the stance limb permits ankle dorsiflexion to control
forward progression of the tibia, while knee and hip extension and pelvic tilt and internal rotation take place.

be processed and displayed as a function of time or as a

percent of the stride period. Derived data include joint
angles, angular velocities, acceleration, and limb segment rotation.

Quantitative Gait Analysis

Visual analysis of gait is routinely done as part of the
clinical assessment of patients. This type of analysis does
not provide quantitative information and has many limitations because of the complexity and speed of the
events that occur during walking, coupled with deviations and possible compensations that occur in pathologic gait. Three primary components of quantitative
gait analysis that can be recorded are kinetics (the analysis of forces that produce motion), polyelectromyography (poly-EMG) or dynamic EMG (the analysis of muscle activity), and kinematics (the analysis of motion and
the resulting temporal and stride measurements). Gait
analysis that is useful in the clinical evaluation of patients and in choosing treatment options must provide
measured parameters that correlate with the functional
capacity of the patient, supply additional and more relevant information than clinical examination, be accurate
and repeatable, and must result from a test that does
not alter the natural performance of the patient.

Kinematics
Temporal and Spatial Descriptive Measures
To characterize gait, basic variables concerning the
temporal-spatial sequencing of stance and swing phases
can be measured (Figure 2). These data can be obtained
by measuring the distances and timing involved in the
foot-floor contacts. Temporal-spatial footfall patterns
are the end product of the total integrated locomotor
movement. Techniques to obtain these data include the
use of simple ink and paper to foot switches and other
more sophisticated measuring systems. By comparing information from the two legs, measures of symmetry can
be obtained to determine the extent of unilateral impairment.

Motion Analysis
Kinematic data provide a description of movement
without regard to the force generating it (Figure 3). Earlier techniques included photographic and cinematographic analysis. Other techniques include the use of accelerometers and electrogoniometry. Modern systems
involve the use of high-speed video recording or specialized optoelectronic apparatus in which passive (such as
retroreflective markers) or active optical sources (such
as light emitting infrared diodes) are attached to the
subject and serve as markers.
In optoelectronic systems, the spatial coordinates of
the markers are generated directly by the computer after the system has been calibrated. Kinematic information can be used to provide coordinate data, which can

American Academy of Orthopaedic Surgeons

Kinetics
Kinetic analysis involves study of the forces that develop during walking. Ground reaction forces are generally measured using a triaxial force platform. It is preferred that two platforms, placed adjacent to each other,
be used so that the forces transmitted through the contact surface for each foot can be recorded simultaneously and independently. The reaction forces are divided into their orthogonal components, and plotted as
a function of time or as a function of the stride time percentage. For comparison to standards, the measured
ground reaction forces are often normalized and reported as a percentage of body mass.
Two orthogonal components of force define a timevarying force vector. In some laboratories the vertical/
sagittal vector is displayed using laser optics or computer technology in real time and superimposed on the
image of the walking subject (Figure 4).
The magnitude of the ground reaction forces and
their relationship to anatomic joint centers are the factors that determine moments or torque about a joint,
which indicate the direction and magnitude of joint rotation. Internal forces generated by muscles, tendons,
and ligaments act to control these external forces.

Electromyographic Activation Patterns in

Human Gait
Because of the redundant relationship between muscles
about a joint, there is no unique association between a
particular joint movement and the pattern of muscle
forces giving rise to that movement. In normal locomotion, the gravitational forces are carefully controlled by
opposing muscle forces to yield a smooth and energyefficient movement pattern. The EMG signal can be
used as an indication of neurologic control over muscle
activation. Superficial muscles can be studied using
bipolar surface electrodes. For deep muscles or to differentiate between adjacent muscles, indwelling Tefloncoated wire electrodes are inserted through a hypodermic needle.
EMG timing patterns from a patient can be compared to normative data (including both a mean and
standard deviation) (Figure 5) to identify deviations
from normal EMG patterns. Because EMG patterns are
very sensitive to walking velocity, data from a patient
walking at a slow speed should not be compared with
data from an able-bodied control patient who walks at a
higher velocity and with a natural cadence.
A particular muscle may be overactive or underactive during a given segment of the gait cycle. When such

379

Biomechanics of Gait

Orthopaedic Knowledge Update 8

Figure 2 An example of gait temporal-spatial data that depicts measurement of symmetry and timing.

380

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 31 Biomechanics of Gait

Figure 3 Normal three-dimensional sagittal kinematic gait data obtained with CODA mpx30 motion tracking system (Charnwood Dynamics, Leicestershire, England). Normalized
gait cycle; 0 = initial contact, vertical line swing phase, 100 = next initial contact.

Figure 5 Representative raw EMG data for gastrocnemius during walking. Normalized
gait cycle; 0 = initial contact, vertical line swing phase; 100 = next initial contact.
Solid horizontal bar represents normative data.

Figure 4 Force line visualization system in the AP and medial lateral views obtained
using the DIGIVEC system (BTS, Milan, Italy).
deviations are observed, they should be carefully correlated with the measured kinematics. When interpreting
dynamic EMG data, it is important to distinguish between cause and effect. If there is a clinical correlation
between the EMG pattern and the observed kinematics,
then a fairly confident diagnostic conclusion may be
drawn regarding the cause of an observed gait deviation.

Clinical Applications of Gait Analysis

Pathologic gait can result from a variety of clinical conditions that can be classified into three major etiologic
categories: structural (musculoskeletal deformities such
as limb amputation), joint and soft-tissue pathology (arthritis or soft-tissue contractures), and neurologic disor-

American Academy of Orthopaedic Surgeons

ders (pathology of the peripheral and/or central nervous

system). Gait analysis can be used to evaluate the dynamic basis for an observed gait deviation, to objectively assess the impact of various treatment interventions, and to develop objective selection criteria for
different treatment options. Instrumented gait analysis
and accurate clinical diagnoses can be used to guide
specific recommendations regarding surgery, therapeutic
exercises, walking aids, and mechanical and electrical
orthoses. Examinations can be performed before and after a therapeutic intervention to help assess the effectiveness of the treatment. Such interventions would
include surgical reconstruction, application or modification of an orthosis or special shoe, pharmacologic treatments (such as chemodenervation with botulinum
toxin), realignment of a prosthetic limb, or the use of a
walking aid (such as a cane).
Careful assessment can lead to the development of
rational criteria for specific surgical interventions such
as tendon releases and transfers. For example, an
equinovarus foot posture can result from several distinct
dynamic EMG patterns (such as overactivity of the
gastrocnemius-soleus complex along with the anterior
and/or posterior tibialis) that can best be differentiated
with gait analysis and dynamic EMG. In patients with
arthritis, data from gait analysis can be used to help se-

381

Biomechanics of Gait

Orthopaedic Knowledge Update 8

Figure 6 A, Patient with swing phase equinovarus ankle foot posture second to upper motor neuron syndrome. B, Dynamic EMG confirms overactive tibialis posterior and
gastrocnemius-soleus complex during the swing phase of gait. Normalized gait cycle; 0 = initial contact, vertical line indicates swing phase; 100 = ipsilateral initial contact.

lect patients who would benefit from joint arthroplasty

and to assess the impact of such surgery on gait biomechanics. In patients with chronic neurologic impairment,
gait analysis together with selective nerve and motor
point blocks can be used to differentiate fixed contractures (static deformity) from spasticity and muscle overactivity (dynamic deformity), allowing more appropriate
treatment selection.

Pathologic Gait
Functional gait deviations may be applicable to many
conditions rather than just to a specific disease. From a
functional perspective, gait deficiencies can be categorized based on their timing with respect to the gait cycle. During the stance phase, an abnormal base of support and limb instability may make walking unsteady
and energy inefficient, and possibly painful. Inadequate
limb clearance and advancement during the swing phase
will interfere with balance and energy efficiency.

Abnormal Base of Support

The ankle/foot posture is critical in the interface with
the walking surface during the stance phase. Ankle plantar flexion, inversion or eversion, and toe flexion or ex-

382

tension can all interfere with normal gait. An inadequate base of support can result in instability of the
entire body; therefore, the correction of the abnormal
ankle/foot posture by conservative, interventional, or
surgical methods is essential.
Equinovarus deformity is one of the most common abnormal lower limb postures seen in patients with neurologic disorders. Contact with the ground occurs with the
forefoot first (with decreased or absent heel contact), resulting in the weight being borne primarily on the lateral
border of the foot, which can produce an unstable base of
support. Limited ankle dorsiflexion will prevent forward
progression of the tibia over the stationary foot, resulting
in knee hyperextension and interference with terminal
stance and preswing and loss of the propulsive phase of
gait. During the swing phase, there is a sustained plantarflexed and inverted posture of the foot resulting in difficulty with limb clearance. Results of dynamic poly-EMG
show that prolonged activation of the gastrocnemiussoleus complex is the most common cause of sustained
plantar flexion. Inversion is the result of the abnormal activities of the tibialis posterior and/or tibialis anterior in
combination with long toe flexors and the gastrocnemiussoleus complex group (Figure 6).

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 31 Biomechanics of Gait

Figure 7 Polio survivor with weak left knee extensors that require a knee ankle-foot
orthosis to stabilize the knee joint. Note line of force through the knee joint of the
orthotic device.

Figure 8 Patient with insufficient hip abductor musculature that produces a compensated gluteus medius gait. Note lateral trunk lean toward the stance limb.

An abnormal base of support also occurs in foot

drop. The use of a mechanical device (such as an anklefoot orthosis) during the stance and swing phases may
help to control the abnormal ankle posture.
Studies have shown that botulinum toxin injected
into the hyperactive musculature can provide selective
time-limited relief of spasticity. Surgical interventions
such as gastrocnemius-soleus complexsoleus lengthening, split tibialis anterior, tendon transfer, and lengthening of the tibialis posterior and of the toe flexors to attain a balanced foot posture may be necessary.

ment that occurs during the early stance phase. This abnormality may be evident in a transfemoral amputee
and can interfere with the ability to ambulate or may result in hyperextension of the knee joint as a compensatory mechanism. The dynamic poly-EMG shows shortened or uncoordinated activities of the quadriceps
musculature. Occasionally, increased activities of the
knee flexors also are found. A shoe with a soft heel, a
molded ankle-foot orthosis (set in a few degrees of
plantar flexion), or a knee-ankle-foot orthosis with posterior offset knee joints or stance phase stabilization
joints all can provide improved knee stability by positioning the ground reaction force anterior to the knee
joint center. For the above-the-knee amputee, changes
in alignment to improve knee stability, the use of an articulated prosthetic foot, or the use of a mechanical
knee lock is required (Figure 7).
Knee hyperextension during the stance phase may
occur as the result of spasticity of the ankle plantar flex-

Abnormal Limb Stability

Knee flexion or hyperextension during the early stance
phase caused by ligamentous instability, degenerative
joint disease, muscle weakness, or flaccidity can make
walking unsteady and increase the risk for falling. The
patient is unable to control the normal knee flexor mo-

American Academy of Orthopaedic Surgeons

383

Orthopaedic Knowledge Update 8

Biomechanics of Gait

Figure 9 Stiff-knee gait EMG and three-dimensional data. Overactive rectus femoris
is the cause of reduced knee flexion in swing phase when compared with normal data.
Normalized gait cycle; 0 = initial contact, vertical line indicates swing phase;
100 = ipsilateral initial contact.

ors or knee extensors, a plantar flexion contracture, or

as a compensatory mechanism for quadriceps weakness.
This abnormal posture of the knee prevents normal forward advancement of the tibia during the stance phase
and restricts contralateral limb advancement. Correction of the knee deformity can be achieved by decreasing the ankle plantar flexor or knee extensor spasticity,
if present, by the use of selected botulinum toxin injection into the gastrocnemius-soleus complex or by appropriate surgical intervention through gastrocnemiussoleus complex lengthening. Compensation for the
ankle deformity can be achieved with the use of a heel
lift to accommodate the ankle equinus. The use of a
knee-ankle-foot orthosis with offset knee joints and hy-

384

perextension control is indicated to compensate for

knee extensor weakness when present.

Trendelenburg Gait
Insufficient hip abductor musculature or mechanical deficiency of the hip joint caused by pain, degenerative
changes, malalignment, or a nerve injury can result in a
gluteus medius gait pattern (Figure 8). During the
stance phase, the patient will have an exaggerated ipsilateral trunk lean (compensated gluteus medius gait) in
an attempt to stabilize the pelvis. Some patients will be
unable to compensate and have a pelvic drop of the
swinging limb resulting in a noncompensated gluteus
medius gait. The use of a cane held by the contralateral

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

hand should result in a marked improvement in this
pathologic gait pattern.

Limb Clearance and Advancement

Limb clearance and advancement occurs during the
swing phase of gait. When limb clearance is inadequate,
limb advancement is compromised. The most common
causes of limb clearance difficulties are lack of adequate
hip flexion, and inadequate knee flexion and/or ankle
dorsiflexion. It is important to recognize that synchronized lower limb motion during the swing phase is essential to produce adequate limb clearance.
Stiff-knee gait pattern can result from orthopaedic
intervention intended to provide joint stability or relieve pain (knee fusion) or may be found in patients
with spasticity. The patient is unable to flex the knee adequately, creating a large moment of inertia, which increases the energy required to initiate the swing phase
of the involved limb. This condition requires the patient
to use ipsilateral hip, trunk, and contralateral limb compensatory motions. Even if the ankle-foot system has an
appropriate position, early swing toe drag can occur and
can be corrected only by increasing hip flexion, increasing the contralateral limb length, or generating knee
flexion.
In patients with a spastic stiff knee, a dynamic polyEMG will show increased activity in the quadriceps
muscles as a group. In patients who have had a stroke,
preferential activation of the rectus femoris and vastus
intermedius with or without hamstrings co-contraction
is found (Figure 9). Lack of momentum caused by decreased walking speed can be another possible cause for
stiff-knee gait. Quadriceps chemodenervation, selective
surgical releases, or tendon transfer may be useful treatments.
In some patients, inadequate hip flexion is also a
cause of abnormal limb clearance. This condition effectively prevents physiologic shortening of the limb, producing a swing phase toe drag. Decreased muscle
strength or delayed activation of the iliopsoas (as shown
with EMG) is a primary cause of this gait deviation.
Strengthening of the iliopsoas, if possible, or a contralateral shoe lift to facilitate limb clearance can be used to
treat this gait deviation.
Increased hip adduction can interfere with ipsilateral
and contralateral limb advancement. Increased activity
of the hip adductor musculature or imbalance in
strength between the hip abductor and adductor muscle
groups is the main cause of this type of abnormal gait.
Percutaneous phenolization of the obturator nerve or
obturator neurectomy (with or without adductor tenotomy) is the most common treatment.
Incomplete knee extension during the late swing and
early stance phases, which result from joint derangement or hamstrings spasticity, is another important gait

American Academy of Orthopaedic Surgeons

Chapter 31 Biomechanics of Gait

deviation. This condition interferes with ipsilateral limb
advancement as the knee will be flexed and the limb is
not able to easily reach the ground. Contralateral limb
clearance also will be affected as a decrease in total
functional height will occur, requiring increased hip and
knee flexion to avoid foot drag. Total knee joint replacement or hamstring release may be useful as a treatment
option.
Hip retraction or lack of pelvic rotation affecting the
involved limb during the gait cycle may interfere with
limb advancement, resulting in a shortened step length
because of limited hip motion. Total hip arthroplasty or
iliopsoas lengthening may address this gait deviation.

Annotated Bibliography
Al-Zahrani KS, Bakheit AM: A study of the gait characteristics of patients with chronic osteoarthritis of the
knee. Disabil Rehabil 2002;24:275-280.
The kinematic and kinetic parameters of gait and the pattern of activation of four lower limb muscles were examined
during walking at a self-selected pace on level ground in this
study. The spatiotemporal parameters of gait were also computed in 58 patients with severe osteoarthritis of the knee and
a control group of 25 age-matched healthy people. The patients with osteoarthritis had a significantly reduced walking
speed; shorter stride length; a more prolonged stance phase of
the gait cycle; less range of motion at the hip, knee and ankle
joints; and generated less moments and powers at the ankle
and more moments at the knee than the control group. It was
concluded that the observed gait abnormalities were caused
by instability of the knee joint in the stance phase. This finding
may have important clinical implications for the rehabilitation
of patients with severe osteoarthritis of the knee.

Esquenazi A, Mayer NH, Keenan MA: Dynamic polyelectromyography, neurolysis, and chemodenervation
with botulinum toxin A for assessment and treatment of
gait dysfunction. Adv Neurol 2001;87:321-331.
This review article describes evaluation techniques using
gait analysis and possible treatment options for patients with
gait dysfunction resulting from upper motor neuron syndrome.
Treatment interventions ranging from focal injections of botulinum toxin to surgery are described.

Fantozzi S, Benedetti MG, Leardini A, et al: Fluoroscopic and gait analysis of the functional performance in
stair ascent of two total knee replacement designs. Gait
Posture 2003;17:225-234.
This article reviews stair ascent kinematics and kinetics of
two types of knee joints (mobile bearing or posterior stabilized) using three-dimensional fluoroscopy and gait analysis
techniques. Statistical significant correlation was found between knee flexion at foot strike and the position of the midcondylar contact points and between maximum knee adduction moment and corresponding trunk tilt. Results of this

385

Biomechanics of Gait
study suggested that a combined evaluation technique is more
useful than fluoroscopic assessment of the knee alone.

Fuchs S, Tibesku CO, Frisse D, Laass H, Rosenbaum D:

Quality of life and gait after unicondylar knee prosthesis are inferior to age-matched control subjects. Am J
Phys Med Rehabil 2003;82:441-446.
A total of 17 patients were examined at an average
follow-up of 21.5 months after implantation of unicondylar
sledge knee prostheses. Patients had clinical evaluation, threedimensional gait analysis, surface EMG investigation of the
lower limb, and quality-of-life assessment using the Short
Form-36 health questionnaire. Results were compared with a
control group of 11 healthy individuals. Significantly poorer
results were found for the patient group in the Hospital for
Special Surgery score, the Knee Society score, the patella
score, and the Visual Analog Scale for pain. Significant differences also were found in the level of physical functioning, role
limitation because of physical complications, and the presence
of pain. EMG activities during gait were significantly lower in
the patient group, except for the rectus femoris and the tibialis
anterior. Gait analysis showed a significant difference between
the two groups for ground reactive forces and stride length;
maximum knee extension and flexion did not vary significantly.

Fuller DA, Keenan MA, Esquenazi A, Whyte J, Mayer

N, Fidler-Sheppard R: The impact of instrumented gait
analysis on surgical planning: Treatment of spastic
equinovarus deformity of the foot and ankle. Foot Ankle
Int 2002;23:738-743.
This article reports the results of a prospective investigation of the impact of instrumented gait analysis on surgical
planning for the treatment of equinovarus foot deformity in
patients with spasticity. The patients in the study had instrumented gait analysis and poly-EMG data collection using a
standard protocol. The agreement between the surgical plans
of two surgeons were compared before and after the gait
study. Results showed that instrumented gait analysis can produce higher agreement between surgeons in surgical planning
for patients with spastic equinovarus deformity of the foot and
ankle.

McGibbon CA, Krebs DE: Compensatory gait mechanics in patients with unilateral knee arthritis. J Rheumatol
2002;29:2410-2419.
Ankle, knee, hip, and low back mechanical energy expenditures and compensations during gait were characterized in
13 elderly patients with unilateral knee osteoarthritis and a
control group of 10 age-matched healthy people studied during preferred and paced speed gait. Patients with knee osteoarthritis had a lower (but not significantly different) walking speed and step length compared to the control group, and
had significantly different joint kinetic profiles.

386

Orthopaedic Knowledge Update 8

Schmalz T, Blumentritt S, Jarasch R: Energy expenditure
and biomechanical characteristics of lower limb amputee gait: The influence of prosthetic alignment and different prosthetic components. Gait Posture 2002;16:255263.
The influence of different prosthetic alignments and components on oxygen consumption and the important biomechanical characteristics of the normal gait pattern of 15 transtibial and 12 transfemoral amputees was studied. Oxygen
consumption while walking on a treadmill was analyzed and
biomechanical parameters during walking on even ground at a
self-selected speed were defined. It was found that variations
of the prosthetic alignment affect the energy consumption of
transfemoral amputees more significantly than transtibial amputees. All investigated variations could be clearly characterized by the sagittal moments acting on the joints of the prosthetic limb during gait.

Stevens PM, MacWilliams B, Mohr RA: Gait analysis of

stapling for genu valgum. J Pediatr Orthop 2004;24:7074.
This article evaluates the effects of stapling or epiphysiodesis of the distal medial femur as a treatment for correcting
genu valgum. Clinical improvement (in appearance, pain, and
function) and objective evidence of kinetic and kinematic improvement was shown. Preoperative and postoperative measurements in a series of patients treated for genu valgum were
compared to document the benefits of normalizing the mechanical axis. Results indicated that knee and hip angles and
knee moments were returned to the normal range (compared
with an age-matched control group) for patients treated with
surgery.

Classic Bibliography
Esquenazi A, Talaty M: Physical medicine and rehabilitation: The complete approach, in Grabois M, Garrison
SJ, Hart KA, Lehmkuhl LD (eds): Normal and Pathological Gait Analysis. New York, NY, Blackwell Science,
2000, pp 242-262.
Gage JR: Gait Analysis in Cerebral Palsy. New York,
NY, Mac Keith Press, 1991.
Inman VT, Ralston HJ, Todd F: Human Walking. Baltimore, MD, William & Wilkins, 1981.
Perry J: Gait Analysis: Normal and Pathological Function. Thorofare, NJ, Slack Inc, 1992.
Perry J, Waters RL, Perrin T: Electromyographic analysis of equinovarus following stroke. Clin Orthop
1978;131:47-53.

American Academy of Orthopaedic Surgeons

Chapter

32

Pelvis and Acetabulum: Trauma

Mark C. Reilly, MD

Pelvic Fractures
Evaluation
Fractures of the pelvic ring frequently result from highenergy injuries. The orthopaedic surgeon should be involved early in the treatment process. Patient evaluation
should begin with information from the injury scene,
and the patients hemodynamic stability assessed while
en route to the emergency department. A physical examination should identify associated integument, neurologic, urologic, and skeletal injuries. A careful evaluation
of the soft tissues surrounding the pelvis should include
an evaluation of the perineum for evidence of swelling,
laceration, or deformity. The patient should be log rolled
to allow for examination of possible open wounds or
subcutaneous degloving injuries. Rectal and vaginal examinations are mandatory and may identify lacerations
in connection with the pelvic ring injury.
Concomitant urologic injuries are present in approximately 15% of patients with pelvic fractures and are
most commonly urinary tract injuries. Physical findings
often associated with urethral injury in men are blood at
the meatus and a high-riding or excessively mobile prostate. Female patients should be examined for vaginal
wall, urethral, or labial lacerations. Hematuria, when
present, is an accurate indicator of urologic injury (particularly bladder injuries). A retrograde cystourethrogram should be done on hemodynamically stable male
patients with displaced anterior pelvic ring injuries before Foley catheter placement. In female patients, catheter placement may be performed without a urethrogram
because the urethra is short and is not often injured.
Retroperitoneal bladder ruptures are generally repaired
at the time of anterior pelvic ring fixation. If no anterior
pelvic ring surgery is performed, these ruptures may be
treated nonsurgically. Although controversy exists concerning the treatment of urethral injuries, multiple studies have shown that early endoscopic primary realignment is associated with an acceptably low rate of
intraoperative morbidity, stricture formation, impotence,
and incontinence.

American Academy of Orthopaedic Surgeons

The trauma AP pelvis radiograph is recommended

as an Advanced Trauma Life Support diagnostic adjunct
for use in the resuscitation of blunt trauma patients and
provides information about the mechanism of injury
that may contribute to the initial treatment protocol.
However, because most blunt trauma patients also will
undergo a CT scan, the utility and cost effectiveness of
the trauma AP radiograph as a resuscitative adjunct has
been questioned. In one study of awake, alert trauma
patients, physical examination was found to be as sensitive as the AP radiograph of the pelvis in identifying unstable pelvic injuries. Another large study found that
physical examination alone would not have identified a
significant number of unstable pelvic ring injuries.
Definitive radiographic evaluation of a patient with
a pelvic ring injury should include the AP radiograph of
the pelvis and the 40 caudad (inlet) and 40 cephalad
(outlet) projections. The CT scan of the pelvis, usually
obtained in conjunction with the trauma abdominal CT
scan, should include cuts of 5 mm or less through the
posterior pelvic ring. Results from the CT scan will help
determine the location and type of posterior pelvic ring
injury, identify compression of neurologic elements, and
highlight subtle ligamentous or bony injuries that may
alter the treatment plan. The lower lumbar spine can be
seen and concurrent injuries may be identified.

Classification
Systems based on the anatomic location of the injury,
mechanism of injury, or stability of the pelvic ring are
used to classify pelvic ring injuries. These classification
systems are usually used together. The anatomic classification system helps to identify all of the injured bony
and ligamentous structures. The mechanism of injury
system aids in fracture pattern recognition and assists in
the early resuscitation and treatment of the patient (Figure 1). Determining the stability of the pelvic ring can
help in the selection of the most appropriate definitive
fixation for the injury.

387

Pelvis and Acetabulum: Trauma

Orthopaedic Knowledge Update 8

Figure 1 Young-Burgess classification of mechanism of injury. A, Lateral compression, grade I. B, Lateral compression, grade II. C, Lateral compression, grade III. D, Anteriorposterior compression, grade I. E, Anterior-posterior compression, grade II. F, Anterior-posterior compression, grade III. G, Vertical shear. (Reproduced from Tornetta P III: Pelvis
and acetabulum: Trauma, in Beaty JH (ed): Orthopaedic Knowledge Update 6. Rosemont, IL, American Academy of Orthopaedic Surgeons, 1999, pp 427-439.)

Initial Treatment
Resuscitation
Pelvic ring injuries may be associated with significant
hemorrhage. The patients response to resuscitation,
which begins during the initial trauma evaluation, will
guide the overall treatment plan. Patients with pelvic
fractures often will require blood replacement in addition to receiving fluid and crystalloid. Patients presenting in shock (systolic blood pressure less than 90 mm
Hg) have mortality rates of up to 10 times of those
found in normotensive patients. In one study, the presence of shock on arrival in the emergency department
and revised trauma score were determined to be the
most useful predictors of mortality and transfusion requirement. The most common direct causes for mortality in patients with pelvic fractures are head and thorax
injuries; however, hemorrhage from pelvic fractures may
be a significant contributing factor to mortality. Hypothermia and coagulopathy frequently contribute to ongoing blood loss and should be treated aggressively if
present. Most pelvic bleeding is venous and can be controlled with mechanical stabilization, prevention of clot
disruption, and treatment of coagulopathy.
External Stabilization
If the pelvic ring is mechanically unstable, external immobilization may be indicated. Initial stabilization for
transport from the injury site may consist of sandbags,
beanbags, or military antishock trousers (MAST). All
devices must be removed for the evaluation of the
trauma patient. The MAST suit should be deflated grad-

388

ually to avoid a sudden expansion in intravascular volume and subsequent hypotension.

In the emergency department, external stabilization
may be an important factor in patient resuscitation. Circumferential pelvic antishock sheeting or the placement
of a pelvic binder is a noninvasive and rapid means for
obtaining pelvic stability and has been shown to achieve
reduction and stabilization of external rotation-type pelvic fractures. Pelvic sheeting has replaced the application of a resuscitative external fixator in many treatment centers because it avoids the delay required to
apply the frame and may be applied earlier in the
course of treatment to prevent ongoing hemorrhage.
The trauma AP radiograph of the pelvis should be reviewed to determine the fracture pattern (before application of the sheet) to ensure that the injury is not a lateral compression injury, which could be further
displaced by such treatment. When used, external fixation pins may be inserted into the medius tubercle portion of the iliac crest or just above the anterior inferior
iliac spine. Skeletal traction also may be indicated as an
additional method for temporary stabilization, particularly for fractures with cranial displacement.
Angiography
Angiography with selected embolization is useful for patients who are not responding to fluid and blood resuscitation. In one study, it was found that hemorrhage in patients with unstable pelvic fractures usually originated
from pelvic sources and was often treatable with embolization. In patients with both pelvic and abdominal
sources of bleeding, mortality was lower in the group of

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

patients treated with angiography followed by laparotomy
compared with a group first treated with laparotomy.
Stable fracture patterns also may be complicated by
arterial bleeding and may be appropriate for treatment
by selective embolization. Arterial bleeding can occur in
up to 18% of patients with mechanically stable pelvic
fractures. In one study however, an intra-abdominal
source of bleeding was found in 85% of patients who
were hemodynamically unstable but had mechanically
stable pelvic fractures.

Definitive Treatment
Most pelvic fractures are mechanically stable injuries
and are often caused by a lateral compression mechanism resulting in an anterior impaction fracture of the
sacrum and pubic rami fractures. If there is less than
1 cm of posterior pelvic ring displacement and no neurologic deficit, these injuries are appropriate for nonsurgical treatment with progressive mobilization. Repeat
radiographs should be obtained after mobilization to
ensure that there has been no further displacement.
Fractures of the processes of the pelvis, such as anterior
superior iliac spine avulsion fractures, do not disrupt the
stability of the pelvic ring and are usually treated nonsurgically unless significant displacement is present.
External Fixation
External fixation as definitive treatment is generally only
appropriate for rotationally unstable injuries. The most
common scenario involves an AP compression injury,
which results in an external rotation of one or both hemipelves. The anterior ring usually fails as a symphysis dislocation or less commonly as fractures of the pubic rami.
The posterior ring injury is incomplete. In this situation,
the external fixator may provide enough anterior stability
to allow the anterior injury to heal. In a lateral compression injury, distraction external fixation with external rotation of the injured hemipelvis has been used with success; this treatment is only required if there is neurologic
compression or unacceptable deformity. Although it may
be used in association with internal fixation for some injuries, external fixation alone is not appropriate for the
treatment of unstable posterior pelvic ring injuries. The
posterior pelvic ring injury may ultimately heal; however,
it will heal in a displaced position and may lead to pelvic
obliquity, pain, and long-term disability.
Internal Fixation
Internal fixation is the most biomechanically stable fixation for the pelvic ring. The implants are situated closer
to the site of injury than in external fixation and may be
optimally located to resist the forces applied to the pelvic ring. Achieving an accurate reduction of the pelvic
ring may be a prerequisite to achieving stable fixation.
Placement of internal fixation for the pelvis may be

American Academy of Orthopaedic Surgeons

Chapter 32 Pelvis and Acetabulum: Trauma

complicated by neurologic or vascular injury, infection,
wound complications, and nonunion or malunion of the
pelvic ring, and loss of reduction.

Anterior Pelvic Ring Injuries

Injuries to the anterior pelvic ring include symphysis
dislocations, pubic body fractures, and pubic rami fractures. These injuries may occur alone, in combination, or
in association with a posterior pelvic ring injury. All
complete dislocations of the symphysis pubis should be
stabilized. Displacements of less than 2.5 cm in conjunction with an intact posterior pelvic ring are incomplete
injuries and can be considered for nonsurgical treatment. Because radiographs of the injury may not reflect
the magnitude of the initial displacement, close
follow-up is warranted if nonsurgical treatment is selected. The symphysis dislocation is most effectively and
efficiently treated with a single plate applied superiorly
through a rectus-splitting Pfannenstiel approach. This
procedure may be done in conjunction with an emergent laparotomy or urologic surgery and adds the least
amount of additional soft-tissue disruption.
Fractures of the superior pubic ramus rarely require
stabilization; even those that occur in association with
symphyseal dislocations are usually treated nonsurgically. Pouparts, Coopers, and the inguinal ligaments
combine with the periosteum of the ramus to provide
stability for these injuries. It is believed that more than
2 cm of residual distraction or fracture gap of the ramus
after treatment of the posterior pelvic injury implies disruption of these soft tissues and fixation may be needed.
Fixation may be achieved with open reduction and plate
osteosynthesis; intramedullary screw fixation or external
fixation also has been used. If the fracture involves the
body of the pubis (medial to the pubic tubercle), the
supporting tissues mentioned above are all lateral to the
site of injury and cannot contribute to stability. Pubic
body fractures are generally treated as symphysis dislocations and require reduction and plate fixation.

Posterior Pelvic Ring Injuries

The posterior pelvic ring is the most important component in overall stability and function of the pelvis. The
anatomic site of injury will determine the surgical approach and the type of fixation used. Unstable posterior
pelvic ring injuries are ilium fractures, sacroiliac (SI)
joint dislocations, sacral fractures, or SI joint fracturedislocations. It is usually best to reduce the posterior
pelvic ring injury first, building to the intact portion of
the pelvis. This is particularly crucial if there are many
sites of pelvic displacement. If there are only two sites
of injury within the pelvic ring, reduction of the anterior
ring may facilitate reduction of the posterior pelvis in
some circumstances; however, beginning treatment with
the posterior injury is still recommended.

389

Pelvis and Acetabulum: Trauma

Orthopaedic Knowledge Update 8

Figure 2 A, Radiograph showing dislocation of the symphysis pubis and incomplete injuries to both SI joints. B, Radiograph at 2 years shows maintenance of the reduction of
the pelvic ring without evidence of SI joint arthrosis.

Ilium Fractures
Ilium fractures typically propagate from the iliac crest
to the greater sciatic notch and are unstable injuries. Although nondisplaced fractures may be treated nonsurgically, displaced fractures require reduction and fixation.
A posterior pelvic approach is useful, although some
fracture patterns may be treated through the lateral
window of the ilioinguinal approach. Fractures that involve only the iliac wing are stable injuries, are often
minimally displaced, and can be treated nonsurgically. If
significant displacement is present, open reduction and
internal fixation may be indicated. These fractures are
generally reduced and fixed through the lateral window
of the ilioinguinal approach. Iliac wing fractures have a
high incidence of local arterial injuries, bowel injury, and
soft-tissue degloving. Plate or screw fixation between
the tables of the ilium can be useful.

Sacroiliac Joint Dislocations

A complete radiographic evaluation including CT scan
is often required to differentiate between incomplete SI
joint injuries and complete SI dislocations. The anterior
pelvic ring and anterior SI ligaments may be significantly disrupted, whereas the posterior SI ligaments remain intact. These injuries are only rotationally unstable
and usually require treatment of only the anterior ring
injury (Figure 2). The posterior ring injury will reduce
indirectly and be stabilized by the anterior ring fixation.
Complete dislocations of the SI joint are vertically unstable injuries and require posterior pelvic ring fixation
in addition to anterior ring fixation (Figure 3).
Reduction and fixation of the SI joint may be done
either open (through an anterior or posterior approach)
or through closed manipulation with percutaneous fixation. The anterior approach to the SI joint allows direct

390

visualization of the cranial aspect of the joint to the pelvic brim, and the ilium can be manipulated through the
placement of clamps on the crest or through the interspinous notch. Excessive retraction or retractors placed
too medially on the sacrum may cause injury to the L5
nerve root. Fixation is achieved through the use of
plates applied with a single screw in the sacrum and
with one or two screws placed into the ilium. The use of
two plates, oriented at 90 to each other, is recommended. In the obese patient, reduction and fixation
through the anterior approach can be very difficult because of the inability to retract the abdominal contents.
Fixation also may be compromised if there is a marginal
fracture of the sacral lip of the SI joint. This fracture can
be identified on the CT scan preoperatively and may
preclude stable plate fixation from the anterior approach. Fixation also can be achieved through iliosacral
screws placed percutaneously while the patient is supine
and while the reduction is assessed and held from the
anterior approach.
Reduction and fixation is facilitated by the use of an
open reduction done through the posterior pelvic approach with the patient prone. The posterior-inferior SI
joint is visualized while the anterior joint is palpated
through the greater sciatic notch. Reduction is performed with a combination of clamps placed between
the ilium and sacrum. Fixation is achieved with the fluoroscopically guided placement of iliosacral lag screws.
The iliosacral screws are inserted through a separate incision with a percutaneous technique; it is rarely possible to insert the screws through the posterior approach
incision. If the reduction of the SI joint can be achieved
with closed manipulation and traction, the joint may be
similarly stabilized with iliosacral lag screws placed with
the patient either prone or supine.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 32 Pelvis and Acetabulum: Trauma

Figure 3 A, Patient with dislocation of the symphysis pubis, incomplete injury to the right SI joint, and complete dislocation of the left SI joint. Note the cranial displacement of
the left hemipelvis. B, Open reduction of the SI joint was performed through a posterior pelvic approach and the joint stabilized with iliosacral screw fixation. Open reduction and
internal fixation of the symphysis pubis followed.

Sacroiliac Fracture-Dislocations
SI fracture-dislocations are a combination of an iliac
fracture and an SI dislocation. The posterior superior
spine and often the posterior iliac crest remain attached
to the sacrum by the posterior SI ligaments. The remaining portion of the ilium dislocates from the sacrum as
the anterior SI ligaments rupture. Fracture-dislocations,
which leave only a small intact iliac fragment, resemble
pure SI dislocations and are treated similarly. Fracturedislocations with a large intact iliac fragment have been
termed crescent fractures and may be large enough to
maintain the integrity of the posterior SI ligaments. In
this situation, interfragmentary fixation of the ilium will
restore skeletal stability and the posterior SI ligaments
will maintain the reduction of the SI joint. If the fragment is small or the integrity of the posterior SI ligaments cannot be ensured, interfragmentary fixation
must be augmented with SI joint fixation. Generally,
rami fractures are the type of anterior ring injury seen
in association with the SI fracture-dislocation; this injury
may be treated nonsurgically if secure posterior fixation
is achieved. Closed reduction and percutaneous fixation
of SI fracture-dislocations has been reported but has
been associated with a significant incidence of fixation
failure. Outcome, as measured by patient satisfaction,
was acceptable.

Sacral Fractures
Most fractures of the sacrum are minimally displaced
and stable. Those associated with lateral compressiontype injuries are often impacted and have a negligible
incidence of subsequent displacement. Displaced and
unstable sacral fractures require reduction and fixation.

American Academy of Orthopaedic Surgeons

Open accurate reduction and internal fixation is recommended, but closed reduction and percutaneous fixation
also has been advocated. Open reduction through a posterior pelvic approach allows direct visualization of the
fracture site and sacral nerve roots. This approach allows for direct decompression of the nerve roots and visualization of the fracture during fixation to ensure that
the fracture is reduced and not overcompressed. Closed
manipulation and percutaneous fixation may increase
the risk for iatrogenic nerve injury if the fracture is not
aligned and is overcompressed. The space available for
safe placement of iliosacral screw fixation is increasingly
compromised with the increasing magnitude of malreduction. In either open or closed reduction, it is imperative to obtain an accurate reduction to ensure safe
fixation.
A subgroup of sacral fractures is the U-shaped fractures in which bilateral transforaminal sacral fractures
are connected by a transverse fracture, usually between
the second and third sacral segments. This condition
represents a complete spinopelvic dissociation and often
occurs with a sacral kyphosis and disruption of the
cauda equina at the level of the transverse sacral fracture. Percutaneous screw fixation has been used without
reduction of the kyphotic deformity; however, reduction
of the deformity and fixation with spinopelvic instrumentation is recommended. Late decompression is reserved for patients with neurologic deficits and no evidence of spontaneous recovery. Midline sagittal sacral
fractures also have been reported. These fractures are
generally vertically stable injuries and are treated with
fixation of the anterior ring injury alone and indirect reduction of the sacral fracture.

391

Orthopaedic Knowledge Update 8

Pelvis and Acetabulum: Trauma

Open Versus Closed Reduction

Outcome

Although percutaneous fixation of the pelvic ring has

been proposed, the most important factor in the treatment of pelvic ring fractures remains the reduction. The
stability of the fracture fixation and the safety of fixation placement has been shown to be compromised with
fracture malreduction. In one study, 20% of patients
with sacral fractures who were treated with closed reduction and percutaneous iliosacral screws had fixation
failures and displacement, and 13% required revision
surgery. CT-guided, computer-, fluoroscopic-, and
endoscopic-assisted insertion techniques for internal fixation have all been described but only a few of these
techniques review the reduction obtained. In one study
of closed reductions, 92% of iliac wing and SI
fracture-dislocations were reduced to within 1 cm of residual displacement. Fractures may be treated with internal fixation using many different techniques, but successful closed reduction of the pelvic ring remains a
challenge.
Closed reduction of the pelvic ring is most successful
when applied early in the postinjury period. Open reduction of posterior pelvic ring injuries has been shown
to result in accurate reductions but increases the risk of
posterior wound complications. Infection rates after
open reduction and internal fixation of the posterior
pelvic ring are reported to be 4%. Significant posterior soft-tissue injury should be considered a relative
contraindication to a formal, open posterior approach.

Outcome after pelvic fracture is more a function of the

effects of associated injuries than of the pelvic ring injury itself. At long-term follow-up after pelvic fracture,
patients reported decreased satisfaction in most of the
categories measured by the Medical Outcomes Study
Short Form-36. Pain, general health, and physical functioning are typically affected. Clinical results appear to
decline with the increasing level of instability caused by
the initial injury. In one study, functional and radiographic results were worse and mortality was higher in
patients with more unstable fracture patterns. Good and
excellent outcomes after rotationally unstable fractures
have been reported in up to 96% of patients, whereas
up to 70% of patients with vertically unstable fractures
reported acceptable results.
Associated neurologic, urologic, and lower extremity
injuries are the most common causes of long-term disability, pain, and impaired function. Resolution of neurologic dysfunction has been seen in up to 50% of patients at long-term follow-up, with the L5 nerve root
being the least likely to regain normal function.
Sexual dysfunction after pelvic fracture has been
noted in women. Dyspareunia was reported in 43% of
female patients who had more than 5 mm of residual
displacement. In another study of functional outcomes
after pelvic fracture, 44% of patients reported significant sexual dysfunction after unstable pelvic injuries. A
study of erectile dysfunction after pelvic fracture found
that 30% of male patients who were sexually active after pelvic fracture reported some degree of erectile dysfunction; those patients sustaining symphyseal disruptions reported the greatest dissatisfaction. Disruption of
the cavernosal nerves has been implicated in this complication.
Urologic injury may result in a significant compromise in patient outcome after pelvic fracture. Urethral
stricture, incontinence, and erectile dysfunction may
complicate the treatment of urethral injuries, particularly delayed perineal reconstruction. Patients who undergo early endoscopic primary realignment have been
shown to have a lower rate of incontinence and impotence when compared with patients who have delayed
open repair. Management of concurrent urologic
trauma may impact the treatment of the patients skeletal injury. In up to 35% of patients, orthopaedic treatment was altered because of urologic intervention.

Iliosacral Screw Fixation

Iliosacral screws are used in the internal fixation of SI
dislocations, fracture-dislocations, and sacral fractures.
The placement of these screws is technically demanding
and requires a thorough knowledge of the threedimensional anatomy of the posterior pelvic ring. Common bony anatomic variants such as transitional vertebrae and hypoplastic first sacral segments may
complicate or preclude the safe placement of iliosacral
screws. Screw malposition has been reported to be as
high as 13% and may be associated with serious complications. In one study, 8% of patients treated with iliosacral screw fixation had errant screws and neurologic
complications. These findings highlight the need for
careful screw placement, adequate imaging, fracture reduction, and surgeon familiarity with the procedure. It
has been suggested that the increased proprioceptive
feedback that is obtained by using an oscillating drill
rather than a threaded guide-wire may allow for safer
and more accurate screw insertion. Some surgeons recommend the use of electromyogram monitoring during
screw insertion. If this device is used, the anode must be
located at or beyond the patients midline.

392

Acetabular Fractures
Acetabular fractures are usually the result of high-energy
injuries and are frequently associated with other skeletal,
visceral, or abdominal injuries. The position of the hip at
the time of injury and the direction of impact will determine the fracture pattern. A detailed patient evaluation is
mandatory to identify life-threatening associated injuries,

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 32 Pelvis and Acetabulum: Trauma

Figure 4 The classification system of Letournel and Judet.

complications associated with the acetabular fracture, and
other skeletal injuries requiring treatment.Approximately
one half of patients with an acetabular fracture will have
an injury to another organ system.
Although hemodynamic instability occurs infrequently with isolated fractures of the acetabulum, persistent unexplained blood loss despite resuscitation may
be caused by vascular injury. Fractures involving the
greater sciatic notch may injure the superior gluteal artery, requiring angiography and selective embolization.
Neurologic injury is frequently associated with fractures
of the acetabulum and may be present in up to 20% of
patients. The peroneal division of the sciatic nerve is the
most frequently injured. Closed reduction of associated
hip dislocations should be performed as quickly as possible to reduce the risk of osteonecrosis of the femoral
head. Persistent subluxation of the hip may be caused
by either the fracture displacement or from intraarticular fracture fragments and should be treated with
urgent skeletal traction to prevent the head from wearing against the fracture edge or incarcerated fragment.
Soft-tissue degloving injuries (Morel-Lavall lesions)
may be initially recognized by a fluid wave on palpation

American Academy of Orthopaedic Surgeons

and may be later recognized by the presence of a fluctuant circumscribed area of cutaneous anesthesia and ecchymosis. These injuries should be treated with dbridement and delayed acetabular fixation because of the
significant incidence of positive bacterial culture from
these lesions.

Diagnosis and Classification

The diagnosis of acetabular fractures begins with appropriately positioned, well-penetrated, plain radiographs.
The AP pelvis radiograph and the Judet views (45 obturator and iliac oblique) are needed for accurate interpretation and fracture classification. A CT scan can better define rotational displacements, intra-articular
fragments, marginal articular impactions, and associated
femoral head injuries. A three-dimensional CT reconstruction may be helpful in understanding the relationships between multiple sites of injury, but is not a replacement for plain radiographs. The classification
system of Letournel and Judet, which groups acetabular
fractures into five elementary and five associated fracture patterns, is used to classify the fractures (Figure 4).
The interobserver and intraobserver reliability of the

393

Orthopaedic Knowledge Update 8

Pelvis and Acetabulum: Trauma

Figure 5 Measurement of the posterior and medial roof arcs as measured on the
iliac oblique (A) and the AP (B) radiographs.
classification has been found to be excellent on the basis
of plain radiographs alone; the CT scan did not improve
reliability. However, CT has been reported to be more
accurate than plain radiographs in measuring the true
magnitude of articular displacement.

Nonsurgical Treatment
Fracture displacements of greater than 3 mm are generally treated surgically. Certain fractures, however, may
be amenable to nonsurgical treatment. Roof arc measurements are a means of determining fractures with an
intact weight-bearing dome, which is defined as having
medial, anterior, and posterior roof arcs of greater than
45 as measured on the AP, obturator, and iliac oblique
radiographs. Geometric analysis has shown that the cranial 10 mm of the acetabulum on the CT scan corresponds to the area defined as the weight-bearing dome
by roof arcs (Figure 5). It has been postulated that fractures that do not involve this dome are unlikely to lead
to posttraumatic arthrosis and are candidates for nonsurgical treatment. Prerequisites for nonsurgical treatment of associated acetabulum fractures include both
intact roof arc measurements and congruence of the
femoral head to the intact acetabulum on nontraction
AP and Judet radiographs. Roof arc measurements are
not applicable to associated both-column fractures because there is no intact portion of the acetabulum to
measure. Instead, perfect secondary congruence of an
associated both-column fracture on all three standard
radiographs, taken when the patient is out of traction, is
necessary for nonsurgical treatment. Although a fracture healed with secondary congruence may have an adequate articular surface, the resultant shortening of the
limb and medialization of the hip may not be accept-

394

able. Secondary congruence alone, therefore, is necessary but not a sufficient criterion for nonsurgical treatment. The criteria also do not apply to fractures of the
posterior wall. It is believed that at least 50% to 60% of
the width of the posterior wall on the CT scan must be
intact for satisfactory clinical outcome after nonsurgical
treatment. Smaller fractures of the posterior wall may
allow hip subluxation; stress radiographs taken while
the patient is under anesthesia may be useful in determining whether surgical intervention is required.
Nonsurgical treatment is also appropriate for nondisplaced acetabulum fractures. Although it has been
suggested that percutaneous fixation of nondisplaced
fractures allows earlier mobilization of multiply injured
patients, some physicians believe that nondisplaced fractures are unlikely to displace even with early mobilization. CT or fluoroscopic-guided percutaneous fixation
remains investigational in the treatment of acetabular
fractures.
In addition to fracture location and displacement,
patient-related factors such as age, preinjury activity
level, functional demands, and medical comorbidities
must be considered when determining whether a patient
is best served by surgical or nonsurgical treatment. Nonsurgical treatment of elderly or infirm patients, with
planned subsequent arthroplasty if symptomatic arthritis develops, may be appropriateparticularly if the
fracture displacement is minimal.

Surgical Treatment
Open anatomic reduction and internal fixation is the
treatment of choice for displaced fractures of the acetabulum. The goal of surgical treatment is to obtain an
anatomic reduction of the articular surface while avoiding complications. This treatment restores the contact
area between the femoral head and the acetabulum,
produces a stable painless joint, and maximizes the potential for long-term survival of the hip (Figure 6). Clinical outcome is correlated with the quality of the articular reduction. The results of perfect reductions (less than
1 mm of residual displacement) are superior to those of
imperfect (1 to 3 mm) and poor (greater than 3 mm) reductions at long-term follow-up. Other factors associated with poor outcomes are femoral head injuries and
postoperative complications.

Surgical Approach
The choice of surgical approach is determined by the fracture pattern. A single surgical approach is generally selected with the expectation that the fracture reduction and
fixation can be completely performed though the one approach.The most commonly used surgical approaches are
the Kocher-Langenbeck and the ilioinguinal approaches.
The extended iliofemoral approach is an extensile approach developed to allow maximal simultaneous access

American Academy of Orthopaedic Surgeons

Chapter 32 Pelvis and Acetabulum: Trauma

Orthopaedic Knowledge Update 8

Figure 6 A, Associated transtectal transverse plus posterior wall acetabular fracture in an 18-year-old woman. B, AP and Judet radiographs at 3 years after injury. The patients
hip is rated 6,6,6 on the modified DAubigne and Postel scale.

to both columns of the acetabulum. It is most often used

in associated fracture patterns that are surgically treated
more than 21 days after injury or on certain transverse or
both-column pattern fractures with complicating features
that are not amenable to treatment by either of the two
more limited approaches. Modifications of all three approaches have been described; however, long-term clinical outcomes of large numbers of patients are not available for comparison.
Closed reduction and percutaneous screw fixation of
acetabular fractures has not been shown to achieve
comparable articular reductions when compared with
traditional surgery. Because there is a strong correlation
between accuracy of reduction and clinical outcome,
such techniques should be considered investigational
until appropriate results and long-term outcome data
are available.

Posterior Wall Fractures

Posterior wall fractures are the most common type of acetabular fracture. Although they account for nearly one
third of all acetabular fractures, patients with this type of
fracture have a disproportionate number of poor outcomes. Suboptimal outcomes have been reported in up to
32% of patients with these injuries despite perfect reductions in 92% to 100% of the fractures. Risk factors for
poor outcome are delay in reduction of associated hip dislocation, age older than 55 years at the time of injury,
intra-articular comminution, and osteonecrosis. When
postoperative CT is added to the routine evaluation of the
posterior wall fracture, however, the association between
fracture reduction and clinical outcome is reinforced. The
CT scan may be more accurate (particularly in the evaluation of multifragmentary posterior wall fractures) in determining true residual fracture displacement and identifying small articular malreductions.Articular reduction of
the posterior wall fracture as measured by CT was found
to strongly correlate with long-term outcome.

American Academy of Orthopaedic Surgeons

Complications
The primary complication after fracture of the acetabulum is posttraumatic arthrosis. Although symptomatic
arthritis after acetabular fracture is generally treated
with arthroplasty, arthrodesis and osteotomy remain viable treatment options. Posttraumatic arthritis is more
common after poor articular reductions than after a perfect reduction. Evidence shows that, if arthritis develops
after a perfect reduction, the onset is later and the progression slower than arthritis that develops after a poor
reduction.
Heterotopic ossification is related to the degree of
soft-tissue disruption, from either the injury or the surgical approach. Other factors associated with the formation of heterotopic ossification include head injury, prolonged mechanical ventilation, and male gender. Use of
an extensile approach also contributes to the formation
of heterotopic ossification and is probably caused by the
amount of muscle dissection and elevation from the ilium. Most patients who develop heterotopic ossification
after acetabular fracture do not have functional restrictions of their hip motion. Prophylactic treatments for
heterotopic ossification include 6 weeks of indomethacin use, single-dose external beam radiotherapy, or a
combination of both treatments. In a direct comparison
of irradiation with indomethacin use, no difference was
shown in the development of heterotopic bone. In the
same study, 38% of the patients who were not treated
with prophylaxis developed clinically significant heterotopic ossification when compared with 7% in those patients who received some form of prophylaxis. Other
prospective randomized studies have failed to confirm
the efficacy of indomethacin use compared with no prophylaxis. Because of concerns about the use of irradiation in young adults, prophylaxis with indomethacin is
preferred by many physicians. One study, however,
showed an increased incidence of long bone nonunion

395

Pelvis and Acetabulum: Trauma

in patients treated with indomethacin for concurrent acetabular or pelvic fractures.
Deep venous thrombosis and pulmonary embolism
are common complications after pelvic or acetabular
fractures treated without prophylaxis. Chemoprophylaxis with low molecular weight heparin or warfarin sodium may reduce the incidence of thromboembolic disease, particularly in association with mechanical
prophylaxis. Duplex ultrasound is typically used preoperatively to identify patients with venous thrombosis,
however, it is limited in its ability to detect proximal
thrombi. Despite earlier studies that suggested that
magnetic resonance venography was more sensitive
than ultrasound at detecting proximal thrombi, an evaluation of contrast CT and magnetic resonance venography showed a significant false positive rate in both studies. This finding was confirmed by invasive contrast
venography. If thrombi are present, the placement of an
inferior vena caval filter is recommended before fracture surgery.
Iatrogenic neurologic injury has been reported in
2% to 15% of patients who were surgically treated for
acetabular fractures. Intraoperative neurologic monitoring has been recommended but there is no evidence
that the routine use of monitoring lowers the incidence
of iatrogenic injury. A direct comparison of monitored
and nonmonitored acetabular fracture surgeries used
both sensory and motor pathway monitoring but failed
to show a difference in the incidence of iatrogenic nerve
injury. Most orthopaedic trauma surgeons use prophylaxis to help prevent heterotopic ossification and deep
venous thrombosis; however, nerve monitoring is not
routinely used.

Annotated Bibliography
Pelvic Fractures
Eastridge BJ, Starr A, Minei JP, OKeefe GE, Scalea
TM: The importance of fracture pattern in giving therapeutic decision-making in patients with hemorrhagic
shock and pelvic ring disruptions. J Trauma 2002;53:446451.
In 86 patients with pelvic fracture with persistent hemodynamic instability, abdominal hemorrhage was responsible for
hypotension in 85% of stable pelvic fractures. Hemorrhage
was from pelvic sources in 59% of patients with unstable fracture patterns. Patients with unstable fracture patterns had a
higher mortality (60%) when celiotomy was performed before
angiography when compared with patients in which angiography was performed first (25% mortality).

Gonzalez RP, Fried PQ, Bukhalo M: The utility of clinical examination in screening for pelvic fractures in blunt
trauma. J Am Coll Surg 2002;194:121-125.
In this study, 2,176 trauma patients were evaluated.
Ninety-seven patients (4.5%) were diagnosed with a pelvic
fracture. Clinical examination was found to be as sensitive as

396

Orthopaedic Knowledge Update 8

the AP pelvic radiograph at identifying pelvic fracture in the
awake, alert trauma patient.

Griffin DR, Starr AJ, Reinert CM, Jones AL, Whitlock

S: Vertically unstable pelvic fractures fixed with percutaneous iliosacral screws: Does posterior injury pattern
predict fixation failure? J Orthop Trauma 2003;17:399405.
Sixty-two patients were treated with closed reduction and
percutaneous iliosacral screw fixation for posterior pelvic ring
injuries. Results show that sacral fractures were significantly
more likely to displace than other posterior lesions. Twenty
percent (6 of 30) of sacral fractures displaced and 67% of
those required revision fixation surgery.

Kabak S, Halicik M: Tuncel M Avsarogullari L, Baktir

A, Bastruk M: Functional outcome of open reduction
and internal fixation for completely unstable pelvic ring
fractures (type C): A report of 40 cases. J Orthop
Trauma 2003;17:555-562.
This study involved 40 patients with type C pelvic fractures. At 1-year follow-up, sexual dysfunction was found in
44% of patients and correlated with anxiety disorder and major or moderate depression. Seventy-two percent of patients
had returned to work at their original jobs, and of those who
did not, an increased incidence of depression was found. Persistent pelvic pain was reported by 25% of patients.

Mayher BE, Guyton JL, Gingrich JR: Impact of urethral

injury management on the treatment and outcome of
concurrent pelvic fractures. Urology 2001;57:439-442.
In 61 patients with combined pelvic and lower urinary
tract injuries, urologic treatment affected the orthopaedic
treatment choices in 35% of patients. Long-term suprapubic
catheterization precluded surgical treatment in four patients;
three had poor results. The authors recommend early endoscopic realignment for urethral injuries and improved communication and cooperation between subspecialty groups.

Miller MT, Pasquale MD, Bromberg WJ, Wasser TE,

Cox J: Not so FAST. J Trauma 2003;54:52-59.
In this study, 359 patients with blunt abdominal injury
were evaluated with focused assessment with sonography for
trauma (FAST) and contrast CT of the abdomen and pelvis.
FAST resulted in an underdiagnosis of intra-abdominal injury
with a false negative rate of 6% (27% of the false negatives
required laparotomy). The authors recommended that CT
with contrast remain the gold standard for the evaluation of
patients with suspected blunt abdominal injury.

Moudouni S, Tazi K, Koutani A, Ibn Attya A, Hachimi

M, Lakrissa A: Comparative results of the treatment of
post-traumatic rupture of the membranous urethra with
endoscopic realignment and surgery. Prog Urol 2001;11:
56-61.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

In this study, 29 men with urethral injuries had primary endoscopic urethral realignment. At an average follow-up of 5.6
years, improved rates of continence, potency, and stricture formation were found when compared with rates for patients
who had delayed open urethroplasty. The authors recommended early realignment as an effective and safe technique.

Pehle B, Nast-Kolb D, Oberbeck R, Waydhas C, Ruchholtz S: Significance of physical examination and radiography of the pelvis during treatment in the shock emergency room. Unfallchirurg 2003;106:642-648.
In this study, 979 blunt trauma patients were evaluated for
pelvic instability. Physical examination alone had a sensitivity
of 44% and a specificity of 99% for detecting pelvic fracture.
Surgically significant pelvic injury could not be reliably ruled
out by examination alone.

Reilly MC, Bono CM, Litkouhi B, Sirkin M, Behrens

FF: The effect of sacral fracture malreduction on the
safe placement of iliosacral screws. J Orthop Trauma
2003;17:88-94.
In a cadaveric model of a zone 2 sacral fracture, increasing
cranial displacement of the hemipelvis was found to correlate
with a decrease in the space available for the safe placement
of iliosacral screws. Space available for safe screw placement
was insufficient at displacements greater than 1 cm.

Rommens PM, Hessmann MH: Staged reconstruction of

pelvic ring disruption: Differences in morbidity, mortality, radiologic results, and functional outcome between
B1, B2/B3, and C-type lesions. J Orthop Trauma 2002;16:
92-98.
A review of functional outcome of 122 patients with surgically treated pelvic ring injuries was done. Mortality was higher
in patients with type C injuries than in those with type B
(15% versus 5%). Higher rates of anatomic reductions were
found in B1 (open-book) injuries than in lateral compression
injuries (B2, B3) or C-type injuries. Good or excellent outcomes were obtained in 74% of patients with B1 injuries, 92%
with B2/B3 injuries, and 71% of those with C injuries.

Starr AJ, Griffin DR, Reinert CM, et al: Pelvic ring disruptions: Prediction of associated injuries, transfusion
requirement, pelvic arteriography, complications, and
mortality. J Orthop Trauma 2002;16:553-561.
In a review of 325 trauma patients with pelvic ring injury,
the presence of shock on arrival in the emergency department
was associated with increased mortality, transfusion requirements, and injury severity score. Mortality of patients presenting in shock was 57%. The authors were unable to identify an
association between fracture classification and outcome or
fracture presence and/or type of associated injuries.

Acetabular Fractures
Beaule PE, Dorey FJ, Matta JM: Letournel classification
for acetabular fractures: assessment of interobserver

American Academy of Orthopaedic Surgeons

Chapter 32 Pelvis and Acetabulum: Trauma

and intraobserver reliability. J Bone Joint Surg Am 2003;
85-A:1704-1709.
The intraobserver and interobserver reliability of the
Letournel-Judet classification system for acetabular fractures
was assessed by surgeons who studied with Letournel, surgeons who specialize in acetabular fracture surgery, and general orthopaedic trauma surgeons. The reliability of the classification was excellent in the first two groups. Use of the CT
scan in addition to the plain radiographs did not increase the
reliability of the classification system.

Burd TA, Hughes MS, Anglen JO: Heterotopic ossification prophylaxis with indomethacin increase the risk of
long-bone nonunion. J Bone Joint Surg Br 2003;85:700705.
Patients receiving indomethacin for heterotopic ossification prophylaxis were compared with those receiving external
radiation therapy. The 38 patients receiving indomethacin had
a statistically significant increase in the incidence of long bone
fracture nonunion compared with the 38 patients receiving external radiation therapy (26% versus 7%). No difference in
the efficacy of both methods of prophylaxis was found in the
authors previous study.

Burd TA, Lowry KJ, Anglen JO: Indomethacin compared with localized irradiation for the prevention of
heterotopic ossification following surgical treatment of
acetabular fractures. J Bone Joint Surg Am 2001;83:
1783-1788.
In this study, 166 patients were treated surgically for a
fracture of the acetabulum. Seventy-eight patients received external beam radiotherapy, 72 received 6 weeks of indomethacin, and 16 patients received no prophylaxis. Grade 3 or 4 heterotopic ossification developed in 7% of the treated groups
and 38% of the untreated group. No difference between the
two treated groups was identified.

Haidukewych GJ, Scaduto J, Herscovici D Jr, Sanders

RW, DiPasquale T: Iatrogenic nerve injury in acetabular
fracture surgery: a comparison of monitored and unmonitored procedures. J Orthop Trauma 2002;16:297301.
This article is a retrospective review of acetabular fracture
surgery performed with and without somatosensory evoked
potential or electromyography nerve monitoring. The use of
intraoperative nerve monitoring did not decrease the rate of
iatrogenic nerve palsy. Seven of 10 iatrogenic nerve injuries in
the monitored group had previously had normal intraoperative monitoring.

Moed BR, Carr SE, Gruson KI, Watson JT, Craig JG:
Computed tomographic assessment of fractures of the
posterior wall of the acetabulum after operative treatment. J Bone Joint Surg Am 2003;85-A:512-522.
In this study, 67 patients with surgically treated posterior
wall acetabular fractures were evaluated for radiographic and
functional outcome at a mean of 4 years after injury. Use of

397

Pelvis and Acetabulum: Trauma

postoperative CT scanning to document reduction revealed a
strong correlation between quality of reduction and functional
outcome. Residual displacement after reduction of posterior
wall fractures was more accurately determined on the CT scan
than on the plain radiographs.

Moed BR: WillsonCarr SE, Watson JT: Results of operative treatment of fractures of the posterior wall of the
acetabulum. J Bone Joint Surg Am 2002;84-A:752-758.
In the largest published study to date, the authors present
the results of surgical treatment of 100 patients with posterior
wall acetabular fractures at a mean follow-up of 5 years after
injury. Good or excellent results were obtained in 89% of patients. Risk factors for unsatisfactory outcome were a delay in
reduction of hip dislocation of greater than 12 hours, age older
than 55 years, the presence of intra-articular comminution, and
the development of osteonecrosis.

Stover MD, Morgan SJ, Bosse MJ, et al: Prospective

comparison of contrast-enhanced computed tomography versus magnetic resonance venography in the detection of occult deep pelvic vein thrombosis in patients
with pelvic and acetabular fractures. J Orthop Trauma
2002;16:613-621.
A prospective comparison of magnetic resonance venography and CT venography as screening examinations for pelvic
deep venous thrombosis was performed. Invasive contrast
venography was used as the confirmatory study. The false positive rate for magnetic resonance venography was 100% and
for CT venography was 50%. The authors question the use of
either test as the sole means of screening for pelvic deep
venous thrombosis after pelvic or acetabular fracture.

Classic Bibliography
Bucholz RW: The pathological anatomy of Malgaigne
fracture-dislocations of the pelvis. J Bone Joint Surg Am
1981;63:400-404.
Copeland CE, Bosse MJ, McCarthy ML, et al: Effect of
trauma and pelvic fracture on female genitourinary, sex-

398

Orthopaedic Knowledge Update 8

ual, and reproductive function. J Orthop Trauma 1997;
11:73-81.
Dalal SA, Burgess AR, Siegal JH, et al: Pelvic fracture
in multiple trauma: Classification by mechanism is the
key to pattern of organ injury, resuscitative requirements, and outcome. J Trauma 1989;29:981-1000.
Kellam JF, McMurtry RY, Paley D, Tile M: The unstable
pelvic fracture: Operative treatment. Orthop Clin North
Am 1987;18:25-41.
Letournel E: Acetabular fracture: Classification and
management. Clin Orthop 1980;151:81-106.
Letournel E, Judet R: Fractures of the Acetabulum, ed 2.
Berlin, Germany, Springer-Verlag, 1993.
Matta JM: Fractures of the acetabulum: Accuracy of reduction and clinical results in-patients managed operatively within three weeks after the injury. J Bone Joint
Surg Am 1996;78:1632-1645.
Matta JM, Anderson LM, Epstein HC, Hendricks P:
Fractures of the acetabulum: A retrospective analysis.
Clin Orthop 1986;205:230-240.
Matta JM, Siebenrock KA: Does indomethacin reduces
heterotopic bone formation after operations for acetabular fractures?: A prospective randomized study. J Bone
Joint Surg Br 1997;79:959-963.
Routt ML Jr, Simonian PT, Mills WJ: Iliosacral screw
fixation: Early complications of the percutaneous technique. J Orthop Trauma 1997;11:584-589.
Slatis P, Huittinen VM: Double vertical fractures of the
pelvis: A report on 163 patients. Acta Chir Scand 1972;
138:799-807.
Tornetta P III, Matta JM: Outcome of operatively
treated unstable posterior pelvic ring disruptions. Clin
Orthop 1996;329:186-193.

American Academy of Orthopaedic Surgeons

Chapter

33

Hip: Trauma
George J. Haidukewych, MD
David J. Jacofsky, MD

Introduction
Fractures and dislocations around the hip remain
among the most common injuries and are challenging to
treat. With the ever-growing elderly population with osteopenic bone, the number of fractures continues to increase proportionately. Additionally, younger patients
may sustain various fractures and dislocations around
the hip as a result of high-energy trauma; these injuries
can threaten the vascularity of the femoral head and the
long-term prognosis of the hip joint.

Hip Dislocations
Hip dislocations typically result from high-energy
trauma, such as a motor vehicle accident. Associated injuries are common, and have been reported in more
than 70% of patients. Hip dislocations are generally
classified as either anterior or posterior. A postreduction 3-mm-cut CT scan is mandatory, even if plain films
appear normal. Small osteochondral intra-articular fragments and acetabular and proximal femoral fractures
must be excluded by CT scan after closed reduction. If
closed reduction cannot be achieved, a CT scan obtained before surgery may guide the surgeon in selecting the surgical approach and evaluating appropriate
treatment of associated fractures.

Anterior Hip Dislocations

Anterior hip dislocations are extremely rare. They can
be subdivided as superior or inferior in relationship to
the pubic ramus; however, treatment is the same. The
patient may present with the extremity held in an abducted externally rotated figure-of-4 position after a
high-energy injury. Femoral head fractures are commonly associated with anterior hip dislocations and can
involve impaction or osteochondral injury of the femoral head on the acetabular rim. Closed reduction requires complete muscle relaxation, traction, extension,
and gentle internal rotation. If closed reduction is unsuccessful, then open reduction should be performed,
usually through an anterior approach. Forceful closed
reduction attempts are not recommended because of

American Academy of Orthopaedic Surgeons

the possibility of iatrogenic femoral neck or acetabular

fracture.

Posterior Hip Dislocations

Posterior hip dislocations account for most hip dislocations (approximately 90%). These injuries are typically
caused by an axial force to the flexed hip, such as the
hip striking a dashboard. Bone quality and the position
of the limb at the time of impact determines whether an
associated acetabular fracture or a simple dislocation
occurs. Traumatic posterior hip subluxations, without
dislocation, have also been reported with the pathognomonic MRI findings of iliofemoral ligament rupture, hemarthrosis, and marginal posterior acetabular wall avulsion fracture. The treatment of isolated posterior
dislocations involves emergent closed reduction or, if
necessary, open reduction. Open reduction of irreducible posterior dislocations usually proceeds through the
Kocher-Langenbeck approach. Identification and careful protection of the sciatic nerve is recommended because it may be tented by the displaced femoral head.
Dislocations with associated posterior wall fractures are
treated as indicated based on fragment size, displacement, and hip stability. Postreduction treatment of a
simple hip dislocation, regardless of direction, involves
early mobilization with gait aids as needed for patient
comfort. Patients should avoid hyperflexion of the hip
for posterior dislocations and extension and external rotation for anterior dislocations.

Complications
A substantial subset of patients will remain persistently
symptomatic after treatment of hip dislocation; however, good to excellent results have been reported in
about 70% of patients. Posttraumatic arthritis has been
reported in more than 15% of patients in several longterm studies. Osteonecrosis of the femoral head can occur in approximately 10% of hip dislocations. The risk
of osteonecrosis increases with the presence of an associated fracture of the acetabulum, probably because of
the more extensive soft-tissue injury. Osteonecrosis has

399

Orthopaedic Knowledge Update 8

Hip: Trauma
also been reported to occur after traumatic hip subluxations. Early reduction of simple dislocations and
fracture-dislocations has been suggested to lower the
rate of osteonecrosis. Sciatic nerve injuries, most commonly associated with posterior dislocations, have been
documented in as many as 8% to 19% of patients.

Femoral Head Fractures

Femoral head fractures have been categorized by Pipkin
into four types based on location of the fracture fragment in relation to the fovea centralis and presence of
associated fractures. Type 1 fractures are inferior to the
fovea, type 2 fractures are superior to the fovea, type 3
also involve a femoral neck fracture, and type 4 also involve a fracture of the acetabulum. Patients with type 1,
type 2, and type 4 femoral head fractures should undergo emergent closed reduction of the hip dislocation
with postreduction CT scanning to evaluate fracture displacement. In general, femoral head fractures are
treated based on fragment location, size, displacement,
and hip stability. A nondisplaced or minimally displaced
Pipkin type 1 fracture can be managed nonsurgically.
Simple excision of a small or comminuted displaced Pipkin type 1 fracture can usually be done because this
type of fracture is located below the weight-bearing
dome of the femoral head; larger fragments may require
surgical fixation. Pipkin type 2 fractures require accurate anatomic reduction and stable internal fixation. Titanium countersunk screw fixation is preferred to allow
subsequent MRI if needed. The anterior Smith-Peterson
approach is generally preferred because it provides improved visualization for reduction and internal fixation
and a lower complication rate as compared with a posterior approach. A trochanteric flip approach may be
added. Pipkin type 3 fractures are extremely rare and
usually occur in younger patients. These fractures should
be treated with internal fixation of the femoral neck and
femoral head fracture. In older patients with poor bone
quality and low functional demands, prosthetic replacement is probably more predictable and is generally preferred. Pipkin type 4 fractures are treated based on the
location of the femoral head fracture and the type of associated acetabular fracture. The most common clinical
scenario is a posterior wall acetabular fracture associated with a small, displaced, inferior (infrafoveal) femoral head fracture. This combination of injuries may be
treated through a Kocher-Langenbeck approach with
excision of the inferior femoral head fragment and simultaneous internal fixation of the posterior acetabular
wall fracture. A larger (suprafoveal) femoral head fracture in this situation may require an anterior exposure,
for femoral head fracture fixation, and a posterior exposure for posterior acetabular wall fixation or the use of
an extensile approach.

400

Hip Fractures
General Considerations and Risk Factors
Although hip fractures typically occur in elderly, osteopenic patients, often after a low-energy fall, these injuries also occur in younger active patients, usually as a
result of high-energy trauma. The number of hip fractures that occur annually continues to rise in proportion
to the increasing elderly population. Decision making
regarding treatment is based on fracture pattern, patient
age, associated injuries, and medical comorbidities.

Clinical and Radiographic Evaluation

In the frail, elderly population with multiple medical comorbidities, preoperative medical evaluation and optimization is important, along with attention to associated
injuries that often involve the ipsilateral upper extremity. Elderly patients with hip fractures should be brought
to surgery as soon as medically optimal; the benefits of
early mobilization cannot be overemphasized. In the
younger patient population, life-threatening injuries, if
present, should be managed first, and then the hip fracture should be treated in an urgent fashion.
Most hip fractures will be evident on AP and lateral
radiographs. However, occult or stress fractures of the
femoral neck may require additional imaging studies for
diagnosis. MRI can be helpful in not only quickly determining whether a fracture is present but also ruling out
other potential causes of hip pain, such as pubic ramus
fractures or osteonecrosis. MRI can also provide information about fracture location (femoral neck or intertrochanteric) and fracture verticality. Radionuclide bone
scanning can also be useful in this setting, especially if
symptoms have been present for several days; however,
it cannot provide as much information as MRI.

Femoral Neck Fractures

Classification
Multiple classification schemes for femoral neck fractures exist; the Garden classification is the most common. Although the Garden classification comprises four
types, most surgeons group these fractures into those
that are displaced (stage III and IV) or nondisplaced
(stage I and II) because treatment decisions and prognosis are also grouped in this manner. The Orthopaedic
Trauma Association classification has also been used.
Although more cumbersome, it distinguishes subcapital,
transcervical, and basicervical fractures and also takes
into account other potentially important variables such
as fracture verticality. The Pauwels classification has divided femoral neck fractures based on increasing
amounts of fracture verticality measured from the horizontal as type 1, less than 30; type 2, 30 to 50; and
type 3, more than 50. These fractures are believed to
behave in biomechanically distinct ways based on the
increase in shear forces imparted at the fracture site by

American Academy of Orthopaedic Surgeons

Chapter 33 Hip: Trauma

Orthopaedic Knowledge Update 8

Figure 1 AP view (A), and lateral view (B) of a valgus impacted femoral neck fracture treated with three cannulated cancellous screws.
increasing degrees of fracture verticality. Some authors
have recommended the use of a fixed angle internal fixation device for higher shear angle (more vertical) transcervical and basicervical fractures, and cancellous
screws alone for fractures with lower shear angle (more
horizontal) transcervical fractures based on this theoretical concern.

Nondisplaced Femoral Neck Fractures

Treatment of nondisplaced fractures is the same regardless of patient age, with internal fixation typically involving multiple parallel cannulated cancellous screws.
Nonsurgical treatment is reserved for only the most
frail, essentially nonambulatory patients with prohibitive medical comorbidity and minimal discomfort from
the injury. Subsequent displacement of nondisplaced or
valgus impacted fractures treated nonsurgically has
been demonstrated in nearly 40% of patients in a recent
series, accompanied by increased rates of osteonecrosis
and nonunion. Recent cadaveric data demonstrated superiority of three screws over two screws when used to
stabilize subcapital fractures in an inverted triangular
configuration. It is important to place the screws near
the cortex of the femoral neck to allow host bone to
support the shafts of the screws and avoid varus, shortening, and external rotation displacement (Figure 1).
Rates of nonunion of less than 5% and osteonecrosis of
less than 10% have been reported with this technique.
Early postoperative mobilization is encouraged.

Displaced Femoral Neck Fractures: Young Patients

A displaced femoral neck fracture in the young patient
should be treated expeditiously. In theory, fracture dis-

American Academy of Orthopaedic Surgeons

placement can kink any vessels that have not been disrupted by the injury. Additionally, intracapsular tamponade can occur because of fracture hematoma that may
impede blood flow to the femoral head. Clinical data
have documented lower rates of osteonecrosis with
early treatment. A gentle closed reduction attempt is
reasonable; however, multiple aggressive attempts at
closed reduction are not indicated because of potential
damage to the remaining femoral head vascularity or
potential fracture comminution. If closed reduction is
excellent, internal fixation should be performed as fracture pattern (verticality) dictates. If closed reduction
cannot be achieved, then open reduction is indicated.
No generally accepted guidelines exist on what constitutes an acceptable femoral neck fracture reduction. In
general, anatomic reduction is recommended. A slight
valgus reduction is acceptable; however, any varus
should be avoided. Review of a radiograph of the contralateral hip can assist the surgeon in determining the
neck-shaft angle that is anatomic for the patient. Typically the Watson-Jones or Hardinge approaches allow
direct visualization of the fracture fragments, anatomic
reduction, and internal fixation, usually with multiple
parallel cannulated cancellous screws. If the fracture exhibits high verticality and a tendency to shear intraoperatively, screws alone are not recommended and a fixed
angle device should be used. In one series, transcervical
shear fractures exhibited a high failure rate when
treated with screws alone. More data are needed to determine the ideal fixation device for the vertical femoral
neck fracture.
The role of capsulotomy in the treatment of femoral
neck fractures remains controversial. Original fracture

401

Hip: Trauma

Figure 2 Cemented bipolar hemiarthroplasty.

displacement probably determines the fate of the femoral head. Capsulotomy is recommended because it is relatively simple to perform, exposes the patient to minimal additional risk, and may reduce the intracapsular
tamponade effect. In the young patient, efforts are focused on preservation of the femoral head and avoiding
arthroplasty at a young age. More data are needed to
clearly determine if capsulotomy significantly alters the
prognosis for young patients with displaced femoral
neck fractures.

Displaced Femoral Neck Fractures: Older Patients

Prosthetic replacement has been favored in the United
States for the treatment of displaced femoral neck fractures in older patients because of the challenges of
achieving stable proximal fragment fixation in osteopenic bone, the need for a predictable surgery with a
low fixation failure rate, and the need for early, full
weight-bearing mobilization. A failure rate of nearly
40% has been recently reported in ambulatory elderly
patients with displaced femoral neck fractures treated
with internal fixation. Fixation failure rates of 30% to
40% have been consistently reported over multiple series over the past few decades. In sharp contrast, multiple series evaluating the outcome of prosthetic replace-

402

Orthopaedic Knowledge Update 8

ment in these patients have consistently demonstrated
predictable pain relief, functional improvement, and a
low revision rate for a patient population that cannot
tolerate a prolonged convalescence and multiple surgeries associated with fixation failure. Once prosthetic arthroplasty has been chosen, further controversy surrounds the selection of the type of arthroplasty, unipolar
or bipolar, hemiarthroplasty, or total hip arthroplasty,
and the type of fixation, cemented or cementless.
Good to excellent results can be expected with either cemented or cementless newer generation arthroplasties. Risks of cementless arthroplasty include femoral fracture, prosthesis subsidence, and anterior thigh
pain. Careful attention to accurate prosthetic sizing and
appropriate seating on the calcar is essential. Cementation of the prosthesis places the patient at risk for intraoperative death or embolization of marrow content during cementation. This risk may be reduced by canal
venting and gentle cement pressurization. The available
literature suggests generally better outcomes with cemented arthroplasties. In the nonambulatory patient,
when the procedure is performed predominantly for
pain control, first-generation cementless (Austin-Moore
type) prostheses can be used.
Considerable data exist comparing unipolar with bipolar bearings for elderly patients with displaced femoral neck fractures. Short- to mid-term follow-up studies
demonstrate no clear difference in morbidity, mortality,
or functional outcome. Longer-term follow-up suggests
a lower revision rate for bipolar bearings (Figure 2).
This finding is not surprising, because patients who live
longer are probably more active, and acetabular erosion
is a time-dependent phenomenon. More data are
needed to clearly determine the long-term superiority of
one design over the other.
Total hip arthroplasty has been classically recommended for patients with displaced fractures and symptomatic ipsilateral degenerative change of the hip. This
combination of pathology, however, is very rare. Recent
studies have expanded the indications to include active
elderly patients with displaced femoral neck fractures
and otherwise normal hip joints because of the more
predictable pain relief and better function total hip arthroplasty provides, when compared with hemiarthroplasty. The main complication of hip arthroplasty performed in this setting is dislocation, with rates averaging
10% across multiple series. Of those hips that dislocate,
approximately 25% have recurrent dislocation. A recent
meta-analysis demonstrated a mean dislocation rate approximately seven times greater for total hip arthroplasty compared with hemiarthroplasty in this setting.
Considering the large number of patients with displaced
femoral neck fractures treated annually, the potential
societal and economic impact of such dislocations can
be substantial. Use of the anterolateral approach decreases the dislocation rates as does the selective use of

American Academy of Orthopaedic Surgeons

Chapter 33 Hip: Trauma

Orthopaedic Knowledge Update 8

Figure 4 Failed fixation and cut-out of a reverse obliquity fracture treated with a
sliding hip screw.

Figure 3 A well-placed sliding hip screw with deep and central position of the lag

Intertrochanteric Hip Fractures

screw into the femoral head.

Classification

larger diameter femoral heads, and should be considered when performing total hip arthroplasty in this setting.

Complications
Nonunion is rare in the younger patient, with most series reporting nonunion rates of less than 10%. Secondary surgeries such as valgus-producing osteotomies are
successful in ultimately achieving union, probably because of the excellent bone stock and healing potential
in the young patient. Valgus intertrochanteric osteotomies convert the shear forces of a vertical fracture line
to compressive forces by increasing fracture horizontality. Nonunion is more common in the older patient, with
rates averaging less than 5% for nondisplaced fractures
and to over 30% for displaced fractures. Nonunion in
the older patient is typically treated with hip arthroplasty.
Rates of posttraumatic osteonecrosis have averaged
10% for nondisplaced fractures and 25% for displaced
fractures. Not all patients with osteonecrosis will be
symptomatic and require further treatment. The treatment of symptomatic posttraumatic osteonecrosis varies
with patient age and osteonecrosis grade.

American Academy of Orthopaedic Surgeons

Various classifications for intertrochanteric fractures

have been proposed, but none has been widely accepted. Commonly, fractures are described by the number of parts and the presence of certain fracture characteristics that indicate greater instability. For example,
a large posteromedial fragment, a reverse obliquity configuration, or subtrochanteric extension are commonly
considered fracture features that result in more unstable fractures. Intertrochanteric fractures are treated
similarly regardless of patient age. Choice of internal
fixation device should be based on fracture pattern.

Sliding Hip Screw

The most important variables under a surgeons control
when treating an intertrochanteric hip fracture include
correct device selection based on fracture pattern, accurate reduction, and placement of the lag screw into the
center of the femoral head. Most intertrochanteric fractures can be treated successfully with a sliding hip screw.
This device offers the advantages of a simple, predictable surgical technique, and a long clinical history of
successful results. An increasing rate of failure and hardware cutout with poor implant placement or poor reduction has been documented. The tip to apex distance
has been described as a guide to accurate screw placement, and should be less than 25 mm. Ideally, a centercenter position with the lag screw within 1 cm of the

403

Hip: Trauma

Orthopaedic Knowledge Update 8

Figure 5 A, Four-part comminuted intertrochanteric fracture with reverse obliquity. B, Postoperative view after treatment with an intramedullary hip screw.
subchondral bone on both AP and lateral views is preferred (Figure 3). The sliding hip screw should never be
used for fractures with reverse obliquity, because in this
situation this device does not allow controlled collapse
and fracture compression, but allows shear across the
fracture site with medial displacement of the distal fragment, excessive sliding, and eventual lag screw cutout
(Figure 4). In one series, a 56% failure rate was noted
for reverse obliquity fractures treated with a sliding hip
screw. For intertrochanteric fractures with reverse obliquity, either a 95 fixed angle device (such as the 95 dynamic condylar screw and the condylar blade plate) or a
cephalomedullary device is recommended. A recent
prospective randomized series documented superior
outcomes of intramedullary techniques over a 95 dynamic condylar screw (Figure 5).
Studies have compared the results of sliding hip
screws with two-hole sideplates to conventional fourhole sideplates for both stable and unstable fractures.
No difference in clinical outcomes was noted. The
shorter sideplates offer the advantage of less soft-tissue
dissection. Newly designed percutaneously applied
plates have not demonstrated a clear advantage over
traditional open sliding hip screw techniques in early
studies.

404

Intramedullary Devices
Multiple randomized, nonrandomized, prospective, and
retrospective studies have compared intramedullary devices with sliding hip screws for the fixation of intertrochanteric fractures. Although the concept of treating
these fractures through small incisions and avoiding the
usually bloody dissection of the vastus lateralis necessary for sideplate placement is appealing, the literature
has not demonstrated the clear advantage of intramedullary devices to justify their routine use. In earlier series that evaluated outdated nail designs, higher complication rates, including iatrogenic femur fractures, were
reported. Contemporary intramedullary implant designs
have addressed many of these concerns and may allow
more minimally invasive fracture management techniques. A recent prospective, randomized series of 400
patients comparing the redesigned gamma nail to a sliding hip screw demonstrated a higher (but not statistically significant) rate of complications with the gamma
nail. The authors concluded that the routine use of the
gamma nail cannot be recommended. Currently, intramedullary devices may be most suitable for fractures
with reverse obliquity or high subtrochanteric or intertrochanteric fractures with subtrochanteric extension.
More data are needed to define which fractures derive

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

the most benefit from intramedullary fixation and
whether further design refinements will decrease the
number of complications noted in earlier series, thereby
making intramedullary fixation more widely applicable.

Primary Prosthetic Replacement

The role of primary prosthetic replacement for intertrochanteric fractures remains controversial. The potential
advantages of primary prosthetic replacement in the
face of an unstable intertrochanteric fracture in a patient with severely osteopenic bone include relatively
predictable pain relief, early mobilization, and the fact
that revision rates may be lower. The disadvantages include the more extensive nature of the surgical procedure, the frequent necessity to use calcar replacing, longstem cemented implants in medically frail patients, and
the fact that often comminuted, osteopenic trochanteric
fragments need to be stabilized with some form of internal fixation. It should be noted from the literature that
the overwhelming majority of well-reduced intertrochanteric hip fractures treated with properly selected
and accurately implanted internal fixation devices will
heal predictably without complication. Additionally,
should failure occur, prosthetic replacement for salvage
of failed internal fixation has demonstrated excellent
durability and predictable pain relief. Prosthetic replacement as the acute treatment of intertrochanteric fractures should be reserved for patients with pathologic
fractures as a result of neoplasm, neglected fractures
with deformity and poor bone stock precluding internal
fixation, or for patients in whom internal fixation attempts have failed. Additionally, prosthetic replacement
may be a reasonable therapeutic option for patients
with severe ipsilateral symptomatic degenerative joint
disease, or certain unfavorable fracture patterns associated with poor bone quality.

Subtrochanteric Fractures
The subtrochanteric area of the femur experiences some
of the highest biomechanical stresses in the human
body. In general, the subtrochanteric region is considered the anatomic region immediately below the lesser
trochanter to the proximal aspect of the femoral isthmus. Various classification systems, including the
Russell-Taylor classification, have been proposed based
on the location of the fracture relative to the lesser trochanter and the presence of the fracture line extension
into the piriformis fossa. More proximal subtrochanteric
fractures often involve extension of the fracture line
into the piriformis fossa, which influences internal fixation device selection. Multiple fixation methods have
been described, including the use of the sliding hip
screw, dynamic condylar screw, or angled blade plate for
more proximal (high) subtrochanteric fractures, and interlocking cephalomedullary nailing for more distal

American Academy of Orthopaedic Surgeons

Chapter 33 Hip: Trauma

(low) subtrochanteric fractures. In general, if the piriformis fossa is intact, a nailing technique is preferred. If the
lesser trochanter is intact and the proximal fragment is
of sufficient length, then fixation into the femoral head
is usually not necessary and a standard antegrade nail
with locking into the lesser trochanter is adequate. With
shorter proximal fragments without piriformis fossa involvement, cephalomedullary reconstruction nailing
with interlocking screw fixation into the femoral head
and neck is preferred. A preoperative CT scan may be
helpful to evaluate the integrity of the proximal fragment when plain radiographs are difficult to interpret
because of factors such as patient size or proximal fragment rotation. If plating techniques are chosen for comminuted fractures, the vascularity of the medial bony
fragments should be carefully preserved by avoiding dissection and periosteal stripping in this region. Areas of
comminution should be bridged and stable proximal
and distal fixation should be obtained while maintaining
correct limb alignment, length, and rotation. Such indirect reduction techniques have demonstrated high union
rates but are technically demanding. Although plating
techniques are associated with potentially lower rates of
malalignment, dissection and blood loss can be substantial. Biomechanically, intramedullary fixation of subtrochanteric fractures is generally preferred. Nails provide
load-sharing stability that allows early postoperative
weight bearing that plated fractures typically cannot tolerate. This condition may be advantageous in multiple
trauma patients with other extremity injuries. Nailing
short subtrochanteric fractures can be challenging because of the flexed, abducted, and externally rotated position of the proximal fragment. Great care should be
taken to avoid varus proximal fragment malalignment,
which can occur with conventional antegrade nails
placed through a starting point that is too lateral. The
use of newer nails designed for entry through the tip of
the greater trochanter may facilitate access to the intramedullary canal and result in less potential malalignment for these challenging fractures; however, there are
few data to substantiate this speculation.

Pathologic Fractures of the Proximal Femur

Metastatic disease commonly involves the proximal femur, affecting the femoral neck in 50%, subtrochanteric
region in 30%, and intertrochanteric region in 20%. Decision making regarding whether some form of prophylactic internal fixation or hip arthroplasty is appropriate
is based on the anatomic location of the lesion, size of
the lesion, the extent of bony destruction, and anticipated life expectancy of the patient. In general, lesions
that cause more than 50% destruction of a cortex or
those that present with functional pain should be stabilized. Complete fractures should be treated surgically in
all but the most infirm patients. High-quality preopera-

405

Hip: Trauma
tive radiographs of the acetabulum and entire length of
the femur are mandatory to evaluate for ipsilateral lesions. Other sites of bony pain should also be evaluated
by plain films and bone scintigraphy. Appropriate preoperative workup in consultation with an oncologist is
recommended. A fracture resulting from a solitary
pathologic lesion of the proximal femur requires a
pathologic tissue diagnosis before internal fixation or
prosthetic replacement, even in a patient with a history
of cancer. These preoperative studies will help avoid the
rare, but potentially disastrous complication of internal
fixation of a primary malignancy. If necessary, a CT scan
of the proximal femur and the acetabulum can be obtained to further evaluate for proximal lesions. The entire femur should be protected by the internal fixation
device, typically a third-generation cephalomedullary
nail. Lesion debulking and methacrylate augmentation
of the fixation construct may be necessary for larger lesions. If extensive involvement of the proximal femur
precludes predictable and durable internal fixation, then
prosthetic replacement can provide functional improvement and pain relief in this cohort. Modular, so-called
tumor prostheses are available to manage bony deficiency and restore leg length and hip stability. Pathologic fractures of the femoral neck, head, and intertrochanteric region often require treatment with prosthetic
replacement. Medical comorbidities are quite common
and multidisciplinary management with a medical oncologist, radiation oncologist, or nutritionist is recommended. Postoperative radiation to the entire construct
and surgical bed, after the surgical wound has healed, is
recommended to minimize the chance of tumor progression and implant failure. Should fixation failure occur,
conversion to hip arthroplasty has been shown to predictably improve function and relieve pain; however,
these reconstructions are plagued by a high rate of postoperative infection.

Annotated Bibliography
Hip Dislocations and Femoral Head Fractures
Ganz R, Gill TJ, Gautier E, Ganz K, Krugel N, Berlemann U: Surgical dislocation of the adult hip: A technique with full access to the femoral head and acetabulum without the risk of avascular necrosis. J Bone Joint
Surg Br 2001;83:1119-1124.
The authors describe a safe surgical approach for hip dislocation in 213 hips with no avascular necrosis. This approach,
known as the trochanteric flip, is useful for multiple degenerative and traumatic disorders of the hip joint.

Moorman CT III, Warren RF, Hershman EB, et al: Traumatic posterior hip subluxation in American football.
J Bone Joint Surg Am 2003;85:1190-1196.
The authors discuss the clinical presentation, MRI findings, suggested treatment, and outcomes of eight football play-

406

Orthopaedic Knowledge Update 8

ers with traumatic hip subluxation. Two of eight patients developed osteonecrosis and required total hip arthroplasty.

Sahin V, Karakas ES, Aksu S, Atlihan D, Turk CY, Halici

M: Traumatic dislocation and fracture-dislocation of the
hip: A long term follow-up study. J Trauma 2003;54:520529.
Forty-seven patients with hip dislocation and fracturedislocation were followed for a mean of 9.6 years. Seventy-one
percent had medium to very good results. Sixteen percent developed posttraumatic degenerative joint disease, and 9.6%
developed osteonecrosis. Early reduction improved outcomes.

Femoral Neck Fractures

Bartonicek J: Pauwels classification of femoral neck
fractures: Correct interpretation of the original.
J Orthop Trauma 2001; 15:358-360.
This article discusses Pauwels classification and gives the
correct interpretation of the values.

Bhandari M, Devereaux PJ, Swiontkowski MF, et al: Internal fixation compared with arthroplasty for displaced
fractures of the femoral neck: A meta-analysis. J Bone
Joint Surg Am 2003;85:1673-1681.
The authors evaluated published trials between 1969 and
2002 on the treatment of displaced femoral neck fractures in
patients age 65 years or older. Arthroplasty provided a significantly lower rate of revision surgery (P = 0.0003) but was associated with greater blood loss, longer surgical time, and a
trend toward higher mortality in the first 4 months after surgery (not significant).

Haidukewych GJ, Israel TA, Berry DJ: Long-term survivorship of cemented bipolar hemiarthroplasty for fracture of the femoral neck. Clin Orthop 2002;403:118-126.
The results of 212 patients older than age 60 years treated
with cemented bipolar hemiarthroplasty are reported. Overall
10-year survivorship free of reoperation for any reason was
94%. Only one patient was revised for acetabular cartilage
wear. More than 90% of patients had no or minimal pain at
follow-up, and the dislocation rate was less than 2%.

Jain R, Koo M, Kreder HJ, Schemitsch EH, Davey JR,

Mahomed NN: Comparison of early and delayed fixation of subcapital hip fractures in patients sixty years of
age or less. J Bone Joint Surg Am 2002;84:1605-1612.
Delayed treatment of subcapital fractures was associated
with a higher rate of osteonecrosis; however, this complication
did not significantly affect functional outcome at follow-up.

Keating JF, Masson M, Scott N, et al: Randomized trial

of reduction and fixation versus bipolar hemiarthroplasty versus total hip arthroplasty for displaced subcapital fractures in the fit older patient. 70th Annual Meeting Proceedings., Rosemont, IL, American Academy of
Orthopaedic Surgeons, 2003, p 96.

American Academy of Orthopaedic Surgeons

Chapter 33 Hip: Trauma

Orthopaedic Knowledge Update 8

The authors demonstrate superior functional outcomes for
fractures treated with total hip arthroplasty when compared
with open reduction and internal fixation or hemiarthroplasty.

Maurer SG, Wright KE, Kummer FJ, Zuckerman JD,

Koval KJ: Two or three screws for fixation of femoral
neck fractures? Am J Orthop 2003;32:438-442.

Intertrochanteric Hip Fractures

Adams CI, Robinson CM, Court-Brown CM, McQueen
MM: Prospective randomized controlled trial of an intramedullary nail versus dynamic screw and plate for intertrochanteric fractures of the femur. J Orthop Trauma
2001;15:394-400.

Three screws demonstrated less fracture displacement

than two screws in axial loading in embalmed cadaveric specimens with subcapital osteotomies.

Four hundred patients were randomized for either a sliding hip screw or a gamma nail. The group with the gamma nail
had a higher rate of reoperations and complications. The authors concluded that the routine use of an intramedullary device is not recommended.

McKinley JC, Robinson CM: Treatment of displaced intracapsular hip fractures with total hip arthroplasty:
Comparison of primary arthroplasty with early salvage
arthroplasty after failed internal fixation. J Bone Joint
Surg Am 2002;84:2010-2015.

Haidukewych GJ, Berry DJ: Hip arthroplasty for salvage of failed treatment of intertrochanteric hip fractures. J Bone Joint Surg Am 2003;85:899-904.

The authors found better outcomes and fewer complications with primary arthroplasty than when arthroplasty was
performed after internal fixation failure in a matched pair
case-controlled study.

The durability and predictable functional improvement of

hip arthroplasty for failed treatment of intertrochanteric fractures is documented in 60 patients. Long-stem, calcar-replacing
prostheses are commonly required. The greater trochanter was
a persistent source of discomfort in a substantial subset of patients.

Ong BC, Maurer SG, Aharonoff GB, Zuckerman JD,

Koval KJ: Unipolar versus bipolar hemiarthroplasty:
Functional outcome after femoral neck fracture at a
minimum of thirty-six months of follow-up. J Orthop
Trauma 2002;16:317-322.

Haidukewych GJ, Israel TA, Berry DJ: Reverse obliquity of fractures of the intertrochanteric region of the
femur. J Bone Joint Surg Am 2001;83:643-650.

No difference in functional outcome at midterm follow-up

was noted between unipolar and bipolar cemented hemiarthroplasties.

Rogmark C, Carlsson A, Johnell O, Sernbo I: Primary

hemiarthroplasty in old patients with displaced femoral
neck fracture: A 1-year follow-up of 103 patients aged
80 years or more. Acta Orthop Scand 2002;73:605-610.

The authors demonstrated a 56% failure rate when the

sliding hip screw was used for a reverse obliquity fracture. The
95 blade plate had the lowest failure rate. Few patients were
treated with intramedullary fixation during the study period.
The authors concluded that the sliding hip screw is contraindicated for reverse obliquity fractures.

Treatment with primary hemiarthroplasty showed a lower

failure rate than treatment with internal fixation. There was no
difference in length of hospital stay or mortality.

Janzig HM, Howben BJ, Brandt SE, et al: The Gotfried

Percutaneous Compression Plate versus the Dynamic
Hip Screw in the treatment of peritrochanteric hip fractures: Minimal invasive treatment reduces operative
time and postoperative pain. J Trauma 2002;52:293-298.

Sharif KM, Parker MJ: Austin Moore hemiarthroplasty:

Technical aspects and their effects on outcome, in patients with fractures of the neck of femur. Injury 2002;
33:419-422.

One hundred fifteen patients randomized to either the

percutaneous compression plate or a sliding hip screw were
reviewed. The patients with the percutaneous plate had a
lower surgical time and less pain, but more mechanical complications.

This retrospective study evaluated radiographic findings

with outcomes and revision rates for 243 patients with a mean
age of 81 years treated with Austin-Moore hemiarthroplasty.
Undersizing of the prosthetic head and poor seating of the
prosthesis on the calcar were associated with loosening and residual pain.

Kosygan KP, Mohan R, Newman RJ: The Gotfried percutaneous compression plate compared with the conventional classic hip screw for the fixation of intertrochanteric fractures of the hip. J Bone Joint Surg Br 2002;
84:19-22.

Tanaka J, Seki N, Tokimura F, Hayashi Y: Conservative

treatment of Garden stage I femoral neck fracture in
elderly patients. Arch Orthop Trauma Surg 2002;122:
24-28.
Nonsurgical treatment of Garden stage I femoral neck
fractures showed a 39% nonunion rate.

American Academy of Orthopaedic Surgeons

One hundred eleven patients were prospectively randomized to the percutaneous compression plate or a sliding hip
screw. The percutaneous plate was associated with less blood
loss and fewer transfusions, but longer surgical time. There was
no difference in the number of complications or fracture healing.

407

Hip: Trauma
Sadowski C, Lubbeke A, Saudan M, Riand N, Stern R,
Hoffmeyer P: Treatment of reverse oblique and transverse intertrochanteric fractures with use of an intramedullary nail or a 95 degree screw-plate: A prospective, randomized study. J Bone Joint Surg Am 2002;84:
372-381.
Intramedullary fixation demonstrated a lower rate of fixation failure than the 95 dynamic condylar screw.

Subtrochanteric Fractures
Vaidya SV, Dholakia DB, Chatterjee A: The use of a dynamic condylar screw and biologic reduction techniques
for subtrochanteric femur fractures. Injury 2003;34:123128.
Thirty-one patients were treated with indirect reduction
techniques. The authors reported a 100% union rate; 6.4% of
patients had a malunion.

Classic Bibliography
Alho A, Benterud JG, Solovieva S: Internally fixed femoral neck fractures: Early prediction of failure in 203
elderly patients with displaced fractures. Acta Orthop
Scand 1999;70:141-144.
Asnis SE, Wanek-Sgaglione L: Intracapsular fractures of
the femoral neck: Results of cannulated screw fixation.
J Bone Joint Surg Am 1994;76:1793-1803.
Barquet A, Francescoli L, Rienzi D, Lopez L:
Intertrochanteric-subtrochanteric fractures: Treatment
with the long Gamma nail. J Orthop Trauma 2000;14:
324-328.
Baumgaertner MR, Curtin SL, Lindskog DM, Keggi JM:
The value of the tip-apex distance in predicting failure
of fixation of peritrochanteric fractures of the hip.
J Bone Joint Surg Am 1995;77:1058-1064.
Booth KC, Donaldson TK, Dai QG: Femoral neck fracture fixation: A biomechanical study of two cannulated
screw placement techniques. Orthopedics 1998;21:11731176.
Calder SJ, Anderson GH, Jagger C, Harper WM, Gregg
PJ: Unipolar or bipolar prosthesis for displaced intracapsular hip fracture in octogenarians: A randomized
prospective study. J Bone Joint Surg Br 1996;78:391-394.
Chua D, Jaglal SB, Schatzker J: Predictors of early failure of fixation in the treatment of displaced subcapital
hip fractures. J Orthop Trauma 1998;12:230-234.

Orthopaedic Knowledge Update 8

Hammer AJ: Nonunion of the subcapital femoral neck
fracture. J Orthop Trauma 1992;6:73-77.
Kinast C, Bolhofner BR, Mast JW, Ganz R: Subtrochanteric fractures of the femur: Results of treatment with
the 95 degree condylar blade plate. Clin Orthop 1989;
238:122-130.
Koval KJ, Sala DA, Kummer FJ, Zuckerman JD: Postoperative weight-bearing after a fracture of the femoral
neck or an intertrochanteric fracture. J Bone Joint Surg
Am 1998;80:352-356.
Kyle RF, Gustilo RB, Premer RF: Analysis of six hundred and twenty-two intertrochanteric hip fractures.
J Bone Joint Surg Am 1979;61:216-221.
Lee BP, Berry DJ, Harmsen WS, Sim FH: Total hip arthroplasty for the treatment of an acute fracture of the
femoral neck: Long-term results. J Bone Joint Surg Am
1998;80:70-75.
Lu-Yao GL, Keller RB, Littenberg B, Wennberg JE:
Outcomes after displaced fractures of the femoral neck:
A meta-analysis of one hundred and six published reports. J Bone Joint Surg Am 1994;76:15-25.
Marchetti ME, Steinberg GG, Coumas JM:
Intermediate-term experience of Pipkin fracturedislocations of the hip. J Orthop Trauma 1996;10:455461.
Maruenda JI, Barrios C, Gomar-Sancho F: Intracapsular
hip pressure after femoral neck fracture. Clin Orthop
1997;340:172-180.
Parker MJ, Blundell C: Choice of implant for internal
fixation of femoral neck fractures: Meta-analysis of 25
randomised trials including 4,925 patients. Acta Orthop
Scand 1998;69:138-143.
Parker MJ, Pryor GA: Gamma versus DHS nailing for
extracapsular femoral fractures: Meta-analysis of ten
randomised trials. Int Orthop 1996;20:163-168.
Pipkin G: Treatment of grade IV fracture-dislocation of
the hip: A review. J Bone Joint Surg Am 1957;39:10271042.
Robinson CM, Saran D, Annan IH: Intracapsular hip
fractures: Results of management adopting a treatment
protocol. Clin Orthop 1994;302:83-91.

Dreinhofer KE, Schwarzkopf SR, Haas NP, et al: Isolated traumatic dislocation of the hip: Long-term results
in 50 patients. J Bone Joint Surg Br 1994;76:6-12.

Russell TA, Taylor JC: Subtrochanteric fractures of the

femur, in Browner BD, Jupiter JB, Levine AM, Trafton
PG (eds): Skeletal Trauma, ed. 2. Philadelphia, PA, WB
Saunders, 1997.

Garden RS: Malreduction and avascular necrosis in subcapital fractures of the femur. J Bone Joint Surg Br
1971;53:183-197.

Sanders R, Regazzoni P: Treatment of subtrochanteric

femur fractures using the dynamic condylar screw.
J Orthop Trauma 1989;3:206-213.

408

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 33 Hip: Trauma

Siebenrock KA, Muller U, Ganz R: Indirect reduction

with a condylar blade plate for osteosynthesis of subtrochanteric femoral fractures. Injury 1998;29(suppl 3):C7C15.

Thompson VP, Epstein HC: Traumatic dislocation of the

hip: A survey of two hundred and four cases covering a
period of twenty-one years. J Bone Joint Surg Am 1951;
33:746-778.

Stannard JP, Harris HW, Volgas DA, Alonso JE: Functional outcome of patients with femoral head fractures
associated with hip dislocations. Clin Orthop 2000;377:
44-56.

Upadhyay SS, Moulton A, Srikrishnamurthy K: An analysis of the late effects of traumatic posterior dislocation
of the hip without fractures. J Bone Joint Surg Br 1983;
65:150-157.

Stewart MJ, Milford LW: Fracture-dislocation of the hip:

An end-result study. J Bone Joint Surg Am 1954;36:315342.

van Doorn R, Stopert JW: The long Gamma nail in the

treatment of 329 subtrochanteric fractures with major
extension into the femoral shaft. Eur J Surg 2000;166:
240-246.

Swiontkowski MF, Thorpe M, Seiler JG, et al: Operative

management of displaced femoral head fractures: Casematched comparison of anterior versus posterior approaches for Pipkin I and Pipkin II fractures. J Orthop
Trauma 1992;6:437-442.

American Academy of Orthopaedic Surgeons

Wiss DA, Brien WW: Subtrochanteric fractures of the

femur: Results of treatment by interlocking nailing. Clin
Orthop 1992;283:231-236.

409

Chapter

34

Hip, Pelvic Reconstruction, and

Arthroplasty
Javad Parvizi, MD, FRCS
James J. Purtill, MD

Scope of Pathology

Radiographic Examination

With the increasing life span of the general population,

the incidence of osteoarthritis (OA) of the hip is on the
rise. More than 300,000 joint arthroplasties are currently
performed annually in the Unites States. By year 2030,
this number is expected to exceed 1 million. The two
major categories of conditions that give rise to OA of
the hip are dysplasia and nondysplasia. Dysplasia, commonly with anterolateral deficiency, may account for up
to 40% of OA cases in the United States and over 80%
of OA cases in Japan and Italy. In a distinct category of
patients, there is no identifiable cause for OA, or socalled idiopathic arthritis. There is emerging evidence
that subtle morphologic abnormalities around the hip,
resulting in femoroacetabular impingement, may be a
contributing factor in many instances.

A good quality AP radiograph of the pelvis and AP and

lateral radiographs of the hip should be ordered as part
of the initial evaluation. The radiographs should be examined for evidence of arthritis, fracture, osteonecrosis,
and other morphologic abnormalities that could account
for the patients symptoms. Signs of dysplasia should also
be sought (Figure 1).
Subtle signs of dysplasia include isolated anterior acetabular deficiency (often missed), overloading of the
rim (os acetabuli or hypertrophic labrum), and rim ossification (double shadow on AP radiograph). An estimate of acetabular version can be determined by defining the outline of the posterior and anterior walls
(Figure 2). In a normal anteverted acetabulum the anterior wall lies medial to the posterior wall and does not
cross it. Retroversion is characterized by crossover of
the anterior and posterior wall markings on a true AP
radiograph (coccyx pointing toward the symphysis pubis
with a distance of 1 to 2 cm between them).
Functional views with the hip in abduction or adduction should be considered before redirectional osteotomies are performed. Cross-sectional studies may be ordered to define the extent of a particular pathology or
confirm its presence. CT is useful for defining bony
anatomy and has an invaluable role in the evaluation of
pelvic fracture. MRI can be useful for confirmation of
osteonecrosis or transient osteoporosis, and when combined with gadolinium arthrogram it is useful for evaluating labral pathology. Imaging of the articular cartilage
remains inadequate at present. Patients with dysplasia
often have a hypertrophic labrum. Labral tears are common in patients with femoroacetabular impingement
(Figure 3).

Clinical Evaluation
The evaluation of a patient with hip pain should begin
with a thorough history. It is particularly important to
confirm or rule out the hip as the cause of the patients
symptoms. Various intra-abdominal, spinal, and other
pathologies may present as hip pain. True hip pain usually presents in the groin, anterior thigh, buttock, or
even the knee region. As part of the history, the suitability of the patient for surgical intervention should be determined. Physical examination of the patient includes
assessment of gait, limb length, and range of motion,
palpation of various regions around the hip, and a complete neurovascular examination. The skin should be examined to ensure no sources of infection exist. Provocative tests such as the impingement test (pain with
flexion, adduction, and internal rotation), apprehension
test (feeling of the hip popping out of the socket with
extension: sign of anterior deficiency), Patrick test
(groin pain with hip in figure-of-4 position ), and Stinchfield test (pain during resisted straight leg raise) are indicative of hip pathology.

American Academy of Orthopaedic Surgeons

Femoroacetabular Impingement
The widely accepted theory implicating axial overload
for the onset of OA of the hip fails to provide a satisfac-

411

Hip, Pelvic Reconstruction, and Arthroplasty

Orthopaedic Knowledge Update 8

Figure 1 A, AP radiograph of a patient with dysplasia of the right hip. Note the superior inclination of the weight-bearing region (high Tonnis angle), lateralized hip center of
rotation, anterolateral deficiency of the femoral head coverage, and coxa valga. B, A schematic presentation of the same hip demonstrating the various radiographic measurements that can be used to evaluate dysplasia.

Figure 3 The magnetic resonance arthrogram shows the labral destruction and a
Figure 2 AP radiograph of a patient with bilateral acetabular retroversion is evident
by the crossover (Reynolds) sign of the anterior and posterior walls on the left side. The
patient has undergone reverse periacetabular osteotomy to correct the retroversion.
Note that the anterior and posterior wall marking meet at the point of sourcil and do
not cross.

tory explanation for development of arthritis in young

patients with apparently normal skeletal structures and
intra-articular pressures. Femoroacetabular impingement has been proposed as a possible etiologic mechanism for the development of OA in this patient group.
The condition occurs either as a result of morphologic
abnormality involving the femur (cam) (Figure 4), the
acetabulum (pincer) (Figure 5), or both. Excessive and
supraphysiologic demand on the hip may precipitate

412

secondary ossicle on the femoral neck that resulted from linear contact between the
femoral neck and the acetabular rim during flexion (femoroacetabular impingement).

symptoms. The abnormal contact between the femoral

neck and acetabular rim leads to labral injury, particularly in the anterosuperior zone of the acetabulum (Figure 4). The labral tear as seen on arthroscopy or magnetic resonance arthrogram is frequently accompanied
by chondral damage. The typical patient with femoroacetabular impingement is young and has groin pain that is
exacerbated by activity or long periods of sitting (such as
while driving). Examination of the hip often reveals limitation of motion, particularly internal rotation and abduction in flexion. The impingement test, which is almost
always positive, is performed with the patient supine. The

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 34 Hip, Pelvic Reconstruction, and Arthroplasty

Figure 4 A, Schematic presentation shows the mechanism for cam impingement when nonspherical portion of the femoral head abuts against the acetabular rim during hip
flexion. B, The AP radiograph appears normal. C, The nonspherical femoral head leading to reduced offset at the neck and predisposition to cam-type impingement is visible on
the lateral radiograph. D, The magnetic resonance arthrogram confirms labral tear and chondral injury resulting from impingement. (Reproduced with permission from Ganz R,
Parvizi J, Beck M, Leunig M, Ntzli H, Siebenrock KA: Femoroacetabular impingement: A cause for osteoarthritis of the hip. Clin Orthop 2003;417:112-120.)

hip is internally rotated as it is passively flexed to about

90 and adducted. Flexion and adduction lead to the approximation of the femoral neck and the acetabular rim.
Forceful additional internal rotation creates a sharp pain
when there is a chondral and/or labral lesion.
The radiographs often appear normal at first glance.
However, upon detailed review some abnormalities may
become apparent. This includes the presence of a bony
prominence usually in the anterolateral head and neck
junction that is best seen on lateral radiographs (Figure
4, C), reduced anterior offset of the femoral neck and
head junction, and changes on the acetabular rim such
as os acetabuli or double line that are seen with rim ossification. Close scrutiny of the femoral neck may reveal
the presence of a herniation pit that is indicative of impingement. Morphologic changes affecting the acetabulum and/or the proximal femur such as retroversion, rel-

American Academy of Orthopaedic Surgeons

ative anterior overcoverage, coxa profunda, protrusio

acetabuli, coxa vara, or extreme coxa valga may only become apparent upon systematic examination of the
plain radiographs. Magnetic resonance arthrogram using
radial sequences is very useful for diagnosis. If conservative treatment is unsuccessful, femoroacetabular impingement can be treated by surgical dislocation of the
hip and improvement of femoral head-neck contour.
The acetabular rim can also be trimmed. A reverse periacetabular osteotomy may be performed in patients
with excessive retroversion and inadequate posterior
coverage (posterior rim lies medial to a vertical line
drawn through the center of the head).

Hip Arthroscopy
Hip arthroscopy is being used more often for diagnostic
as well as therapeutic treatment of intra-articular pa-

413

Orthopaedic Knowledge Update 8

Hip, Pelvic Reconstruction, and Arthroplasty

Figure 5 A, Schematic presentation shows the mechanism of pincer impingement

that arises from linear contact between the acetabular rim and the femoral head-neck
junction. B, The femoral head may have normal morphologic features and the impingement is the result of acetabular abnormality (overcoverage, rim ossification, retroversion). C, The first structure to fail in this situation is the acetabular labrum as seen on
the magnetic resonance arthrogram. The persistent anterior abutment with chronic leverage of the head in the acetabulum may result in chondral injury in the posteroinferior acetabulum. Note the prominence on the femoral neck that is the result of persistent impingement of the neck against deep acetabulum. (Reproduced with permission
from Ganz R, Parvizi J, Beck M, Leunig M, Ntzli H, Siebenrock KA: Femoroacetabular
impingement: A cause for osteoarthritis of the hip. Clin Orthop 2003;417:112-120.)

thology of the hip. Indications include loose bodies, acetabular labral tears, cartilage flaps, and synovitis.
Standard radiographs are unable to adequately detect acetabular labral tears and other subtle intra-

414

articular hip pathology. MRI as well as magnetic resonance arthrogram of the hip improves diagnostic ability.
In addition, hip arthroscopy may be used diagnostically
in patients with mechanical symptoms or suspected
intra-articular pathology.
Hip arthroscopy may be performed with the patient
supine or in a lateral decubitis position. Three portals,
anterior, anterolateral, and posterolateral, have been described. Access to the hip joint is accomplished by the
use of cannulated trochars. Traction is applied to the leg
undergoing surgery with the use of a fracture table. A
well-padded peroneal post minimizes risk to the peroneal structures, especially the pudendal nerve. Fluoroscopy is necessary for proper cannula placement.
Acetabular labral tears are the most common indication for hip arthroscopy. These may result from an acute
hyperflexion or twisting injury to the hip. In addition to
acute injuries, acetabular labral pathology may be associated with early degenerative arthritis. Arthritic
changes of the anterosuperior region of the acetabulum
may be associated with detachment of the labrum from
this region of the joint.
Patients with labral tears typically have intermittent
groin pain. Mechanical symptoms such as locking, catching, or clicking are common. Physical examination may
demonstrate an increase in groin pain with maximum
flexion and internal rotation of the hip. In the absence
of significant osteoarthritis, arthroscopic acetabular labral dbridement has resulted in relief of mechanical
symptoms and a decrease in groin pain in a substantial
number of patients. Patients with dysplasia may commonly have labral tears but typically are not ideal candidates for arthroscopic surgery.
Intra-articular loose bodies have a wide range of etiologies including trauma, synovial chondromatosis, and
ligamentum teres rupture. Hip arthroscopy provides less
invasive access to the loose bodies of the hip joint than
standard open techniques. There are several reports of
successful removal of intra-articular bullets with hip arthroscopy.
Complications related to hip arthroscopy are rare
and occur in fewer than 2% of patients in most large series. Complications include nerve injury (lateral femoral
cutaneous, pudendal and femoral), instrument breakage,
portal hematoma, septic arthritis, articular cartilage
damage, trochanteric bursitis, and extravasation of arthroscopic fluid.

Hip Arthrodesis
Hip arthrodesis traditionally has been considered for
the very young patient (especially large male laborers)
with severe unilateral hip arthritis. As hip replacement
durability has improved and as patients have become
aware of the benefits associated with hip replacement,
fewer patients are willing to accept arthrodesis. How-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

ever, satisfactory results in terms of hip pain relief and
functional improvement may be expected. Compared
with individuals who have not undergone surgery, patients who have undergone arthrodesis of the hip have
an approximate increase in energy expenditure during
ambulation of 30%.
Abnormal gait resulting from hip fusion causes increased stress on the lumbar spine, and other joints of
the lower extremities. Low back pain, ipsilateral knee
pain and instability, and contralateral hip arthritis develop in as many as 60% of patients after hip arthrodesis. Pain in the low back or knee typically develops
within 25 years of hip fusion and may become disabling
over the long term.
Patients with associated contralateral hip, low back,
or knee problems may require conversion of the hip fusion to total hip arthroplasty. Patients undergoing ipsilateral knee replacement may gain better knee function
if the hip arthrodesis is converted to total hip arthroplasty (THA) before knee replacement. Low back pain
may improve after conversion to THA. Because many
patients with arthrodesis will eventually require conversion to THA, this eventual possibility should be taken
into account during surgery. Suggested surgical techniques include preservation of the abductor musculature, placing the hip in 20 to 25 of flexion, and plate
fixation. Because ipsilateral knee and back pain are
more common in patients whose hips are fused in abduction, neutral abduction is suggested.

Osteotomy
Osteotomies around the hip are important biologic
treatment modalities for patients younger than age 40 to
50 years because they may preserve the host hip joint.
For symptomatic patients with structural hip abnormalities and mild or no arthritis, osteotomy may provide
long-term pain relief and improve function.

Pelvic Osteotomy
Pelvic osteotomy is indicated for treatment of symptomatic dysplasia in young, active patients. Pelvic osteotomy
has traditionally been classified into reconstructive and
salvage osteotomy. The latter (shelf and Chiari) does not
provide articular cartilage coverage for the femoral
head. Salvage osteotomies have been abandoned in favor of THA, except in rare circumstances. Reconstructive osteotomies rely on redirection of the acetabulum
to provide better coverage for the femoral head. It is
believed that improvement in femoral head coverage
halts or retards the progression of the degenerative process in most of these patients. In a recent review of
more than 800 patients receiving Ganz or Bernese osteotomy, the procedure failed in 42 patients and THA
was required at a mean of 6.8 years after the osteotomy.
Function was not affected in the patients who did not

American Academy of Orthopaedic Surgeons

Chapter 34 Hip, Pelvic Reconstruction, and Arthroplasty

require THA. The results of most pelvic osteotomies
have been encouraging. The Bernese osteotomy is preferred by most reconstructive surgeons for treatment of
dysplasia in adults because it allows a large degree of
correction, permits joint medialization (reducing the
joint forces), does not breach the weight-bearing posterior column (allowing early ambulation), does not violate the abductors, and is associated with a relatively
low incidence of complications and morbidities.

Femoral Osteotomy
Femoral osteotomy, a once popular option for treatment
of dysplasia, has been mostly abandoned in favor of pelvic osteotomy. The current indications for femoral osteotomy include severe deformity of the proximal femur, treatment of femoral neck nonunions in young
patients, and patients with osteonecrosis. Varusproducing osteotomy (rotation of the proximal femur
into varus) is a valuable option used in isolation or in
combination with pelvic osteotomy for treatment of
symptomatic coxa valga.

Osteonecrosis of the Femoral Head

The incidence of osteonecrosis is not known with certainty. Osteonecrosis is more common in males and typically affects patients in their late 30s or early 40s. Osteonecrosis of the femoral head is bilateral in
approximately 50% of cases.
Symptoms include pain in the groin region that is exacerbated with ambulation and activity. Some patients
with early disease may be asymptomatic; therefore, a diagnosis may require a high index of suspicion. MRI or
bone scan may show changes in the femoral head before
plain radiographs.
Osteonecrosis of the femoral head occurs as a result
of altered blood supply to the femoral head, which can
result in bone necrosis, subcortical fracture, collapse,
and eventual destruction of the hip joint. In some situations, the etiology of vascular compromise is clear (such
as hip dislocation or femoral neck fracture with traumatic disruption of the blood vessels). In addition to direct blood flow disruption, other associated causes of
osteonecrosis include corticosteroid use, ethanol abuse,
and hypercoagulable states. Osteonecrosis is associated
with hip trauma in 10% of patients. More than one third
of patients have idiopathic osteonecrosis. Patients receiving medical treatment for the virus that causes acquired immunodeficiency syndrome apparently are at
risk for osteonecrosis of the hip and other joints.
The widely used classification system for osteonecrosis of the femoral head is a modified version of one proposed by Ficat. In stage zero, osteonecrosis changes of
the hip are noted only on MRI in an asymptomatic patient. Stage one osteonecrosis is seen in a symptomatic
patient with positive MRI findings but no evidence of

415

Orthopaedic Knowledge Update 8

Hip, Pelvic Reconstruction, and Arthroplasty

abnormalities on plain radiographs. In stage two, radiographs reveal sclerotic bone of the femoral head without subchondral bone collapse. In stage three, subchondral bone collapse (crescent sign) is seen on plain
radiographs. In stage four, there is collapse of the femoral head with secondary degenerative changes of the hip
joint. Poor prognostic factors for osteonecrosis of the
femoral head include high degree of femoral head involvement (greater than 30%) and continued corticosteroid use.
Surgical treatment of osteonecrosis of the femoral
head depends on the stage at the time of diagnosis.
Stage four is best managed with hip replacement
whereas precollapse stages (zero, one, and two) may be
initially treated with joint-preserving measures (core decompression or vascularized fibular graft). The treatment of stage three disease in middle-aged and older
patients usually is THA; in younger patients, jointsparing procedures of hemiresurfacing arthroplasty may
be considered. In recent years bone grafting of the femoral head by elevating the articular cartilage flap (trapdoor approach) has been attempted for patients with
stage two and three disease.
Core decompression involves drilling multiple holes
through the avascular portion of the femoral head under fluoroscopic guidance. Vascularized bone grafting,
on the other hand, involves the use of a vascularized fibular graft with microsurgical anastomosis to local blood
vessels. The vascularized graft is harvested from the central portion of the fibula. It is then inserted through a
drill hole into the avascular zone of the femoral head.
One recent study noted a survival rate of 67% for postcollapse osteonecrosis of the femoral head treated using
this technique.

Total Hip Arthroplasty

Cemented Femoral Component
The best choice for fixation of femoral components continues to be a subject of debate. Several studies have reported excellent mid-term and long-term survivorship
with cemented femoral components, particularly with
the use of modern cementation techniques. There are a
number of important factors that affect the outcome of
arthroplasty using a cemented femoral stem. Some of
these factors are patient selection, geometry and surface
finish of the implant, femoral neck design of the implant, material used for the implant, and surgical technique. Overall, younger patients, who are more active,
have an increased risk of cemented femoral component
failure than older patients. Men have an approximate
twofold higher risk of failure than women. Torsional and
axial stability in the cement mantle is an important determinant of the success of cemented femoral components. Numerous designs of femoral stems have been
successful in providing stability. Some designs aim to

416

provide stability without possibility of subsidence (calcar resting stems), whereas others allow controlled subsidence of the stem within the cement mantle over time.
Surface finish of the component also contributes to the
stability of the stem. The roughened surfaces allow better cement interdigitation and minimize subsidence. The
smooth stems, on the other hand, permit taper-slip subsidence. There is conflicting evidence in the literature
regarding which type of surface finish will provide better long-term results. One of the primary mechanisms
for failure of cemented stems is initiation and propagation of cracks in the cement mantle through preexisting
pores. Modern generation cementing techniques with
emphasis on porosity reduction, good pressurization of
the cement to obtain a uniform cement mantle, and optimal interdigitation into cancellous bone is believed to
be a critical determinant of success of cemented femoral
components. The long-term performance of cemented
femoral stems is likely to be influenced by a combination of these factors.

Cementless Femoral Component

The premise that cement was responsible for the failure
of hip arthroplasties led to the development of cementless hip components to decrease the incidence of aseptic
loosening following THA. In recent years, there has
been a shift toward the use of cementless femoral components in North America, particularly in young patients. Various factors are known to influence the performance of cementless femoral components, such as
geometry, surface finish, and the extent of coating. Early
clinical studies have shown mixed results for cementless
hip femoral components. Some femoral stem designs
had high failure rates, whereas others have enjoyed
higher rates of success. Durable fixation and a low rate
of thigh pain have been reported for different stem designs, including proximally- and extensively-coated
stems with different geometries. The value of coating
the femoral stem with hydroxyapatite is debatable; it
enhanced fixation in one design but has had no influence on the performance of other femoral components.

Cementless Cups
Durable results for porous-coated cups have been reported by various centers. A recent study reported the
15-year outcome of 120 primary THAs performed with
a cementless cup in a relatively young patient population. More than half of the patients were still alive after
15 years. No cup was revised for loosening. The linear
polyethylene wear was 0.15 mm/yr. Pelvic osteolysis was
observed in 6.9% of the surviving patients. The durability of fixation was excellent and was superior to that associated with cups that had been inserted with cement
by the same surgeon.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 34 Hip, Pelvic Reconstruction, and Arthroplasty

Revision of the Femoral Component

Various factors influence the choice of femoral stem
during revision surgery, including the extent of bone
loss, age and physical demands of the patient, reason for
revision, and status of the femoral canal. Because of a
relatively high failure rate, cemented femoral stems are
used with decreasing frequency during revision arthroplasty and cemented stems are mostly reserved for the
elderly and low-demand patients. Although revision of a
failed cementless femoral implant with cement provided
pain relief and improved function for most patients, the
rate of loosening at the time of intermediate-term
follow-up was high. Bone removal at the time of the initial implantation of the cementless stem and bone loss
caused by subsequent failure of the cementless implant
often left little intramedullary cancellous bone, which
may explain the high rate of loosening observed in the
first decade after revision. Most surgeons consider cementing of the femoral component (without impaction
bone grafting) during revision surgery only if an adequate amount of cancellous bone exists in the canal to
allow cement interdigitation. Extraction of a previously
cemented femoral component usually leads to the loss
of cancellous bone and reduces the fixation of cemented
femoral stems under this circumstance. Some surgeons
may cement a femoral component into a preexisting, intact, well-fixed cement mantle.
Most of the femoral components used during revision surgery are inserted without cement. Proximally
coated monoblock stems have a high failure rate in revision surgery and therefore are rarely used, but modular
proximally-coated stems have had a higher success rate.
Extensively porous-coated cylindrical stems are used
frequently for revision arthroplasty and reportedly are
associated with excellent outcome in appropriately selected patients. Ten-year survivorship free of revision for
these stems was reported to be 89% in one study. Cylindrical extensively-coated stems have a higher failure
rate when used in Paprosky type IIIB or IV defects or
with poor cortical bone in the isthmus. Impaction grafting is preferred by some surgeons for these patients.
Fluted modular stems that rely on diaphyseal fixation
have been popular in Europe and are gaining popularity
in North America. However, fluted conical stems may
be subject to subsidence. Extended osteotomy may allow better press fit for these stems (Figure 6). In revision surgery, a stem of adequate length should be used.
Resistance to torsion progressively increases in association with increasing depth of insertion. A cadaveric
study evaluated the rotational stability of two different
revision stem designs (plasma sprayed cylindrical stem
and fluted cylindrical stems) at varying depths of insertion (20 mm, 40 mm, and 60 mm). There was no difference in rotational stability of the two stems. Importantly, the actual length of bone-implant contact was

American Academy of Orthopaedic Surgeons

Figure 6 Fluted stems should have 5 to 8 cm of diaphyseal fixation. With the natural
curvature of the femur, proper implantation of these stems to obtain the required diaphyseal contact is difficult and may lead to subsidence. Extended femoral osteotomy,
with or without transverse reduction osteotomy allows better implantation of these
stems.

found to be around 50% of the presumed values at all

three depths. Stems with good diaphyseal fixation and
poor proximal bone support can fracture (Figure 7).

Revision of the Acetabular Component

Cementless fixation is the preferred technique of acetabular fixation during revision arthroplasty. Excellent
mid-term and long-term survivorship free of failure has
been reported. Therefore, the use of hemispherical cementless cups is preferred whenever possible. Severe
bone loss leading to distortion of acetabular architecture that may in turn compromise fixation, or native
bone-to-metal contact that is essential for osseointegration may preclude the use of conventional cementless
hemispherical cups. The recent introduction of cups
made of the trabecular metal (such as tantalum) possibly with better osseointegration potential has led surgeons to implant these cups when the bone-to-metal
contact surface has been small (less than 50% of the
surface area). Large or so-called jumbo cementless cups
may be used for patients with severe but contained
bone defects with an intact acetabular rim. The use of
jumbo cups has been reported to result in excellent outcomes, with over 90% survivorship at 10 years, provided
that there is sufficient rim and posterior column support. Bone grafting of cavitary or segmental defects is
valuable in restoring the bone stock. When placement of
an uncemented hemispherical cup is not possible, reconstruction cages may be used, but because of lack of biologic fixation potential, loosening and component fracture occur in some patients.
Isolated liner exchange to address severe wear of
polyethylene is an acceptable method of treatment.
Some studies of revision surgery for isolated liner exchange have noted a high dislocation rate.

417

Orthopaedic Knowledge Update 8

Hip, Pelvic Reconstruction, and Arthroplasty

Figure 7 Cantilever failure for cementless stem. The stem is well fixed distally. Because of poor proximal bone support and repeated loading the stem has fractured at
the weakest point (neck and body junction).

Infection
Although the management of deep periprosthetic infections has become more successful over the past decade,
it is still one of the most challenging complications of
joint arthroplasty surgery. The most important issue in
management of deep periprosthetic infections is prevention. Prophylactic antibiotic use is likely the most important factor for reducing the incidence of deep infection
from 9% 30 years ago to the present rate of between
1% to 2% for primary and 2% to 4% for revision arthroplasty. Antibiotics should be administered 30 minutes before the skin incision and for 24 hours after surgery. However, the orthopaedic community should not
rely solely on the use of antibiotics. A clean air environment (vertical laminar flow, body exhaust system, limiting traffic flow and personnel), effective skin preparation (iodine or povidine with alcohol) and draping (use
of adhesive iodine), efforts to reduce skin bacteria
(shaving, iodine showers, treatment of skin lesions),
careful attention to surgical technique, and expeditious
execution of the surgery are very important principles.
Studies have shown that contamination of instruments

418

(occurring during set up), suction tips (in 50% of affected patients), splash basins (in 70% of patients), and
contamination arising from glove perforations (in 100%
of patients after 3 hours) are common.
Treatment options are antibiotic suppression alone,
dbridement and antibiotics, prosthesis removal with
one- or two-staged reimplantation, or resection arthroplasty. Antibiotic suppression without surgery may be
indicated for medically infirm patients with susceptible
organisms, well-fixed components, and the ability to tolerate oral antibiotics. Antibiotic suppression is contraindicated in patients with resistant organisms. Deep
periprosthetic infections presenting within 4 weeks of
the initial arthroplasty or a late hematogenous infection
presenting with brief history of symptoms such as pain
and swelling, or the inciting events that lead to infection
may respond to treatment with dbridement, retention
of the prosthesis, and antibiotics. The success of this procedure depends on host-related factors (immune status,
age, soft tissue); organism-related factors (type, susceptibility, response to antibiotics); and surgical factors (interval for presentation before surgery, extent of dbridement, soft-tissue coverage). The early success of
dbridement and antibiotics, in appropriately selected
patients, is greater than 70%, but the results deteriorate
over time so that by 2 years 56%, and by 5 years only
26% of patients remained infection free in one study.
Resection arthroplasty with one-stage or two-stage reimplantation is the treatment of choice for most patients
presenting with deep periprosthetic infections. Some authorities advocate removal of the components and insertion of new prosthesis under the same anesthesia (onestage reimplantation) for patients with low virulence
and sensitive organisms, or for patients unable to tolerate multiple procedures. The success (patients who are
infection free) of one-stage reimplantation varies between 73% to 92%.

Wear
Wear of the bearing surfaces has become the most important factor limiting the longevity of most hip arthroplasties. Table 1 summarizes the overall advantages and
disadvantages of current bearing choices. Age and activity level are among the most important predictors of
wear. Increased body mass may have a protective effect
on wear as increased body mass index has been associated with reduced activity. The mean number of gait cycles per year as measured with a pedometer is approximately 1.2 million. The mean wear for metal femoral
head on conventional polyethylene is 0.14 to 0.2 mm per
year. Computerized radiographic wear analysis programs have been developed. These techniques are accurate to a variance in the range of 0.25 to 0.41 mm.
Multiple factors can influence the radiographic measurement of polyethylene wear, including the quality of

American Academy of Orthopaedic Surgeons

Chapter 34 Hip, Pelvic Reconstruction, and Arthroplasty

Orthopaedic Knowledge Update 8

Table 1 | Bearing Surfaces

Overall Advantages and Disadvantages of Current Bearing Choices
Bearing Combination
Alumina-on-alumina

Potential Advantages
Usually very low wear
High biocompatibility

Cobalt-chromium on cobalt-chromium

Usually very low wear

Can self-polish moderate surface scratches
Lower wear of polyethylene than with conventional
metal-polyethylene
Some additional protection against third body
abrasion
No short-term or long-term oxidation

Ceramic-on-polyethylene

Polyethylene sterilized with ethylene oxide or gas

plasma
Polyethylene sterilized with gamma in low oxygen
Cross-linked, thermally-stabilized polyethylene

Some cross-linking, some wear reduction

Minimal polyethylene wear rate
No short-term or long-term oxidative degradation

Potential Disadvantages
Sometimes high wear
Component fracture
Higher cost
Technique-sensitive surgery
Question of long-term local and systemic
reactions to metal debris and/or ions
Component fracture
Difficulty of revision (that is if Morse taper is
damaged)
Higher cost
No cross-linking so does not minimize
polyethylene wear
Polyethylene wear not minimized
Residual free radicals (long-term oxidation?)
Newest of low wear bearing combinations, only
early clinical results available
Questions remain regarding optimum level and
optimum method for thermal stabilization

(Reproduced from McKellop HA: Bear surfaces in total hip replacements: State of the art and future developments. Instr Course Lect 2001;50:165-179.)

the radiographs, polyethylene creep, and the manufacturing tolerances of the shell and the liner. To surmount
the problem of wear, various improvements have been
achieved.

Highly Cross-Linked Polyethylene

Highly cross-linked polyethylene has demonstrated
markedly decreased wear in vitro compared with conventional polyethylene. Retrieval analysis evaluating the
microscopic damage to the surfaces did not detect any
difference in the quantity or the quality of surface damage between the conventional and the highly crosslinked polyethylene. Multidirectional scratching was the
most prevalent pattern of damage on all retrieved liners.
Recently, in vivo measurements of highly cross-linked
polyethylene are becoming available. In one study, the
cross-linked polyethylene was associated with a 65% reduction in the two-dimensional linear wear rate, a 54%
reduction in three-dimensional wear rate, and 38% reduction in volumetric wear rate. The mean wear rate for
cross-linked polyethylene is 0.07 mm/yr. Early polyethylene wear measurement accuracy is hampered by a
bedding-in phenomenon (in part, polyethylene creep)
over the first 2 years after implantation when femoral
head penetration rates are elevated; thereafter, femoral
head penetration rates revert to the steady state wear
rate.

American Academy of Orthopaedic Surgeons

Ceramic
Ceramic-on-ceramic articulations have been used extensively in Europe. Several series have demonstrated good
midterm clinical results with alumina on alumina bearings. The wear rate appears to be very low and risk of
ceramic fracture has been markedly reduced compared
with early studies. Fractures have not been completely
eliminated, however, and the development of improved
ceramics continues.

Metal-on-Metal
Because of a relatively high rate of failure, the first generation of metal-on-metal hips were largely supplanted
by metal-on-polyethylene prostheses. However, newer
generations of metal-on-metal prosthesis have become
available and in vitro studies suggest that the metal-onmetal bearing surface has markedly superior wear characteristics to the metal-on-polyethylene surface. Metalon-metal surfaces do not pose a risk of fracture. The
major concern about metal-on-metal prostheses is that
they appear to be associated with some elevation of systemic metal ion levels. In theory, elevated ion levels
might cause organ toxicity, mutagenicity, and carcinogenicity, but convincing evidence of these effects has not
been seen in any clinical studies. Metal-on-metal surfaces have self-healing ability; that is, smaller surface
scratches may polish out with time. Surface micropitting,

419

Hip, Pelvic Reconstruction, and Arthroplasty

thought to be related to the presence of smaller carbides, has also been observed with some first- and
second-generation metal-on-metal prostheses. Although
micropitting was not associated with high wear rates,
this complication represents an area for potential improvement. Hard-on-hard surfaces (metal-on-metal and
ceramic-on-ceramic) probably have better wear characteristics with large femoral heads and these large diameter bearings may also reduce the risk of intra-articular
impingement and dislocation. However, any intraarticular impingement with hard-on-hard surfaces can
create greater potential problems than with metal-onpolyethylene bearings.

Osteolysis
Improving bearing surfaces and reducing wear particles
hopefully will reduce particle-induced periprosthetic osteolysis in the future; currently, osteolysis continues to
be a major long-term complication of hip arthroplasty.
Osteolysis generally develops and progresses in the absence of clinical symptoms. Hence, close monitoring of
hip arthroplasties to detect and treat osteolysis is important. Radiographs underestimate the extent of osteolysis, particularly in the periacetabular region. CT with
special software has been used for quantification of pelvis osteolysis. Osteolysis related to particulate debris occurs as a result of phagocytosis of wear debris, which in
turn leads to activation of inflammatory cells (macrophages) and ultimately recruitment of osteoclasts. The
cytokine OPG/RANKL/RANK is believed to be an important mediator of differentiation of osteoclasts and
their interaction with osteoblasts. Osteoprotegrin
(OPG) decreases osteoclast differentiation by working
through a receptor-ligand interaction (RANKL). Tumor
necrosis factor-, a cytokine released in response to
phagocytosis of wear particles, is found to have a role in
stimulating osteoclast formation.

Dislocation
In one study, the cumulative risk of any dislocations was
2.2% at 1 year, 3.8% at 10 years, and 6% at 20 years.
The 10-year risk was 3.2% for anterolateral approach
and 6.8% for posterolateral approach. One third to two
thirds of dislocations can be treated by closed reduction
without further complications. Dislocation is more common in association with the following factors: posterolateral approach, smaller femoral head size, trochanteric
nonunion, obesity, alcoholism, neuromuscular conditions, and revision surgery.

Thromboembolic Disease
The ultimate goal of prophylaxis is to prevent fatal pulmonary embolism. Deep venous thrombosis may also
lead to postthrombotic limb syndrome with swelling, ulceration, and extremity pain. The chemical method of

420

Orthopaedic Knowledge Update 8

prophylaxis is preferred in North America. Coumadin
has been shown to be effective in preventing fatal pulmonary embolism. Low molecular weight heparin is also
an effective method of prophylaxis but carries a higher
rate of bleeding complications according to most studies. Mechanical methods such as pneumatic compression
devices have been shown to be effective for reducing
the prevalence of thrombosis after THA. These devices
are believed to achieve their efficacy by increasing
blood flow and localized fibrinolysis (systemic fibrinolysis, once believed to be important, could not be demonstrated by a recent study). Oral anticoagulation agents
that do not require serologic monitoring have recently
been tested in clinical trails for prevention of deep
venous thrombosis following total knee arthroplasty,
with promising results.

Periprosthetic Fractures After Total Hip

Arthroplasty
Periprosthetic fractures following THA occur in a variety of different clinical situations. They may occur intraoperatively, during the postoperative period, or many
years after hip replacement. Periprosthetic fractures
may occur as a result of trauma or secondary to late deleterious effects of hip replacement components on the
surrounding bone (such as osteolysis or stress shielding).
The incidence of intraoperative periprosthetic fractures after THA is higher using cementless rather than
cemented hip replacement components. The location of
intraoperative fractures related to cementless components depends on the type of prosthesis used. Wedge-fit,
tapered stems tend to cause proximal femur fractures
and cylindrical fully porous-coated stems tend to cause
a distal split in the femoral shaft.
Several classification systems for periprosthetic femur fractures exist. The Vancouver classification of
periprosthetic fractures takes into consideration the stability of the prosthesis as well as the location of the fracture and the quality of the surrounding bone. Type A
fractures occur in the trochanteric region. Type B fractures occur around the stem or just below it. Patients in
subgroup B1 have a well-fixed stem, in B2 the stem is
loose, and in B3 the stem is loose and the proximal bone
is of poor quality or severely comminuted. Type C fractures occur well below the prosthesis.
Management of periprosthetic fractures depends on
the location of the fracture as well as the quality of
bone stock and stability or fixation of the hip arthroplasty components. Type A periprosthetic femur fractures commonly are associated with osteolysis, which
may require treatment of the process causing osteolysis.
Type B1 periprosthetic fractures almost always are
treated surgically, usually with open reduction and internal fixation. Cortical strut allografts, cerclage cables, and

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

locking plates with unicortical screws may be necessary
to achieve adequate fixation. Type B2 periprosthetic femur fracture (femoral component is loose) usually are
treated with revision of the femoral component and fixation of the fracture fragment. Type B3 periprosthetic
fractures require femoral component revision. Usually,
either proximal femoral allograft or proximal femoral
replacement is necessary. Type C periprosthetic femur
fractures usually are treated with open reduction and internal fixation. The hip replacement components can
usually be left alone.
Periprosthetic acetabular fractures are less common
than femoral fractures. With cementless acetabular components, a tight interference fit is achieved by underreaming the acetabulum. Underreaming by more than
2 mm may create excess stress in acetabular bone, resulting in intraoperative fracture. If acetabular fracture
is noted, additional cup fixation with screws may be
needed. Osteolysis may result in significant weakening
of the retroacetabular bone. Fracture through this weakened bone can cause acute failure of the acetabular
component.
Prevention of periprosthetic fractures requires recognition of risk factors. Adequate surgical exposure during primary and revision procedures is necessary. Controlled osteotomy is preferable to damaging weak
proximal femoral bone during retraction or implant removal. Special care must be taken when using cementless implants in patients with poor bone stock, such as
those with rheumatoid arthritis or severe osteoporosis.
Cortical perforations must be bypassed with the implant, structural bone graft, or fracture plate to a sufficient degree to avoid a stress riser. Careful, regular
follow-up of arthroplasty patients is necessary to evaluate for failure of implants. Elective revision of failed implants with osteolysis and mechanical failure of components may prevent some periprosthetic fractures.

Annotated Bibliography
Femoroacetabular Impingement
Beck M, Leunig M, Parvizi J, Boutier V, Wyss D, Ganz
R: Anterior femoroacetabular impingement: Part II.
Midterm results of surgical treatment. Clin Orthop 2004;
418:67-73.
The outcome of surgical dislocation and osteoplasty of the
femur and the acetabulum in 19 patients with a mean age of
36 years (range, 21 to 52 years) was reported. The follow-up
averaged 4.7 years (range, 4 to 5.2 years). Using the Merle
dAubigne hip score, 13 hips were rated excellent to good. In
the hips without subluxation of the head into the acetabular
cartilage defect, no additional joint space narrowing occurred.
According to this study, surgical dislocation with correction of
femoroacetabular impingement yields good results in patients
with early degenerative changes not exceeding grade 1 osteoarthrosis.

American Academy of Orthopaedic Surgeons

Chapter 34 Hip, Pelvic Reconstruction, and Arthroplasty

Ganz R, Gill TJ, Gautier E, et al: Surgical dislocation of
the adult hip: A technique with full access to femoral
head and acetabulum without the risk of avascular necrosis. J Bone Joint Surg Br 2001;83:1119-1124.
The authors describe a novel technique for surgical dislocation of the hip, based on detailed anatomic studies of the
blood supply. The approach involves anterior dislocation
through a posterior approach with a trochanteric flip osteotomy. The external rotator muscles are not divided and the medial femoral circumflex artery is protected by the intact obturator externus.

Ganz R, Parvizi J, Beck M, Leunig M, Ntzli H, Siebenrock KA: Femoroacetabular impingement: A cause for
osteoarthritis of the hip. Clin Orthop 2003;417:112-120.
The authors, based on clinical experience with more than
600 surgical dislocations of the hip, allowing in situ inspection
of the damage pattern and the dynamic proof of its origin, propose femoroacetabular impingement as a mechanism for the
development of early OA for most nondysplastic hips.

Hip Arthroscopy
Clarke MT, Arora A, Villar RN: Hip arthroscopy: complications in 1054 cases. Clin Orthop 2003;406:84-88.
Complications in a large series of hip arthroscopies are reviewed.

McCarthy JC, Noble PC, Schuck MR, Wright J, Lee J:

The role of labral lesion to development of early degenerative hip disease. Clin Orthop 2001;393:25-37.
Arthroscopic findings in a large series of hip arthroscopies
are correlated with cadaveric specimens to support the idea
that acetabular labral tears and degenerative arthritis are part
of a continuum of disease.

Osteonecrosis of the Femoral Head

Berend KR, Gunneson EE, Urbaniak JR: Free vascularized fibular grafting for the treatment of postcollapse
osteonecrosis of the femoral head. J Bone Joint Surg
Am 2003;85:987-993.
A survival rate of 67% is noted for postcollapse osteonecrosis of the femoral head treated with free vascularized fibula
grafting.

Total Hip Arthroplasty

Boucher HR, Lynch C, Young AM, Engh CA Jr, Engh C
Sr: Dislocation after polyethylene liner exchange in total hip arthoplasty. J Arthroplasty 2003;18:654-657.
Twenty-four patients undergoing an isolated polyethylene
liner exchange for wear or osteolysis with retention of the acetabular shell and femoral stem were assessed. At a mean 56month follow-up, six hips (25%) had dislocated. Of these, two
underwent repeat surgery for recurrent dislocation; one had
three dislocations, one had two dislocations, and two had single dislocations. It was concluded that polyethylene liner exchanges, with or without femoral head exchange for wear or

421

Hip, Pelvic Reconstruction, and Arthroplasty

Orthopaedic Knowledge Update 8

osteolysis, are associated with a high risk of dislocation and

possible decrease in function.

precoating is not detrimental to successful fixation of primary

hybrid THA.

Davis CM III, Berry DJ, Harmsen WS: Cemented revision of failed uncemented femoral components of total
hip arthroplasty. J Bone Joint Surg Am 2003;85:12641269.

Laupacis A, Bourne R, Rorabeck C, Feeny D, Tugwell P,

Wong C: Comparison of total hip arthroplasty performed with and without cement: a randomized trial.
J Bone Joint Surg Am 2002;84:1823-1828.

Forty-eight consecutive hips in which a failed primary cementless femoral component was revised with use of cement
at the Mayo Clinic. Rate of loosening at the time of
intermediate-term follow-up was higher than that commonly
reported after revision of failed cemented implants with use of
cement and also was higher than that commonly reported after revision with use of cementless extensively porous-coated
implants.

The study reported the intermediate-term outcome of a

prospective, randomized clinical trial in which 250 patients
were randomly assigned to receive a hip prosthesis with cement or the same prosthesis designed for insertion without cement. At 6.3-year follow-up, the number of stem revisions in
the group that had fixation with cement was significantly
higher than that in the group that had fixation without cement
(P < 0.002).

Gaffey JL, Callaghan JJ, Pedersen DR, Goetz DD, Sullivan PM, Johnston RC: Cementless acetabular fixation at
fifteen years: A comparison with the same surgeons results following acetabular fixation with cement. J Bone
Joint Surg Am 2004;86:257-261.

Parvizi J, Sharkey PF, Hozack WJ, Orzoco F, Bissett GA,

Rothman RH: Prospective matched-pair analysis of
hydroxyapatite-coated and uncoated femoral stems in
total hip arthroplasty: A concise follow-up of a previous
report. J Bone Joint Surg Am 2004;86:783-786.

The outcome of THA using Harris-Galante I cementless

acetabular component and cemented femoral stem in 120 patients at 13 to 15 years was evaluated. No acetabular component had been revised because of aseptic loosening, and no acetabular component had migrated. Among the 70 hips with at
least 13 years of radiographic follow-up, 5 had pelvic osteolysis
and 3 had revision of a well-fixed acetabular component because of pelvic osteolysis secondary to polyethylene wear. The
mean linear wear rate was 0.15 mm/yr.

Klapach AS, Callaghan JJ, Goetz DD, Olejniczak JP,

Johnston RC: Charnley total hip arthroplasty with use
of improved cementing techniques: A minimum twentyyear follow-up study. J Bone Joint Surg Am 2001;83:
1840-1848.
Of the 91 hips in the 82 patients who were alive at a minimum of 20 years, authors reported five (5%) revisions for
aseptic loosening of the femoral component. The rate of failure when radiographic signs of loosening were included was
4.8%. Adequate filling of the femoral canal with cement was
found to be associated with improved survival of the femoral
component.

Lachiewicz PF, Messick P: Precoated femoral component in primary hybrid total hip arthroplasty: Results at
a mean of 10-year follow-up. J Arthroplasty 2003;18:1-5.
This study reports the midterm results of a precoated femoral component used in primary hybrid THA. Of an original
cohort of 98 hips undergoing THA performed by one surgeon,
75 hips in 65 patients were prospectively followed up for 7 to
12 years. All hips had the same porous-coated acetabular component and a precoated femoral component (with an oval
cross-section) implanted using bone cement. There was no
femoral component loosening or revision. The authors concluded that if used in this manner in this patient population,

422

This prospective study reporting the results of THA using

tapered cementless femoral component with and without hydroxyapatite coating in a matched-pair group of 52 patients
noted no difference in osseointegration, revision rate, or radiolucency between the femoral components in each group at
9.8-year follow-up.

Patel JV, Masonis JL, Bourne RB, Rorabeck CH: The

fate of cementless jumbo cups in revision hip arthroplasty. J Arthroplasty 2003;18:129-133.
This study reported 5-year minimum results of cementless
oversized cups used in revision hip arthroplasty, with significant associated bone defects. Forty-three porous-coated jumbo
cups were used to treat acetabular defects in revision hip arthroplasty in 42 patients with a mean age of 63 years (range,
25 to 86 years). Morcellized allograft only was used in 27 hips,
and bulk allograft was used in 8. At a mean follow-up of 10
years, two acetabular components were revised for aseptic
loosening and graft resorption. Dislocation occurred in two
hips. A satisfactory 92% Kaplan-Meier shell survival rate was
seen at 14 years.

Teloken MA, Bissett G, Hozack WJ, Sharkey PF, Rothman RH: Ten to fifteen-year follow-up after total hip arthroplasty with a tapered cobalt-chromium femoral
component (tri-lock) inserted without cement. J Bone
Joint Surg Am 2002;84:2140-2144.
Excellent 15-year results were reported in a study of 49
patients receieving tapered cobalt-chromium, collarless,
proximally-coated tapered stems. No stems were revised for
loosening, while two were judged to be loose radiographically.
Most importantly, no stem that was judged to be boneingrown at 2 years afer surgery progressed to loosening.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Infection
Hanssen AD, Osmon DR: Evaluation of a staging system for infected hip arthroplasty. Clin Orthop 2002;403:
16-22.
A previously reported staging system for prosthetic joint
infection was evaluated in 26 consecutive patients with an infected hip arthroplasty. Six patients were treated by a definitive resection arthroplasty whereas the remaining 20 patients
received delayed insertion of another hip arthroplasty. In 4 of
the 20 patients (20%) receiving a new prosthesis, reinfection
developed. The only common variable among the patients who
had reinfection was the use of a massive femoral allograft at
reconstruction. The authors concluded that although the concept of a staging system for treatment of an infected hip arthroplasty is promising, the number of patients required to
evaluate the use of a staging system will require a multicenter
collaborative study.

Wear
Alberton GM, High WA, Morrey BF: Dislocation after
revision total hip arthroplasty: An analysis of risk factors and treatment options. J Bone Joint Surg Am 2002;
84:1788-1792.
Data were obtained from 1,548 revision arthroplasties in
1,405 patients at the Mayo Clinic. The dislocation rate was
7.4%. Larger femoral head and elevated acetabular liners reduced the incidence while trochanteric nonunion was a significant risk factor for subsequent dislocation.

Barrack RL, Cook SD, Patron LP, Salkeld SL, Szuszczewicz E, Whitecloud TS III: Induction of bone ingrowth
from acetabular defects to a porous surface with OP-1.
Clin Orthop 2003;417:41-49.
To evaluate the role osteoinductive bone proteins may
play in enhancing bone ingrowth, six canines had bilateral
THAs with a cementless press-fit porous-coated acetabular
component. The osteogenic protein-treated defects healed
more completely than allograft bone-treated or empty defects
and achieved a bone density equivalent to the intact acetabulum. Bone ingrowth also occurred to a significantly higher degree in the osteogenic protein group compared with allograft
or empty defects, achieving a degree of ingrowth equivalent to
the intact acetabulum controls. The osteogenic bone protein
was successful in achieving complete defect healing and inducing extensive ingrowth from the defect into the adjacent porous coating.

Berry DJ, Von Knoch M, Schleck CD, Harmsen WS: The

cumulative long-term risk of dislocation after primary
Charnley total hip arthroplasty. J Bone Joint Surg Am
2004;86:9-14.
The cumulative risk of dislocation in 5,459 patients undergoing cemented Charnley hip arthroplasty at one institution
was calculated. There were 320 (4.8%) dislocations. The cumulative risk of a first-time dislocation was 1% at 1 month and
1.9% at 1 year and then rose at a constant rate of approximately 1% every 5 years to 7% at 25 years for patients. Multi-

American Academy of Orthopaedic Surgeons

Chapter 34 Hip, Pelvic Reconstruction, and Arthroplasty

variate analysis revealed that the relative risk of dislocation
for female patients (as compared with male patients) was 2.1
and that the relative risk for patients who were 70 years old or
older (as compared with those who younger than 70 years old)
was 1.3. Three underlying diagnoses, osteonecrosis of the femoral head; acute fracture or nonunion of the proximal part of
the femur; and inflammatory arthritis, were associated with a
significantly greater risk of dislocation than OA.

Hui AJ, McCalden RW, Martell JM, MacDonald SJ,

Bourne RB, Rorabeck CH: Validation of two and threedimentional radiographic techniques for measuring
polyethylene wear after total hip arthroplasty. J Bone
Joint Surg Am 2003;85:505-511.
This study sought to validate two in vivo radiographic
wear measurement techniques by comparing their results with
those obtained directly from retrieved specimens. There was
good agreement between the wear estimates made with both
in vivo techniques and the measurements of the retrieved
polyethylene liners made with the coordinate measuring machine. Two-dimensional wear analysis (based on AP radiographs) accounted for most of the polyethylene wear, while
one technique of three-dimensional wear analysis demonstrated some additional wear in the lateral plane.

Macaulay W, Westrich G, Sharrock N, et al: Effect of

pneumatic compression on fibrinolysis after total hip arthoplasty. Clin Orthop 2002;399:168-176.
This prospective randomized clinical trial investigated the
possible enhanced systemic fibrinolysis mechanism of venous
thrombosis prevention by pneumatic compression after THA
in 50 patients. Serum determinations of antigen of tissue plasminogen activator and plasminogen activator inhibitor-1 were
done using enzyme-linked immunosorbent assays. The data
did not support the enhancement of systemic fibrinolysis
mechanism for lowering thromboembolic risk after THA by
pneumatic compression devices.

Martell JM, Verner JJ, Incavo SJ: Clinical performance

of a highly cross-linked polyethylene at two years in
total hip arthoplasty: A randomized prospective trial.
J Arthroplasty 2003;18:55-59.
The 2-year results for a prospective randomized trial comparing highly cross-linked with standard polyethylene in total
hip replacement revealed a significant reduction in two- and
three-dimensional linear wear rates (42% and 50%) in the
highly cross-linked group (P = 0.001 and P = 0.005). Forty-six
hips were available for radiographic analysis at 2- and 3-year
follow-up. Femoral bearings were 28-mm cobalt-chromium,
with the polyethylene insert randomly selected at the time of
implantation to be highly cross-linked polyethylene or standard polyethylene.

Ulrich-Vinther M, Carmody EE, Goater JJ, Soballe K,

OKeefe RJ, Schartz EM: Recombinant adenoassociated virus-mediated osteoprotegrin gene therapy

423

Hip, Pelvic Reconstruction, and Arthroplasty

inhibits wear debris-induced osteolysis. J Bone Joint
Surg Am 2002;84:1405-1412.
The role of gene therapy using a recombinant adenoassociated viral vector expressing OPG, a natural decoy protein that inhibits osteoclast activation and bone resorption, to
inhibit wear debris-induced osteolysis in a mouse calvarial
model was investigated. A single intramuscular injection of the
vector efficiently tranduced myocytes to produce high levels
of OPG. The OPG effectively inhibited wear debris-induced
osteoclastogenesis and osteolysis.

Classic Bibliography
Berry DJ, Harmsen WS, Ilstrup D, Lewallen DG, Cabanela ME: Survivorship of uncemented proximally
porous-coated femoral components. Clin Orthop 1995;
319:168-177.
Callaghan JJ, Brand RA, Pedersen DR: Hip arthrodesis:
A long-term follow-up. J Bone Joint Surg Am 1985;67:
1328-1335.

Orthopaedic Knowledge Update 8

McDonald DJ, Fitzgerald RH Jr, Ilstrup DM: Two-stage
reconstruction of a total hip arthroplasty because of infection. J Bone Joint Surg Am 1989;71:828-834.
Mont MA, Carbone JJ, Fairbank AC: Core decompression versus nonoperative management for osteonecrosis
of the hip. Clin Orthop 1996;324:169-178.
Reynolds D, Lucas J, Klaue K: Retroversion of the acetabulum: A cause of hip pain. J Bone Joint Surg Br 1999;
81:281-288.
Rittmeister M, Starker M, Zichner L: Hip and knee replacement after longstanding hip arthrodesis. Clin
Orthop 2000;371:136-145.
Steinberg ME, Bands RE: ParryS, Hoffman E, Chan T,
Hartmean KM: Does lesion size affect the outcome in
avascular necrosis? Clin Orthop 1999;367:262-271.

Charnley J, Eftekhar N: Postoperative infection in total

prosthetic replacement arthroplasty of the hip joint with
special reference to the bacterial content of the air in
the operating room. Br J Surg 1969;56:641-649.

Urbaniak JR, Coogan PG, Gunneson EB, Nunley JA:

Treatment of osteonecrosis of the femoral head with
free vascularized fibular grafting: A long-term follow-up
study of one hundred and three hips. J Bone Joint Surg
Am 1995;77:681-694.

Ganz R, Klaue K, Vinh TS, Mast JW: A new periacetabular osteotomy for the treatment of hip dysplasias: Technique and preliminary results. Clin Orthop 1988;232:2636.

Waters RL, Barnes G, Husserl T, Silver L, Liss R: Comparable energy expenditure after arthrodesis of the hip
and ankle. J Bone Joint Surg Am 1988;70:1032-1037.

Gie GA, Linder L, Ling RS, Simon JP, Slooff TJ, Timperley AJ: Impacted cancellous allografts and cement for
revision total hip arthroplasty. J Bone Joint Surg Br
1993;75:14-21.

Younger TI, Bradford MS, Magnus RE, Paprosky W: Extended proximal femoral osteotomy: A new technique
for femoral revision arthroplasty. J Arthroplasty
1995;10:329-338.

424

American Academy of Orthopaedic Surgeons

Chapter

35

Femur: Trauma
William M. Ricci, MD

The Winquist and Hansen classification system for femoral shaft fractures is based on the diameter of bone that
is comminuted (Figure 1). Type I fractures have a small
area of comminution, with greater than 75% of the diameter of the bone remaining in continuity. Type II fractures
have increased comminution, but with at least 50% of the
diameter intact.Type III fractures have less than 50% cortical contact. Type IV fractures are defined as having no
abutment of the cortices at the level of the fracture to prevent shortening. Type I and II fractures are axially stable,
whereas type III and IV fractures are both axially and rotationally unstable. Rotational stability for less comminuted fractures is determined by the amount of comminution and obliquity of the fracture, with more transverse
fracture patterns being less rotationally stable.Axially stable fractures are more amenable to earlier weight bearing, especially after intramedullary (IM) nailing.
The AO/Orthopaedic Trauma Association (OTA) classification system is also commonly used, especially for
comparative investigations (Figure 2). Fractures of the
femoral shaft are designated as 32. Type 32A fractures
are simple (without comminution), type 32B are comminuted but maintain some degree of cortical continuity between the proximal and distal shaft fragments, and type
32C fractures have complete loss of continuity between
the proximal and distal fragments. Further subtypes represent increasing fracture complexity.
The location of the fracture along the length of the
shaft is usually described as being of the proximal, middle, or distal one third. There is some overlap between
subtrochanteric fractures and proximal one third shaft
fractures. Fractures located within 5 cm of the lesser trochanter are considered to be in the subtrochanteric region.

juries. The mechanism of injury should heighten the suspicion for other particular injuries. Motor vehicle crashes,
especially those with dashboard impact, have a high incidence of associated knee pathology (up to 60%), including ligamentous injuries, meniscal injuries, and bone contusions. Knee stability should be evaluated with the
patient under anesthesia and immediately after bony stabilization. Patients with persistent knee pain should be
evaluated for occult internal derangement and bone contusion. Falls from a height can be associated with other injuries that are common after axial loading such as calcaneal fractures and spinal compression fractures. Visceral,
chest, and head trauma should always be a consideration
in patients with high-energy femoral shaft fractures.
Bleeding at the site of the fracture is usually self-limited,
but several hundred milliliters of blood can be lost. In patients with bilateral femoral shaft fractures or those with
other long bone fractures, the cumulative bleeding associated with these fractures can become clinically significant. These patients should be monitored closely for anemia and hemodynamic changes.Associated neurovascular
injury is uncommon, but patients with diminished or
asymmetric pulses should be carefully evaluated for this
type of injury. Neurologic deficit associated with penetrating trauma may require acute surgical exploration. Although found in only 2.5% to 6% of patients with femoral
shaft fractures, associated femoral neck fractures have a
high incidence (> 30%) of misdiagnosis. All patients with
femoral shaft fractures should have AP and lateral radiographic views of the entire femur and a separate evaluation of the femoral neck with at least AP and lateral radiographic views. CT scans, which are often performed to
evaluate the abdomen and pelvis, can be useful to diagnose nondisplaced associated femoral neck fractures and
have been advocated as routine screening in patients at
high risk.

Evaluation

Treatment

Fractures of the femur are usually associated with relatively high-energy trauma.Accordingly, patients should be
carefully and systematically evaluated for associated in-

Nonsurgical Treatment
Nonsurgical treatment has a very limited role for adult
patients with femoral shaft fractures. Severely debili-

Femoral Shaft Fractures

Classification

American Academy of Orthopaedic Surgeons

425

Femur: Trauma

Orthopaedic Knowledge Update 8

Figure 1 Winquist and Hansen classification of comminuted femoral shaft fractures. (Reproduced from Poss R (ed): Orthopaedic Knowledge Update 3. Park
Ridge, IL, American Academy of Orthopaedic Surgeons, 1990, pp 513-527.)
tated, nonambulatory patients, including those with
paraplegia or those with contraindications to anesthesia,
can be treated with skeletal traction for 6 weeks followed by cast-brace application. Skeletal traction is frequently used if a delay in surgical treatment is expected
to be greater than 12 to 24 hours. Traction through the
distal femur or proximal tibia (provided there are no
ligamentous knee injuries) may be used. Distal femoral
traction provides improved stability and comfort by
avoiding traction through the knee joint.
Intramedullary Nailing
Reamed locked antegrade IM nailing through the piriformis fossa remains the gold standard for treatment of
femoral shaft fractures. Healing rates as high as 99%
with low complication rates have been achieved with
this treatment. Patients with multiple injuries who are
treated with early fracture stabilization (within 24
hours) have an improved prognosis, decreased mortality,
and fewer pulmonary complications (adult respiratory
distress syndrome, fat embolism syndrome, pneumonia,
and pulmonary failure). The advantage of early stabilization is therefore magnified in patients with chest
trauma. Prospective randomized trials have shown that
reamed nail insertion provides better healing rates than
nonreamed insertion. Increased IM pressures and fat
embolization during the reaming process has made
reaming controversial in patients with chest and lung injury. Evidence indicates that the clinical relevance of
marrow content embolization during the reaming process is negligible and is outweighed by the benefits of
reaming on the healing process. Nonetheless, sharp
reamers, proper reamer design, and slow passage of the
reamer can decrease IM pressures and fat embolization

426

Figure 2 The AO classification of femur fractures. Simple (A), Wedge (B), Complex (C).
(Reproduced with permission from Mller ME, Nazarian S, Koch P, Schatzker J (eds):
The Comprehensive Classification of Fractures of Long Bones. Berlin, Germany,
Springer-Verlag, 1990.)

American Academy of Orthopaedic Surgeons

Chapter 35 Femur: Trauma

Orthopaedic Knowledge Update 8

and should be used, especially for patients with associated chest or lung injury.
Supine positioning on a fracture table with skin traction through the distal limb is the most common method
for antegrade nailing. Antegrade nailing without traction on a radiolucent table can reduce surgical time and
can reduce the incidence of rotational malalignment because it allows for better assessment of the contralateral
limb. The lateral decubitus position offers improved access to the piriformis fossa but is associated with higher
rates of malalignment. Nail insertion through the tip of
the greater trochanter (as introduced by Kntcher) is
technically easier than through the piriformis fossa, especially in obese patients. However, varus malalignment
and iatrogenic fracture comminution are concerns with
the use of this starting point. Implants specifically designed for trochanteric insertion, with a proximal lateral
bend, can reduce these risks. Cerclage wire fixation of
displaced fragments should be avoided to prevent excessive iatrogenic soft-tissue stripping at the fracture site
that can lead to healing complications. Reaming across
segmental fracture fragments should be done with great
care to avoid spinning and therefore stripping of these
fragments.
Retrograde nailing has evolved as a viable alternative to antegrade nailing when the proper technique is
used. The insertion site should be in the intracondylar
notch approximately 1 cm anterior to the posterior cruciate ligament origin and the nail should be inserted beneath the articular surface. More complications related
to the knee have been found after retrograde nailing
and more complications related to the hip have been
found after antegrade nailing. The relative importance
of these complications on functional outcome remains
unknown. Knee stiffness and septic arthritis have not
been shown to be significant problems after retrograde
nailing. Retrograde nailing has the added benefit of improved fracture alignment of distal shaft fractures, decreased surgical time, and decreased blood loss. Current
indications for retrograde nailing include the clinical situations in which proximal access to the femur (for antegrade nailing) is either impossible or not desired. Routine use of retrograde nailing for treatment of isolated
femoral shaft fractures is limited by the unknown longterm effects on the knee.
External Fixation
External fixation as definitive treatment for acute femoral shaft fractures is limited to the pediatric population.
In adults, it can be useful for temporary fixation when
IM nailing is not advised, such as in patients with severely contaminated open wounds (especially when repeated access to a contaminated IM canal is indicated)
and in patients with associated vascular injury, for whom
time constraints preclude IM nailing.

American Academy of Orthopaedic Surgeons

Plating
Because of increased complication rates compared with
IM nailing, plate fixation of acute femoral shaft fractures is reserved for pediatric patients and for adults in
whom IM nailing is either impossible or undesirable,
such as in those with ipsilateral femoral neck and shaft
fractures, small IM canals, associated vascular injury, and
periprosthetic fractures about IM implants. Minimally
invasive methods such as indirect reduction techniques
and submuscular plating have been advocated to reduce
soft-tissue disruption and maximize healing potential
during plating of femoral shaft fractures. Plates with
locking screws have theoretic advantages in osteoporotic bone, but the specific benefits and indications
of such devices for femoral shaft fractures are yet to be
determined.

Special Situations
Open Fractures
Open fractures of the muscle-surrounded femur are
much less common (5% to 20%) than those of the subcutaneous tibia. Because of the presence of this large
protective soft-tissue envelope, open fractures are often
associated with significant soft-tissue trauma. Small skin
wounds can disguise more significant deep muscle and
periosteal injury. All open fractures of the femoral shaft
(except gunshot wounds) should be emergently treated.
Wounds should be extended for evaluation of the deep
soft tissues. All nonviable soft tissues and bone should
be dbrided. Serial dbridements at 24- to 48-hour intervals are indicated with higher-grade open injuries. Although closure of contaminated wounds should be
avoided, there is controversy on whether clean wounds
should be left open or closed between serial dbridements. Immediate IM nailing of open femoral shaft fractures is indicated except for the most severely injured
patients. Provisional external fixation is useful when repeat irrigation and dbridement of a contaminated IM
canal is necessary. IM nailing can be done when the canal has been sufficiently cleansed. Intravenous antibiotics should be initiated when the patient presents for
treatment and should be continued until definitive
wound closure takes place. Routine wound culture is
not indicated.
Gunshot Fractures
Fractures of the femur resulting from gunshot wounds
are technically open fractures; however, they can usually
be treated as closed injuries. The entry and exit wounds
should be dbrided locally at the level of skin and subcutaneous tissue. The deeper tissues do not require formal irrigation and dbridement; therefore, fracture stabilization can follow standard treatment protocols for
closed fractures. High-velocity gunshot wounds and
shotgun blasts at close range are exceptions to this

427

Orthopaedic Knowledge Update 8

Femur: Trauma
method of treatment because of severe soft-tissue compromise. In these instances, the fractures should be
treated like other high-grade open injuries.
Vascular and Nerve Injuries
Femoral shaft fractures associated with either vascular or
nerve injury are relatively uncommon (< 1%) and are usually associated with penetrating trauma. Bony stabilization, either definitive or provisional, with attention to obtaining proper limb length should be performed before
neurovascular repair. The most expeditious stabilization
method is usually external fixation, which can be safely
converted to IM nailing within 2 weeks without an increased risk of infection related to pin tracts. Great care
should be taken to avoid disruption of the soft-tissue repair during the secondary nailing procedure. Another expeditious alternative is nailing with interlocking deferred
until after neurovascular repair.
Compartment Syndrome
Compartment syndrome associated with femoral shaft
fracture is uncommon. A heightened index of suspicion
should accompany injuries with a crushing mechanism,
prolonged compression, vascular injury, systemic hypotension, and coagulopathy. When a clinical diagnosis
is made, fasciotomy should be performed emergently.
Compartment pressure measurements can be used as an
adjunct to clinical diagnosis, especially in obtunded patients.
Obese Patients
It is estimated that approximately 30% to 40% of adults
in the United States are obese. Difficulty in obtaining a
proper starting point for antegrade nailing in obese patients has been recognized, and is responsible for the increased number of complications when nailing with an
entry site through the piriformis fossa is performed. Better results have been obtained with nailing through the
tip of the greater trochanter, especially with newer implants that have a proximal lateral bend designed for
this insertion site. Patient obesity is a relative indication
for retrograde nailing.
Floating Knee (Ipsilateral Associated Tibial Fracture)
Femoral shaft fractures associated with tibial fractures
(the floating knee) are usually caused by high-energy injury mechanisms. Good results, similar to those found
after high-energy isolated injury, have been obtained
with retrograde nailing of the femur followed by antegrade nailing of the tibia through a single anterior knee
approach.
Ipsilateral Proximal Femur and Shaft Fractures
Femoral shaft fractures associated with femoral neck or
intertrochanteric fractures are challenging injuries to
treat. The femoral neck component of such injuries is

428

the highest priority for optimal, but not necessarily initial, stabilization. Separate treatment with retrograde
nailing or plating of the shaft combined with standard
fixation of the proximal fracture is associated with the
best results. Reduction of these femoral neck fractures
can be difficult without an intact shaft. Provisional fixation of the femoral neck before retrograde nailing can
be done using guidewires for cannulated screws to help
avoid further displacement during retrograde nailing.
Control of the shaft component, obtained after locked
retrograde nailing, facilitates reduction of the proximal
fracture either with manual traction or subsequent
placement of the limb in traction on a fracture table. Simultaneous treatment of the proximal and shaft fractures using a single IM device in reconstruction mode is
another alternative, but is technically more difficult and
is associated with a higher complication rate, especially
when applied for an associated femoral neck fracture.
Femoral neck fractures, when associated with shaft fractures, are most often vertically oriented and have very
little inherent stability. A sliding hip screw construct
with a derotation screw may provide improved biomechanics over cannulated lag screws for these fractures.
Decompression of the hip capsule has been advocated
to decrease the risk of osteonecrosis and formal open
reduction of displaced femoral neck fractures is indicated if an anatomic reduction cannot be achieved by
closed means.

Complications
Malalignment
Fractures of the middle third of the shaft have a low incidence of angular malalignment (2%), whereas fractures of the proximal and distal thirds of the shaft are at
highest risk of malalignment (30% and 10%, respectively). Antegrade nailing can facilitate improved reduction for proximal fractures, and retrograde nailing can
be used for distal fractures. All patients should be evaluated for rotational symmetry compared with the uninjured limb before leaving the operating room. When rotational malalignment is identified, immediate
correction should be performed.
Delayed Unions and Nonunions
Dynamization for nonunited femoral shaft fractures has
been less successful than for tibial fractures and should
be reserved for axially stable fracture patterns to avoid
limb shortening. Reamed nailing (exchange nailing with
the presence of a prior nail) is the treatment of choice
for femoral shaft nonunions, particularly in the absence
of angular deformity. Nonunion repair with compression
plate osteosynthesis with judicious use of autologous
bone graft is an effective alternative, especially when exchange nailing has failed or when deformity correction
is necessary.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 35 Femur: Trauma

Disability
Pain and functional disability after femoral shaft fracture can occur regardless of the mode of treatment. Antegrade femoral nailing can be associated with hip disfunction and pain in up to 40% of patients. Heterotopic
ossification and prominent implants increase the incidence of these complications. Thorough irrigation of the
surgical wound and the use of tissue protectors may reduce heterotopic ossification. Whenever possible, the
fixation devices should be countersunk beneath bone to
minimize related pain and muscle disfunction. Adequate
rehabilitation with attention to abductor, quadriceps,
and hamstring strengthening can reduce muscle disfunction. Injury to the patellofemoral articulation can be
avoided with retrograde nailing by countersinking the
nail beneath the articular surface. Retrograde nails
should be locked with at least two distal interlocking
bolts, especially for axially unstable fractures, to avoid
migration of the nail into the knee joint.
Other Complications
Use of the hemilithotomy position for antegrade femoral nailing increases compartment pressures in the nonoperated leg. Prolonged use of this position should be
avoided to prevent contralateral leg compartment syndrome, especially in patients with injury to the contralateral limb. Excessive and prolonged traction against
a perineal post should be avoided to minimize the risk
of pudendal and sciatic nerve injury from compression
and stretch, respectively.

Supracondylar/Intracondylar Femur Fractures

Classification
According to the AO/OTA classification, supracondylar/
intracondylar femur fractures are designated as type
33 (Figure 3). Type 33A fractures are extra-articular,
type 33B are partially articular (with a portion of the articular surface remaining in continuity with the shaft),
and type 33C fractures are completely articular (no part
of the articular surface remains in continuity with the
shaft). Subtypes of 33A fractures, 33A1 to 33A3 represent increasing comminution of the metaphyseal fracture. Subtypes of 33B fractures represent unicondylar
sagittal splits of the lateral femoral condyle (33B1), medial femoral condyle (33B2), or fracture in the frontal
plane, the so-called Hoffa fracture (33B3). Subtypes of
33C fractures represent simple articular with simple
metaphyseal fractures (33C1), simple articular with
comminuted metaphyseal fractures (33C2), and fractures with articular comminution (33C3).

Treatment
Nonsurgical
Nonsurgical treatment may be indicated for nondisplaced extra-articular supracondylar fractures, for pa-

American Academy of Orthopaedic Surgeons

Figure 3 The Mller classification system of supracondylar/intracondylar femur fractures. (Reproduced with permission from Mller ME, Nazarian S, Koch P, Schatzker J
(eds): The Comprehensive Classification of Fractures of Long Bones. Berlin, Germany,
Springer-Verlag, 1990.)
tients who are nonambulatory, or for those who are not
candidates for surgery. Long leg casts or cast-braces can
be used in such circumstances, but they are associated
with poor results when treating displaced fractures in
ambulatory patients.
Surgical
Most supracondylar and intracondylar distal femur fractures are amenable to surgical fixation with either plate
osteosynthesis or IM nailing. Surgical goals include anatomic reconstruction of the articular surface with restoration of limb length, alignment, and rotation with stable fixation to allow early mobilization and knee range
of motion. The more distal the fracture and the more
intra-articular involvement, the more amenable such
fractures are to plate and screw osteosynthesis. As is
true of most articular fractures, the articular reduction
and fixation is of paramount importance. After this objective is accomplished, the articular segment is attached
to the shaft with attention to proper length, alignment,

429

Femur: Trauma
and rotation. To promote an uneventful union, indirect
fracture reduction techniques including percutaneous or
submuscular plating can be used. With such minimally
invasive techniques, the use of bone grafts, even in the
presence of metaphyseal comminution, is not routinely
necessary. Plates that offer a fixed-angle construct, such
as 95 blade plates, 95 condylar screws, and newer locking plate devices, are indicated when treating fractures
with metaphyseal comminution. These constructs minimize the risk of varus collapse seen with traditional
nonfixed-angle devices. Blade plates are technically the
most demanding, requiring proper insertion in all three
plates simultaneously, but they offer the ability to treat
very distal fractures. Dynamic condylar screw fixation
requires slightly more distal bone for adequate purchase
of the condylar screw, but the screw is technically easier
to insert than blade plates because of the ability to control plate position in the sagittal plane. The newer fixedangle plates, designed specifically for the distal femur,
provide relative technical ease to insert compared with
blade plates and offer multiple distal and proximal locking options. Threads on the outer diameter of the screw
head engage and lock into threaded screw holes in the
plate. Surgeons using such fixed-angle constructs should
be familiar with the unique properties of such systems.
Standard nonlocking screws can be used in some of
these systems to lag fracture fragments and to compress
plate to bone and should be inserted before locking
screws. Another alternative to help avoid varus in comminuted metaphyseal fractures is supplemental medial
plating. This technique has the disadvantage of requiring
increased soft-tissue disruption.
Retrograde IM nailing, with associated screw fixation of simple intra-articular components, also has been
successful. The advantage of IM nails for this application is the minimal dissection of the surrounding soft tissues. Newer nail designs with very distal interlocking
holes allow nailing of fractures with small (4 to 5 cm)
distal fragments. Static distal interlocking with multiple
oblique interlocks should be used to enhance stability
and reduce the risk of nail migration into the knee.
Intra-articular fractures should be treated with appropriate anatomic reduction and stabilization (usually with
screws placed such that they do not interfere with subsequent retrograde nailing) before IM nailing. Short, supracondylar nails provide a stress riser at their tip in the
diaphyseal portion of the bone and increase the risk of
subsequent periprosthetic fracture. Long retrograde
nails, therefore, are preferable in most instances.

Special Considerations
Elderly patients with osteopenia and distal femur fractures represent a significant treatment challenge. Polymethylmethacrylate cement can be used to augment
screw fixation in the distal fragment segment. The newer

430

Orthopaedic Knowledge Update 8

locking plate designs that offer multiple fixed-angle
locking screws offer a theoretic advantage in patients
with osteopenia. The ideal fixation device and construct
in this group of patients remains unproven.

Annotated Bibliography
Femoral Shaft Fractures
Bellabarba C, Ricci WM, Bolhofner BR: Results of indirect reduction and plating of femoral shaft nonunions
after intramedullary nailing. J Orthop Trauma 2001;15:
254-263.
This article reviews a consecutive study of 23 patients with
femoral shaft nonunion after IM nailing. All patients were
treated with indirect plating techniques and judicious use of
autologous bone graft. Twenty-one of the 23 nonunions healed
without further intervention at an average follow-up of 12
weeks.

The Canadian Orthopaedic Trauma Society: Nonunion

following intramedullary nailing of the femur with and
without reaming. J Bone Joint Surg Am 2003;85:20932096.
This multicenter, prospective, randomized trial of 224 patients was conducted to compare reamed and nonreamed femoral nailing. Nonunion occurred in 7.5% of patients in the
nonreamed group and only 1.7% of patients in the reamed
group.

Robinson CM, Alho A, Court-Brown C: Femur. London,

England, Arnold, 2003.
This is one of four books in a series dedicated to musculoskeletal trauma. It is a contemporary, well-referenced, and
complete guide to treating every aspect of femur trauma.

Watson JT, Moed BR: Ipsilateral femoral neck and shaft

fractures: Complications and their treatment. Clin
Orthop 2002;399:78-86.
A retrospective review of 13 patients who had healing
complications after surgical treatment of ipsilateral femoral
neck and shaft fractures is presented. Lag screw fixation of the
neck with reamed IM nailing of the shaft were associated with
the fewest complications.

Supracondylar/Intracondylar Femur Fractures

Bellabarba C, Ricci WM, Bolhofner BR: Indirect reduction and plating of distal femoral nonunions. J Orthop
Trauma 2002;16:287-296.
The results of a prospective study of 20 patients with nonunion of the distal femur are presented. Repair was done with
plate fixation using indirect reduction techniques; in 45% of
patients, autologous cancellous bone graft was included. All 20
nonunions healed without further intervention at an average
follow-up of 14 weeks. The authors concluded that contemporary plating techniques are effective in the treatment of distal
femoral nonunions.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Marti A, Fankhauser C, Frenk A, Cordey J, Gasser B:
Biomechanical evaluation of the less invasive stabilization system for the internal fixation of distal femur fractures. J Orthop Trauma 2001;15:482-487.
This is a biomechanical study using cadaver femurs that
compared the Less Invasive Stabilization System (LISS) plate
(Synthes USA, Paoli, PA) to conventional condylar buttress
plate and dynamic condylar screw plate for fixation of distal
femur fractures. The LISS construct had less irreversible deformation and a higher elastic deformation.

Prayson MJ, Datta DK, Marshall MP: Mechanical comparison of endosteal substitution and lateral plate fixation in supracondylar fractures of the femur. J Orthop
Trauma 2001;15:96-100.
This article presents a review of a biomechanical evaluation using synthetic femur supracondylar fracture model comparing lateral plate fixation to lateral plate fixation with endosteal substitution. Specimens with endosteal substitution
showed decreased motion at the fracture site in both torsion
and axial loading.

Classic Bibliography
Bhandari M, Guyatt GH, Tong D, Adili A, Shaughnessy
SG: Reamed versus nonreamed intramedullary nailing
of lower extremity long bone fractures: A systematic
overview and meta-analysis. J Orthop Trauma 2000;14:
2-9.
Bolhofner BR, Carmen B, Clifford P: The results of
open reduction and internal fixation of distal femur fixation using a biologic (indirect) reduction technique.
J Orthop Trauma 1996;10:372-377.
Brumback RJ, Reilly JP, Poka A, Lakatos RP, Bathon
GH, Burgess AR: Intramedullary nailing of femoral
shaft fractures: Part 1. Decision-making errors with interlocking fixation. J Bone Joint Surg Am 1988;70:14411452.

American Academy of Orthopaedic Surgeons

Chapter 35 Femur: Trauma

Brumback RJ, Uwagie-Ero S, Lakatos RP, Poka A, Bathon GH, Burgess AR: Intramedullary nailing of femoral shaft fractures: Part II. Fracture-healing with static
interlocking fixation. J Bone Joint Surg Am 1988;70:
1453-1462.
Clatworthy MG, Clark DI, Gray DH, Hardy AE:
Reamed versus unreamed femoral nails: A randomized,
prospective trial. J Bone Joint Surg Br 1998;80:485-489.
Kempf I, Grosse A: Beck G: Closed locked intramedullary nailing: Its application to comminuted fractures of
the femur. J Bone Joint Surg Am 1985;67:709-720.
Kuntscher G: The intramedullary nailing of fractures.
Clin Orthop 1968;60:5-12.
Ostrum RF, Agarwal A, Lakatos R, Poka A: Prospective
comparison of retrograde and antegrade femoral intramedullary nailing. J Orthop Trauma 2000;14:496-501.
Ricci WM, Bellabarba C, Evanoff B, Herscovici D, DiPasquale T, Sanders R: Retrograde versus antegrade
nailing of femoral shaft fractures. J Orthop Trauma
2001;15:161-169.
Tornetta P, Tiburzi D: Reamed versus nonreamed anterograde femoral nailing. J Orthop Trauma 2000;14:15-19.
Winquist RA, Hansen ST, Clawson DK: Closed intramedullary nailing of femoral fractures. J Bone Joint
Surg Am 1984;66:529-539.
Wolinsky PR, McCarty E, Shyr Y, Johnson K: Reamed
intramedullary nailing of the femur: 551 cases. J Trauma
1999;46:392-399.
Yang KH, Han DY, Park HW, Kang HJ, Park JH: Fracture of the ipsilateral neck of the femur in shaft nailing:
The role of CT in diagnosis. J Bone Joint Surg Br
1998;80:673-678.

431

Chapter

36

Knee and Leg: Bone Trauma

Michael T. Archdeacon, MD, MSE

Patellar Fractures

Treatment

Mechanism of Injury

Minimally displaced or nondisplaced fractures with an

intact extensor mechanism and less than 1 to 2 mm of
step-off can generally be treated nonsurgically. Full
weight bearing in a long leg cylinder cast or a locked
knee brace are reasonable treatment options. Incisions
are typically midline longitudinal; however, in large
traumatic disruptions, transverse incisions can be incorporated to minimize flap development. Open wounds
should be incorporated into the surgical incisions to
help prevent flap devascularization. Digital palpation or
direct inspection through the traumatic arthrotomy or
the fracture planes can facilitate obtaining and assessing
the reduction of the joint surface. Maintenance of the
length-tension ratio of the extensor mechanism is a secondary goal. Fluoroscopic AP and lateral radiographs of
the patella can show articular incongruency.
Fixation techniques include the modified tension
band and/or cerclage wire, tension band through cannulated compression screws, or independent lag screws.
These techniques can be supplemented with a protective wire around the superior pole of the patella and
through a drill hole at the tibia tubercle. For severely
comminuted fractures with extensive articular comminution, partial patellectomy may be a reasonable alternative. If patellar tendon advancement is performed, it
is critical to reattach the tendon to the remaining central portion of patella to maintain an extensor mechanism that is congruous in the patellofemoral articulation.
The cannulated compression screw technique offers
the advantages of a compression screw as well as those
of a cerclage wire. Parallel, longitudinal guidewires are
passed through the patella and sequentially replaced
with cannulated compression screws. Sizes from 3.5 to
6.5 mm can be used. A tension band construct is placed
through the cannulated screws and tightened over the
anterior surface of the patella. Ideally, the screw is
slightly shorter than the length of the patella, which prevents a stress riser in the tension wire at the tip of the
screw and allows the tension effect of the wire to fur-

Patellar fractures most commonly occur as a result of a

direct or indirect trauma to the knee. The direct mechanism involves an impact, typically by a dashboard during a motor vehicle crash or a fall onto the knee. The indirect mechanism results from an abrupt quadriceps
contraction, which can result in an avulsion-type fracture of the inferior or superior pole or a transverse patellar fracture. An associated retinaculum injury is common.

Diagnosis
Suspicion for patellar fracture is raised based on the
mechanism of injury and the patients clinical history
and physical examination. The ability to extend the knee
against gravity can be limited or impossible, and an extensor lag may be noted. It is imperative to confirm that
an open fracture or traumatic arthrotomy has not occurred. Confirmation can be achieved with a saline retention test by injecting 30 to 60 mL of saline into the
knee and observing extravasation of fluid from the
wound. Radiographs including AP and lateral views of
the patella usually will show the fracture.

Classification
Fractures are typically classified as transverse, stellate,
or vertical. Transverse fractures, classified as displaced
or nondisplaced, can be avulsion injuries at either pole,
or true fractures occurring anywhere along the length of
the patella. Stellate fractures include a spectrum of injury from minimally displaced, multifragment fractures
through displaced, comminuted, high-energy injuries.
Vertical fractures are believed to occur from direct compression and knee hyperflexion. Finally, a subset of patellar fractures has been reported in patients who have
undergone bone-patella-bone autograft donation for anterior cruciate ligament (ACL) reconstruction. It is hypothesized that an accelerated rehabilitation protocol
puts the patient at risk for a transverse fracture pattern.

American Academy of Orthopaedic Surgeons

433

Orthopaedic Knowledge Update 8

Knee and Leg: Bone Trauma

Table 1 | Schatzker Classification of Tibial Plateau

Fractures

are completely displaced. A fourth type, the comminuted tibial eminence fracture, has been described.

Type

Characteristics

Treatment

I
II

Pure split facture of the lateral plateau

Split fracture of the lateral plateau with a depression component
Pure depression injuries to the lateral plateau
Medial tibial plateau fracture*
Bicondylar fractures, usually the result of a highenergy injury
Metadiaphyseal dissociation with a unicondylar
or bicondylar fracture

Surgical intervention has been advised for type III and IV

fractures because a high incidence of ACL incompetence
has been reported with nonsurgical treatment. An open
surgical technique with reduction and internal fixation has
been historically advocated. A parapatellar arthrotomy is
useful with compression lag-type fixation. The current favored techniques include arthroscopically-assisted and arthroscopic reduction and internal fixation. This technique
is accomplished using standard arthroscopic portals with
an accessory portal placed high anterolaterally or anteromedially to manipulate the fracture. The fracture is reduced and secured with percutaneous Kirschner wires followed by cannulated screw fixation (with or without a
soft-tissue washer). Postoperatively, active-assisted range
of motion is initiated immediately in a hinged brace. The
patient is allowed to bear weight as tolerated in extension.
A rehabilitation protocol, similar to that used for an ACL
injury, is followed.

III
IV
V
VI

*Commonly associated with a concomitant vascular injury

Comcomitant injuries including knee dislocations, vascular or neurologic injury, and compartment syndrome are common

ther compress the bony surfaces. The technique is ideal

for transverse fractures and useful for comminuted fractures because the screws can be driven in multiple
planes and wires can be used to cerclage the multiple
fragments.
Postoperatively, patients are treated in a hinged knee
brace, which remains locked in extension until the softtissue wound heals, typically in 7 to 10 days. If stable internal fixation is achieved, then gentle active-assisted
and passive flexion arcs are initiated with wound healing. Active extension is delayed for 6 weeks. When a
partial patellectomy has been done or less stable constructs are achieved, the patient is maintained in extension for approximately 6 weeks before initiating flexion
arcs.
Complications include loss of fixation and/or reduction, infection, delayed union, malunion, refracture, arthrofibrosis, and extensor lag. Posttraumatic arthritis of
the patellofemoral joint can occur.

Tibial Spine Fractures

Classification
Tibial spine or tibial eminence fractures are terms used
to describe an injury at the intercondylar region of the
tibial plateau. The tuberculum intercondylare mediale is
the site of insertion of the ACL. No ligamentous insertions occur on the lateral portion of the tibial eminence.
The mechanism of injury for tibial spine fractures is often a simple, low-energy fall onto an outstretched leg. In
terms of an avulsion injury, the ACL, although intact, is
typically functionally incompetent. Occasionally, the
fracture will extend into the weight-bearing portion of
the articular surface of the medial tibial plateau.
Type I fractures are nondisplaced, type II fractures
are partially displaced or hinged, and type III fractures

434

Tibial Plateau Fractures

Mechanism of Injury
Tibial plateau fractures typically have a bimodal age distribution. The spectrum of injury includes high-energy,
axial load, or impact injuries with a significant softtissue component through low-energy medial-lateral impact injuries to the tibial plateau. The high-energy injuries are more common in the young trauma patient
population, and impact injuries are more common in
elderly patients with osteopenic bone.

Classification
Tibial plateau fractures have been classified by Schatzker into six types (Table 1). The AO/Orthopaedic
Trauma Association classifications are also widely recognized. Type A fractures are extra-articular with increasing severity from subtypes 1 through 3. Type B
fractures are unilateral plateau injuries ranging from the
pure split injuries through split-depression fractures.
Type C injuries are bicondylar injuries increasing in severity from subtypes 1 through 3.

Imaging Evaluation
Imaging studies include orthogonal views of the fulllength tibia and oblique views of the knee. CT may delineate the extent of depressed fragments and clarify
fracture planes. These studies can be useful for minimally invasive surgical techniques and reduction maneuvers. More recently, MRI of tibial plateau fractures has
been advocated. This modality is useful in the diagnosis
of associated soft-tissue pathology, including meniscal
and ligamentous injuries.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 36 Knee and Leg: Bone Trauma

drome is necessary when treating patients with highenergy trauma.
Current surgical treatment strategies include open
reduction and internal fixation (ORIF) with adjunctive
techniques including limited open reduction, arthroscopically-assisted reduction and fixation, and augmented internal fixation with resorbable bone cements.
External fixation including temporary spanning fixators,
hybrid and/or fine-wire fixation, and combined limited
ORIF are acceptable techniques, particularly for patients with significant soft-tissue injury.

Figure 1 Periarticular tibial plateau fixation with raft or subchondral screws. The
screws maintain the articular reduction, but offer no stability for the metadiaphyseal
component of the injury. (Reproduced with permission from Karunakar MA, Egol KA,
Peindl R, Harrow ME, Bosse MJ, Kellam JF: Split depression tibial plateau fractures: A
biomechanical study.J Orthop Trauma 2002;16:172-177.)

Treatment
Nonsurgical treatment is recommended for low-energy
tibial plateau fractures, which are stable to varus-valgus
stress, as well as for nonambulatory patients and for
those not medically fit for surgery. Nonsurgical treatment is indicated for minimally displaced split fractures
or depression injuries with less than 1 cm of depression
and is particularly relevant for injuries that occur deep
to the meniscal tissue. Treatment should include use of a
hinged knee brace and mobilization. Active-assisted and
passive range of motion is initiated immediately, and
weight bearing is typically delayed for 8 to 12 weeks.
Surgical treatment of closed tibial plateau fractures
is indicated based on two criteria. First, the congruency
of the joint surface must be evaluated. Although articular congruency should be the goal of treatment, articular
displacement up to 10 mm has been accepted. Second,
joint stability must be assessed. If instability exists because of depression or condyle subluxation, then surgical intervention should be considered. Additionally,
high-energy injuries involving the metadiaphyseal junction or unstable bicondylar fractures should be considered for surgical intervention. Other indications include
open fractures, associated neurovascular injuries, compartment syndrome, and floating knee injuries.
A staged treatment protocol has been advocated for
high-energy injuries with significant soft-tissue damage.
The first step is to ensure immediate stability and ligamentotaxis with spanning external fixation to allow for
appropriate preoperative planning. Second, provisional
fixation allows the soft-tissue envelope to heal so that
future surgical reconstruction can proceed with minimal
complications. A high index of suspicion for internal degloving injuries, vascular injuries, and compartment syn-

American Academy of Orthopaedic Surgeons

Open Reduction and Internal Fixation

ORIF for tibial plateau fractures has evolved to include
implants and techniques that permit adequate periarticular stabilization without extensive exposures and because of advances in soft-tissue injury treatment, including staged reconstruction of high-energy injuries.
Limited ORIF necessitates ligamentotaxis to show
which fragments can be reduced without direct surgical
intervention or with minimal surgical intervention. Indirect reduction is accomplished with spanning external
fixation or a femoral distractor across the knee joint. As
major fragments are reduced with ligamentotaxis, articular depression and malreduction can be treated through
limited percutaneous wounds or small open incisions.
This procedure is guided fluoroscopically or arthroscopically. The use of tamps, elevators, and joystick Kirschner
wires can assist in manipulating fragments into acceptable reduction positions. If the articular surface is adequately reduced and the overall alignment of the limb is
obtained through ligamentotaxis, periarticular reduction
clamps can be used to close the unicondylar or bicondylar components onto the elevated articular surface, thus, stabilizing the articular block. Raft screws
or lag screws are then used across the subchondral region of the plateau (Figure 1).
After the articular surface is restored, the metaphyseal void created by elevating depressed fragments is
supported by making a small osteotomy below the articular surface or using a window in the fracture plane to
pack the void with autogenous or allogenic bone graft
or bone graft substitute. Finally, the restoration of the
articular block to the metaphysis-diaphysis is stabilized
with a periarticular plate construct.
With the recent advances in locking fixed-angle
plates, a single device can be used to stabilize unicondylar and bicondylar injuries and metadiaphyseal injuries.
A limited incision into the anterior compartment followed by submuscular dissection allows for percutaneous plating of the proximal tibia. After the submuscular
interval has been exposed, a locking, periarticular plate
can be passed along the shaft of the tibia, and provisionally secured with fixed-angle screws, wires, or bone
clamps. The overall alignment is assessed with fluoroscopy, and if acceptable, locking screws can be placed

435

Knee and Leg: Bone Trauma

Orthopaedic Knowledge Update 8

be placed in the posterior medial interval to buttress
and stabilize these fragments. Patients are treated with
immediate active-assisted and passive range-of-motion
exercises postoperatively. A hinged brace is used for
protection against varus-valgus stress. Weight bearing is
delayed for 8 to 12 weeks.
Arthroscopically-Assisted Reduction
Several authors have advocated arthroscopy as an adjunctive technique in the treatment of tibial plateau
fractures. Stated advantages include assessment and
treatment of associated intra-articular ligamentous and
meniscal injuries. A direct assessment of the articular reduction can be obtained with arthroscopy. A potential
complication associated with arthroscopically-assisted
reduction and fixation of tibial plateau fractures involves the extravasation of arthroscopy fluid through
the fracture planes and into the lower extremity compartments. Compartment syndrome has been reported
after using this technique. Arthroscopic fluid pumps
should not be used or should be kept at low pressure in
these circumstances. No studies have shown superior
outcomes for tibial plateau fractures treated with adjunctive arthroscopy.

Figure 2 Percutaneous fixation of a bicondylar tibial plateau fracture using a fixedangle plate-screw construct for metadiaphyseal stability.

throughout the plate. Various available plates have options for locking screws throughout the plate or in the
periarticular region. The locking screws or fixed-angle
screws are advantageous in the reconstructed subchondral surface and in metadiaphyseal dissociations because these regions are principally susceptible to high
shear stress. Additionally, these constructs can maintain
the overall alignment of the metadiaphyseal region and
the proximal tibia with minimal surgical dissection or
trauma (Figure 2).
In patients with high-energy bicondylar fractures, a
secondary incision may be required on the posteromedial aspect of the tibial plateau. This type of injury is
typically treated with an incision over the posterior
compartment, elevating the pes anserine tendons anteriorly and entering the interval between the gastrocnemius and the plateau. This interval allows for manipulation and periarticular clamp placement of the posterior
medial fragment. A 3.5-mm dynamic compression plate
or a fixed-angle lateral plate for the opposite limb can

436

Augmented Internal Fixation

Recent developments in bone graft substitutes have
made available calcium phosphate cements and other
bone substitutes, which are well suited for the compressive loading environment of the tibial plateau metaphysis. This adjunctive technique involves open or limited
internal fixation of the tibial plateau articular surface
augmented with calcium phosphate or calcium sulfate
bone cements, injected either percutaneously or through
surgical wounds to fill metaphyseal voids created with
reduction of periarticular fragments. Care must be taken
not to allow extravasation of the cement into the joint
through articular fractures. Screws must be placed before cement injection or while the cement is in the
moldable phase. After the cement cures, screw placement can fracture or crumble the calcium phosphate cements.
External Fixation
External fixation can be used for definitive stabilization
of unicondylar, bicondylar, and metadiaphyseal fractures of the tibial plateau. Previously mentioned reduction techniques and limited periarticular fixation can be
used in a similar manner for definitive external fixation
treatment. Circular fine wire or hybrid external fixation
frames are used to associate the articular block to the
metadiaphyseal segment rather than using locking,
fixed-angle plates. These frames realign the axis of the
limb in both the AP and lateral planes and secure fixation of the metadiaphyseal component in patients with
high-energy injuries. These external fixation techniques

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Table 2 | Open Fracture Classification as Modified by the

Lower Extremity Assessment Project
Type

Characteristics

Wounds measure less than 1 cm

Generally inside-out injuries, low energy, minimal
periosteal injury
Wounds measure 1 to 10 cm
Minimal damage to periosteum and soft tissues
A fracture resulting from a high-energy injury with
extensive damage to soft tissue, including
periosteal stripping (wound size less critical)
A fracture resulting from a high-energy injury as
with a type IIIA fracture but requiring rotational
flap coverage or free-tissue transfer
A fracture resulting from a high-energy injury as
with a type IIIA or IIIB fracture and resulting in
vascular injury requiring repair

II
IIIA

IIIB

IIIC

are particularly well suited to patients who have fractures with significant soft-tissue injuries.

Tibial Shaft Fractures

Classification
Tibial shaft fractures are classified based on the anatomy of the injury, or by the energy imparted at the time
of fracture. The anatomic classification is based on the
location and the fracture configuration. Position is defined as proximal, midshaft, or distal. Configuration is
classified as simple fractures (such as transverse or spiral fractures) and more extensive patterns (such as butterfly fragments and comminuted fractures). The AO
classification advocates a higher level of classification
for an increasing severity of injury. Type A fractures are
simple fractures that are spiral, oblique, or transverse.
Type B fractures result from higher energy dissipation
at the level of the injury and are classified as spiral,
bending, or fragmented wedges. Complex fractures or
type C fractures are multiple spiral fractures, segmental
fractures, and highly comminuted fractures.
Open fracture classification is based on the work of
Gustilo and Anderson and has been modified in the recent multicenter Lower Extremity Assessment Project
study (Table 2).

Mangled Extremity
Recently, several limb salvage indices have been reported in the literature. However, there is no consensus
on which index is the most useful, and whether any of
these indices are reliable to predict outcomes. The absence of validation of these indices led to the development of the Lower Extremity Assessment Project study

American Academy of Orthopaedic Surgeons

Chapter 36 Knee and Leg: Bone Trauma

in an attempt to provide criteria for the care of the
mangled lower extremity. Analysis of the prospectively
collected data revealed that none of the previously developed scoring systems showed any validated clinical
utility. Although a high specificity confirmed that limb
salvage with low index scores could be predicted, a low
sensitivity failed to support any index as a valid predictor of amputation. Currently, no limb salvage index has
been statistically confirmed to be reliable in the evaluation and treatment of patients with severely mangled
lower extremities. Therefore, current recommendations
emphasize early assessment of the extent of injury,
along with frank discussions with the patient to explain
the potential risks relating to the functional, social, and
economic outcomes of limb salvage compared with amputation.

Open Fractures and Soft-Tissue Injury

Care for a patient with an open fracture should include
the administration of prophylactic systemic antibiotics
in the emergency department. Treatment should consist
of a first-generation cephalosporin with the addition of
an aminoglycoside for type III open fractures. Penicillin
should be administered to patients with massively contaminated wounds when a concern exists for clostridial
infection. The regimen is continued for 24 to 72 hours
after a clean wound bed has been obtained with surgical
dbridement. Usually antibiotic coverage is extended
for 48 hours after each subsequent surgical procedure.
Wound coverage or closure of soft tissues is advocated,
ideally within 5 to 7 days after the injury. The use of this
protocol results in a 4% to 10% overall infection rate,
with a 10% to 20% incidence of deep infection from
type III open fractures.

Compartment Syndrome
In 1% to 10% of tibial fractures or lower extremity crush
injuries, elevated intracompartmental pressures are
known to occur. If an acute compartment syndrome is not
treated emergently with surgical decompression, irreversible neurologic damage and myonecrosis will occur.
Awareness of the possibility of compartment syndrome is
critical and is primarily based on clinical examination.
Hallmark symptoms include a tense or increasingly tense
lower extremity, incapacitating pain that is not in proportion to the severity of the injury, worsening pain over time,
and the most reliable criteriapain with passive stretch
of the ankle or toes. Symptoms of hypesthesia may indicate progressive neurologic injury. The presence of pulses
is not a reliable factor for excluding a diagnosis of compartment syndrome.
When clinical examination is not reliable secondary
to head injury, intoxication, or sedation, compartment
pressure monitoring has been advocated. Criteria include absolute compartmental pressures, with critical

437

Knee and Leg: Bone Trauma

values ranging from 30 to 45 mm Hg. A less inconsistent
parameter is the absolute difference between the patients diastolic blood pressure and the intracompartmental pressure. The critical value appears to be at the
point when the intracompartmental pressure is within
30 mm Hg of the diastolic pressure.
Compartment syndrome is treated with emergent fasciotomy and decompression of all four compartments of
the lower leg. A dual incision medial-lateral technique or
single incision lateral technique are acceptable. Incisions
approximately 15 to 18 cm in length are required to adequately decompress lower extremity intracompartmental pressure. Postoperative treatment consists of dressing
changes followed by delayed primary closure or skin
grafting at 3 to 7 days after decompression.

Treatment
Most low-energy tibial shaft fractures can be treated in
a closed manner with reduction and application of a
long leg cast followed by functional fracture bracing. Parameters accepted for closed treatment vary; however,
general recommendations are the presence of less than
1 cm of shortening, less than 5 of angulation in any
plane, and rotational deformity limited to 5 after immobilization. Although closed treatment is perfectly acceptable for tibial shaft fractures, caution should be exercised after immobilization with more proximal and
distal fractures. These fractures are more difficult to
control with functional bracing, and the imposed stability of a periarticular fracture brace may limit the functional range of motion in the adjacent joints.
Plate Fixation
With the widespread use of intramedullary (IM) nailing
for tibial fractures, ORIF has been reserved for fractures in the proximal or distal third or fourth of the
tibia. The rationale for these recommendations, when
using ORIF, include a concern for soft-tissue devitalization, the increased risk of infection in tibial fractures
compared with IM nailing, and the disadvantages of a
load-bearing device compared with a load-sharing IM
nail. However, plate fixation of tibial shaft fractures is
still a viable option, particularly in concomitant periarticular fractures for which IM nailing may be very difficult, or in fractures in which an open wound would allow easy access for plating with minimal further
dissection. Percutaneous plating techniques, which limit
soft-tissue dissection, and fixed-angle locking plates are
being used more often. These techniques can provide
the theoretical benefits of external fixation with minimal soft-tissue and fracture site disruption and avoidance of the associated IM injury associated with reaming and nailing.

438

Orthopaedic Knowledge Update 8

External Fixation
External fixation is an acceptable treatment method for
open tibial fractures. Its advantages include minimal
soft-tissue disruption, rapid application, and the ability
to control difficult fracture patterns in a stable manner.
Disadvantages include pin tract infections, delayed
union in open fractures, and a higher incidence of malalignment and malunion when compared with IM nailing. The overall risk of infection between external fixation and IM nailing is similar. When significant softtissue injury is involved, external fixation is a safe and
reliable option. However, in the presence of a healed
open fracture wound, exchange of an external fixator
for an IM nail is reasonable if this is accomplished
within 10 to 14 days of fixator application.
Intramedullary Nailing
Currently, fractures not amenable to closed treatment
are most commonly treated with statically locked IM
nailing. In open fractures, IM nailing has been shown to
be safe and effective with a relatively low risk of infection compared with other treatment modalities. Current
data indicate that reamed versus unreamed IM nailing
produce no essential difference in infection or nonunion
rates. The most reasonable option is the ream to fit
model in which the IM canal is reamed approximately 1
mm larger than the cortical isthmus of the tibia and a
nail one size smaller is placed. Statically locked nailing
is advocated; however, in stable shaft fractures for which
one segment is controlled through the isthmus of the
tibia, dynamic nailing is acceptable.
The indications for tibial nailing have expanded as
the interlocking screw configurations have improved
and techniques have been developed to treat more
proximal and distal fractures. Modalities such as blocking screws and unicortical reduction plates have been
advocated as adjunctive techniques in proximal and distal metadiaphyseal fractures treated with IM nails (Figure 3). Additional techniques to prevent malreduction
for proximal third tibial fractures include nail placement
in the extended position, a slightly more posterior entry
portal, a parapatellar arthrotomy, an IM nail with a less
acute proximal bend to prevent metadiaphyseal procurvatum deformity, and an external fixator or femoral distractor to maintain reduction during nailing.
Complications
Infection rates following IM nailing of open tibial fractures range from 4% to 20%, with the highest incidence
in the type III open fractures. However, aggressive
dbridement, early stabilization, early prophylactic antibiotics, and meticulous soft-tissue care combined with
early coverage all help reduce and maintain a low incidence of infection.
Nonunion has historically been attributed to bone loss,
excessive motion at the fracture site, and a tenuous soft-

American Academy of Orthopaedic Surgeons

Chapter 36 Knee and Leg: Bone Trauma

Orthopaedic Knowledge Update 8

Annotated Bibliography
Patellar Fractures
Stein DA, Hunt SA, Rosen JE, Sherman OH: The incidence and outcome of patella fractures after anterior
cruciate ligament reconstruction. Arthroscopy 2002;18:
578-583.
In eight patients, the diagnosis of patellar fractures was
made after 618 ACL bone-patellar-bone autograft reconstructions. Five of the injuries were the result of indirect trauma
and three were the result of direct injury. All patients regained
a full flexion arc. These outcomes were consistent with the remaining population of patients with ACL reconstructions. The
authors determined that there were minimal residual sequelae
after postoperative patellar fracture following ACL reconstruction.

Tibial Spine Fractures

Osti L, Merlo F, Liu S, Bocchi L: A simple modified arthroscopic procedure for fixation of displaced tibial eminence fractures. Arthroscopy 2000;16:379-382.
Ten consecutive adult patients who underwent arthroscopic fixation of displaced tibial spine fractures using an ACL
guide to reduce the fracture are described. Arthroscopically
placed pins were sequentially replaced with metallic suture
wire placed over the tibial eminence and exiting out the anterior cortex of the tibia and tied over a screw. The advantages
of this technique include stable fixation, easy device removal,
and avoidance of injury to the ACL.

Tibial Plateau Fractures

Figure 3 A unicortical reduction plate can be used to maintain reduction in proximal
third tibial fractures treated with IM nails.

tissue envelope and/or blood supply. A recent observational study showed that the most reliable predictors of
revision for tibial nonunion include an open fracture, a
fracture gap after fixation, and a transverse fracture pattern. Recent data also have shown that patients who
smoke have a higher risk of nonunion and delayed union.
All delayed unions or nonunions in open tibial fractures
should be considered as potentially infected. Staged protocols are most reasonable in this situation. Current recommendations for the treatment of tibial nonunions include dynamization by locking screw removal, exchange
reamed nailing, compression plate fixation, external fixation with or without fibular osteotomy, posterolateral
bone grafting, and/or the use of adjunctive bone stimulators. After 6 to 9 months, dynamization is unlikely to result in union. Other complications include anterior knee
pain and failure of fixation. Implant failure generally results from fatigue failure of the locking screws, which can
be removed or exchanged if causing painful symptoms.

American Academy of Orthopaedic Surgeons

Hung SS, Chao EK, Chan YS, et al: Arthroscopically

assisted osteosynthesis for tibial plateau fractures.
J Trauma 2003;54:356-363.
Thirty-one patients with tibial plateau fractures had
arthroscopically-assisted reduction. More than 50% of the patients had concomitant interarticular injury including 44%
with meniscal injuries, 38% ACL injuries, and 20% osteochondral collateral ligament injuries.

Keating JF, Hajducka CL, Harper J: Minimal internal

fixation in calcium phosphate cement in the treatment
of fractures of the tibial plateau. J Bone Joint Surg Br
2003;85:68-73.
Forty-nine lateral tibial plateau fractures were treated with
limited ORIF and augmented with subchondral and metaphyseal resorbable calcium phosphate bone cement. Thirty-three
of 44 patients were rated as having an excellent reduction at
1-year postoperative follow-up. In seven of the patients, slight
loss of reduction (less than 3 mm) was noted; however, it did
not require any further action. One infection was found. The
authors advocate calcium phosphate bone cement as an adjunctive alternative to bone grafting in tibial plateau fractures.

Lobenhoffer P, Schulze M, Gerich T, Lattermann C,

Tscherne H: Closed reduction/percutaneous fixation of

439

Knee and Leg: Bone Trauma

tibial plateau fractures: Arthroscopic versus fluoroscopic
control of reduction. J Orthop Trauma 1999;13:426-431.
Thirty-three patients with unicondylar tibial plateau injuries were reviewed; 10 had arthroscopically assessed reduction.
The remaining 23 patients had the reduction judged fluoroscopically. The authors concluded that there was no difference
in arthroscopic compared with fluoroscopic assessment of reduction in terms of outcome after limited internal fixation of
unicondylar tibial plateau fractures.

Shepherd L, Abdollahi K, Lee J, Vangsness T Jr: The

prevalence of soft-tissue injuries in nonoperative tibial
plateau fractures as determined by magnetic residence
imaging. J Orthop Trauma 2002;16:628-631.
Twenty nonsurgically treated tibial plateau fractures were
evaluated with MRI; 80% had meniscal tears and 40% had
complete ligamentous disruptions. Meniscal injuries were associated with lateral condylar fractures or bicondylar fractures
and less with medial condylar fractures. The authors concluded
that the use of MRI for tibial plateau fractures can result in
the diagnosis of many soft-tissue injuries.

Orthopaedic Knowledge Update 8

In this study of 105 patients with 110 open tibial fractures
treated with external fixation or IM nailing, smoking was
noted to have a deleterious effect on healing. A union rate of
84% occurred in patients who smoked compared with 94% of
those who did not smoke. A higher incidence of delayed
unions and nonunions occurred in the smoking group compared with the nonsmoking group.

Ricci WM, OBoyle M, Borrelli J, Bellabarba C, Sanders

R: Fractures of the proximal third of the tibial shaft
treated with intramedullary nails and blocking screws.
J Orthop Trauma 2001;15:264-270.
In 12 patients with proximal third tibial shaft fractures
treated with IM nailing and blocking screws, the authors concluded that posterior and lateral blocking screws in the proximal fracture reduced procurvatum and valgus malalignment,
respectively. These complications are associated with the nailing of proximal third tibial shaft fractures. The authors concluded that blocking screws for proximal third tibial fractures
are a useful adjunctive technique to control reduction during
IM nailing.

Tibial Shaft Fractures

Bhandari M, Tornetta P III, Sprague S, et al: Predictors
of reoperation following operative management of fractures of the tibial shaft. J Orthop Trauma 2003;17:353361.
Two hundred patients with tibial shaft fractures were evaluated for predictors of reoperation within 1 year of the index
procedure. Variables that were prognostic for reoperation included an open fracture wound, lack of cortical continuity between the fracture ends after fixation, and a transverse fracture pattern.

Bosse MJ, MacKenzie EJ, Kellam JF, et al: A prospective evaluation of the clinical utility of the lower extremity injury severity scores. J Bone Joint Surg 2001;83:
3-14.
The Lower Extremity Assessment Project (LEAP Study)
was a National Institutes of Health investigation to evaluate
limb salvage versus amputation in severe lower extremity injuries. An open fracture classification system was clearly defined
by the LEAP authors, so that final grading was determined at
the time of definitive closure or amputation. Analysis of prospective data for 556 high-energy, lower extremity injuries revealed that none of the tested limb salvage indices demonstrated any validated clinical utility. Additionally, a high
specificity confirmed that limb salvage with low index scores
could be predicted, but a low sensitivity failed to support any
index as a valid predictor of amputation. Currently, no limb
salvage index has been statistically confirmed to be reliable in
the evaluation and treatment of patients with severely mangled lower extremities.

Harvey EJ, Agel J, Selznick HS, Chapman JR, Henley

MB: Deleterious effect of smoking on healing of open
tibia-shaft fractures. Am J Orthop 2002;31:518-521.

440

Schmitz MA, Finnegan M, Natarajan R, Champine J: Effect of smoking on tibial shaft fracture healing. Clin
Orthop 1999;365:184-200.
In a study of 146 tibial fractures treated either surgically
or nonsurgically, absolute union rate was not significantly different between smokers and nonsmokers. However, time to
union was significantly delayed in patients who smoked with
average time to healing at 136 days for nonsmokers and 269
days for smokers. In patients treated nonsurgically, these differences were not significant.

Toivanen JA, Vaisto O, Kannus P, Latvala K, Honkonen

SE, Jarvinen J: Anterior knee pain after intramedullary
nailing of fractures of the tibial shaft: A prospective,
randomized study comparing two different nailinsertion techniques. J Bone Joint Surg Am 2002;84-A:
580-585.
In this randomized study of 50 patients undergoing either
transtendinous or paratendinous tibial nailing, 21 patients in
both groups had an average 3-year follow-up. No difference
was noted in anterior knee pain with either approach. Knee
scoring systems, muscle strength measurements, and functional
tests showed no significant difference between the techniques.
The authors concluded that there were no significant differences in outcomes for patients treated with the transpatellar
versus peritendinous approach for tibial nail insertion.

Classic Bibliography
Bhandari M, Guyatt GH, Tong D, et al: Reamed versus
nonreamed intramedullary nailing of lower extremity
long bone fractures: A systematic overview and metaanalysis. J Orthop Trauma 2000;14:2-9.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Bosse MJ, MacKenzie EJ, Kellam JF, et al: A prospective evaluation of the clinical utility of the lowerextremity injury-severity scores. J Bone Joint Surg Am
2001;83:3-14.
Buehler KC, Green J, Woll TS, Duwelius PJ: A technique for intramedullary nailing of proximal third tibia
fractures. J Orthop Trauma 1997;11:218-223.
Collinge CA, Sanders RW: Percutaneous plating in the
lower extremity. J Am Acad Orthop Surg 2000;8:211216.
Finkemeier CG, Schmidt AH, Kyle RF, et al: A prospective randomized study of intramedullary nails inserted
with and without reaming for the treatment of open and
closed fractures of the tibial shaft. J Orthop Trauma
2000;14:187-193.
Gustilo RB, Anderson JT: Prevention of infection in the
treatment of one thousand and twenty-five open fractures of long bones: A retrospective and prospective
analysis. J Bone Joint Surg Am 1976;58:453-458.
Gustilo RB, Mendoza RM, Williams DN: Problems in
the management of type III (severe) open fractures.
J Trauma 1984;24:742-746.
Henley MB, Chapman JR, Agel J, et al: Treatment of
type II, IIIA, and IIIB open fractures of the tibial shaft:
A prospective comparison of unreamed interlocking intramedullary nails and half-pin external fixators. J Orthop Trauma 1998;12:1-7.
Karunakar MA, Egol KA, Peindl R, Harrow ME, Bosse
MJ, Kellam JF: Split depression tibial plateau fractures:
A biomechanical study. J Orthop Trauma 2002;16:172177.
Keating JF, OBrien PI, Blachut PA, et al: Reamed interlocking intramedullary nailing of open fractures of the
tibia. Clin Orthop 1997;338:182-191.
Krettek C, Miclau T, Schandelmaier P, et al: The mechanical effect of blocking screws (Poller screws) in
stabilizing tibia fractures with short proximal or distal

American Academy of Orthopaedic Surgeons

Chapter 36 Knee and Leg: Bone Trauma

fragments after insertion of small-diameter intramedullary nails. J Orthop Trauma 1999;13:550-553.
MacKenzie EJ, Bosse MJ, Kellam JF, et al: Characterization of patients with high-energy lower extremity
trauma. J Orthop Trauma 2000;14:455-466.
Marsh JL, Smith ST, Do TT: External fixation and limited internal fixation for complex fractures of the tibial
plateau. J Bone Joint Surg Am 1995;77:661-673.
McQueen MM, Gaston P, Court-Brown CM: Acute
compartment syndrome: Who is at risk? J Bone Joint
Surg Br 2000;82:200-203.
Meyers MH, McKeever FM: Fracture of the intercondylar eminence of the tibia. J Bone Joint Surg Am 1970;52:
1677-1684.
Raiken SM, Landsman JC, Alexander VA, et al: Effect
of nicotine on the rate and strength of long bone fracture healing. Clin Orthop 1998;353:231-237.
Reynders P, Reynders K, Broos P: Pediatric and adolescent tibial eminence fractures: Arthroscopic cannulated
screw fixation. J Trauma 2002;53:49-54.
Schatzker J, McBroom R: Tibial plateau fractures: The
Toronto experience 1968-1975. Clin Orthop 1979;138:94104.
Sirkin MS, Bono CM, Reilly MC, Behrens FF: Percutaneous methods of tibial plateau fixation. Clin Orthop
2000;375:60-68.
Watson JT, Coufal C: Treatment of complex lateral fractures using Ilizarov techniques. Clin Orthop 1998;353:97106.
Welch RD, Zhang H, Bronson DG: Experimental tibial
plateau fractures augmented with calcium phosphate cement. J Bone Joint Surg Am 2003;85-A:222-231.
Yetkinler DN, McClellan RT, Reindel ES, Carter D,
Poser RD: Biomechanical comparison of conventional
open reduction internal fixation versus calcium phosphate cement fixation with central depressed tibial plateau fracture. J Orthop Trauma 2001;15:197-206.

441

Chapter

37

Knee and Leg: Soft-Tissue Trauma

Eric C. McCarty, MD
Kurt P. Spindler, MD
Reed Bartz, MD

Collateral Ligament Injury

Injuries to the medial collateral ligament (MCL) occur
much more frequently than to the lateral collateral ligament (LCL). The mechanism of MCL injury is a valgus
force caused by lateral contact to the knee or lower leg,
which either injures the distal femoral physis (in the
skeletally immature patient), MCL (in the young to
middle-aged adult), or the lateral tibial plateau (in the
middle-aged to older adult or senior citizen). Careful
physical examination of the knee can confirm the presence of a Salter-Harris injury to the distal physis in the
adolescent athlete; pain proximal to the normal insertion of the MCL and, more importantly, pain extending
across the femur to the lateral side is common. Radiographs or MRI also can confirm diagnosis. Physical examination determines whether treatment of MCL or
LCL sprains is nonsurgical (majority of tears) or surgical (associated with cruciate and/or posteromedial corner tears). Isolated injuries to the MCL are most common, with knee laxity at 20 to 30 of flexion; however,
these injuries are stable with the knee in full extension
(indicating intact posteromedial capsule). Isolated MCL
injuries associated with anterior cruciate ligament
(ACL) tears are usually treated nonsurgically at the
time of ACL reconstruction. In the relatively rare case
of MCL injuries with laxity in full extension, the posteromedial corner is torn and usually one or both cruciates
have been torn. In this circumstance, MRI to localize
the site of the tear and fully define the associated injuries (especially of the cruciates) is helpful in surgical reconstruction of one or both cruciates and repair of the
MCL and posteromedial capsule tears.
Isolated injuries to the LCL are relatively rare but
can occur with a contact mechanism that causes a complete ACL rupture.
If laxity is present only in 20 to 30 of flexion, a palpable LCL in the figure-of-4 position is found, and no
laxity is present in full extension, then a complete disruption of LCL is not present and the knee can be protected from varus stress with a brace. When significant
laxity is present in full extension, then a complete dis-

American Academy of Orthopaedic Surgeons

ruption of the LCL with associated posterolateral corner tears with involvement of a cruciate ligament is
likely. Primary surgical repair is indicated, usually with
cruciate reconstruction. In combined injuries to the
LCL and posterolateral corner, MRI can help localize
the site of injury (femur versus midsubstance versus fibular head) and identify associated injuries.
Figure 1 outlines the evaluation and treatment approach. Evaluation for growth plate injuries in adolescents, for instability in extension (indicating a posterolateral or medial capsular tear), and for complete
cruciate ligament injuries is essential. Isolated injuries
documented on physical examination are usually treated
nonsurgically with protection from valgus (MCL) and
varus (LCL) forces in the healing phase.

Anterior Cruciate Ligament Injury

The physical examination, imaging, timing of surgical reconstruction, and rehabilitation of the ACL have been
well defined in the literature. However, there has been
intense scientific activity on the choice of autograft, with
several randomized clinical trials comparing patellar
tendon with hamstrings. New techniques for hamstring
fixation have improved stability of the construct and allowed more aggressive rehabilitation.
The most common mechanisms of injury for ACL
tears are noncontact (approximately 70%), sportsrelated (approximately 80%), and recollection of a
pop (approximately 70%). The overwhelming majority of patients do not return to play in most sports after
the injury without surgery. Contact injuries are more
likely to involve MCL injuries. Patients injured while
jumping have a significant increase in intra-articular injuries. Female athletes have a twofold to fourfold increased risk of ACL tears when participating in the
same sports and at the same levels as male athletes.
Physical examination remains the mainstay in diagnosis of ACL (Lachman test) and PCL (posterior
drawer test) tears. A positive Lachman test is the absence of a firm end point with or without a perceived in-

443

Knee and Leg: Soft-Tissue Trauma

Orthopaedic Knowledge Update 8

Figure 1 Evaluation and management of collateral ligament injuries. *Rare unilateral collateral (MCL or LCL) tears with instability in extension indicating posterior corner disruption
without complete cruciate tear.

Figure 2 Cruciate ligament injuries in sports.

*Partial injuries uncommon in ACL tears but
are the majority with PCL tears in sports.

Complete = overwhelming majority of ACL

tears but rare for PCL in sports. Follow-up
MRI documented return of normal ligament
signal on MRI consistent with healing.
crease in excursion. A complete ACL tear is confirmed
by a positive pivot shift test. Radiographs are normal in
more than 95% of young, active athletes. The rare presence of a tiny fleck of bone on the lateral side at the lateral tibial plateau is indicative of a Segond fracture,
which is pathognomonic for an ACL tear. In the middleaged athlete, radiographs obtained while the patient is
standing should be reviewed for evidence of arthritis
(narrowing of joint or cartilage space) or evidence of a
chronic ACL tear (notch stenosis and tibial spine spurs).
MRI is particularly helpful when the diagnosis is in
doubt or for identifying clinically relevant pathology, including collateral tears or bucket handle meniscus tears.
Figure 2 outlines recommended treatment approaches
for a patient whose cruciate ligament is injured during
sports activity. Complete posterior cruciate ligament
(PCL) tears are extremely uncommon during sports activity (relatively low-velocity mechanism of injury), with
the exception of occasional knee dislocations. However,
PCL tears involving one or both collateral ligaments are

444

common during motor vehicle crashes (higher-velocity

mechanism of injury) and are discussed in the section
on multiple ligament injuries.

Anterior Cruciate Ligament Reconstruction

Patients with partial tears of the ACL or PCL who meet
the criteria outlined in Figure 2 can achieve a high level
of function and return to sports after rehabilitation. The
variables that need to be considered in the decision of
whether to reconstruct the ACL-deficient knee without
a complete tear of the collateral ligaments or PCL are
diagrammed in Figure 3. Key factors to consider include
the presence of a repairable meniscus lesion, desired return to competitive sports activity, and involvement in
sports activity involving cutting and pivoting. Prerequisites to ACL reconstruction are nearly full range of motion, good quadriceps tone, normal gait, and limited effusion; all are signs that the knee has recovered from
acute inflammatory trauma of hemarthrosis and that arthrofibrosis is less likely to develop postoperatively. Al-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 37 Knee and Leg: Soft-Tissue Trauma

Figure 3 ACL reconstruction decision making. *IKDC = International Knee Documentation Committee consensus classification;
strenuous, jumping and pivoting sports; moderate, heavy manual work and skiing; light,
light manual work, running; sedentary, activities of daily living. Individualize based on
willingness to change activity levels, occupation, arthritis, other medical conditions.
though the exact type and frequency of supervised rehabilitation for patients undergoing ACL reconstruction
has not been determined, monitored rehabilitation by a
qualified therapy team immediately after surgery is
helpful. Successful home rehabilitation protocols all include preoperative instruction, written materials for patients, regular supervised intervals, and the option for
patients to call for advice. Principles of rehabilitation include immediate range-of-motion exercises, early weight
bearing, closed chain exercises, safe restoration of quadriceps strength, and an emphasis on proprioceptive
training to guide the safe return to sports activity.
There has been debate on the definition of acute
versus delayed or chronic ACL reconstruction. As mentioned previously, a prerequisite consideration for ACL
reconstruction is restoration of the normal activities of
daily living, which indicates that acute inflammatory
trauma has subsided and is important in the prevention
of postoperative arthrofibrosis. Reinjury to an ACLdeficient knee has been shown to increase the frequency
of meniscal tears and articular cartilage injuries, including arthritis. Thus, acute reconstruction can be indicated
for patients who have not been reinjured, and delayed
or chronic rehabilitation can be indicated for patients
experiencing additional episodes of giving way. No exact
time frame is recommended. In contrast, some authors
believe acute reconstruction should take place within 6
weeks of injury.
A patient with chronic ACL deficiency may have developed medial compartment arthritis, especially if partial medial meniscectomy has been performed. Stiffness
and pain are usually the result of arthritis and not instability. Instability from ACL deficiency is associated with
giving way or patients describing a shifting of the
knee. Radiographs should be obtained with the patient
standing, and careful evaluation of the patient while
walking should be performed. A high tibial osteotomy

American Academy of Orthopaedic Surgeons

(HTO) is indicated if a varus thrust is seen with or without medial arthritis because without correction of mechanical axis, the ACL graft will eventually fail because
varus thrusting overstresses the graft. The decision to
perform an HTO should be based on the patients
symptoms and the shift of the weight-bearing axis or
away from a degenerative medial compartment. If arthritic symptoms are believed to be the predominant
complaint, then HTO should be considered before ACL
reconstruction. Unless an experienced surgeon is confident that stable fixation of HTO and ACL graft can be
achieved so that early motion can begin, staging of these
procedures may be more advantageous.
An evidence-based review of ACL ligament surgery
shows that the results of randomized clinical trials support the failure of primary repair, and an improved stability and decreased rate of meniscus reinjury after ACL
reconstruction. Randomized trials have not shown differences in outcomes using autograft patellar tendon
versus hamstring approaches, or arthroscopic surgical
techniques (two-incision [rear entry] versus singleincision [endoscopic]). Thus, ACL reconstructions are
performed to resume short-term function (2 years), especially for participation in sports activities requiring
cutting and pivoting. Whether ACL reconstruction prevents or delays knee osteoarthritis is unknown. Accurate tunnel placement, strong graft choices, solid initial
fixation of grafts within the tunnel, and a rational rehabilitation program are factors that have been shown to
bring about good to excellent short-term results. Placement of the femoral tunnel within 1 to 3 mm over the
top position and tibial tunnel placement behind the intercondylar roof in full extension also are important.
Coupling a proven fixation technique with the use of a
specific graft and with aggressive rehabilitation seems
appropriate. For example, patellar tendon graft with interference screws in young patients allows aggressive re-

445

Knee and Leg: Soft-Tissue Trauma

habilitation. Newer hamstring fixation methods allow
similar rehabilitation. Recently, the use of improved aperture fixation at the femoral tunnel with a cross-pin
and metallic washer at the tibial tunnel have provided
improved biomechanical fixation, resulting in adoption
of similar rehabilitation protocols as patellar tendon
grafts.
Typically, autogenous grafts are harvested from the
ipsilateral knee; however, the use of the contralateral
patellar tendon for ACL reconstruction has been shown
to have good results. Advantages cited are decreased
morbidity on the reconstructed knee and faster patient
recovery.
Graft sources other than autogenous grafts are often
used. Attempts at reproducing the ACL with synthetic
grafts was popular in the mid 1980s. Fueled by the desire to decrease donor morbidity and potentially speed
rehabilitation, alternative graft sources included three
broad categories of synthetic ligament replacements: (1)
permanent replacement of the ACL with a prosthesis
(for example, the Gore-Tex graft; Gore-Tex, Flagstaff,
AZ); (2) use of a stent to augment a biologic graft and
protect it during the early postoperative period (for example, the Kennedy ligament augmentation device; 3M,
St. Paul, MN); and (3) use of a scaffold for appropriate
support for collagen ingrowth to recreate the ligament
(for example, the Leeds-Keio polyester device; Yufo
Seiki, Tokyo, Japan). These methods were largely abandoned by the mid 1990s because of high rates of complications and failures. Complications included an inflammatory response to the graft and graft wear. Despite
these failures, the search for a synthetic graft continues.
Two future options are a biologic scaffold and/or a
bioengineered prosthesis.
Allograft use for knee ligamentous reconstruction
has increased significantly in the past decade. Some of
the main reasons cited for the use of allografts include
decreased surgical morbidity associated with the graft
harvest, a quicker recovery time, smaller incisions, and
decreased surgical time. Inherent disadvantages to using
allograft tissue include the potential risk of disease
transmission, cost, a possibly slower biologic remodeling
process, and a theoretic slow chronic immunologic response to the tissue. The allograft tissue provides the
collagen scaffold for a process of incorporation of host
tissue. The stages of allograft incorporation are similar
to those of autograft; however, the remodeling phase of
the allograft can be one and a half times as long. The
use of allograft tissue for reconstruction of all knee ligaments is becoming more popular. It is appealing for
multiligament injuries in which autograft choices may
be limited. Some studies have shown good results using
allograft tissue for ACL reconstruction, although longterm results are lacking.

446

Orthopaedic Knowledge Update 8

Results/Outcomes of Anterior Cruciate Ligament

Reconstruction
Overall, the intermediate (5 years) or long-term (10
years or longer) results for ACL reconstruction are not
known. Traditional instruments used to judge clinical
outcomes following ACL surgery have been anteroposterior stability, nonvalidated activity level scales, and
nonvalidated scales combining objective with subjective
results into a combined score (International Knee Documenting Committee [IKDC]). Evidence-based medicine has focused on development of validated,
questionnaire-based assessment tools, first in knee osteoarthritis with the Western Ontario and McMaster
Universities Arthritis Index and with the Medical Outcomes Short Form-36. Sport-specific validated measures
of surgical outcome on activity level developed between
1998 and 2001 include the Knee Injury and Osteoarthritis Score and the IKDC questionnaire. The first validated measure of sports activity was developed in 2001.
These validated outcome instruments are currently in
use by multicenter prospective cohort studies to address
clinically relevant questions on putative risk factor assessment for pain, physical function, return to activity
level, and global score. The use of these instruments will
supplement and provide information not available from
smaller stability outcome measures.
Table 1 is a systematic review of nine randomized
controlled trials in the literature. All had equal preoperative groups, the same type of rehabilitation, no continuous passive motion, return to sports at approximately 6
months, and follow-up at 2 to 3 years. Several conclusions are evident from these trials. First, 44% (four of
nine randomized controlled trials) showed from 0.5 to
3.4 loss of range of motion with patellar tendon reconstruction. Second, 43% (three of seven randomized controlled trials) showed hamstring weakness with hamstring reconstruction. Third, 43% (three of seven
randomized controlled trials) of reconstructions using
patellar tendons were more stable by KT-1000 [Medmetric, San Diego, CA] arthrometer testing in the 1.0 to
3.4 mm range. Fourth, 89% (eight of nine randomized
controlled trials) showed no difference in anterior or
patellofemoral pain, but 100% (four of four randomized
controlled trials) of the patellar tendon group had more
kneeling pain. Fifth, there was no difference in nonvalidated measures of return to preinjury function, Tegner
activity level of outcomes by Lysholm, Cincinnati, or
original IKDC 1991. Finally, no validated outcome tools
previously mentioned were used in the best studies to
date on ACL surgery.

Posterior Cruciate Ligament Injury

PCL injuries with or without associated MCL/
posteromedial capsular or LCL/posterolateral capsular
injuries have been investigated using in vitro biome-

American Academy of Orthopaedic Surgeons

Chapter 37 Knee and Leg: Soft-Tissue Trauma

Orthopaedic Knowledge Update 8

Table 1 | Randomized Controlled Trials Outcomes for ACL Reconstruction: Hamstring Versus Patellar Tendon
Kneeling
Pain

XR

Return
Preinjury

Tegner
Activity

Lysholm

Cinn

IKDC

Ns
Ns
(maximum
manual)
PT 1.0 mm Ns
better

Ns

Ns

Ns

PT best

Ham

Ns

Ns

PT pain

Ns

PT3

Ns

PT pain

Ns

Ns

PT3.4

Ns

Ns

Ns

Ham

Ns

PT1.5

Ham 7% PT 1.1 mm Ns
better

PT pain

Ns

Ns

Ns

Ns

PT0.5
Ext

Ns

Ns

Ns

Ns

Ns

Ns

Ns

Ns

Ns

Ns

Ns

PT pain

Ns

Ns

Ns

ROM

ISO

KT(n)

ONeill
( 1996)

Ns

Ns

Anderson
et al
(2001)
Aune et al
(2001)
Eriksson
et al
(2001)
Shaieb
et al
(2002)
Beynnon
et al
(2002)
Feller and
Webster
(2003)
Jansson
et al
(2003)
Ejerhed
et al
(2003)

Ns

Ns

Ns

Ant Knee
Pain

PT > Ham
6 + 24
months
PT 3.4 mm Ns
better

ROM = range of motion; ISO = isokinetic strength hamstring versus patellar tendon; KT(n) = KT 1000 stability anterior to posterior (n) = Newtons; Ant knee pain = anterior or patellofemoral knee
pain; XR = x-rays or radiographs; Cinn = Cincinnati University Knee Rating; IKDC = original subjective and objective scale; PT = patellar tendon; Ham = hamstring; Ns = not significant (P >0.05).

chanical studies. Key decision-making criteria in the

clinical management include magnitude of trauma
(sports activity versus motor vehicle crashes), the degree of posterior laxity in reference to the anterior femur, and associated collateral or ACL tears. The posterior drawer test is performed with the knee flexed and
both hands placed around the knee. The joint line is palpated with the thumbs. The degree of laxity is then
judged by comparing the posterior tibial translation to
the contralateral normal knee with the tibia anterior to
the femur. Partial tears to the PCL have an end point
and laxity with posterior force with the tibia even or
slightly posterior to the femur. Complete tears have no
end point to posterior force, even with internal rotation
and significant displacement posterior to the femur.
Plain radiographs are indicated to rule out bony avulsions, which can be repaired. MRI can be helpful in determining partial versus complete injuries; some fibers
have continuity with partial tears. Partial tears are more
common in low-velocity injuries (especially from sports
activities) and are treated nonsurgically. Patients with

American Academy of Orthopaedic Surgeons

partial PCL injuries have shown healing over time on

follow-up MRI. It has been shown that 1+ or 2+ (5 to 10
mm) posterior laxity is well tolerated even in the highly
competitive athlete. Complete PCL tears are more common from high-velocity injuries resulting from motor
vehicle crashes and are often associated with collateral
injuries and possible knee dislocation. Figure 2 summarizes treatment approaches to sports-related injuries to
the PCL.
When PCL reconstructions are considered, the
choice of surgical approach and grafts should be
matched. Traditionally a single femoral tunnel approach
was used, but several authors have recently proposed
the use of two tunnels. A subject of current debate is
whether a tibial tunnel with the so-called killer curve
versus tibial onlay should be used. Tibial onlay requires
a bone block and either a patellar tendon or Achilles
tendon allograft. If tibial onlay is preferred and autograft tissue is used, then a single femoral tunnel with
interference screws is used. An Achilles tendon allograft
can be fixed as tibial only and either single or dual tun-

447

Knee and Leg: Soft-Tissue Trauma

nels can be used. Reproducible in vitro biomechanical
studies have not identified the best approach or graft
choice to date; however, research is ongoing. No comparative in vivo animal models or clinical studies have
provided a preferred technique. The surgeon should
choose the safest, most reproducible approach and technique to improve posterior stability.

Posterolateral Complex Injury

The posterolateral complex includes the biceps tendon,
iliotibial band, popliteus tendon, popliteofibular ligament, arcuate ligament, and LCL. The functional and
clinical significance of these closely interrelated structures is not yet completely understood. The posterolateral structures act synergistically with the PCL to resist
posterior translation and external and varus rotation of
the knee. The biceps femoris, a dynamic lateral stabilizer
of the knee, inserts on the fibular head. The iliotibial
band is an anterolateral stabilizer. The primary function
of the popliteus tendon is to externally rotate the femur,
unlocking the knee to allow flexion when the knee is
loaded. The popliteofibular ligament attaches the popliteus tendon to the posterior fibular head and the anterior lateral femoral condyle, providing resistance to posterior translation and external rotation of the tibia. The
arcuate ligament courses from the fibula styloid process
to the lateral femoral condyle, with limbs to the popliteus tendon and fascia overlying the popliteus muscle.
The LCL is the primary static restraint to varus stress
on the knee, and secondary restraint to external rotation
of the tibia.
The most common mechanisms of injury to the posterolateral structures are a direct blow to the anteromedial tibia and hyperextension with a varus twisting force.
Physical examination for evaluation of the posterolateral corner include the tibia external rotation test, in
which the degree of external rotation of the tibia in relation to the femur of each lower extremity is tested at
both 30 and 90 of knee flexion. An increase in external rotation (of at least 10), found only at 30 of knee
flexion, is indicative of an isolated injury to the posterolateral corner. Associated injury to the PCL yields increased external rotation at 90 of flexion. The posterolateral external rotation test is similar to the tibia
external rotation test, except a coupled force of posterior translation and external rotation is applied to the
knee. The test is positive if posterolateral subluxation of
the proximal tibia is noted at 30 of flexion. The reverse
pivot-shift test, which is not specific for injury to the
posterolateral corner, is performed by applying a valgus
stress to the knee while externally rotating the foot and
extending the knee from 90 of flexion. A positive test is
elicited when a palpable jerk is noted during reduction
of the posteriorly subluxated lateral tibia plateau. This
test has been reported to be positive in up to 35% of

448

Orthopaedic Knowledge Update 8

asymptomatic individuals. The external rotation recurvatum test is performed in a supine patient by grasping
the great toes of both feet and lifting both lower limbs
off the examining table. Posterolateral corner injury is
indicated by recurvatum or varus deformity of the knee
and external rotation of the tibia. Other tests for injury
to the posterolateral corner include the posterolateral
drawer test, adduction stress test, and dynamic posterior
shift test.
Plain film radiography may reveal an avulsion fracture of the fibular head from pull of the biceps femoris
or LCL. MRI has also proved to be useful for evaluation of these multiligamentous injuries and should include imaging of the entire fibular head.
For partial injuries to the posterolateral corner, including grade I and II instability with a good end point,
nonsurgical treatment is recommended. This entails a
3-week period of immobilization with the knee in full
extension followed by progressive range-of-motion and
strengthening exercises. For acute injuries, primary repair of the posterolateral structures offers the best
chance for a successful clinical outcome. Avulsion fractures of the fibular head can be repaired directly. Repair
can be augmented with allograft or autogenous tissue.
For chronic injuries, various techniques for reconstruction have been recommended with varied clinical results. Graft reconstructions recreating the popliteus tendon and LCL seem to fare the best. Postoperative
rehabilitation after surgical treatment of injuries to the
posterolateral corner involves protection from external
rotation and varus stress with a hinged knee brace for a
minimum of 2 months, followed by gradual range-ofmotion and strengthening exercises.

Knee Dislocation
Knee dislocation is a devastating injury that usually results from high-energy trauma. Any three-ligament injury should be considered and treated as a knee dislocation. With improvements in surgical technique and
instrumentation, the results of surgical treatment have
surpassed those of conservative methods and are now
the primary form of treatment of the dislocated knee.
Knee dislocation is classified primarily by the direction of the dislocated tibia in relation to the femur (anterior, posterior, medial, lateral, and rotatory). Determining the direction of the dislocation provides
information about the likelihood of associated neurovascular injury. Injury to the popliteal artery is more
likely with posterior dislocation, whereas injury to the
common peroneal nerve is more likely with posterolateral dislocation.
Associated injuries with knee dislocation are common. The incidence of vascular injury with all dislocations has been estimated at 32%. The popliteal artery is
injured either by a stretching mechanism secondary to

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

tethering of the vessel at the adductor hiatus or by direct contusion by the posterior tibia plateau. Thrombus
formation secondary to intimal damage can occur several days after injury. Decreased pulses or vascular compromise should never be attributed to arterial spasm,
and additional investigation is warranted. Injury to the
popliteal vein is rare. Injury to the peroneal nerve, or
less commonly the tibial nerve, has been estimated to
occur in 20% to 30% of knee dislocations. The incidence
of associated fractures, usually of the tibia plateau or
the distal femur, has been estimated at 60%.
The diagnosis of a dislocated knee is usually obvious; however, it is crucial to remember that a dislocated
knee can reduce spontaneously. A spontaneously reduced knee is characterized by gross instability and extensive soft-tissue swelling. With a suspected knee dislocation, immediate evaluation of the vascular status of
the involved limb is important because limb ischemia of
more than 8 hours duration is likely to result in amputation. Vascular injury has been reported in the presence
of normal pulses, Doppler studies, and capillary refill.
Repeat neurovascular examination should be performed
after any reduction maneuver.
Plain film radiographs should be obtained in two
planes to determine the direction and severity of the
dislocation, as well as for evaluation for any associated
fractures. Slight joint space widening may be indicative
of a spontaneously reduced knee dislocation. Plain film
radiographs should also be evaluated for avulsion fractures, which may alter the surgical plan. Repeat plain radiographs should be obtained after any reduction maneuver to ensure satisfactory alignment of the joint.
With a suspected arterial injury, immediate arteriography and vascular consultation are warranted. Although arteriograms have been used historically as a
screening tool for arterial injury, it is a low yield study in
the presence of a normal vascular examination. With no
evidence of vascular compromise on physical examination, foregoing an arteriogram and performing regular,
detailed neurovascular examinations in an inpatient setting is accepted treatment. If vascular compromise is
noted, immediate surgical intervention is indicated. Formal angiography is bypassed to lessen ischemic time.
After the initial survey and acute treatment, MRI is
useful for evaluation of ligament and meniscus status,
occult fracture, and capsular disruption. MRI gives essential information for preoperative planning for ligament reconstruction, such as the number of allografts
that will be needed.
Timely closed reduction is indicated for any dislocated knee. Physical and radiographic examination
should be performed immediately after reduction to
confirm proper reduction and evaluate postreduction
neurovascular status. The presence of the dimple sign
is indicative of an irreducible posterolateral dislocation,

American Academy of Orthopaedic Surgeons

Chapter 37 Knee and Leg: Soft-Tissue Trauma

but it is a contraindication to closed reduction because
of the risk of skin necrosis.
Indications for acute surgery after knee dislocation
include vascular injury, open fracture, open dislocation,
irreducible dislocation, and compartment syndrome.
Restoration of vascular integrity takes precedence over
other injuries. Open injuries require immediate surgical
irrigation and dbridement.
Improved surgical techniques and postoperative rehabilitation protocols, and the use of allografts have improved the clinical results for surgical treatment of knee
dislocation. Factors to consider in the timing of surgical
treatment of knee dislocations include the vascular status of the patient, the presence of open wounds, the success of closed reduction, the stability of the joint after
reduction, and the presence of associated injuries.
There has been debate over the timing of surgical procedures after knee dislocation. Two reasons cited for delaying surgery include allowing a period of vascular monitoring and reducing the risk of arthrofibrosis. Multiple
ligament injuries may require staged procedures. Posterolateral structures, capsular structures, and avulsion fractures can be repaired acutely. Combined ACL and PCL
reconstruction can be performed in delayed fashion with
acceptable results. Graft options for ligament reconstruction in the multiligament-injured knee include patellar
tendon, hamstring, and quadriceps tendon autografts,
and/or allografts. The literature does not support the use
of one graft type over another or the use of a particular
technique for reconstruction. After combined reconstruction of the cruciate ligaments, the knee has traditionally
been kept in extension with no weight bearing for 6 weeks.
Postoperative rehabilitation programs after a knee dislocation should allow range-of-motion exercises as soon as
the integrity of soft-tissue repair, vascular repair, and ligament reconstruction permit.

Meniscal Injury
The primary function of the meniscus is to evenly distribute the weight-bearing load across the knee joint.
The menisci transmit approximately 50% of the load
with the knee in extension, and close to 90% of the load
at 90 of knee flexion. With flexion past 90, most of the
force is transmitted through the posterior horns. The lateral meniscus has been shown to transmit a greater percentage of the load compared with the medial meniscus.
When meniscal integrity is lost, abnormal articular
contact stresses result, leading to potential increased
wear of the articular cartilage and early degenerative
changes. The more meniscal tissue that is lost, the
greater the loss of contact surface area and the greater
the increase in peak local contact stresses. Thus, the primary goal of treatment of a meniscal tear is to maintain
as much healthy meniscus tissue as possible.

449

Knee and Leg: Soft-Tissue Trauma

Figure 4 Zone of meniscal vascularity: Red zone = rich in vascular plexus, Red/White
zone = transition between vascular zone and avascular zone, White zone = avascular
zone. (Reproduced with permission from Miller, MD, Warner JJP, Harner CD: Meniscal
repair, in Fu FH, Harner CD, Vince KG (eds): Knee Surgery. Baltimore, MD, Williams &
Wilkins, 1994, p 616.)
In addition to some biomechanical differences in
medial and lateral menisci function, there are several
anatomic variances. The medial meniscus is semicircular
with disparate insertions, whereas the lateral meniscus is
more circular with closely approximated insertions. The
medial meniscus is wider posteriorly than anteriorly,
whereas the lateral meniscus has posterior and anterior
segments that are close to equal in width. The medial
meniscus is also more firmly attached to the knee capsule whereas the lateral meniscus is loosely attached.
A key aspect of the meniscal anatomy is its vascularity, which is one of the critical elements in healing of a
meniscal repair. The most peripheral 20% to 30% of the
medial meniscus and the peripheral 10% to 25% of the
lateral meniscus are consistent in vascularity (Figure 4).
Branches from the superior, inferior, and lateral geniculate arteries supply this vascular zone. Because of its
rich blood supply, this area is commonly referred to as
the red zone, and it is an area that has greater healing
potential than the inner portions of the meniscus with
less or no vascular supply. The inner third of the meniscus is avascular and is referred to as the white zone. This
area is nourished by synovial fluid diffusion and repairs
usually do not heal well in this zone. The area (middle
third) of meniscus between the red and white zones is
known as the red/white zone. Because this area does
have some blood supply, it has the potential for healing,
particularly in the young patient. The area in the
posterolateral aspect of the lateral meniscus in front of
the popliteus tendon, however, is a watershed area and
even its peripheral third is relatively hypovascular.
There are several patterns/types of meniscal tears,
each with potential ramifications on healing. Vertical
longitudinal tears are common and often can be repaired, especially if located in the peripheral third of the
meniscus. If these tears extend in length circumferen-

450

Orthopaedic Knowledge Update 8

tially around the meniscus they will become bucket handle tears. Bucket handle tears may become unstable,
and the fragment can displace into the knee, causing
true locking of the knee. Most bucket handle tears can
be repaired and will heal if the fragment is not degenerative and deformed and if the tear is in a vascular area.
Radial tears start in the central portion and can propagate toward the periphery. In general, these are not receptive to repair because the circumferential hoop fibers are disrupted and much of the tear is in an
avascular zone. Oblique tears often occur at the junction
of the posterior and middle thirds of the meniscus and
can often be repaired. Flap (parrot beak) tears are meniscal tears that begin as a radial type and then extend
circumferentially around a portion of the meniscus.
These tears often will have a large flap component that
causes significant mechanical symptoms and cannot be
repaired. Horizontal tears occur more frequently with
age, are often associated with meniscal cysts, and typically are not repairable. Complex tears are also more
common in older patients, are typically a combination of
the types of tears described previously, and occur in
multiple planes. Most often these tears are seen in the
posterior horn and are best treated with excision. Tears
are sometimes evident in the discoid meniscus. Incidental discoid menisci found at arthroscopy are not treated.
If a tear is evident in a classic discoid meniscus, then an
excision of the tear is performed with saucerization of
the meniscus. Peripheral detachments of the discoid meniscus are repaired.
Some meniscal tears, depending on the symptoms
they incur, can be treated nonsurgically. These include
(1) longitudinal tears that are stable (displaced < 3 mm)
and less than 5 to 8 mm in length; (2) partial tears that
are stable; (3) shallow radial tears (< 3 mm in depth);
and (4) tears with a favorable natural history, which includes small lateral meniscal tears with a concurrent
ACL reconstruction.
A partial meniscectomy is indicated for tears in the
avascular (white) zone and radial, oblique, and flap
tears. This procedure may also be performed for any
tear that has caused significant injury to the body of the
meniscus, such as a complex tear, numerous cleavage
tears, an alteration in the contour of the meniscal body,
or degenerative tears, which may render any repair effort as futile. Injuries to the body of the meniscus may
damage the structural integrity of the meniscus and the
vascularity may be in doubt. Additionally, it is difficult
to hold together degenerative tears with the various meniscal repair techniques. If excision of a meniscal tear is
performed, the concept of preserving as much viable
meniscal tissue as possible remains applicable. The goal
is to remove only a torn, degenerated, or abnormally
shaped meniscal fragment and thus perform a partial
meniscectomy and not a total meniscectomy. The possible adverse effects a total meniscectomy on the knee

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

joint have been well documented in the literature. The
excision of a tear or degenerative tissue will retain functional meniscus. Long-term results after partial meniscectomy seem to indicate satisfactory results, but deterioration of articular cartilage may still occur.
Degenerative changes on radiographs and a decrease in
function have been shown to occur earlier in patients
undergoing partial lateral meniscectomy than in those
undergoing partial medical meniscectomy. Additionally,
the results of partial meniscectomies are better in patients with acute traumatic tears than in those who have
degenerative tears.
If a repair is performed, it is important to consider
the numerous factors involved in healing, such as location and vascularity of the tear, the type of the tear, and
quality of the meniscal tissue. The stability of the knee is
important because repairs are more successful in a stable knee. Also, those repairs concurrently performed
with an ACL reconstruction have a higher rate of healing, probably because of the various growth factors in
the associated hemarthrosis. Chronicity is also a factor.
In general, results of meniscal repair are better acutely
after tear (< 8 weeks after injury) than a chronic tear
secondary to the potential deterioration of the meniscal
tissue with time. Patient age is also a factor. Younger patients have higher healing rates and axial alignment.
Those patients with a varus alignment have a lower
healing rate for medial repairs.
Meniscal repair can be done by either an open or an
arthroscopic technique. Although the open technique
has good results, most surgeons now prefer arthroscopic
techniques. The traditional gold standard for meniscal
repair is the inside-out repair, which is performed with
either absorbable or nonabsorbable sutures placed arthroscopically with long needles and retrieved through a
small incision on the outside of the capsule. The
outside-in technique involves placing a monofilament
suture through a spinal needle percutaneously into the
knee joint and retrieving it through one of the arthroscopic portals. The all-inside technique is performed
through the anterior knee portals and uses numerous
devices and variations of techniques to stabilize the meniscus. Most of the devices are absorbable stents, and
some are a combination of suture attached to the stent.
A cited theoretical advantage to the all-inside technique
is a decreased risk of a neurovascular injury that may
occur with the use of the needles in the other techniques. All of the meniscal repair techniques have
shown reasonable rates of clinical success (73% to 98%)
as determined by the absence of symptoms and the patients return to activities. In studies of second-look arthroscopy, healing rates have ranged from 45% to 91%.
Most of the studies on meniscal repair have been retrospective.
Devices and techniques continue to be developed
for meniscal repair. Research is focusing more on the bi-

American Academy of Orthopaedic Surgeons

Chapter 37 Knee and Leg: Soft-Tissue Trauma

ologic issues of healing and techniques to stimulate or
augment that healing. In the past, techniques such as
trephination of meniscal tissue, rasping, and fibrin clot
have been used to augment the repair of the meniscus.
More recently, investigations have targeted growth factors and gene therapy to enhance the healing of the meniscus.
If a total meniscectomy or subtotal meniscectomy
previously has been performed, a meniscal substitute is
reasonable. Because genetically engineered or other
similar biologic replacements are still experimental, a viable option for the meniscectomized knee is a meniscal
allograft. Meniscal allografts have been used for over a
decade with reasonable results. With proper technique,
the grafts have been shown to heal in the peripheral repair site and at its bony insertion. Grafts placed without
a bony base, either a plug or block, have been found to
have a higher failure rate. A symptomatic patient with
lack of a functional meniscus is a candidate for a meniscal allograft. Factors that contribute to the failure of a
meniscal allograft and thus are considered contraindications include patients with grade IV chondral changes,
ligamentous instability, and knee malalignment.

Patellofemoral Disorders
Patellofemoral pain continues to be an enigma. Conditions causing patellofemoral symptoms include patellofemoral pain syndrome, tracking disorders, instability/
dislocation, and chondromalacia.
The initial patient history should determine if the
chief symptom is pain or instability. The patient should
be questioned about a history of trauma, instability, and
activities that initiate or increase symptoms (often ascending or descending stairs). The typical patient with
patellofemoral symptoms is a female adolescent or
young adult. The onset of anterior knee pain is usually
insidious but can be acute. The symptoms of pain are
usually worse with activity.
Physical examination must include assessment of
lower extremity alignment, quadriceps angle, patellar
tracking and mobility, crepitus, patellar apprehension,
and specific areas of tenderness.
Standard radiographic evaluation of the knee should
include AP, lateral, axial, and long-standing (for alignment) views. The lateral view, which should be obtained
with the knee in at least 30 of flexion, is used to assess
the position of the patella in relation to the patellar tendon (the Insall-Salvati ratio). The average ratio is 1.02,
with a value over 1.2 indicative of patella alta and a
value less than 0.8 indicative of patella baja.
Various techniques for axial views of the patellofemoral joint are used to evaluate trochlear morphology
and patellar tilt. The Merchant/sunrise view is obtained
with the knee flexed 45 and the x-ray tube angled 30
from horizontal. The sulcus and congruence angles are

451

Knee and Leg: Soft-Tissue Trauma

both measured on the Merchant view. Patellar tilt and
congruence angles can also be measured using CT midpatellar transverse images with the knee flexed 15, 30,
and 45. MRI is useful for the evaluation of articular
cartilage status.
Nonsurgical treatment continues to be the primary
treatment for patients with a new onset of anterior knee
pain. A well-supervised rehabilitation program is augmented by anti-inflammatory medication and local modalities. The goal of the rehabilitation program is to
reduce symptoms by emphasizing quadriceps and hamstring stretching, strengthening, and endurance.
As biomechanical studies have shown that patellofemoral contact pressures are lowest between 0 and 30
of knee flexion, closed-chain, short-arc knee extensions
performed within a pain-free arc of motion are the
mainstay exercises for quadriceps strengthening. Patellar bracing and the McConnell taping technique have
been used as an adjunct to a rehabilitation program.
Lateral patellar compression syndrome is characterized by anterior knee pain that increases with patellar
loading activities and a patella centered without radiographic tilt. The syndrome has been associated with an
increased Q angle (the angle formed by a line drawn
from the middle of the tibial tuberosity to the middle of
the patella to the anterior-superior iliac spine) and a
tight lateral retinaculum. Patients sometimes experience
subjective episodes of instability, which are likely caused
by quadriceps inhibition. Routine radiographic studies
yield little useful information. If nonsurgical treatment
fails and the patient has a tight lateral retinaculum, arthroscopic lateral release has yielded successful results
in 60% to 91% of patients.
Patellar tilt is similar clinically to lateral patellar
compression syndrome with the exception that axial imaging reveals patellar tilt within the trochlear groove.
Initial treatment is nonsurgical, followed by arthroscopic lateral release if no improvement is noted. Potential complications following arthroscopic lateral release
include hemarthrosis and prolonged quadriceps weakness.
Patellar instability can be classified as subluxation or
dislocation. Acute dislocation often occurs with a twisting movement of the knee. Predisposing factors for patellar dislocation include patella alta, generalized ligamentous laxity, lateral femoral condyle hypoplasia,
lateral insertion of the patellar tendon, and an increased
Q angle. Dislocation results from injury to the medial
patellofemoral ligament. The patient may note that the
patella initially lies on the lateral aspect of the knee,
then relocates with knee extension. There is often an associated hemarthrosis. Physical examination frequently
reveals tenderness over the medial retinaculum and the
medial femoral epicondyle. MRI is useful for evaluating
injury to the medial patellofemoral ligament and assess-

452

Orthopaedic Knowledge Update 8

ing for associated injuries, such as osteochondral injuries
to the medial facet of the patella or the lateral femoral
condyle.
As studies of nonsurgical treatment have reported
redislocation rates ranging from 15% to 44%, there has
been renewed interest in surgical treatment of patellar
dislocation. Surgery is also considered for treatment of
associated injuries with an acute patellar dislocation,
such as removal, fixation, or dbridement of osteochondral injuries. In the young athletic patient, direct repair
of the medial patellofemoral ligament to its attachments
to the adductor tubercle and vastus medialis obliquus
has been advocated. A lateral release can be performed
if concomitant patellar tilt is appreciated. For the skeletally mature patient experiencing recurrent episodes of
patellar subluxation or recurrent dislocation despite
nonsurgical treatment, a distal patellar realignment procedure, such as a Fulkerson or Elmslie-Trillat procedure,
should be considered. The Fulkerson procedure is an
oblique osteotomy of the tibial tuberosity posterolaterally to anteromedially resulting in an anteromedial
transfer of the tuberosity, decreased patellofemoral contact forces, and realignment of the extensor mechanism
medially. The Elmslie-Trillat procedure is done by performing the osteotomy laterally to medially, preserving
a distal pedicle of the tuberosity and resulting in a medial transfer of the tuberosity.
Patellofemoral chondromalacia continues to be an
enigma because studies have shown 9% to 69% of patients with patellofemoral pain have normal cartilage in
the patellofemoral joint at the time of arthroscopy. Nonsurgical treatment of patellofemoral chondromalacia
consists of activity modification and anti-inflammatory
medication to reduce inflammation followed by extensor mechanism strengthening, particularly of the vastus
medialis obliquus. Surgical treatment of patellofemoral
chondromalacia falls into two categories: (1) treatment
aimed at relieving symptoms by decompressing the patellofemoral joint or correcting malalignment, and (2)
surgery aimed at directly addressing the chondromalacia. Surgical treatment of patellofemoral chondromalacia secondary to increased compression caused by malalignment or a tight lateral retinaculum includes lateral
release and anteromedialization of the tibial tubercle.
One study has suggested that release of adhesions in the
interval of soft tissue between the anterior proximal
tibia and the patellar tendon can decompress the patellofemoral joint. Patellar chondroplasty is used to decrease the mechanical symptoms by using a motorized
shaver to smooth or remove regions of cartilage fibrillation or loose flaps. Uniformly successful results with patellar chondroplasty have not been reported. For regions of grade IV chondromalacia, marrow stimulation
techniques, such as microfracture or subchondral drilling, have been advocated. Results with marrow stimulation techniques have been unpredictable.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Knee Plica
Three synovial plicae are commonly described: suprapatellar, medial shelf, and infrapatellar. The clinical significance of plicae remains debatable. One hypothesis suggests that these synovial remnants have the potential to
undergo an inflammatory process, causing them to become thickened and fibrotic. Repetitive contact between a fibrotic synovial plica and articular cartilage of
the knee can lead to cartilage degeneration.
The patient with a pathologic synovial plica reports
anteromedial knee pain and often reports an episode of
trauma. Other symptoms include swelling, a sense of
subpatellar tightness, and tenderness medial to the patella. Physical findings may include a thickened, palpable, tender cord medial to the patella. A palpable snap
may be elicited with knee flexion.
Initial treatment of a suspected pathologic plica is
nonsurgical, with nonsteroidal anti-inflammatory medication, hamstring and quadriceps stretching and
strengthening, and local modalities. Arthroscopy remains the most reliable method for diagnosis of a
pathologic synovial plica. Successful results with arthroscopic resection and removal are reliable when a thickened plica is the only pathologic finding in a symptomatic knee.

Extensor Mechanism Disruption

Disruption of the extensor mechanism occurs during a
sudden eccentric contraction of the quadriceps muscles
with the foot planted and the knee flexed. Disruption
usually occurs through the musculotendinous junction
of the quadriceps or through a diseased quadriceps or
patellar tendon. Early recognition of this injury is crucial because results of early surgical repair of both
quadriceps and patellar tendon ruptures are more favorable than late repair or reconstruction.
Patients with extensor mechanism disruption usually
present with a history of trauma, pain, swelling, and a
loss of extensor function. Physical examination may reveal a palpable defect above the patella with quadriceps
rupture and below the patella with patellar tendon rupture. A low-lying patella is associated with quadriceps
rupture, whereas a high-riding patella is associated with
patellar tendon rupture. Extensor lag is commonly
noted.
A lateral plane radiograph is a valuable diagnostic
tool for evaluation of extensor mechanism disruption. A
patellar fracture needs to be excluded. Patella alta is associated with patellar tendon disruption, whereas patella baja indicates quadriceps disruption. MRI is also
useful in identifying the location of the disruption if
clinical data are uncertain.
With acute ruptures of the extensor mechanism, direct repair can achieve favorable results. Delay in surgical repair is the factor that most significantly diminishes

American Academy of Orthopaedic Surgeons

Chapter 37 Knee and Leg: Soft-Tissue Trauma

results of surgical treatment of extensor mechanism disruption secondary to the retraction and contracture of
tissue that can occur. Chronic disruption of the extensor
mechanism is often associated with soft-tissue retraction, necessitating tendon lengthening or even allograft
soft-tissue augmentation.

Advances in Knee Arthroscopy

There are many recent advances in arthroscopic instrumentation and techniques including video and display
equipment that use high-definition television cameras
for high-definition video display. Images can now be
projected onto glasses that the surgeon wears that weigh
only 8 oz, thus avoiding the need for large video monitors. New digital capture devices can transfer video sequences from an arthroscopic procedure directly to a
digital video disk or picture-archiving and communications system, which can link them directly to the patients medical records and imaging. The second generation of voice-activated software is now being developed
for improved use of hands-free instrumentation. Newer
types of arthroscopic instrumentation include new forms
of biodegradable implants and improved knotless fixation devices. Research continues with biologic solutions,
such as meniscal healing with growth factors and gene
therapy. Additionally, bioengineering continues to advance with research involving meniscal, ligamentous,
and articular cartilage scaffolds.

Annotated Bibliography
Anterior Cruciate Ligament Injury
Anderson AF, Snyder R, Lipscomb AB, et al: Anterior
cruciate ligament reconstruction: A prospective randomized study of three surgical methods. Am J Sports Med
2001;29:272-279.
A prospective randomized study was done to ascertain the
differences in results of three surgical methods for ACL reconstruction (autogenous bone-patellar tendon-bone graft; semitendinosus and gracilis tendon graft reconstruction with an
extra-articular procedure; semitendinosus and gracilis tendon
graft reconstruction alone). The authors concluded that ACL
reconstruction with a semitendinosus and gracilis patellar tendon autograft may have similar subjective results, but the patellar tendon autograft may provide better long-term stability.

Aune AK, Holm I, Risberg MA, et al: Four-strand hamstring tendon autograft compared with patellar tendonbone autograft for anterior cruciate ligament reconstruction: A randomized study with two-year follow-up.
Am J Sports Med 2001;29:722-728.
In this prospective, randomized study, 72 patients with subacute or chronic rupture of the ACL were assigned at random
to receive autograft reconstruction with either gracilis and
semitendinosus tendon or patellar tendon-bone. Sixty-one patients (32 with hamstring tendon grafts and 29 with patellar

453

Knee and Leg: Soft-Tissue Trauma

tendon grafts) were assessed after 24 months. Anterior knee
pain was not significantly different between the two groups,
but pain with kneeling was present in the group with patellar
tendon reconstructions.

Beynnon BD, Johnson RJ, Fleming BC, et al: Anterior

cruciate ligament replacement: Comparison of bonepatellar tendon-bone grafts with two-strand hamstring
grafts. A prospective, randomized study. J Bone Joint
Surg Am 2002;84:1503-1513.
Results from this study indicated that, after 3-year followup, ACL replacement with a bone-patellar tendon-bone autograft was superior to replacement with a two-strand
semitendinosus-gracilis graft; knee laxity, pivot-shift grade and
strength of the knee flexor muscles were improved.

Ejerhed L, Kartus J, Sernert N, et al: Patellar tendon or

simitendinosus tendon autografts for anterior cruciate
ligament reconstruction: A prospective randomized
study with a two-year follow-up. Am J Sports Med 2003;
31:19-25.
According to this study, the semitendinosus tendon graft
can be considered an equivalent option to the bone-patellar
tendon-bone graft for ACL reconstruction.

Eriksson K, Anderberg P, Hamberg, et al: A comparison

of quadruple semitendinosus and patellar tendon grafts
in reconstruction of the anterior cruciate ligament.
J Bone Joint Surg Br 2001;83:348-354.
In this randomized study of 164 patients with unilateral instability of the ACL, arthroscopic reconstruction with a patellar tendon graft using interference screw fixation or a quadruple semitendinosus graft using endobutton fixation was done.
The authors concluded that medium-term outcomes between
the two methods were similar.

Feller JA, Webster KE: A randomized comparison of

patellar tendon and hamstring tendon anterior cruciate
ligament reconstruction. Am J Sports Med 2003;31:564573.
In this randomized study of 65 patients, patellar tendon
and hamstring tendon autografts both resulted in satisfactory
functional outcomes; morbidity was increased in the patients
who received patellar tendon grafts, and knee laxity and radiographic femoral tunnel widening was increased in the patients
who received hamstring tendon.

Orthopaedic Knowledge Update 8

Peterson RK, Shelton WR, Bomboy AL: Allograft versus autograft patellar tendon anterior cruciate ligament
reconstruction: A 5-year follow-up. Arthroscopy 2001;17:
9-13.
A prospective nonrandomized study to compare the longterm results of allograft versus autograft central one third
bone-patellar tendon-bone reconstruction of the ACL showed
no statistically significant differences in the presence of pain,
giving way, effusion, Lachman and pivot shift results, or arthrometer measurements.

Shaieb MD, Chang SK, Marumoto JM, Richardson AB:

A prospective randomized comparison of patellar tendon versus semitendinosus and gracilis tendon autografts for anterior cruciate ligament reconstruction.
Am J Sports Med 2002;30:214-220.
In this randomized study, 70 patients with patellar tendon
or hamstring tendon autografts for single-incision ACL reconstruction were evaluated at 2-year follow-up. Hamstring tendon autografts performed similarly to patellar tendon grafts;
however, more patients who received patellar tendon grafts
had patellofemoral pain and loss of motion.

Shelbourne KD: Contralateral patellar tendon autograft

for anterior cruciate ligament reconstruction. Instr
Course Lect 2002;51:325-328.
This article reviews the use and benefits of using the contralateral patellar tendon for ACL reconstruction.

Siebold R, Buelow JU, Bos L, Ellermann A: Primary

ACL reconstruction with fresh-frozen patellar versus
Achilles tendon allografts. Arch Orthop Trauma Surg
2003;123:180-185.
The authors retrospectively evaluated the clinical outcome
of 251 fresh-frozen patellar versus Achilles tendon allografts
for primary ACL reconstruction with a mean follow-up of 3
years. The authors concluded that satisfactory clinical results
can be achieved with the use of allografts for primary ACL reconstruction. They indicated that the Achilles tendon-bone allograft seemed to be advantageous over the bone-tendonbone allograft for ACL reconstruction because the failure rate
was significantly lower.

Posterior Cruciate Ligament Injury

Oakes D, Markolf K, McWilliams J, Young C, McAllister
D: A biomechanical comparison of tibial inlay and tibial
tunnel posterior cruciate ligament reconstruction techniques: Analysis of graft forces. J Bone Joint Surg Am
2002;84:938-944.

Jansson KA, Linko E, Sandelin J, et al: A prospective

randomized study of patellar versus hamstring tendon
autografts for anterior cruciate ligament reconstruction.
Am J Sports Med 2003;31:12-18.

This study examines the biomechanics of the tibial tunnel

technique compared with the use of a tibial inlay in reconstruction of the PCL.

In this randomized study of 99 patients (89 of whom were

available for 2-year follow-up), results were similar for patellar and hamstring tendon autografts for ACL reconstruction.

Posterolateral Complex Injury

454

Kanamori A, Lee JM, Haemmerle MJ, Vogrin TM,

Harner SD: A biomechanical analysis of two reconstruc-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

tive approaches to the posterolateral corner of the knee.
Knee Surg Sports Traumatol Arthrosc 2003;11:312-317.
This article reviews a biomechanical study comparing the
effects of biceps tenodesis with those of popliteofibular ligament reconstruction. Ten human cadavers were tested intact
and in the posterolateral corner-condition, deficient, with each
type of reconstruction.The authors concluded that the popliteofibular ligament reconstruction more closely reproduces the
primary function of the posterolateral corner compared with
the biceps tenodesis.

McGuire DA, Wolchok JC: Posterolateral corner reconstruction. Arthroscopy 2003;19:790-793.

This article describes a posterolateral corner reconstruction procedure that uses allograft and interference screw fixation. The authors recommend the use of this procedure in conjunction with PCL reconstruction to restore rotatory and
posterior knee instability in the multiligament-injured knee.

Knee Dislocation
Liow RY, McNicholas MJ, Keating JF, Nutton RW: Ligament repair and reconstruction in traumatic dislocation
of the knee. J Bone Joint Surg Br 2003;85:845-851.
This study compared 8 knee dislocations that were treated
acutely (<2 weeks after injury) and 14 dislocations that were
treated at least 6 months after injury. Both groups of knees
were treated with a combination of repair or reconstruction of
all injured ligaments. Although differences were small, the outcome in terms of overall knee function, activity levels, and anterior tibial translation were better for the patients whose
knees were reconstructed within 2 weeks of injury.

Shelbourne KD, Carr DR: Combined anterior and posterior cruciate and medial collateral ligament injury:
Nonsurgical and delayed surgical treatment. Instr
Course Lect 2003;52:413-418.
This article reviews management principles of the multiple
ligament-injured knee. Four treatment principles are stressed:
(1) medial-side injuries can heal with proper nonsurgical treatment; (2) PCL tears with grade II laxity or less can heal with
similar long-term results as grade I injuries; therefore, surgery
may not be indicated; (3) PCL laxity greater than grade II and
a soft end point should be considered for semiacute reconstruction; and (4) ACL injuries in combination with medial or
PCL injuries can initially be treated nonsurgically and reconstructed at a later date as symptoms dictate.

Chapter 37 Knee and Leg: Soft-Tissue Trauma

and Fairbank changes in 92.9% of the radiographs. The authors noted that the extent of the resection is a significant factor.

McCarty EC, Marx RG, DeHaven KE: Meniscus repair:

Considerations in treatment and update of clinical results. Clin Orthop 2002;402:122-143.
An updated review of meniscus repair techniques and
published results is presented.

Rankin CC, Lintner DM, Noble PC, Paravic V, Greer E:

A biomechanical analysis of meniscal repair techniques.
Am J Sports Med 2002;30:492-497.
Various methods that are used in the repair of meniscal
tears were evaluated in bovine meniscus, including a biodegradable meniscal implant without sutures, a suture anchor
device, and horizontal and vertical mattress sutures. Suture
techniques were stronger at all levels of testing. The ultimate
strength of repair was strongest for the vertical sutures (202
7 N) and lowest for the arrow and suture anchor device (95.9
8 N and 99.4 8 N, respectively).

Shelbourne KD, Carr DR: Meniscal repair compared

with meniscectomy for bucket-handle medial meniscal
tears in anterior cruciate ligament-reconstructed knees.
Am J Sports Med 2003;31:718-723.
This article presents a retrospective review of 155 patients
with isolated bucket handle medial meniscal tears and ACL
tears. Outcomes from meniscal repair were not superior to
those from partial removal. Patients with repaired degenerative tears had significantly lower subjective scores than those
with nondegenerative tears.

Venkatachalam S, Godsiff SP, Harding ML: Review of

the clinical results of arthroscopic meniscal repair. Knee
2001;8:129-133.
A retrospective review of 62 arthroscopic meniscal repairs
is presented. Early repair within 3 months of the tear gave better results (91%) than if performed later (58%). Suture repair
alone yielded better results (78%) than meniscal arrows or a
suture anchor device (56%). Healing rates of atraumatic meniscal tears were much lower than for traumatic tears (42%
versus 73%). The isolated atraumatic medial meniscal tear appeared to heal poorly (33% healing) and was believed to be
better treated by meniscectomy.

Patellofemoral Disorders
Meniscal Injury
Bonneux I, Vandekerckhove B: Arthroscopic partial lateral meniscectomy long-term results in athletes. Acta
Orthop Belg 2002;68:356-361.
A retrospective case-control study of arthroscopic partial
meniscectomy for isolated lesions of the lateral meniscus is
presented. An 8-year follow-up on 31 knees found deterioration of results with decreased Tegner scores (7.2 down to 5.7)

American Academy of Orthopaedic Surgeons

Aderinto J, Cobb AG: Lateral release for patellofemoral

arthritis. Arthroscopy 2002;18:399-403.
A retrospective study of 53 patients who underwent lateral
retinacular release for symptomatic patellofemoral arthritis is
presented. Four patients underwent total knee replacement
within 18 months after the lateral release. Of the remaining
patients, 80% had a reduction in their preoperative pain, 16%
were unchanged, and 4% were worse.

455

Knee and Leg: Soft-Tissue Trauma

Bizzini M, Childs JD, Piva SR, Delitto A: Systematic review of the quality of randomized controlled trials for
patellofemoral pain syndrome. J Orthop Sports Phys
Ther 2003;33:4-20.
Based on the results of trials exhibiting a sufficient level of
quality, treatments that were effective in reducing pain and
improving function in patients with patellofemoral pain syndrome included acupuncture, quadriceps strengthening, resistive bracing, and a combination of exercises with patellar taping and biofeedback.

Schneider F, Labs K, Wagner S: Chronic patellofemoral

pain syndrome: Alternatives for cases of therapy resistance. Knee Surg Sports Traumatol Arthrosc 2001;9:290295.
The results of a prospective randomized study comparing
20 patients treated with proprioceptive neuromuscular facilitation and 20 patients treated with a special training program
using a resistance-controlled knee splint are presented. The
knee splint proved more effective than proprioceptive neuromuscular facilitation for treating patellofemoral pain syndrome resistant to conservative therapy.

Knee Plica
Irha E, Vrdoljak J: Medial synovial plica syndrome of
the knee: A diagnostic pitfall in adolescent athletes.
J Pediatr Orthop B 2003;12:44-48.

Orthopaedic Knowledge Update 8

study with a long-term followup period. Clin Orthop
1991;264:255-263.
Arnoczky SP, Warren RF: Microvasculature of the human meniscus. Am J Sports Med 1982;10:90-95.
Burks RT, Metcalf MH, Metcalf RW: Fifteen year
follow-up of arthroscopic partial meniscectomy. Arthroscopy 1997;13:673-679.
Burks RT, Schaffer JJ: A simplified approach to the tibial attachment of the posterior cruciate ligament. Clin
Orthop 1990;254:216-219.
Daniel DM, Stone ML, Dobson BE, Fithian DC, Rossman DJ, Kaufman KR: Fate of the ACL-injured patient:
A prospective outcome study. Am J Sports Med 1994;22:
632-644.
Fairbank TJ: Knee joint changes after meniscectomy.
J Bone Joint Surg Br 1948;30:664-670.
Fulkerson JP: Anteromedialization of the tibial tuberosity for patellofemoral malalignment. Clin Orthop 1983;
177:176-181.
Fulkerson JP: Patellofemoral pain disorders: Evaluation
and management. J Am Acad Orthop Surg 1994;2:124132.

A prospective study of clinical criteria for the diagnosis of

medial plica syndrome is presented. Two physical examination
tests that improve the accuracy of the clinical examination for
diagnosis of medial plica syndrome are described.

Indelicato PA: Non-operative treatment of complete

tears of the medial collateral ligament of the knee.
J Bone Joint Surg Am 1983;65:323-329.

Extensor Mechanism Disruption

Maenpaa H, Lehto MU: Patellar dislocation: The longterm results of nonoperative management in 100 patients. Am J Sports Med 1997;25:213-217.

Richards DP, Barber FA: Repair of quadriceps tendon

ruptures using suture anchors. Arthroscopy 2002;18:556559.
The authors state that this is the first report in the English
language literature of a technique using suture anchors to attach the quadriceps tendon to bone.

Classic Bibliography
Ahmad CS, Kwak D, Ateshian GA, Warden WH, Steadman JR, Mow VC: Effects of patellar tendon adhesion
to the anterior tibia on knee mechanics. Am J Sports
Med 1998;26:715-724.
Ahmed AM, Burke DL: In-vitro measurement of static
pressure distribution in synovial joints: Part I. Tibial surface of the knee. J Biomech Eng 1983;105:216-225.
Andersson C, Odensten M, Gillquist J: Knee function
after surgical or nonsurgical treatment of the acute rupture of the anterior cruciate ligament: A randomized

456

ONeill DB: Arthroscopically assisted reconstruction of

the anterior cruciate ligament: A prospective randomized analysis of three techniques. J Bone Joint Surg Am
1996;78:803-813.
Seebacher J, Inglis A, Marshall J, Warren R: The structure of the posterolateral aspect of the knee. J Bone
Joint Surg Am 1982;64:536-541.
Shelbourne KD, Davis TJ, Patel DV: The natural history
of acute, isolated, nonoperatively treated posterior cruciate ligament injuries: A prospective study. Am J Sports
Med 1999;27:276-283.
Tenuta JJ, Arciero RA: Arthroscopic evaluation of meniscal repairs: Factors that effect healing. Am J Sports
Med 1994;22:797-802.
Veltri DM, Warren RF: Posterolateral instability of the
knee. Instr Course Lect 1995;44:441-453.

American Academy of Orthopaedic Surgeons

Chapter

38

Knee Reconstruction and

Replacement
Mark I. Froimson, MD

Clinical Evaluation
Patients presenting for orthopaedic care of the knee
complain primarily of pain and functional decline,
marked by difficulty in walking, climbing stairs, and arising from a seated position. In addition, deformity and
instability may be contributing symptoms that influence
the choice of treatment. A detailed history is essential in
determining the impact the knee symptoms have had on
quality of life, and should specifically detail the level of
impairment. It has been widely recognized that a major
source of failure of knee surgery is the inability to live
up to unreasonable patient expectations. As a result, it is
important for the surgeon to document with specificity
the presenting symptoms and objective measures of
knee function and performance. Tools that allow for
clarification of such patient expectations can help direct
care and can provide a baseline against which postintervention outcomes can be compared.
General health assessment questionnaires, including
the Western Ontario and McMaster Universities Osteoarthritis Index and the Short Form 36, are being used with
increasing frequency to evaluate the impact of knee arthritis and subsequent treatment. These validated measures provide information on the effect of knee procedures on the patients general sense of well-being and
have been shown to offer consistent correlation with clinical outcome measures.

Radiographic Evaluation
Radiographs, including AP, 45 weight-bearing, lateral,
and Merchant views, remain the essential diagnostic modality for evaluating the painful knee. Three-joint standing films define the mechanical and anatomic axis of the
limb and can assist in surgical planning. Presenting radiographs are often inadequate, and although repeating
films can be inefficient and costly, using unsatisfactory
imaging studies to plan treatment carries far greater
risk. Supine images of the knee, as are routinely obtained in a primary care or emergent setting, can severely underestimate the degree of joint space narrowing and thus the associated cartilage loss.

American Academy of Orthopaedic Surgeons

A second and equally common mistake resulting

from poor-quality diagnostic radiographs is the overutilization of more sophisticated and expensive studies.
MRI is a sensitive tool in the detection of pathology in
the diseased knee, but is not specific for articular cartilage abnormalities, unless certain sequences are requested. Both T1-weighted, fat-suppressed threedimensional, spoiled gradient-echo technique and T2weighted fast spin-echo technique are essential to
enhance accuracy of detection of articular cartilage lesions. Cartilage has higher signal intensity than fluid on
T1 images and a lower intensity on T2 images. Meniscal
abnormalities that are readily diagnosed on MRI can be
thought to represent the dominant pathology, prompting
referral for arthroscopic intervention.

Nonsurgical Care
Many patients with symptomatic knee arthritis will respond to a period of nonsurgical treatment, and most
patients expect a major surgical intervention as a final
option reserved appropriately for disease and symptoms
that are unremitting despite judicious medical management. Although ambulatory aides are often deemed an
unacceptable solution for arthritic symptoms, a properly
used cane can provide significant functional improvement and pain relief by resting and unloading the joint.
A cane can offer a temporary respite from the pain associated with ambulation and can allow for the initiation of other modalities.
The use of braces is also common, but there are few
supporting studies to document their efficacy. In addition to the placebo impact of brace wear, nonsupportive
sleeves have been thought to provide relative joint unloading by acting as a containment device for the soft
tissues, thereby providing a type of hydraulic support.

Joint-Preserving Surgical Procedures

Arthroscopy
The role of arthroscopic surgery in the treatment of
knee arthritis has been the subject of considerable controversy. Knee arthroscopy is one of the most commonly

457

Orthopaedic Knowledge Update 8

Knee Reconstruction and Replacement

of subsequent knee replacement has been assumed to be

negligible, but this has not been adequately studied.

Osteotomy

Figure 1 Opening wedge valgus osteotomy of the tibia with plate fixation.
performed procedures for the diseased knee because of
its ease of application and proven efficacy. Its success in
the treatment of the arthritic knee is directly proportional to the degree of mechanical symptoms present
preoperatively and inversely proportional to the severity of the underlying arthritis. Although the underlying
disease process of cartilage degradation has not been
shown to be favorably impacted by arthroscopic intervention, the symptoms associated with the secondary,
mechanically significant lesions such as loose bodies,
meniscal tears, and unstable cartilaginous flaps can be
successfully addressed. Arthroscopic treatment of the
arthritic knee is less likely to be effective in the presence of malalignment that causes overloading of the
most diseased portion of the knee.
A recent and widely discussed randomized comparison of knee arthroscopy with lavage and sham surgery
failed to demonstrate superiority of arthroscopy over
sham surgery in a population of men with advanced arthritis. A separate review of arthroscopic dbridement
performed in a large group of patients older than age 50
years with a variety of diagnoses found that within 3 years,
18% of patients needed total knee arthroplasty (TKA),
suggesting overutilization of the index procedure. The effect of previous arthroscopic knee surgery on the results

458

Malalignment resulting in focal overload of either the

medial or lateral compartment of the knee can be successfully managed with osteotomy to restore the mechanical axis of the knee. This strategy has the distinct
advantage of joint preservation that is particularly attractive for younger, more active patients. Osteotomy is
most likely to succeed when the disease affects predominantly one compartment, and there is therefore healthy
tissue onto which to redirect the load. Despite this observation, no advantage has been demonstrated with arthroscopic evaluation of the knee done at the same setting as osteotomy, except to treat mechanically
significant lesions. To maintain a joint line perpendicular
to the weight-bearing axis, and to avoid joint line obliquity, varus malalignment is most commonly corrected
through proximal tibial osteotomy, whereas valgus malalignment is most commonly corrected through supracondylar femoral osteotomy. When the degree of valgus
deformity is not severe, and has resulted from proximal
tibial deformity such as may occur following tibial plateau fracture, opening wedge varus-producing tibial osteotomy has yielded acceptable results.
Valgus-Producing Tibial Osteotomy
High tibial osteotomy requires careful patient selection
and exacting surgical technique. The ideal patient is
younger than age 50 years and active, with high functional demands likely to place undesirable stress on an
arthroplasty. The procedure works best when performed
early enough that symptoms and cartilage loss predominate on the medial side and there is no suggestion of involvement of either of the other two compartments, either clinically or radiographically. Contraindications
generally include inflammatory arthritis, poor flexion
(< 90), flexion contracture, ligament instability, and tricompartmental arthritis. The procedure is less successful
in smokers and in patients age 60 years or older, when
the degree of deformity increases beyond 10 and when
the involvement of the remaining compartments increases.
Valgus correction can be achieved by medial lengthening rather than lateral shortening, using either an
opening wedge osteotomy and interposition plating or a
medial external fixator with gradual distraction. The advantages of correction on the medial side include more
anatomic restoration with resultant improvement in ligament stability, and the ability to more finely tune the
correction. The main disadvantage of this technique is
the risk of nonunion and loss of correction (Figure 1).

American Academy of Orthopaedic Surgeons

Chapter 38 Knee Reconstruction and Replacement

Orthopaedic Knowledge Update 8

Figure 2 Distal femoral varus osteotomy (A), with medial closing wedge and plate fixation (B).

Varus-Producing Femoral Osteotomy

Valgus deformity may result in isolated lateral compartment disease and symptoms. Younger, active patients
with preserved range of motion and no evidence of disease in the remaining compartments are candidates for
realignment through varus femoral osteotomy. Ideally,
deformity should be less than 15, without flexion contracture or inflammatory disease. Correction to physiologic valgus (4 to 6) is obtained through removal of a
medial wedge, verified radiographically, and the fragments are rigidly fixed with a blade plate, applied medially (Figure 2).

Total Knee Arthroplasty

Replacement of the knee joint should be viewed as an
elective procedure that is indicated when there is evidence of advanced disease resulting in failure of the
joint to function satisfactorily. The three key elements
considered in the decision to proceed with TKA include
debilitating symptoms, failure of such symptoms to respond to less invasive treatment, and medical suitability
of the patient to respond to surgery.

Patient and Disease Variables Affecting Outcome

TKA is a very successful procedure, with survivorship
exceeding 90% at 10 years, 80% at 15 years, and 75% at
20 years. The probability of the implant surviving at
these follow-up intervals can be influenced by patientrelated factors including age, gender, and primary diagnosis, and by prosthetic design. Survivorship is adversely

American Academy of Orthopaedic Surgeons

impacted in patients younger than 55 years of age, in

males, and when the primary diagnosis is osteoarthritis.
Patient age of 70 years or older, rheumatoid arthritis,
and cemented fixation are factors that increase the longterm survivability of the implant. These factors likely
have this effect because of the decreased demand on the
implant, rather than specific physiologic factors. In fact,
when the impact of specific underlying diseases is examined, unique considerations become apparent.
Obesity
Obesity confers both increased surgical complexity, with
risk of improper implantation, and greater demand on
the implant. Malalignment in the presence of obesity is
not well tolerated, but is more likely caused by difficulties of exposure. Well-aligned, well-fixed implants fare
as well in the heavy patient as in the general population.
Wound complications are more common and can be
minimized with careful surgical technique.
Juvenile Rheumatoid Arthritis
Juvenile rheumatoid arthritis can result in severe joint
destruction and the need for reconstructive surgery at a
very young age. The level of preoperative function in
this group of patients is poor and is generally compromised by coexisting disease of adjacent joints. Altered
immune function, along with the impact of diseasemodifying agents, result in relatively high rates of infection. Additionally, stiffness remains a problem postoperatively, leading to reoperation in 8% in one study.

459

Orthopaedic Knowledge Update 8

Knee Reconstruction and Replacement

Hemophilic Arthropathy
Hemophilic arthropathy develops as a consequence of repeat hemarthrosis secondary to coagulopathy, and most
commonly affects the knee. Loss of function and pain are
severe, and the advent of adequate management of clotting makes TKA a viable option. These patients are usually relatively young and may have associated immunosuppression caused by human immunodeficiency virus
(HIV). Surgery is complicated by high rates of infection,
independent of the HIV status of the patient. As a result,
the likelihood of survival of implants in these patients is
reduced, with at least 10% failing within 5 years. Stiffness
and limited functional recovery are also seen with greater
frequency, further compromising outcomes.
Osteonecrosis
Osteonecrosis of the knee may occur in younger patients secondary to corticosteroid or alcohol use or in
elderly patients as a spontaneous occurrence. Although
nonsurgical treatment may be attempted, the majority
of patients with this condition ultimately experience
joint failure requiring knee arthroplasty. The compromised vascularity of the bone and the potential for significant involvement of large subchondral regions impacts the surgical approach to these patients.
Preoperative MRI can assist in determining the amount
of periarticular bone involved with disease.
Patellofemoral Arthritis
Isolated patellofemoral arthritis that is recalcitrant to
treatment can be successfully managed with TKA in
older patients. Functional results are superior to patellectomy or patellofemoral arthroplasty, and are equivalent to TKA done for tricompartmental arthritis. Lateral
release is commonly required when preoperative patellar tilt is present. For younger patients with patellofemoral arthritis, nonarthroplasty treatment options may be
considered.

Impact of Prior Surgery on Subsequent TKA

TKA may be used to treat posttraumatic arthritis that develops following fixation of tibial plateau fractures.
Unique considerations include retained hardware, osseous defects, previous incisions, and a compromised softtissue envelope and result in higher complication rates,
higher revision rates, and less satisfactory outcomes than
primary knee replacements. Previous incisions should be
incorporated whenever possible, or the standard incision
should be adjusted to maintain an optimal skin bridge.
When the ability of the skin to heal is in question, preoperative skin incisions can be useful to confirm healing potential. Hardware can be removed at the time of surgery,
but excessive soft-tissue undermining should be avoided
to prevent skin necrosis. If hardware is extensive and removal will devascularize the soft tissues, removal may be

460

staged as a separate procedure.Adjunctive measures such

as bone grafting and lateral release are commonly required, and constrained prostheses may be necessary because of impairment of the soft-tissue envelope. Infection
is a common complication, and may be heralded by persistent postoperative drainage. Although satisfactory results can be obtained, and pain and function can be substantially improved, patients should be counseled on the
compromised outcome expected when TKA arthroplasty
is performed following previous fracture fixation.
Prior osteotomy, both high tibial and distal femoral,
also can compromise the outcome of a subsequent knee
replacement. Osteotomy may result in bone defects or
deformity, necessitating bone grafts and/or offset stems.
Patella baja is common following tibial osteotomy and
can lead to increased tension on the tendon insertion
during exposure.
TKA after femoral osteotomy may result in inferior
outcomes when compared with primary knee replacement. Ligament instability may occur related to the
intra-articular correction of extra-articular deformity
and may require use of components with increased constraint. Relative postoperative varus of the femoral
component may occur as a result of prior deformity, but
its incidence can be reduced with the use of extramedullary alignment verification.
Conversion of a fused knee is a controversial indication for knee arthroplasty and if undertaken can present
unique challenges. Although a hinged or constrained
prosthesis has typically been recommended, appropriate
preservation of the soft-tissue sleeve can allow for successful reconstruction with a standard posterior stabilized design in some patients.

Surgical Technique
Optimal success of TKA can be obtained by accurate
restoration of the mechanical axis, good fit and fixation
of the implant to host bone, and careful attention to
soft-tissue balance. Modern knee systems provide instrumentation that allows for reproducible approaches
to prosthesis implantation. Both intramedullary and extramedullary alignment systems have been shown to be
accurate, and bone preparation has been facilitated
through the use of precise finishing guides that are well
fixed to bone and incorporate multiple cuts in one step.
Given these tools, attention to soft-tissue balancing has
received increased attention as the sometimes overlooked yet essential component of knee replacement
success.
The well-functioning knee must be balanced with
equal tibiofemoral space in both flexion and extension,
producing essentially equal tension in the medial and
lateral soft-tissue envelopes following reconstruction.
Knees that are too tight exhibit unsatisfactory stiffness,
manifested by flexion contracture and/or decreased flex-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

ion, depending on whether it is too tight in extension
and/or flexion. Excess release can produce functional
impairment as a result of instability. Instrumented measures and computer-assisted surgery offer the promise
of providing objective data on ligament balance against
which subjective surgical experience can be quantified,
but to date, such tools have not been validated.
Femoral rotation plays an important role in determining soft-tissue tension and ligament balance in flexion, and has received considerable attention. Because
the tibial articular surface is sloped into approximately
3 of varus, an asymmetric resection of the posterior
condyles is required to obtain medial and lateral balance of the flexion space at 90, if the tibia is cut perpendicular to its long axis. Several anatomic references
have been proposed to establish the optimum amount
of femoral external rotation. Femoral rotation can be
determined from one of three femoral axes, or it can be
oriented parallel to the tibial cut surface following appropriate soft-tissue releases. The first approach allows
femoral preparation to proceed before soft-tissue balancing, whereas the latter approach requires initial tibial
resection and soft-tissue balancing. The three femoral
referencing axes include a line 3 to 5 externally rotated relative to the posterior femoral condyles, the epicondylar axis, and a line perpendicular to the notch and
trochlear axis (Figure 3).

Design Issues
Fixation
TKA has been performed successfully using methylmethacrylate cement for fixation and biologic fixation
of the implant to host bone. Whereas cemented fixation
can be achieved in nearly all bone types and patient
profiles, bone ingrowth around the knee is less predictable. Thus, although loosening can occur with either
mode of fixation, early loosening, caused by failure of
bone ingrowth, is more common with cementless fixation. Strategies to achieve cementless fixation in the
tibia include the use of porous surfaces augmented with
pegs, stems, or screws. Micromotion at the screw/
baseplate interface has been implicated as a source of
particulate debris, with the screw holes serving as pathways for particle migration. Although successful longterm results have been documented for cementless arthroplasty, higher rates of failure resulting from
osteolysis and loosening have been reported. Hybrid fixation, in which the femoral component is inserted without cement and the tibia and patella are cemented, has
been proposed as a compromise, but results with this
technique have not been consistent. Similarly, cementing
only the metaphyseal surface of the tibial component
and press fitting the stem or keel has been associated
with a higher rate of early implant loosening than full
cementation of the tibial component.

American Academy of Orthopaedic Surgeons

Chapter 38 Knee Reconstruction and Replacement

Figure 3 References for femoral rotation: AP axis (a), epicondylar axis (b), posterior
condylar axis (c), and tibial cut surface (d).

Posterior Cruciate Ligament

Total knee components have been successfully used that
allow for posterior cruciate ligament (PCL) retention,
sacrifice, or substitution. Each strategy carries certain
advantages and challenges, and there is no consensus regarding which approach provides superior outcomes.
Long-term success has been reported with each design.
Proponents of PCL retention describe the potential
for more physiologic femoral rollback, accurate joint
line restoration, bone preservation, and the proprioceptive role of the ligament as distinct advantages. Because
the PCL may play a role in the deformity, some studies
advocate limiting its preservation to those knees with
minimal angular malalignment and flexion contracture.
Balancing of the ligament is advocated, with appropriate recession recommended when the trial components
exhibit anterior tibial liftoff (booking open) in flexion.
If the PCL is left too tight, it has been shown to result in
posterior femoral subluxation, as well as asymmetric
posterior polyethylene wear and associated osteolysis.
Evidence from fluoroscopic kinematic studies suggests
that the PCL, when retained, even in the absence of
clinical problems, likely does not function physiologically, and paradoxical anterior femoral sliding rather
than posterior rollback can occur in some patients. At
the other extreme, overrecessing of the tight PCL can
result in late failure with subsequent flexion instability.
This latter syndrome is a recently recognized source of
TKA failure in patients with well-fixed implants and is
characterized by instability, effusion, and pain during

461

Knee Reconstruction and Replacement

Orthopaedic Knowledge Update 8

fered include greater intraoperative flexibility and the
potential for simple revision of a worn polyethylene
bearing surface without concomitant need to address a
well-fixed component. This advance in design has been
associated with new problems. Motion between the
metal base plate and the polyethylene was an unanticipated consequence of this concept and has been shown
to produce polyethylene wear debris associated with osteolysis. Attention has been directed to the role of backside wear, with an attendant redesign of the locking
mechanisms. Some designers, recognizing the inability to
completely eliminate micromotion, advocate polishing
of the tibial base plate to minimize the effects of such
motion.

Figure 4 An unresurfaced patella articulating with anatomic femur is shown. The patient is asymptomatic at 10-year follow-up.
weight bearing on the flexed knee. Diagnosis is apparent through clinical examination with demonstrated laxity at 90 of flexion in the unloaded knee.
When the PCL is sacrificed, its function can be substituted for by a cam and post mechanism or by increasing the anterior lip of a conforming tibial polyethylene.
Both designs counteract the posterior tibial subluxation
resulting from sacrifice of the PCL. Better range of motion has been postulated to occur with the cam and post
mechanism because of enforced femoral rollback. These
prostheses carry the unique risk of dislocation, a rare
complication resulting from collateral ligament laxity, allowing the femoral cam to jump anteriorly over the
post. The subluxation height is the amount of laxity required to allow for such clearance. Deep dish tibial components can possess subluxation heights equal to cam
and post mechanisms, and because the mechanism is far
anterior, the likelihood of dislocation is less.
The cam and post mechanism has been identified recently as an articulation with the potential to produce
polyethylene wear debris. One retrieval analysis identified evidence of adhesive and abrasive wear, as well as
fatigue, in a wide spectrum of implant designs from several different manufacturers. Kinematic analysis has revealed that anterior impingement of the polyethylene
against the femoral component can occur with several
designs when the knee hyperextends. When the femoral
component is inserted in relative flexion, or there is increased posterior tibial slope, this phenomenon is accentuated. Aseptic loosening and osteolysis have been correlated with post wear and damage and underscore the
importance of the design and proper implantation of
this type of knee replacement.
Modularity
Modularity has been introduced as a standard design
feature of metal-backed tibial components in most total
knee systems in current use. The main advantages of-

462

Mobile Bearing Design

An alternative approach is to design the undersurface of
the polyethylene insert as an articulating surface, with
macromotion expected. By allowing these mobile bearings to rotate, increased articular conformity can be
achieved throughout the arc of motion. The theoretical
advantage of this design has yet to be demonstrated at
long-term follow-up, with equivalent survivorship expected with mobile and fixed bearing designs. Wear and
osteolysis have been reported to occur with mobile
bearing designs, demonstrating that this phenomenon is
clearly multifactorial. As with fixed bearing designs, results with mobile bearing implants are durable well into
the second decade. Unique problems include bearing
fracture and dislocation, complications that are minimized with appropriate attention to soft-tissue balancing, alignment, and kinematics.

Patellar Resurfacing
Although patellar resurfacing is considered an integral
component of TKA for the majority of North American
surgeons, the procedure remains controversial and has
been the topic of considerable study. Patellar complications remain one of the most common sources of problems after total knee replacement, prompting some surgeons to advocate avoiding this potential by leaving the
host patella unresurfaced. Some studies have shown a
higher prevalence of anterior knee pain in patients with
unresurfaced patellae, whereas other well-designed
studies have failed to identify statistically significant differences between the two groups. Certainly, many patients with a native patella articulating with an anatomically designed femoral component will achieve an
excellent result (Figure 4). Revision rates have been
shown to be either equivalent or higher following knees
without patellar resurfacing, although results following
such reoperations can vary. Patients with anterior knee
pain caused by an unresurfaced patella fare well with
secondary resurfacing. Serious complications with a significant adverse impact on the ultimate reconstruction,

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 38 Knee Reconstruction and Replacement

Figure 5 A, Preoperative AP radiograph demonstrating medial compartment arthritis. B, Postoperative AP radiograph at 3-year follow-up. C, Postoperative lateral radiograph at
3-year follow-up.
such as patella fracture or component loosening, are
more common following patella resurfacing and are difficult to treat, often resulting in inferior outcomes. Thus,
the consensus has emerged that knees without patellar
resurfacing are at a somewhat increased risk for anterior knee pain, but are at a decreased risk for serious
patellar complications.

Unicompartmental Arthroplasty
Unicompartmental arthroplasty is an alternative to
TKA or osteotomy when the arthritis predominantly affects one compartment of the knee (Figure 5). In such
instances, it is possible to resurface the diseased compartment and restore knee alignment that allows for
load to be shared between the replaced and unreplaced
compartments. Although this procedure lost favor after
evidence of inferior survivorship data of several early
series, newer techniques and patient demand has driven
a resurgence of interest, and survivorship of greater
than 90% at 10 years has been documented.
Patient selection and surgical technique are essential
elements of a successful outcome. Pain that is well localized to the compartment exhibiting disease responds
better to treatment than diffuse or global pain, and although the status of the patellofemoral joint has not
been consistently correlated with success, the presence
of pain in the lateral or patellofemoral joint preoperatively is a predictor of persistent pain after surgery. Contraindications include inflammatory arthritis, severe
fixed deformity, previous opposite compartment meniscectomy, and tricompartmental arthritis. Correction of
deformity must allow appropriate load transfer to prevent premature failure, but overcorrection adversely impacts the retained compartment, and also can lead to
early failure. Recommended correction of the varus
knee has ranged from 1 to 5 of postoperative valgus.

American Academy of Orthopaedic Surgeons

Advantages of unicompartmental arthroplasty may

include quicker recovery, fewer short-term complications, and better functional outcome. There is a high rate
of short- to mid-term satisfaction, but long-term survivorship has not been comparable to TKA when measured by revision rates.
Failure of a unicompartmental arthroplasty may occur as a result of implant wear, loosening or subsidence,
or progression of symptomatic arthritis in the lateral or
patellar compartments. Different failure mechanisms
may predominate depending on design, with fixed bearing designs exhibiting more component failure and mobile bearing designs failing because of disease progression. Overcorrection may lead to deterioration of the
compartment to which load is redirected, whereas undercorrection risks overloading the implant. Patellar impingement against the femoral implant can be symptomatic and lead to revision. Patellar arthritis generally is
well tolerated in unicompartmental arthroplasty. Revision of a unicompartmental arthroplasty has been
shown to be less complex than revision TKA. Nevertheless, defects in the replaced compartment may necessitate the use of augments, bone grafting, and/or stems.
Newer instrumentation has increased the appeal of
the unicompartmental arthroplasty by allowing for small
incisions with attendant lower patient morbidity, including less pain and shorter hospital stays. Care must be
taken when using a minimally invasive approach to be
certain that final implant fixation and alignment, keys to
longevity of the construct, are not compromised. The
long-term results of these techniques are not yet available, and common sense suggests that the incision size
should not be the dominant outcome measure of this
technique. Similarly, as this procedure gains popular appeal, its application to younger patients awaits longterm follow-up of success in this population.

463

Knee Reconstruction and Replacement

Orthopaedic Knowledge Update 8

Image Guidance

Medial Collateral Ligament Injury

The desire to increase the accuracy of prostheses insertion has led to emerging efforts to use guidance systems
to monitor and aid in implantation. Few preliminary
data are available on these techniques and no long-term
clinical data yet supports the widespread use of such
technology.

Unanticipated intraoperative disruption of the medial

collateral ligament during routine primary TKA has
been typically treated with conversion to a prosthesis
that provides varus/valgus restraint. Repair or reattachment has been shown to be an equally viable alternative, with a key advantage being the ability to continue
with the planned primary knee prosthesis. Following repair, patients wear braces for 6 weeks, but are allowed
full range of motion. Knee scores and range of motion
at follow-up are equivalent to those of knees without
this complication, and revision has not been required.

Complications
Infection
Infection is a devastating problem that is best prevented. Attention to careful surgical technique and softtissue handling minimizes would healing problems. Laminar air flow and prophylactic antibiotics have been
shown to reduce infection. The impact of dedicated surgical teams and reduced surgical time, while intuitive,
are not documented. Immunosuppression, diabetes,
smoking, prior surgery, and obesity are known risk factors. Antibiotic-impregnated cement has been shown to
lower the incidence of infection and may be considered
for high-risk patients. Persistent postoperative drainage
is worrisome and should be treated aggressively.

Thromboembolic Disease
In the absence of effective prevention, thromboembolic
disease will occur with great frequency following TKA,
with historical data suggesting rates as high as 50%. Despite consensus that some form of prophylaxis is recommended in the perioperative management of these patients, controversy remains regarding the optimal
prophylaxis regimen. Coumadin and low molecular
weight heparin are two agents commonly used to reduce
the incidence of thromboembolic disease, and although
low molecular weight heparin therapy has been associated with lower rates of venographically documented
deep venous thrombosis, (DVT), enthusiasm for its use
has been tempered by the higher associated hemorrhagic complications. Early initiation of low molecular
weight heparins in the postsurgical period corresponds
to reduced rates of DVT but also with higher rates of
bleeding. Use of aspirin as a preventive strategy is controversial, with advocates citing no difference in the occurrence of fatal pulmonary embolism, the most serious
thromboembolic complication, as rationale for its use.
The risk of DVT following TKA is highest in the early
postoperative period, suggesting that prolonged prophylaxis is likely unnecessary. Physical modalities including
compression stockings, pneumatic compression devices,
continuous passive motion machines, and early mobilization are useful adjuncts in the prevention of occurrence, but have not been proven to substitute for pharmacologic prophylaxis. Routine monitoring for clinically
silent disease is a widespread practice but has not been
shown to be beneficial.

464

Extensor Mechanism Failure

Rupture of the patellar tendon is one of the most compromising complications that can occur following knee
arthroplasty. Nonsurgical treatment and primary surgical reconstruction have failed to provide a satisfactory
solution and are likely to result in significant compromise in functional outcome. Reconstruction with Achilles tendon or extensor mechanism allograft is technically demanding with the potential to salvage this
problematic complication. Fresh frozen graft is preferred because of an improved ability of both bone and
tendon to incorporate. Secure fixation is essential, with
emphasis on a sturdy bone graft, keyholed into a wellprepared tibial bed, augmented by cortical screw
fixation. Nonabsorbable suture fixation of the allograft
tendon onto a broad base of healthy quadriceps musculature allows reliable incorporation. This reconstructive
technique reliably reduces extensor lag and improves
active range of motion at short-term follow-up.
Fracture of the patella may occur as a result of compromised circulation, overaggressive resection, maltracking or overt trauma, and may present as an insidious finding or with a sudden change in knee function. The
prevalence of this complication is less than 1%, but its occurrence can significantly diminish the function of the
knee. It is important to assess three aspects of the injury:
whether the patellar component is loose, whether the extensor mechanism is partially or completely disrupted,
and the extent of remaining patellar bone. Nonsurgical
treatment is successful when the patellar component remains well fixed and the extensor mechanism is not completely dysfunctional. Patients with displacement of the
fracture should be carefully assessed for remaining extensor mechanism function, which can be surprisingly satisfactory. A period of nonsurgical care may be preferred,
with the possibility of later reconstruction if necessary,
given the high incidence of failure and complications with
surgical treatment. Surgical treatment is reserved for
those patients exhibiting marked extensor mechanism disruption or gross patellar loosening, and generally results
in compromised function and high revision rates.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Arthrofibrosis
The stiff total knee is a common source of failure and
remains an unsolved problem. The best predictor of
postoperative range of motion is preoperative motion.
Failure to achieve at least 90 of preoperative motion
compromises patient satisfaction. When arthrofibrosis is
suspected early, it can be managed with manipulation
under epidural anesthesia and aggressive physical therapy. Late treatment of stiffness is less likely to respond
to manipulation with increased risk of periprosthetic
fracture. Surgical correction often is desired by the unhappy patient, but is unpredictable. Correction of preoperatively identified malalignment, improper positioning, or incorrect component sizing may be successful in
selected patients. Lysis of scar tissue, combined with exchange and reduction in polyethylene thickness, although appealing because of its apparent simplicity, has
an unacceptably high rate of failure.

Periprosthetic Fracture
Periprosthetic fracture following TKA can occur following minor or substantial trauma and presents a challenge to restoration of knee function. Although the
prevalence is low, occurring in less than 2% of patients,
treatment of this event carries a high rate of complications. Common risk factors include conditions that create osteoporosis, stress shielding, femoral notching, osteonecrosis, and wear-related osteolysis. Treatment is
directed at maintaining alignment and fracture stability,
with early range of motion essential to preventing stiffness. In high-risk patients, nonsurgical treatment may be
chosen even when immobilization is predicted to result
in poor motion or malalignment. Key factors in surgical
decision making include fracture displacement, stability
of the prosthesis, and quality of the bone.
Failed prostheses demonstrating implant loosening
accompanying or predating the fracture necessitates revision arthroplasty. In this setting, associated bone loss
may necessitate bulk allograft reconstruction. The implant is cemented to the allograft, with diaphyseal fixation to host bone achieved using long stems. Collateral
ligaments are preserved with bone fragments and then
fixed to the allograft, but laxity usually necessitates articular constraint.
Displaced fractures associated with well-fixed implants are best treated with reduction and fixation. If
the intercondylar notch is open, retrograde intramedullary nailing through a transarticular approach allows fixation without periosteal devascularization. Techniques
using fixed angle devices and locked screws are evolving
and are effective for treatment of many fractures. Flexible intramedullary nails introduced both medially and
laterally offer a less invasive but also a less rigid method
of achieving fixation. The key to success under these dif-

American Academy of Orthopaedic Surgeons

Chapter 38 Knee Reconstruction and Replacement

ficult circumstances is to provide rigid fixation and early
motion while minimizing surgical complications.

Revision Total Knee Arthroplasty

Evaluation of Pain
Successful revision of a painful, failed TKA is dependent
on an accurate evaluation of the cause of failure. Revision
of the painful total knee without specific identification of
the source of pain is less likely to result in a pain-free reconstruction and is not recommended. Systematic evaluation of such patients assists in the identification of both
intrinsic (knee-related) and extrinsic sources of pain. Failure of the index knee replacement to provide initial pain
relief or to alter the character of the pain indicates a second, unrecognized source of knee pain. Despite the widely
taught precept that knee pain can originate outside of the
knee, patients continue to present to tertiary referral centers with pain in a recently replaced knee whose origin is
a diseased hip or lumbar spine. Protocols to evaluate the
painful TKA, therefore, generally begin with clinical and
radiographic extra-articular assessment. Pain well distal or
proximal to the knee suggests referred or radicular involvement.A benign knee examination further raises concern about unrecognized pathology. At times, diagnostic
injection of a remote location will be required to show a
patient that the problematic knee is not the primary
source of pain.
Aspiration is an important component in the evaluation of the suspect knee and should be routinely obtained. When a well-obtained culture is positive it is
highly predictive, but failure to recover an organism
preoperatively does not exclude infection and may result from suppression because of partial treatment or
sampling error. Cell count of the obtained synovial fluid
is useful. One recent study suggests that the threshold
for a confirmatory result should be reduced from the
traditional level of 25,000 white blood cells to a lower
value of 2,500 for chronic low-grade prosthetic infection. Repeat aspiration can increase accuracy in problematic cases.

Preoperative Planning
Failure of TKA results in rapid acceleration of symptoms, functional decline, and the need for revision arthroplasty. Several factors, including lower levels of general patient health, decreased soft-tissue integrity, and
bone loss encountered during revision arthroplasty, contribute to the increased challenge of obtaining a successful outcome in this setting. Revision knee systems offer
a wide array of reconstructive options to restore mechanical integrity to the knee. Despite these advances,
complications following revision surgery are much
higher than following primary surgery, approaching
25%. Infection, extensor mechanism dysfunction, insta-

465

Orthopaedic Knowledge Update 8

Knee Reconstruction and Replacement

tal femoral resection or buildup and the tension of the

posterior capsule. Thus, an imbalanced reconstruction
usually requires adjustment of the femoral implant position with regard to its distal proximal level or its AP
dimension. Adjustment of tibial articular height will result in equal changes in the flexion and extension spaces
and is not recommended when an imbalance exists (Figure 6).

Selective Component Retention

Figure 6 Balancing the flexion and extension spaces in revision surgery. A, Factors
impacting the flexion space: I, tibial resection level; II, polyethylene thickness; III, tibial
slope; IV, AP dimension of the femoral component; V, AP placement of the femoral
component. B, Factor impacting the extension space. I, tibial resection level; II, polyethylene thickness, III,. distal femoral resection; IV, distal femoral augments; V, posterior capsule.

bility, fixation failure and periprosthetic fracture can all

compromise outcome.
Total knee revision can be indicated for gross loosening, fracture, instability, infection, malalignment, wear,
osteolysis or extensor mechanism disruption, and
knowledge of the specific etiology assists preoperative
planning. Appreciation of previous incisions and approaches is essential and will assist in planning for extensile exposure. A midline incision is preferred, but
when multiple previous incisions are present, the most
lateral incision that permits exposure should be used.
Existing implants should be ascertained, and if one or
more components remains well fixed, the surgeon must
understand the options for incorporating this component into the final construct and/or how best to remove
it without sacrificing bone. In addition, assessment
should be made of bone damage, remaining bone stock,
and extensor mechanism integrity. Collateral ligament
integrity will determine the level of constraint required.
Revision surgery demands careful attention to gap
balancing and joint line restoration. One useful reference for joint line restoration is its relationship to the
fibula on the contralateral knee, generally 1.5 cm proximal to the tip of the fibula. The surgeon should understand the options available to selectively alter either the
extension gap, the flexion gap, or both, and the impact
such decisions will have on the joint line. Elevation of
the joint line results in patellar impingement and reduced quadriceps strength. The flexion gap is impacted
by tibial resection level and polyethylene thickness, tibial slope, and by the AP dimension of the femoral implant. The extension gap is also determined by the
height of the tibial articular surface, but also by the dis-

466

Clinically significant polyethylene wear and associated

osteolysis is an increasingly common indication for revision TKA. Modular tibial components allow for the isolated replacement of a worn polyethylene insert in those
patients with intact prosthesis fixation to bone. Unfortunately, the success of such isolated polyethylene exchanges has not been as high as anticipated. Whether
these revisions are performed for instability or wear, the
incidence of failure requiring additional surgery is 30%
to 40%. Unappreciated malalignment or inadequate
soft-tissue balance that may have predisposed the implant to initial wear or instability will remain uncorrected with this approach. Failures present as pain, recurrent instability, recurrent wear, stiffness, or infection.
Failure rates are substantially higher than those of routine revision surgery, leading to the recommendation
that caution be exercised when considering this strategy.
In addition, retrieval analysis has demonstrated that the
locking mechanism of modular tibial base plates degrades over time, further raising questions about the advisability of introducing a new insert into such a failing
device.

Patellar Failure
Failure of the patellar component is one of the most
common indications for revision TKA and can occur
alone or in combination with failure of other components. When isolated patellar failure is the indication for
revision, a high rate of failure has been recently reported. These poor results are thought to be caused by
unrecognized component malalignment, evolving patellar osteonecrosis, and inability to restore bone stock.
Treatment of patellar bone loss that precludes the
ability to obtain fixation of a new patellar component
traditionally has been with patellectomy or dbridement
with retention of the patellar bone remnant, both of
which lead to extensor lag and weakness caused by loss
of patellar height. Bone grafting the residual patellar
shell within a soft-tissue pocket secured to the surrounding tissue has been advocated to restore bone
stock, improve tracking, and enhance extensor mechanism leverage. Early results in a small group of patients
have been encouraging.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 38 Knee Reconstruction and Replacement

Management of Bone Loss

Joint-Preserving Surgical Procedures

Septic loosening and osteolysis can result in significant

bone loss that must be addressed at the time of secondary reconstruction. Preoperative evaluation may underestimate the extent of osteolysis and the need for restoration of bone stock. Defects can be addressed with
either metal augments and substitutes or with a variety
of bone graft techniques. Contained defects are managed with morcellized allograft and long stem prostheses to offload the periarticular reconstruction. When defects are uncontained, structural allografts are useful,
and often essential, tools with which the surgeon should
be familiar. Noncircumferential defects can be managed
with bulk grafts acting to fill the defect, fixed to host
bone with cancellous screws and bypassed with canal
filling press fit or cemented stems. Impaction grafting
can also be used to restore bone stock around the knee,
by converting uncontained defects into contained defects. Metallic mesh is fashioned to restore the missing
aspect of the bone and morcellized graft can be impacted into this shell. Preliminary results suggest remodeling and incorporation of bone that is superior to that
obtained with bulk graft. Circumferential defects are
best treated with allograft prosthesis composites with a
step cut junction between allograft and host bone and
overlaid, if possible, with a vascularized shell of remnant
host bone. Intermediate-term results of structural allograft for complex knee reconstruction are promising,
with maintenance of fixation, reliable union at the
allograft-host interface and minimal evidence of graft
resumption. In one series, repeat revision was required
for flexion instability, loosening, or infection in 13% of
patients at a mean of 70 months following the index
procedure.

Moseley JB, Petersen NJ, Menke TJ, et al: A controlled

trial of arthroscopic surgery for osteoarthritis of the
knee. N Engl J Med 2002;347:81-88.

Annotated Bibliography
Clinical Evaluation
Mancuso CA, Sculco TP, Wickiewicz TL, et al: Patients
expectations of knee surgery. J Bone Joint Surg Am
2001;83:1005-1012.
Patient expectations for anticipated knee surgery include
symptom relief and functional improvement. Specific surveys
are offered and validated to help surgeons and patients to
communicate regarding shared goals of surgery.

Nonsurgical Care
Leopold SS, Redd BB, Warme WJ, Wehrle PA, Pettis
PD, Shott S: Corticosteroid compared with hyaluronic
acid injections for the treatment of osteoarthritis of the
knee: A prospective, randomized trial. J Bone Joint Surg
Am 2003;85:1197-1203.
A randomized comparison revealed no difference in effectiveness between corticosteroid and hyaluronic acid when
measured 6 months following treatment. Women were less
likely to respond to treatment.

American Academy of Orthopaedic Surgeons

This randomized trial of arthroscopic surgery for arthritis

failed to demonstrate any measurable advantage of arthroscopic dbridement or lavage over placebo surgery with respect to pain in the knee or function when measured 1 or
2 years after treatment

Wai EK, Kreder HJ, Williams JI: Arthroscopic debridement of the knee for osteoarthritis in patients fifty years
of age or older: Utilization and outcomes in the province of Ontario. J Bone Joint Surg Am 2002;84:17-22.
A review of over 6,000 patients with 3-year follow-up after
arthroscopy for arthritis revealed that 18% had undergone total knee replacement. Age was a predictor of subsequent knee
replacement, with patients older than 70 years most likely to
require additional surgery. These data suggest that arthroscopy
is overutilized in elderly patients.

Total Knee Arthroplasty

Argenson JN, Chevrol-Benkeddache Y, Aubaniac JM:
Modern unicompartmental knee arthroplasty with cement: A three to ten-year follow-up study. J Bone Joint
Surg Am 2002;84:2235-2239.
Unicompartmental arthroplasty results in 94% survivorship at 10 years when used for the treatment of unicompartmental noninflammatory tibiofemoral arthritis. Failure can occur as a result of progression of arthritis or polyethylene wear.

Barrack RL, Bertot AJ, Wolfe MW, Waldman DA, Milicic M, Myers L: Patellar resurfacing in total knee arthroplasty: A prospective, randomized, double-blind
study with five to seven years of follow-up. J Bone Joint
Surg Am 2001;83:1376-1381.
No difference in the occurrence of anterior knee pain is
seen whether or not the patella was resurfaced, and there were
no specific clinical indicators of anterior knee pain.

Mont MA, Rifai A, Baumgarten KM, Sheldon M, Hungerford DS: Total knee arthroplasty for osteonecrosis.
J Bone Joint Surg Am 2002;84:599-603.
Osteonecrosis can be successfully treated with total knee
replacement in a cohort of young patients when cemented fixation and adjunctive stems were used. Survivorship was 97%
at a mean of 108 months.

ORourke MR, Callaghan JJ, Goetz DD, Sullivan PM,

Johnston RC: Osteolysis associated with a cemented
modular posterior-cruciate-substituting total knee design: Five to eight-year follow-up. J Bone Joint Surg Am
2002;84:1362-1371.
A modular posterior stabilized knee prosthesis demonstrated excellent function at 5- to 8-year follow-up, but osteolysis was present in 16%. No osteolysis was seen when an all

467

Orthopaedic Knowledge Update 8

Knee Reconstruction and Replacement

polyethylene component was used. Two revisions demonstrated tibial post impingement and backside wear of the polyethylene.

Parvizi J, Lajam CM, Trousdale RT, Shaughnessy WJ,

Cabanela ME: Total knee arthroplasty in young patients
with juvenile rheumatoid arthritis. J Bone Joint Surg Am
2003;85:1090-1094.
TKA in patients with juvenile rheumatoid arthritis can be
successful, but poor preoperative function, involvement of
multiple joints, and compromised immune system lead to increased rates of postoperative stiffness and infection.

Puloski SKT, McCalden RW, MacDonald SJ, Rorabeck

CH, Bourne RB: Tibial post wear in posterior stabilized
total knee arthroplasty: An unrecognized source of
polyethylene debris. J Bone Joint Surg Am 2001;83-A:
390-397.
Retrieval analysis of posterior stabilized knee replacement
identified that the tibial post can exhibit significant wear and
damage as a result of impingement. This phenomenon was described for several implants and may be related to flexion of
the femoral component, or hyperextension of the knee.

Rand JA, Trousdale RT, Ilstrup DM, Harmsen WS: Factors affecting the durability of primary total knee prostheses. J Bone Joint Surg Am 2003;85-A:259-265.
Total knee replacement is a successful procedure with
long-term durability. Overall, survivorship at 10 years was
91%, 84% at 15 years, and 78% at 20 years. Factors that favorably affect durability were identified and include age over 70
years, rheumatoid arthritis, cemented fixation, female gender,
and retention of the PCL.

Saleh KJ, Sherman P, Katkin P, et al: Total knee arthroplasty after open reduction and internal fixation of fractures of the tibial plateau: A minimum five-year
follow-up study. J Bone Joint Surg Am 2001;83-A:11441148.
Prior open reduction and internal fixation of the tibial plateau poses significant challenges for subsequent knee replacement. Wound complications and infection can compromise
outcome, and bony defects may dictate reconstruction methods.

Tanzer M, Smith KU, Burnett S: Posterior stabilized versus cruciate retaining total knee arthroplasty: Balancing
the gap. J Arthroplasty 2002;17:813-819.
A blinded, prospective randomized comparison of these
two designs using identical surgical technique reveals no appreciable difference in functional outcome at 2-year follow-up.

Wang JW, Wang CJ: Total knee arthroplasty for arthritis

of the knee with extra-articular deformity. J Bone Joint
Surg Am 2002;84-A:1769-1774.

468

Extra-articular deformity can be corrected using intraarticular bone resection and ligament balancing when coronal
plane deformity is less than 20 in the femur and less than 30
in the tibia.

Complications
Crossett LS, Sinha RK, Sechriest VF, Rubash HE: Reconstruction of a ruptured patellar tendon with Achilles
tendon allograft following total knee arthroplasty.
J Bone Joint Surg Am 2002;84-A:1354-1361.
Rupture of the patellar tendon can be successfully managed by reconstruction using an Achilles tendon allograft.
Technical details are critical with proper bone attachment to
the tibia and broad soft-tissue repair to the quadriceps. Recurrent failure is minimized with this approach.

Dennis DA: Periprosthetic fractures following total

knee arthroplasty. Instr Course Lect 2001;50:379-389.
The management of periprosthetic fractures around knee
arthroplasty is reviewed. Key determinants of treatment include fracture displacement, bone quality, and status of implant fixation. Closed treatment, open reduction and fixation,
and component revision are the primary options.

Leopold SS, McStay C, Flafeta K, Jacobs JJ, Berger RA,

Rosenberg AG: Primary repair of intraoperative disruption of the medial collateral ligament during total knee
arthroplasty. J Bone Joint Surg Am 2001;83-A:86-91.
Following intraoperative injury to the medial collateral ligament, primary repair with use of primary, nonconstrained implants resulted in satisfactory results similar to knees without
this occurrence. Conversion to constrained implants was found
to be unnecessary.

Ortiguera CJ, Berry DJ: Patellar fracture after total

knee arthroplasty. J Bone Joint Surg Am 2002;84-A:532540.
Patellar fracture occurs in less than 1% of cases. Nonsurgical treatment is effective for nondisplaced fractures with intact
extensor mechanism. Surgical treatment for displaced fractures or loose patella result in a high rate of complications.

Revision Total Knee Arthroplasty

Babis GC, Trousdale RT, Morrey BF: The effectiveness
of isolated tibial insert exchange in revision total knee
arthroplasty. J Bone Joint Surg Am 2002;84-A:64-68.
Revision knee replacement consisting of isolated exchange
of the polyethylene results in a high rate of failure, with 25%
requiring rerevision. Preoperative instability resulted in the
highest failure rate.

Christensen CP, Crawford JJ, Olin MD, Vail TP: Revision of the stiff total knee arthroplasty. J Arthroplasty
2002;17:409-415.
The stiff total knee can be successfully revised with complete exposure and revision of all components, with attention

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

to accepted techniques of gap balancing. All patients in this series had improved motion and satisfactory pain relief and
function.

Clatworthy MG, Balance J, Brick GW, Chandler HP,

Gross AE: The use of structural allograft for uncontained defects in revision total knee arthroplasty: A
minimum five-year review. J Bone Joint Surg Am 2001;
83-A:404-411.
Structural allograft used for the treatment of uncontained
bone loss around the knee demonstrated a 72% survivorship
at 10-year follow-up. Failure occurred as a result of infection
or nonunion.

Hansen AD: Bone grafting for severe patellar bone loss

during revision knee arthroplasty. J Bone Joint Surg Am
2001;83:171-176.
Severe patellar deficiency encountered during revision
TKA that precludes reinsertion of a prosthetic component was
treated with bone grafting inside a soft-tissue envelope, with
improved tracking and restoration of patellar thickness.

Mason JB, Fehring TK, Odum SM, Griffin WL, Nussman DS: The value of white blood cell counts before revision total knee arthroplasty. J Arthroplasty 2003;18:
1038-1043.
Analysis of synovial fluid prior to revision TKA is useful
in differentiating septic from nonseptic knees. White blood cell
counts over 2,500/mm3 with over 60% polymorphonucleocytes
is highly suggestive of infection.

Saleh KJ, Dykes DC, Tweedie RL, Heck DA: Functional

outcome after total knee arthroplasty revision: A metaanalysis. J Arthroplasty 2002;17:967-977.
In this review of 42 studies on revision arthroplasty, the
procedure was shown to be effective in improving knee symptoms and function. Complications occur in 26% of cases and
rerevision is necessary in 13% of cases for a variety of reasons
including infection, extensor mechanism failure, and loosening.

Sharkey PF, Hozack WJ, Rothman RH, Shastri S, Jacoby

SM: Why are total knee arthroplasties failing today?
Clin Orthop 2002;404:7-13.

Chapter 38 Knee Reconstruction and Replacement

Blount WP: Dont throw away the cane. J Bone Joint
Surg Am 2003;85-A:380.
Callaghan JJ, Squire MW, Goetz DD, Sullivan PM,
Johnston RC: Cemented rotating-platform total knee
replacement: A nine to twelve-year follow-up study.
J Bone Joint Surg Am 2000;82:705-711.
Cartier P, Sanouiller JL, Grelsamer RP: Unicompartmental knee surgery: 10-year minimum follow-up period. J Arthroplasty 1996;11:782-788.
Chandler HP, Tigges RG: The role of allografts in the
treatment of periprosthetic femoral fractures. Instr
Course Lect 1998;47:257-264.
Colizza WA, Insall JN, Scuderi GR: The posterior stabilized total knee prosthesis: Assessment of polyethylene
damage and osteolysis after a 10-year minimum followup. J Bone Joint Surg Am 1995;77:1713-1720.
Coventry MB, Ilstrup DM, Wallrichs SL: Proximal tibial
osteotomy: A critical long-term study of eighty-seven
cases. J Bone Joint Surg Am 1993;75:196-201.
Coyte PC, Hawker G, Croxford R, Wright JG: Rates of
revision knee replacement in Ontario, Canada. J Bone
Joint Surg Am 1999;81:773-782.
Dennis DA, Konistek RD, Stiehl JB, Walker SA, Dennis
KN: Range of motion after total knee arthroplasty: The
effect of implant design and weight-bearing conditions.
J Arthroplasty 1998;13:748-752.
Duff GP, Lachiewicz PF, Kelley SS: Aspiration of the
knee joint before revision arthroplasty. Clin Orthop
1996;331:132-139.
Healy WL, Wasilewski SA, Takei R, Oberlander M: Patellofemoral complications following total knee arthroplasty: Correlation with implant design and patient risk
factors. J Arthroplasty 1995;10:197-201.
Insall JN, Thompson FM, Brause BD: Two-Staged Reimplantation for the Salvage of Infected Total Knee Arthroplasty. J Bone Joint Surg Am 2002;84-A:490.

The indications for revision arthroplasty were reviewed,

and in decreasing order of frequency were: wear, loosening, instability infection, arthrofibrosis, malalignment, and extensor
mechanism failure. More than half of all revisions occur within
2 years of implantation because of infection, instability, malalignment, or failure of fixation. Late revision is more commonly a result of wear, implant loosening, or late instability.

Murray DW: Goodfellow JW. OConnor JJ: The Oxford

medial unicompartmental arthroplasty: A ten-year survival study. J Bone Joint Surg Br 1998;80:983-989.

Classic Bibliography

Partington PF, Sawhney J, Rorabeck CH, Barrack RL,

Moore J: Joint line restoration after revision total knee
arthroplasty. Clin Orthop 1999;367:165-171.

Arima J, Whiteside LA, McCarthy DS, White SE: Femoral rotational alignment, based and the anteroposterior
axis, in total knee arthroplasty in a valgus knee: A technical note. J Bone Joint Surg Am 1995;77:1331-1334.

American Academy of Orthopaedic Surgeons

Naudie D, Bourne RB, Rorabeck CH, Bourne TJ: Survivorship of the high tibial valgus osteotomy: A 10-22year follow up study. Clin Orthop 1999;367:18-27.

Ranawat CS, Flynn WF Jr: Saddler S. Hansraj KK, Maynard MJ: Long-term results of the total condylar knee

469

Knee Reconstruction and Replacement

arthroplasty: A 15-year survivorship study. Clin Orthop
1993;286:94-102.
Ritter MA, Herbst SA, Keating EM, Faris PM, Meding
JB: Long-term survival analysis of a posterior cruciateretaining total condylar total knee arthroplasty. Clin
Orthop 1994;309:136-145.

470

Orthopaedic Knowledge Update 8

Squire MW, Callaghan JJ, Goetz DD, Sullivan PM,
Johnston RC: Unicompartmental knee replacement: A
minimum 15-year follow-up study. Clin Orthop 1999;
367:61-72.
Stern SH, Insall JN: Posterior stabilized prosthesis: Results after follow-up of nine to twelve years. J Bone Joint
Surg Am 1992;74:980-986.

American Academy of Orthopaedic Surgeons

Chapter

39

Foot and Ankle Trauma

Richard Marks, MD

Ankle Fractures
Classification
Ankle fractures may be classified by mechanistic or radiographic criteria. The Lauge-Hansen system consists
of four hyphenated descriptions of the fracture mechanism
(supination-external
rotation,
supinationadduction, pronation-external rotation, pronationabduction) (Figure 1). The first word describes the
position of the foot at the time of injury; the second
word describes the direction of the deforming force, and
hence, the foot. All comprise several stages, based on severity. Supination-external rotation injuries are the most
common, accounting for approximately 85% of all ankle
fractures. The first stage consists of a tear of the anterior
capsule and anterior tibiofibular ligament. In stage 2 the
injury progresses laterally, resulting in an oblique or spiral fracture of the fibula at the level of the plafond.
Stage 3 involves a tear of the posterior capsule or posterior malleolus fracture. Stage 4 consists of a transverse
medial malleolus fracture or tear of the deltoid ligament. The pronation-external rotation injury pattern begins medially, with injury to the deltoid ligament or medial malleolus fracture. The second stage is
characterized by injury of the posterior malleolus or
posterior capsule, whereas the third stage involves a
fracture of the fibula above the level of the plafond,
with disruption of the syndesmosis. Supinationadduction injuries consist of an anterior talofibular ligamentous tear or an avulsion fracture of the lateral malleolar tip, then advance to an oblique, shear-type
fracture of the medial malleolus caused by medial translation of the talus. Pronation-abduction injuries initially
place stress on the medial structures, resulting in deltoid
ligament failure, or avulsion of the distal tip of the medial malleolus. The second stage involves injury to the
posterior complex, and the third stage involves an oblique fracture of the distal fibula caused by shear of the
abducted talus.
The Weber/AO classification is based on the level of
the fibular fracture (Figure 2). Type A fractures occur
distal to the plafond and in more serious injuries involve

American Academy of Orthopaedic Surgeons

oblique, shear fractures of the medial malleolus. They

correspond to supination-adduction injuries. In type B
injuries, the fibular fracture occurs at the level of the
plafond, with varying degrees of obliquity and length of
the fibular fracture. Medial malleolar avulsion or deltoid
ligament disruption is associated with more severe fractures, corresponding to the supination-external rotation
and pronation-abduction injuries. Type C fractures occur
above the level of the plafond, with varying degrees of
syndesmotic involvement, corresponding to the
pronation-external rotation pattern. Most type A fractures can be treated with early mobilization; however, if
medial malleolar displacement is present, casting or surgical intervention may be required. Type B and C fractures are frequently unstable, requiring surgical stabilization. Neither classification scheme is entirely inclusive
or predictive, and the Weber B/AO and supinationexternal rotation patterns have great variability of fracture presentation. Concomitant syndesmotic involvement with widening of the mortise requires evaluation.
The medial clear space should equal the superior clear
space. Any inequality of these measurements or an absolute value of greater than 4 mm is indicative of an unstable fracture. Radiographic measurements, however,
may be influenced by radiographic rotation or magnification.

Medial Malleolar Fractures

Isolated fractures of the medial malleolus occur with
varying degrees of lateral ligamentous involvement.
Transverse fractures below the level of the plafond are
inherently stable and allow for early weight bearing and
mobilization. A transverse fracture at the level of the
plafond or an avulsion fracture with suspected medial
clear space widening may also involve syndesmotic disruption. The Maisonneuve fracture includes syndesmotic
disruption as well as a fracture of the fibular neck. In
addition to surgical treatment of the displaced medial
malleolar fracture, syndesmotic stabilization is also required. Treatment of oblique fractures at the level of the
plafond is dependent on lateral ligamentous involve-

471

Foot and Ankle Trauma

Orthopaedic Knowledge Update 8

Figure 1 A, Schematic diagram and case examples of Lauge-Hansen supination-external rotation and supination-adduction ankle fractures. A supinated foot sustains either an
external rotation or adduction force and creates the successive stages of injury shown in the diagram. The supination-external rotation mechanism has four stages of injury, and
the supination-adduction mechanism has two stages. tib-fib = tibiofibular (Reproduced with permission from Marsh JL, Saltzman CL: Ankle fractures, in Bucholz RW, Heckman JD
(eds): Rockwood and Greens Fractures in Adults, ed 5. Philadelphia, PA, Lippincott Williams & Wilkins, 2001, pp 2001-2090.)

472

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 39 Foot and Ankle Trauma

Figure 1 B, Schematic diagram and case examples of Lauge-Hansen pronation-external rotation and pronation-abduction ankle fractures. A pronated foot sustains either an
external rotation or abduction force and creates the successive stages of injury shown in the diagram. The pronation-external rotation mechanism has four stages of injury, and
the pronation-abduction mechanism has three stages. tib-fib = tibiofibular (Reproduced with permission from Marsh JL, Saltzman CL: Ankle fractures, in Bucholz RW, Heckman
JD (eds): Rockwood and Greens Fractures in Adults, ed 5. Philadelphia, PA, Lippincott Williams & Wilkins, 2001, pp 2001-2090.)

American Academy of Orthopaedic Surgeons

473

Foot and Ankle Trauma

Orthopaedic Knowledge Update 8

lique fractures without comminution. In bimalleolar
equivalent injuries with deltoid disruption, medial malleolar clear space widening may require medial malleolar arthrotomy to clean out entrapped deltoid ligament
remnants. Repair of the deltoid ligament does not improve surgical outcome. Intraoperative arthroscopy of
the ankle has revealed a high incidence of chondral injuries; however, arthroscopy has not improved patient
outcomes. Postoperative immobilization in a functional
brace has not been superior to casting, with a higher incidence of wound complications. Additionally, early motion has not improved functional outcome.

Trimalleolar Fractures

Figure 2 Weber/AO fractures. The staging is completely determined by the level of

fibular fracture. Type A occurs below the plafond, whereas type C starts above the
plafond. (Reproduced from Michelson JD: Ankle fractures resulting from rotational injuries. J Am Acad Orthop Surg 2003;11:403-412.)

ment and medial displacement. Medial shift of greater

than 2 mm requires surgical intervention, typically with
two parallel 4.0-mm lag screws or a lag screw with derotation Kirschner wire. Poor bone quality or significant
comminution may necessitate the use of tension band
wiring. If a vertical shear pattern is noted, the screws
should be placed perpendicular to the shaft, at times in
conjunction with a medial buttress plate to avoid the
risk of proximal migration of the medial malleolar fragment. Stress fracture of the medial malleolus has also
been described, particularly in patients with a varus heel
and repeated adduction moments.

Bimalleolar Fractures
Bimalleolar fractures with displacement or rotational
deformity at the time of injury frequently require open
reduction and internal fixation. If the medial malleolar
fracture occurs below the level of the plafond, closed
treatment may be possible; however, the risk of displacement and the need for frequent reevaluation typically results in surgical intervention, if the patients
medical condition permits. The timing of surgical intervention is dependent on ankle edema and the quality of
the soft tissues. Restoration of lateral malleolar length
and rotation is key to reduction, and any significant lateral shift or shortening of the lateral malleolus can result in alterations of contact characteristics of the ankle
joint. A recent study indicated that both lateral and posterior antiglide plating of the lateral malleolus have
equivalent clinical and radiographic results, with slightly
more peroneal tendon irritation with the antiglide technique. The use of lag screws without plate application
has been successfully reported for management of ob-

474

Fractures involving the posterior malleolus are a result

of posteroinferior tibiofibular ligament avulsion, or impaction from the talus. Frequently, anatomic reduction
of the lateral malleolus and medial malleolus combined
with dorsiflexion of the ankle reduces the posterior malleolar fragment. Fragments with less than 25% involvement of the articular surface or those with less than 2
mm of displacement do not require internal fixation.
Anterior to posterior percutaneous fixation may be possible, although at times a formal posterolateral or posteromedial approach is necessary to free up interposed
soft tissues.

Syndesmotic Complex Injury

Disruption of the inferior syndesmosis ranges from subtle sprain to gross instability seen with pronationexternal rotation and Weber C injuries. Patients with
tenderness over the medial malleolus, proximal fibula,
or interosseous membrane should be evaluated for syndesmotic injury. With subtle injury, MRI or bone scan
may help in the diagnosis. Weight-bearing ankle views
can sufficiently stress the interosseous membrane and
anterior talofibular ligament to allow for evaluation of
the syndesmosis and should be obtained as pain allows.
As previously discussed, medial clear space widening
with loss of tibiofibular overlap indicates deltoid and
syndesmotic disruption. In patients with isolated syndesmotic disruption, reduction with percutaneous internal
fixation is possible if the medial clear space adequately
reduces. If not, a medial arthrotomy is necessary to remove the interposed deltoid ligament. A fully threaded
3.5- or 4.5-mm screw is placed across three or four cortices, without compressive lagging. The screw should be
placed 1.5 to 3.0 mm above the joint and directed 30
anteriorly. Both three- and four-cortex fixation has been
advocated, with four-cortex fixation offering greater stability, but no study has shown superior results with either method of fixation. Although traditionally the syndesmotic screw is placed with the ankle in a neutral
position, the position of the ankle at the time of screw
placement does not affect postoperative range of mo-

American Academy of Orthopaedic Surgeons

Chapter 39 Foot and Ankle Trauma

Orthopaedic Knowledge Update 8

tion. Patients do not bear weight for 8 weeks to allow
for sufficient syndesmotic healing, and screws are removed at 3 months to allow for restoration of normal
distal fibular motion. Bioabsorbable screws have been
shown to have equivalent clinical outcomes compared
with metallic screws and avoid the need for screw removal. For Weber B and C injuries, the fibular fracture
is rigidly fixed, then the syndesmosis is stressed with external rotation of the talus within the mortise with the
ankle in a neutral position. Additionally, the fibula can
be grasped with a towel clip, and distal syndesmotic stability evaluated. Typically, fractures less than 4.5 mm
above the joint do not require a syndesmotic screw, but
this parameter is variable, and intraoperative stressing is
advised for all fibular fractures above the level of the
joint.

Pilon Fractures
Classification
Tibial pilon fractures occur as a result of either a lowenergy rotational mechanism with less comminution and
soft-tissue damage or a high-energy injury secondary to
vertical compression that can result in significant comminution, chondral damage, and soft-tissue involvement.
Foot position at the time of impact can also result in additional injury to the malleoli, talus, and calcaneus. The
Ruedi and Allgower classification remains the most
commonly used for pilon fractures. Type I fractures are
intra-articular with minimal displacement. Type II fractures have significant articular displacement with little
comminution, whereas type III fractures have greater
comminution and metaphyseal involvement. CT is useful in determining the full extent of the injury and helps
with definitive surgical planning.

Treatment
Nonsurgical treatment is possible if articular incongruity
is less than 2 mm and is generally reserved for lowenergy injuries. Many pilon fractures require surgical intervention, which is performed in stages. The application
of an external fixator at the time of injury allows for
restoration of length and partial reduction via ligamentotaxis, while minimizing further soft-tissue damage. The
fixator may be uniplanar or multiplanar, may cross the
ankle and subtalar joints, or be limited to the tibia. Plating of the fibula helps restore length, but is not necessary at the time of initial fixator stabilization. CT is performed after fixator application to better ascertain
fracture configuration and can help determine the
proper surgical approach to limit soft-tissue stripping.
Definitive treatment cannot be undertaken until the soft
tissues have healed and swelling has resolved, usually 10
to 21 days after injury. Treatment goals include restoration of articular congruity and reestablishing proper
length, rotation, and angulation, while avoiding exces-

American Academy of Orthopaedic Surgeons

sive surgical dissection and soft-tissue stripping. Newer

surgical techniques include percutaneous internal fixation, limited open reduction, and subcutaneous plating,
thus avoiding soft-tissue healing complications. Care is
taken to perform a wound closure free of tension. If a
solid construct can be obtained, the external fixator can
be removed at the time of definitive treatment; however, this is generally only possible for type I fractures,
whereas the higher energy level injuries require a prolonged course of fixator usage, frequently 2 to 3 months,
to provide additional stability. If extensive bone loss is
encountered, bone grafting at 6 to 8 weeks after injury
is performed, with consideration given for the application of a bone stimulator.

Hindfoot Fractures
Talus Fractures
The talus acts as a link between the ankle, subtalar, and
transverse tarsal joints. It is devoid of muscle or tendon
attachments, and 70% of its surface is covered by articular cartilage for its five weight-bearing surfaces. Its
blood supply is limited to the nonarticular surfaces;
therefore, vascular compromise may be associated with
fractures, particularly those involving the talar neck.
Fractures are classified according to anatomic location:
body, neck, or head.

Talar Body
Talar body fractures involve the superior articular cartilage and may be classified as coronal, sagittal, or horizontal fractures. These fractures are commonly associated with high-energy ankle fractures. Osteochondral
fractures are also associated with ankle fractures, particularly supination-external rotation fracture patterns.
Other anatomic locations include fracture of the lateral
and posterior processes. Lateral process fractures, also
described as snowboarder fractures, are created by
forced dorsiflexion and external rotation of the foot.
This fracture is commonly missed on initial presentation, but is usually present on plain radiographs of the
ankle. CT is helpful to ascertain the extent of the fracture, which may encompass a significant portion of the
lateral aspect of the posterior facet. If the fragment size
is large and displacement is greater than 2 mm, open reduction and internal fixation is performed. Comminuted
fractures may be initially treated with casting or immediate excision with early range of motion. No prospective study has evaluated the superiority of either treatment method. The posterior process consists of
posteromedial and posterolateral tubercles. Fractures
occur as a result of avulsion of the posterior talotibial
and posterior talofibular ligaments, respectively. The
posterolateral tubercle is more frequently involved, and
because of the close proximity of the flexor hallucis longus tendon in its posterior groove, flexion and extension

475

Foot and Ankle Trauma

Figure 3 Hawkins classification. A, Type I: nondisplaced talar neck fractures.

B, Type II: displaced talar neck fractures, with subluxation or dislocation of subtalar
joint. C, Type III: displaced talar neck fractures with associated dislocation of talar
body from both subtalar and tibiotalar joints. D, Canale and Kelly type IV: displaced
talar neck fracture with associated dislocation of talar body from subtalar and tibiotalar joints and dislocation of head/neck fragment from talonavicular joint. (Reproduced
with permission from Sangeorzan BJ: Foot and ankle joint, in Hansen ST Jr, Swiontkowski MF (eds): Orthopaedic Trauma Protocols. New York, NY, Raven Press, 1993,
p 350.)

of the hallux may exacerbate symptoms. Fractures without significant subtalar involvement are initially treated
with casting. If subtalar joint involvement is associated
with displacement, then surgical reduction is required.
Fragment excision is reserved for symptomatic nonunions without significant subtalar joint involvement.

Talar Neck
Talar neck fractures are grouped according to the
Hawkins classification (Figure 3). Type I is a nondisplaced fracture. Type II involves subluxation or dislocation of the subtalar joint, and type III involves subluxation or dislocation of the subtalar and ankle joints.
Type IV injuries, as described by Canale and Kelly, additionally involve displacement of the talonavicular joint.
Type I fractures can be treated with nonweight-bearing
casting; however, many surgeons prefer surgical treatment to avoid the risk of late displacement. This fracture pattern is amenable to percutaneous internal fixation from a posterolateral insertion site. Type II, III, and
IV fractures require open reduction with internal fixation. Displacement of greater than 1 mm requires surgical care to avoid further vascular and soft-tissue compromise. Irreducible fractures, particularly those with
residual subluxation, dislocation, or threatened soft tissues require emergent surgical intervention. Dual anterolateral and anteromedial incisions are usually re-

476

Orthopaedic Knowledge Update 8

quired, with osteotomy of the medial malleolus
performed if initial exposure is inadequate. Although
the strongest screw configuration is from a posterior to
anterior direction, crossed screws from an anterior to
posterior direction are used for open reductions,
whereas the posterior to anterior approach is reserved
for those fractures treated in a closed manner. Medial
comminution may require bone grafting and medial
plating to prevent subsequent varus malunion. Theoretically, titanium screws offer an advantage if postoperative MRI is to be considered for evaluation of postoperative osteonecrosis, which increases in incidence with
the severity of the fracture. Hawkins sign represents osteopenia that is seen beneath the subchondral surface of
the talar dome and occurs 6 to 8 weeks after the fracture, which is indicative of talar revascularization. If osteonecrosis is suspected, MRI may be useful in making
the diagnosis. If the fracture has healed and no cystic
changes or collapse are noted, progressive, protected
weight bearing can be instituted at 8 weeks, with regular
clinical and radiographic reevaluation. Posttraumatic arthrosis of the ankle and/or subtalar joints is present in
two thirds of all talar fractures. Fusion procedures may
be complicated by varying degrees of osteonecrosis and
may require an extended tibiotalocalcaneal arthrodesis.
The incidence of osteonecrosis increases with degrees of
fracture severity; up to 13% for Hawkins type I fractures, 20% to 50% for Hawkins type II, and 50% to
100% for Hawkins type III. Data are incomplete for
Hawkins type IV fractures because of the infrequent
presentation of these fractures. The presence of subchondral sclerosis, however, does not dictate the requirement for not bearing weight for prolonged periods.
These patients are allowed progressive weight bearing
based on symptoms and radiographic appearance of the
talus.

Subtalar Dislocation
High-energy mechanisms are responsible for most subtalar dislocations and are frequently associated with
open injuries and irreducible dislocation. Overall, the
dislocations are closed in approximately 75% of patients, and medial dislocations, with the foot medially
displaced relative to the hindfoot, occur 65% of the
time. Irreducible dislocations occur in 32% of patients;
the peroneal area blocks reduction with medial dislocation, and the posterior tibial and flexor hallucis longus
and flexor digitorum longus tendons can block reduction with lateral dislocations. These injuries frequently
require emergent open reduction, tendon relocation,
and stabilization. Postreduction CT should be performed to fully ascertain the extent of associated injuries, which occur in virtually all patients with subtalar
dislocations.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 39 Foot and Ankle Trauma

Calcaneal Fractures
The calcaneus is the most frequently fractured tarsal
bone, with most fractures occurring as a result of axial
loading, such as a fall from a height or during a motor
vehicle crash. Given the mechanism of injury, patients
should be evaluated for associated injury of the lumbar
spine, which occurs in approximately 10% of patients
with calcaneal fractures. Axial loading creates an oblique shear fracture caused by impaction of the lateral
process of the talus that results in a superomedial fragment that consists of the sustentaculum (constant
fragment) and a superolateral fragment that has an
intra-articular component. In addition to this primary
fracture line, a secondary fracture component may be
created, based on additional energy imparted and the
position of the foot at the time of injury.
Radiographic evaluation should include standard
foot views, as well as a Harris axial view, which can provide information concerning shortening and varus angulation of the heel. An AP ankle radiograph may also allow for evaluation of lateral wall extrusion and
impingement. Although specialized projections of the
subtalar joint have been described (Broden and Isherwood projections), CT imaging provides the most complete, reliable assessment of these fractures.
Fractures may be classified as either extra-articular
or intra-articular. Avulsion injuries account for most
extra-articular fractures and result in fracture of the anterior process, sustentaculum, or calcaneal tuberosity,
which is secondary to avulsion of the Achilles tendon insertion. Occasionally, oblique fractures that do not involve the subtalar joint are seen as well. Surgical indications are detachment of the Achilles tendon insertion,
or displacement of greater than 2 mm of the sustentaculum or anterior process.
Intra-articular fractures occur in 75% of patients
with calcaneal fractures, and characterization of the degree of displacement and number of posterior facet articular fragments is helpful for treatment recommendations and predictive of treatment outcomes. The Sanders
CT classification system is based on the number and location of articular fragments seen on coronal projections (Figure 4). There are four types, based on the
number of fragments of the posterior facet, with displacement of greater than 2 mm considered significant.
Type I fractures are nondisplaced, regardless of the
number of fragments. Types II, III, and IV have corresponding numbers of displaced articular fragments. In
addition to displacement of the posterior facet, the surgeon must take into account factors such as shortening
and widening of the heel, lateral wall impingement, and
peroneal subluxation/dislocation when making surgical
decisions. Patient factors, such as overall medical condition, peripheral vascular disease, compromised soft tissues, and a history of smoking also need to be taken

American Academy of Orthopaedic Surgeons

Figure 4 Schematic diagram of Sanders classification. (Reproduced with permission

from Sanders R, Fortin P, Pasquale T, et al: Operative treatment in 120 displaced
intra-articular calcaneal fractures: Results using a prognostic computed tomography
scan classification. Clin Orthop 1993;290:87-95.)
into account. Treatment options include no surgery,
closed reduction with manipulation with or without limited internal fixation, open reduction with internal fixation, and primary arthrodesis. Nonsurgical treatment is
limited to patients with a type I fracture pattern or
those with factors or comorbidities that preclude good
surgical outcome. The fracture is immobilized until the
soft-tissue swelling and fracture blisters resolve, then
range-of-motion exercises are instituted. The patient is
to refrain from bearing weight for 10 to 12 weeks.
Closed reduction and manipulation with or without limited internal fixation can be used for those fractures
with mild shortening or lateral impingement, as well as
tongue-type fractures that are amenable to this form of
treatment. As a result, the late complication of subfibular impingement with peroneal involvement can be
avoided and more normal hindfoot anatomy can be restored, while avoiding potential complications associated with open techniques.
Open reduction and internal fixation is indicated for
type II and III fractures. Surgery is delayed until swelling subsides, and a wrinkle sign is present on the lateral aspect of the heel, usually 10 to 14 days after injury.
If a fracture blister is present in the area of the proposed incision, it must resorb and begin to reepithelialize before surgery. The lateral extensile right-angle incision is most commonly used, and although the medial
approach has been used successfully, it is typically reserved for fractures with an irreducible medial fragment. A full-thickness flap is created to maintain softtissue integrity, and the posterior facet is elevated and
reduced. Manipulation of the tuberosity with a Schanz

477

Foot and Ankle Trauma

or Steinmann pin can correct shortening and varus angulation and help with reduction of the posterior facet.
Following provisional reduction maintained by Kirschner wires, a low profile lateral plate and screws are
applied, first stabilizing the posterior facet. Meticulous
closure is performed, and the patient is immobilized in a
bulky posterior and U splint. Sutures are removed 10 to
14 days after surgery, and if the soft tissues allow, early
range of motion is instituted. Patients do not bear
weight for 10 weeks, at which time progressive weight
bearing is begun.
Primary subtalar arthrodesis combined with open reduction and internal fixation to restore anatomic height,
width, and calcaneal pitch is used for type IV fractures.
This approach appears to be superior to initial neglect
of the fracture followed by delayed subtalar fusion or
primary open reduction and internal fixation.
The surgical outcomes of calcaneal fractures correlate with the quality of reduction and number of intraarticular fragments. Two-part fractures are associated
with a better outcome than three-part fractures, and
four-part fractures have uniformly poor outcomes. The
extent of initial injury, initial nonsurgical treatment, and
workers compensation cases are predictors for eventual
subtalar fusion. More than half of patients will require a
supportive device. Bilateral fractures treated surgically
have a worse outcome than unilateral fractures treated
surgically, but have a better outcome than those managed nonsurgically. A recent Canadian study has shown
comparable results in patients treated surgically and
nonsurgically; however, when workers compensation
patients were excluded, those managed surgically had
significantly higher satisfaction scores. Complications
occur in 18% to 40% of surgical cases. Falls from a
height, surgical treatment earlier than 7 days after injury, longer surgical time, and a history of smoking were
factors associated with complications.
Neglected displaced fractures often result in a wide,
shortened heel, problems with shoe wear, subfibular
peroneal impingement, and posttraumatic arthrosis. This
may necessitate decompression of the lateral wall of the
calcaneus, peroneal relocation, and subtalar distraction
arthrodesis to restore proper calcaneal pitch and height.

Midfoot Trauma
Navicular Fractures
The midfoot consists of the navicular, cuboid, the medial, middle, and lateral cuneiforms, as well as the
metatarsal-cuneiform articulations. The midfoot has
constrained motion because of multiple recessed articulations as well as strong ligamentous and capsular attachments. The navicular articulates with the cuneiforms, cuboid, calcaneus, and talus. Coupled motion
between these structures (transverse tarsal joint) provides inversion and eversion of the midfoot and forefoot

478

Orthopaedic Knowledge Update 8

relative to the hindfoot. Given the extensive articular
surfaces of the navicular, vessels enter from the navicular tuberosity and dorsal and plantar surfaces, with a relatively avascular central portion, which has implications
for fracture healing and the development of osteonecrosis.
Fractures can be described as avulsion (chip), tuberosity, or body fractures. Avulsion fractures of the dorsal
lip are the most common and occur secondary to a plantar flexion force with varying degrees of inversion or
eversion. The patient should be evaluated for associated
midfoot or ankle injury. Immobilization is used until
symptoms resolve, and for more severe fractures with a
larger avulsion fragment and soft-tissue swelling, a non
weight-bearing cast should be used. Delayed excision of
fragments is undertaken if they remain symptomatic. If
a fragment includes more than 25% of the articular surface, open reduction and internal fixation is indicated to
avoid midtarsal subluxation and posttraumatic arthrosis.
Navicular tuberosity fractures result from an eversion
force applied to the foot, with resultant contraction of
the posterior tibial tendon and tension on the superficial
deltoid attachment to the navicular. This may also represent a diastasis of the synchondrosis of an accessory
navicular. Oblique foot radiographs obtained with the
foot in 45 of internal rotation may help identify the
fracture, which is differentiated from an accessory bone
by the presence of sharp, irregular cleavage lines. Treatment is based on the severity of symptoms; however,
given the possible late sequelae of posttraumatic flatfoot deformity, cautious treatment of these injuries is
warranted. Injuries with minimal symptoms may be
treated with immobilization and weight bearing as tolerated. In more symptomatic injuries with diastasis less
than 3 mm, a nonweight-bearing cast is applied for 4 to
6 weeks, followed by gradual mobilization of the foot.
Diastasis greater than 3 mm usually requires open reduction and internal fixation, with care taken to repair
the posterior tibial tendon and deltoid ligament attachments.
Body fractures of the navicular are the result of direct axial loading, as occurs with a fall from a height.
Given the strong ligamentous attachments, this probably
requires a plantar flexion/dorsiflexion and rotational
moment in addition to axial loading. Three fracture
types have been described. Type I is a transverse fracture in the coronal plane, with the dorsal fragment less
than 50% of the body, and no angular deformity of the
foot. Type II, the most common, is an oblique fracture,
from a dorsomedial to plantar lateral direction. The lateral fragment is smaller and often comminuted. There
may be an adduction deformity of the foot. Type III
fractures occur as a result of axial loading with lateral
compression that results in central or lateral comminution and midfoot abduction deformity. The cuboid and
anterior calcaneus may also be injured.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 39 Foot and Ankle Trauma

Figure 6 Bony avulsion at the base of the second metatarsal. This fleck sign (arrow)
indicates a Lisfranc injury.

Figure 5 Loss of colinearity between the medial aspect of the middle cuneiform and
medial aspects of the second metatarsal base, as denoted by the black lines, is indicative of a Lisfranc injury.
Type I and II nondisplaced fractures may be treated
with nonweight-bearing cast immobilization for 6 to 10
weeks. Given the severity of type III fractures, these
fractures usually require surgical intervention. Displaced type I fractures are amenable to internal fixation
from the dorsal direction via an anteromedial approach.
Type II and III fractures may require intraoperative external fixator application to assist with reduction; frequently bone grafting may be necessary for any large
lateral defects, and it may be necessary to stabilize the
major medial fragment to the cuneiforms to restore normal midfoot alignment. If extensive comminution is
present, immediate naviculocuneiform arthrodesis
should be considered.

Tarsometatarsal (Lisfranc) Fracture-Dislocations

The Lisfranc joint complex consists of the tarsometatarsal, intermetatarsal, and intertarsal joints. There is inherent osseous stability because of the recessed second
metatarsal base between the medial and middle cuneiforms, as well as the cross-sectional trapezoid shape of
the first three metatarsals and their corresponding cuneiforms. There are also strong capsuloligamentous attachments, particularly on the plantar aspect of the foot,
which provide additional stability. The Lisfranc ligament
runs from the base of the second metatarsal to the me-

American Academy of Orthopaedic Surgeons

dial cuneiform. Injuries of the Lisfranc joint complex

can result from either direct or indirect mechanisms. Direct injuries result from a dorsally applied force, which
will result in plantar displacement of the metatarsals if
the force is applied to the metatarsal base or dorsal
metatarsal displacement if the force is applied to the cuneiforms. There is a high incidence of associated tarsal
fracture, significant soft-tissue destruction, and compartment syndrome. Indirect injuries occur from a combination of axial loading and twisting on an axially loaded,
plantar flexed foot. Because of the mechanism of injury,
relatively weaker dorsal ligaments, and greater mobility
between the first and second metatarsals, displacement
is typically dorsal, with diastasis between the first and
second metatarsals and their corresponding cuneiforms.
AP, internal oblique, and lateral foot radiographs are
obtained to evaluate the tarsometatarsal articulations.
Normal radiographs include colinearity of the medial
aspect of the second metatarsal base and medial cuneiform, the medial aspect of the fourth metatarsal base
with the medial aspect of the cuboid, and the medial aspect of the third metatarsal with the lateral cuneiform.
Any incongruity is pathologic, as is any dorsal displacement of the metatarsals seen on a lateral radiograph
(Figure 5). Diastasis of 2 mm between the medial and
middle cuneiforms is considered pathologic. The fleck
sign represents an avulsion of the Lisfranc ligament off
the base of the second metatarsal (Figure 6). If injury is
suspected despite normal radiographs, weight-bearing
films are obtained once clinically feasible. Stress radiographs, MRI, and CT can also be used to assist in the diagnosis. Lisfranc complex injuries may be classified as

479

Foot and Ankle Trauma

total incongruity, partial incongruity, or divergent, but
because of tremendous variation in mechanism of injury, there is great variation in radiographic presentation.
Nonsurgical treatment is used for nondisplaced fractures and consists of a nonweight-bearing cast for 8
weeks, followed by gradual weight bearing in a removable boot brace. Surgery is performed for injuries with
any displacement seen on radiographs. If anatomic reduction can be obtained by closed means, percutaneous
internal fixation is possible, but interposed Lisfranc ligament remnant can frequently block reduction and result
in the joint springing open once internal fixation is removed. Therefore, open reduction is frequently necessary through a single or dual longitudinal incision as dictated by the injury pattern. Fixation is achieved with
either 3.5- or 4.0-mm screws for the medial (first tarsometatarsal joint) and middle (second, third tarsometatarsal joints) columns and Kirschner wires for the lateral
(fourth, fifth tarsometatarsal joints) column. A severe
abduction mechanism may result in compression of the
cuboid, termed a nutcracker injury. This injury may require lateral plating and bone grafting to avoid residual
abduction. Patients should not bear weight for 8 weeks,
followed by gradual weight bearing in a removable boot
brace. Screws are removed 3 to 4 months after surgery
to avoid screw breakage and to allow restoration of sagittal motion at the tarsometatarsal junction. Alternatively, bioabsorbable screws may also be used, obviating
the need for additional surgery for screw removal. Immediate complete arthrodesis of the midfoot has poor
results, although partial arthrodesis tends to have better
results. Late arthrosis is common, even with anatomic
reduction. Poor results are associated with nonanatomic
reduction, open injury, and extensive comminution.

Metatarsal Fractures
Acute metatarsal fractures may occur as a result of a direct blow, which usually results in a transverse fracture,
or as a result of an indirect twisting, or avulsion mechanism. Metatarsal base fractures are associated with midfoot injury, and these fractures carry a high index of suspicion for additional injury. Compartment syndrome can
be seen in patients with more severe fractures, particularly those resulting from a direct blow. Nonsurgical
treatment is indicated if displacement is less than 3 mm
or angulation is less than 10. Treatment can vary from
the use of a postoperative shoe with weight bearing as
tolerated for a stable fracture to casting with no weight
bearing for an unstable fracture pattern prone to displacement. Surgery is indicated for displacement of
greater than 3 to 4 mm or sagittal displacement of
greater than 10 because this can lead to either direct
overload of a displaced plantar fragment, or transfer
metatarsalgia if dorsal displacement is noted. If associ-

480

Orthopaedic Knowledge Update 8

ated with a Lisfranc ligamentous injury at the tarsometatarsal junction, surgical stabilization is necessary
because of inherent instability. Surgical stabilization
may be achieved with Kirschner wires, compression
screws, or minifragment plating, as the fracture pattern
dictates.
Treatment of fifth metatarsal base fractures is dependent on fracture location. It is helpful to separate
fractures of the base of the fifth metatarsal into three
types, based on location of the fracture. Type I fractures
are true avulsion injuries of the tuberosity, or styloid
process, which occur secondary to contraction of the
long plantar ligament and peroneus brevis insertion;
many are extra-articular. These fractures can typically
be treated with a stiff-soled shoe and weight bearing as
tolerated once symptoms diminish, with many patients
sufficiently asymptomatic after 3 weeks of treatment. If
there is a large intra-articular fragment with displacement or if the base fragment has retracted proximally,
indicative of peroneus brevis retraction, acute surgical
intervention is occasionally necessary. Continued symptoms after nonsurgical treatment are rare and may necessitate removal of the nonunion fragment, with evaluation of the peroneus brevis for possible reattachment
directly to bone, or with the use of a suture anchor.
Type II fractures occur at the metaphysealdiaphyseal junction, approximately 2.5 cm distal to the
base. This is a circulatory watershed region that is subject to nonunion secondary to poor blood supply. The
acute Jones fracture should be treated with either a
nonweight-bearing cast for 6 to 8 weeks, or in a highperformance athlete, with placement of an intramedullary screw, which allows for earlier mobilization. The optimal screw size is largely dependent on the size of the
intramedullary canal. One study has shown comparable
initial and ultimate failure loads for 4.5- and 5.5-mm
partially threaded cannulated screws, whereas another
study compared 5.0- and 6.5-mm partially threaded
screws and found no difference in bending stiffness, but
greater pull-out strength with the 6.5-mm screws. Clinically, failure of fixation has been noted in athletes who
returned to full activities before radiographic evidence
of complete radiographic union.
Type III fractures occur in the diaphysis and are typically stress fractures. Possible etiologic factors include
low arches and associated first metatarsal hypermobility,
as well as cavovarus deformities, both of which can result in abnormally high stresses placed on the lateral
foot. Altered lateral stresses may also be seen in patients with hereditary sensorimotor neuropathy and diabetic neuropathy. Poor blood supply in this region may
also predispose an individual to injury and impair injury
healing, resulting in nonunion. Strict adherence to no
weight bearing is necessary after a proximal diaphyseal
fracture is diagnosed and should be continued until radiographic evidence of fracture healing. Pulsed electro-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

magnetic fields have been shown to accelerate healing.
If sclerosis is noted at the fracture site, this indicates circulatory compromise, and surgical intervention is necessary to remove avascular fibrous tissue, combined with
intramedullary screw fixation. Cancellous bone grafting
may also be necessary. Any contributing structural abnormality should also be addressed.

Complex Regional Pain Syndrome

Complex regional pain syndrome, formerly referred to
as reflex sympathetic dystrophy, is a clinical entity that
develops after a precipitating traumatic event. It involves dysfunction of the sensory, autonomic, and motor
systems and frequently is missed on initial presentation.
Patients with acute complex regional pain syndrome describe antecedent trauma, which may be as severe as a
crush injury or as nominal as a sprain or twisting injury.
The onset of pain may also occur after surgical procedures. In the acute phase, there is an abnormal disproportional painful reaction to a stimulus (hyperpathia),
increased sensitivity to a stimulus (allodynia), hyperhidrosis, and cyanosis. Swelling of the foot is typically
seen, and radiographs will appear normal. The second
phase occurs 2 to 4 months after onset and involves dystrophic and ischemic changes. Hyperpathia and allodynia continue, as does swelling, and hypohydrosis and
pallor are experienced. Radiographs typically show
demineralization. The third phase is characterized by evidence of joint stiffness, muscular atrophy, and contracture. Swelling may persist, with the skin becoming shiny,
dry, and cool. Radiographs show evidence of continued
demineralization. This syndrome may also be characterized into complex regional pain syndrome type I, which
occurs in a generalized area after a traumatic event, or
complex regional pain syndrome type II, which develops
in a discrete area secondary to peripheral nerve injury.
The diagnosis is assisted by the use of a triple-phase
bone scan, which in the delayed phase will show diffuse
radionuclide uptake throughout the foot. This procedure
is best performed within the first 6 months of the appearance of symptoms. MRI will also show typical marrow edema during this time. Treatment consists of a
combination of physical therapy, nerve blocks, and pharmacologic agents, and should be instituted as early as
possible. Physical therapy is designed to maximize motion, in combination with desensitization exercises and
modalities. Sympathetic nerve blocks are used both for
diagnostic and treatment purposes. Failure to respond to
a sympathetic block should bring the diagnosis into
question. Several pharmacologic agents, including antidepressants, anticonvulsants, and calcium channel blockers have been used for the treatment of complex regional pain syndrome, typically coordinated by a
multidisciplinary pain center.

American Academy of Orthopaedic Surgeons

Chapter 39 Foot and Ankle Trauma

Compartment Syndrome
Acute compartment syndrome develops secondary to
local trauma that results in an increase of local tissue interstitial pressure from bleeding and edema from softtissue destruction. As a result, vascular occlusion occurs
and creates myoneural ischemia. Compartment syndrome has been associated with calcaneal fractures, Lisfranc complex injuries, and crush injuries; however, compartment syndrome should be suspected in any injury
mechanism that creates significant swelling. If the ischemic process is left untreated for more than 8 hours,
irreversible myoneural necrosis and fibrosis occur;
therefore, prompt diagnosis and treatment are of great
importance. The diagnosis relies on clinical signs and
measurement of compartment pressures. Pain out of
proportion to injury severity is a classic sign; however, in
the multiply injured patient this symptom is unreliable.
Loss of two-point discrimination and light touch is more
reliable than loss of sensation, whereas pain is exacerbated with passive dorsiflexion of the toes, which places
the intrinsic muscles on stretch. The presence or absence
of pulses and capillary refill is unreliable. The compartments of the foot should be measured in patients suspected of having compartment syndrome. Fasciotomy is
indicated when the compartment pressures exceed 30
mm Hg or is within 30 mm Hg of the diastolic pressure.
Nine compartments have been described: the medial,
lateral, four interosseous, and three central compartments, including the deep central (calcaneal) compartment that contains the quadratus plantae muscle and
the posterior tibial neurovascular bundle. Decompression is performed with dual dorsal incisions, which allow
for decompression of the first and second interosseous
compartments, the medial compartment, and the deep
central compartment. The lateral dorsal incision allows
for decompression of the two lateral interosseous compartments, the superficial and middle central compartments, and the lateral compartment. A single, medially
based fasciotomy has been described, but is technically
more difficult to perform. Sometimes the medially
based incision may be used in addition to the dorsal approach to adequately ensure decompression of the deep
calcaneal compartment. Closure is performed in a delayed fashion, either with primary closure or with splitthickness skin grafts. Exercise-induced compartment
syndrome of the foot has been recognized, most commonly affecting the medial compartment, and has been
shown to follow the same criteria for diagnosis of
chronic exertional compartment syndrome of the leg.

Annotated Bibliography
Ankle and Pilon Fractures
Day GA, Swanson CE, Hulcombe BG: Operative treatment of ankle fractures: A minimum ten-year follow-up.
Foot Ankle Int 2001;22:102-106.

481

Orthopaedic Knowledge Update 8

Foot and Ankle Trauma

In this study, 52% of patients with surgically treated bimalleolar fractures had good or excellent results; 24% of patients
had a poor outcome.

Dickson KF, Montgomery S, Field J: High energy

plafond fractures treated by a spanning external fixator
initially and followed by a second stage open reduction
internal fixation of the articular surface: Preliminary report. Injury 2001;32(suppl 4):SD92-SD98.
This protocol, used to treat patients with highly comminuted fractures, resulted in 81% good or excellent results, a
35% complication rate, and 28% radiographic arthrosis.

Dietrich A, Lill H, Engel T, Schonfelder M, Josten C:

Conservative functional treatment of ankle fractures.
Arch Orthop Trauma Surg 2002;122:165-168.
Ninety percent of patients with an isolated Weber B fracture with less than 1 mm of displacement were treated successfully with a functional brace.

Hasselman CT, Vogt MT, Stone KL, Cauley JA, Conti

SF: Foot and ankle fractures in elderly white women: Incidence and risk factors. J Bone Joint Surg Am 2003;
85:820-824.
Fibular fractures were the most common ankle fracture
(prevalence 57.6%) and occur in younger women with a high
body mass index. Fractures of the fifth metatarsal are the most
common foot fractures in women with low bone mineral density.

Hovis WD, Kaiser BW, Watson JT, Bucholz RW: Treatment of syndesmotic disruptions of the ankle with bioabsorbable screw fixation. J Bone Joint Surg Am 2002;
84:26-31.
Polyevolactic acid screws were used to stabilize syndesmotic disruptions in 33 patients. Twenty-four patients were followed for 34 months after surgery. All patients healed uneventfully, with or without fixation; no patient had osteolysis
or inflammation.

Lamontagne J, Blachut PA, Broekhuyse HM; OBrien

PJ, Meek RN: Surgical treatment of a displaced lateral
malleolus fracture: The antiglide technique versus lateral plate fixation. J Orthop Trauma 2002;16:498-502.
In this study, surgical outcome, complication rate, and requirement for hardware removal was equivalent in the two
groups.

Lehtonen H, Jarvinen TL, Honkonen S, Nyman M, Vihtonen K, Jarvinen M: Use of a cast compared with a
functional ankle brace after operative treatment of an
ankle fracture: A prospective, randomized study. J Bone
Joint Surg Am 2003;85-A:205-211.
The long-term functional outcome was equivalent between
the two groups, with similar fracture healing. The incidence of

482

postoperative wound healing complications was significantly

higher with brace treatment.

Loren GJ, Ferkel RD: Arthroscopic assessment of occult

intra-articular injury in acute ankle fractures. Arthroscopy 2002;18:412-421.
Sixty-three percent of fractures showed chondral injury,
more frequently on the medial aspect of the talus. Pronationexternal rotation fractures had a higher incidence of injury
(70%) than supination-external rotation fractures (46%).
Seventy-five percent of fractures with syndesmotic disruption
had chondral damage.

Pneumaticos SG, Noble PC, Chatziioannou SN, Trevino

SG: The effects of rotation on radiographic evaluation
of the tibiofibular syndesmosis. Foot Ankle Int 2002;23:
107-111.
Comparisons were made between the tibiofibular clear
space, tibiofibular overlap, width of the tibia and fibula, and
medial clear space. The width of the tibiofibular clear space
did not change with rotation, whereas the other parameters
did.

Sinisaari IP, Luthje PM, Mikkonen RH: Ruptured tibiofibular syndesmosis: Comparison study of metallic to
bioabsorbable fixation. Foot Ankle Int 2002;23:744-748.
Poly-L-lactide bioabsorbable screws showed equivalent results with radiographic measurements, range of motion, and
subjective outcome compared with metallic fixation.

Tornetta P III, Creevy W: Lag screw only fixation of the

lateral malleolus. J Orthop Trauma 2001;15:119-121.
Lag screw only fixation of simple oblique fractures, when
compared with plate fixation, resulted in less lateral pain, no
palpable hardware (56% plate), fewer shoe wear restrictions,
no hardware removal requirements (31% plate), and no loss
of fracture reduction.

Tornetta P III, Spoo JE, Reynolds FA, Lee C: Overtightening of the ankle syndesmosis: Is it really possible?
J Bone Joint Surg Am 2001;83-A:489-492.
There was no difference between the values for maximal
dorsiflexion before and after syndesmotic compression in cadaveric specimens fixed with a syndesmotic screw while held
in plantar flexion.

Hindfoot Fractures
Bibbo C, Anderson RB, Davis WH: Injury characteristics and the clinical outcome of subtalar dislocations: A
clinical and radiographic analysis of 25 cases. Foot Ankle
Int 2003;24:158-163.
High-energy mechanisms accounted for 68% of subtalar
dislocations; 75% of dislocations were closed, 65% were medial dislocations, and 32% of dislocations were irreducible. Radiographic evidence of arthrosis was noted in 89% of ankle

American Academy of Orthopaedic Surgeons

Chapter 39 Foot and Ankle Trauma

Orthopaedic Knowledge Update 8

and subtalar joints; the subtalar joint was more frequently
symptomatic.

Bibbo C, Lin SS, Abidi N, et al: Missed and associated

injuries after subtalar dislocation: The role of CT. Foot
Ankle Int 2001;22:324-328.
In all patients, CT scans identified additional injuries
missed on plain radiographs, which altered treatment plans in
44% of patients.

Buckley R, Tough S, McCormack R, et al: Operative

compared with nonoperative treatment of displaced
intra-articular calcaneal fractures: A prospective, randomized, controlled multicenter trial. J Bone Joint Surg
Am 2002;84-A:1733-1744.
Equivalent outcomes were noted between the two groups;
however, when patients receiving workers compensation were
removed from the comparison, significantly better results were
noted with surgical treatment.

Csizy M, Buckley R, Tough S, et al: Displaced intraarticular calcaneal fractures: Variables predicting late
subtalar fusion. J Orthop Trauma 2003;17:106-112.
Initial injury severity, Bohler angle less than 0, workers
compensation patients, heavy laborers, and those fractures initially treated nonsurgically were factors that were more likely
to lead to fusion.

Huefner T, Thermann H, Geerling J, Pape HC, Pohlemann T: Primary subtalar arthrodesis of calcaneal fractures. Foot Ankle Int 2001;22:9-14.
Primary fusion for extremely comminuted calcaneal fractures led to good functional outcome.

Ricci WM, Bellabarba C, Sanders R: Transcalcaneal talonavicular dislocation. J Bone Joint Surg Am 2002;84-A:
557-561.
Dorsal dislocation of the navicular with an associated calcaneal fracture is a severe injury, resulting in severe functional
limitations, osteomyelitis, and amputation.

Schulze W, Richter J, Russe O, Ingelfinger P, Muhr G:

Surgical treatment of talus fractures: A retrospective
study of 80 cases followed for 1-15 years. Acta Orthop
Scand 2002;73:344-351.
Ankle or subtalar arthrosis was noted in two thirds of fractures. Talar necrosis was noted in 11%; 44% had good or excellent function.

Tennent TD, Calder PR, Salisbury RD, Allen PW, Eastwood DM: The operative management of displaced
intra-articular fractures of the calcaneum: A two-centre
study using a defined protocol. Injury 2001;32:491-496.

American Academy of Orthopaedic Surgeons

A 90% satisfaction rate was noted with surgical treatment.

Bilateral injuries resulted in poorer outcome. Delay in operation was associated in a higher infection rate, whereas smoking
did not.

Midfoot Trauma
Fulkerson E, Razi A, Tejwani N: Review: Acute compartment syndrome of the foot. Foot Ankle Int 2003;24:
180-187.
An excellent review of acute foot compartment syndrome
is presented.

Glasoe WM, Allen MK, Kepros T, Stonewall L, Ludewig

PM: Dorsal first ray mobility in women athletes with a
history of stress fracture of the second or third metatarsal. J Orthop Sports Phys Ther 2002;32:560-565.
First ray hypermobility is not associated with stress fracture of the second or third metatarsal.

Kelly IP, Glisson RR, Fink C, Easley ME, Nunley JA: Intramedullary screw fixation of Jones fractures. Foot Ankle Int 2001;22:585-589.
No significant difference was reported between failure
loads of 5.0- and 6.5-mm. screws. Pull-out strength was significantly higher for the 6.5-mm screws.

Korpelainen R, Orava S, Karpakka J, Siira P, Hulkko A:

Risk factors for recurrent stress fractures in athletes.
Am J Sports Med 2001;29:304-310.
Biomechanical factors associated with multiple stress fractures include a high longitudinal arch, forefoot varus, and leglength inequality. Nearly half of female runners reported menstrual irregularity.

Mollica MB, Duyshart SC: Analysis of pre- and postexercise compartment pressures in the medial compartment of the foot. Am J Sports Med 2002;30:268-271.
Normative pressures of the medial foot compartment are
comparable to those in the leg. Previous criteria for diagnosis
of chronic exertional compartment syndrome of the leg may
be used for diagnosis of chronic exertional compartment syndrome of the foot.

Mulier T, Reynders P, Dereymaeker G, Broos P: Severe

Lisfrancs injuries: Primary arthrodesis or ORIF? Foot
Ankle Int 2002;23:902-905.
Primary complete arthrodesis resulted in inferior pain
scores, more stiffness, a loss of longitudinal arch, and higher
rate of sympathetic dystrophy.

Nunley JA, Vertullo CJ: Classification, investigation and

management of midfoot sprains: Lisfranc injuries in the
athlete. Am J Sports Med 2002;30:871-878.
Primary fusion for extremely comminuted calcaneal fractures led to good functional outcome.

483

Foot and Ankle Trauma

Peicha G, Labovitz J, Seibert FJ, et al: The anatomy of
the joint as a risk factor for Lisfranc dislocation and
fracture-dislocation: An anatomical and radiological
case control study. J Bone Joint Surg Br 2002;84:981-985.
A decreased medial depth of the mortise between the medial and middle cuneiforms increases the risk of Lisfranc injury. Lateral depth and second metatarsal length are not risk
factors.

Shah SN, Knoblich GO, Lindsey DP, Kreshak J, Yerby

SA, Chou LB: Intramedullary screw fixation of proximal
fifth metatarsal fractures: A biomechanical study. Foot
Ankle Int 2001;22:581-584.
Initial failure loads and ultimate failure loads were not significantly different for 4.5- and 5.5-mm cannulated screws.

Teng AL, Pinzur MS, Lomasney L, Mahoney L, Havey

R: Functional outcome following anatomic restoration
of tarsal-metatarsal fracture-dislocation. Foot Ankle Int
2002;23:922-926.
Anatomic reduction of the tarsometatarsal joints correlated with normal gait patterns, yet subjective patient outcomes were less than satisfactory.

Theodorou DJ, Theodorou SJ, Kakitsubata Y, Botte MJ,

Resnick D: Fractures of proximal portion of fifth metatarsal bone: Anatomic and imaging evidence of a pathogenesis of avulsion of the planter aponeurosis and the
short peroneal muscle tendon. Radiology 2003;226:857865.
Based on plain radiography, CT, and MRI, fracture of the
base of the fifth metatarsal are reported to be related to avulsion injury of the plantar aponeurosis and peroneus brevis
tendon fibers.

Compartment Syndrome
Fulkerson E, Razi A, Tejwani N: Acute compartment
syndrome of the foot. Foot Ankle Int 2003;24:180-187.
An excellent review of acute foot compartment syndrome
is presented.

Mollica MB, Duyshart SC: Analysis of pre- and postexercise compartment pressures in the medial compartment of the foot. Am J Sports Med 2002;30:268-271.
Normative pressures of the medial foot compartment are
comparable to those in the leg. Previous criteria for diagnosis
of chronic exertional compartment syndrome of the leg may
be used for diagnosis of chronic exertional compartment syndrome of the foot.

Orthopaedic Knowledge Update 8

Benirschke SK, Sangeorzan BJ: Extensive intraarticular
fractures of the foot: Surgical management of calcaneal
fractures. Clin Orthop 1993;292:128-134.
Blotter RH, Connolly E, Wasan A, Chapman MW:
Acute complications in the operative treatment of isolated ankle fractures in patients with diabetes mellitus.
Foot Ankle Int 1999;20:687-694.
Boden SD, Labropolos PA, McCowin P, Lestini WF,
Hurwitz SR: Mechanical considerations of the syndesmosis screw: A cadaver study. J Bone Joint Surg Am
1989;71:1548-1555.
Canale ST, Kelly FB Jr: Fractures of the neck of the talus: Long-term evaluation of seventy-one cases. J Bone
Joint Surg Am 1978;60:143-156.
Coughlin MJ: Calcaneal fractures in the industrial patient. Foot Ankle Int 2000;21:896-905.
Folk JW, Starr AJ, Early JS: Early wound complications
of operative treatment of calcaneus fractures: Analysis
of 190 fractures. J Orthop Trauma 1999;13:369-372.
Glasgow MT, Naranja RJ Jr, Glasgow SG, Torg JS: Analysis of failed surgical management of fractures of the
base of the fifth metatarsal distal to the tuberosity: The
Jones fracture. Foot Ankle Int 1996;17:449-457.
Gourineni PV, Knuth AE, Nuber GF: Radiographic
evaluation of the position of implants in the medial malleolus in relation to the ankle joint space: Anteroposterior compared with mortise radiographs. J Bone Joint
Surg Am 1999;81:364-369.
Hawkins LG: Fractures of the neck of the talus. J Bone
Joint Surg Am 1970;52:991-1002.
Kuo RS, Tejwani NC, Digiovanni CW, et al: Outcome after open reduction internal fixation of Lisfranc joint injuries. J Bone Joint Surg Am 2000;82:609-618.
Lauge-Hansen N: Fractures of the ankle: II. Combined
experimental surgical and experimental-roentgenologic
investigations. Arch Surg 1950;60:957-986.
Marsh JL, Bonar S, Nepola JV, Decoster TA, Hurwitz
SR: Use of an articulated external fixator for fractures
of the tibial plafond. J Bone Joint Surg Am 1995;77:
1498-1509.

Classic Bibliography

OMalley MJ, Hamilton WG, Munyak J: Fractures of the

distal shaft of the fifth metatarsal: Dancers fracture. Am
J Sports Med 1996;24:240-243.

Al-Mudhaffar M, Prasad CV, Mofidi A: Wound complications following operative fixation of calcaneal fractures. Injury 2000;31:461-464.

Myerson M, Manoli A: Compartment syndromes of the

foot after calcaneal fractures. Clin Orthop 1993;290:142150.

484

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Myerson MS: Experimental decompression of the fascial compartments of the foot-basis for fasciotomy in
acute compartment syndromes. Foot Ankle 1988;8:308314.
Quill GE Jr: Fractures of the proximal fifth metatarsal.
Orthop Clin North Am 1995;26:353-361.
Sasse M, Nigg BM, Stefanyshyn DJ: Tibiotalar motion:
Effect of fibular displacement and deltoid ligament
transaction: In vitro study. Foot Ankle Int 1999;20:733737.
Sanders R, Fortin P, DiPasquale T, Walling A: Operative
treatment in 120 displaced intraarticular calcaneal fractures: Results using a prognostic computed tomography
scan classification. Clin Orthop 1993;290:87-95.
Thordarson DB, Motamed S, Hedman T, Ebramzadeh
E, Bakshian S: The effect of fibular malreduction on
contact pressures in an ankle fracture malunion model.
J Bone Joint Surg Am 1997;79:1809-1815.

Chapter 39 Foot and Ankle Trauma

reduction and internal fixation of talar neck fractures.
Foot Ankle Int 1996;17:742-747.
Wyrsch B, McFerran MA, McAndrew M, et al: Operative treatment of fractures of the tibial plafond: A randomized, prospective study. J Bone Joint Surg Am 1996;
78:1646-1657.
Yablon IG, Heller FG, Shouse L: The key role of the lateral malleolus in displaced fractures of the ankle. J Bone
Joint Surg Am 1977;59:169-173.
Yamaguchi K, Martin CH, Boden SD, Labropoulos PA:
Operative treatment of syndesmotic disruptions without
use of a syndesmotic screw: A prospective clinical study.
Foot Ankle Int 1994;5:407-414.
Zufferey P, Boubaker A, Bischof Delaloye A, So AK,
Duvoisin B: Prognostic aspects of scintigraphy and MRI
during the first 6 months of reflex sympathetic dystrophy of the distal lower limb: A preliminary prospective
study of 4 cases. J Radiol 1999;80:373-377.

Thordarson DB, Triffon MJ, Terk MR: Magnetic resonance imaging to detect avascular necrosis after open

American Academy of Orthopaedic Surgeons

485

Chapter

40

Foot and Ankle Reconstruction

Mark E. Easley, MD

Forefoot
Hallux Valgus
Hallux valgus deformity results from progressive deviation and pronation of the great toe and medial deviation
of the first metatarsal (metatarsus primus varus). Hallux
valgus deformity is most common in females and shoewearing societies; there is a high prevalence of bunion
deformity in American women in the fourth to sixth decades of life.
The patient with hallux valgus deformity experiences
difficulty with shoe wear, particularly at the prominent
medial eminence. The patient should be evaluated while
standing to determine the presence of pes planus and
the severity of the deformity. With the patient seated, assessment of medial eminence tenderness, first metatarsophalangeal (MTP) joint range of motion, and hypermobility of the first tarsometatarsal (TMT) joint are
performed. First ray hypermobility remains a diagnostic
challenge because physiologically normal values of first
MTP joint mobility have not been defined. Weightbearing radiographs are required to determine the severity of the bone and joint malalignment, presence of
arthrosis, and other factors that influence surgical treatment. Radiographic assessment of the distal metatarsal
articular angle (DMAA) continues to present diagnostic
challenges, and interobserver reliability for DMAA assessment remains poor (Figures 1 and 2).
Nonsurgical treatment involves shoe stretching at
the medial eminence or shoes with a wider toe box. Surgical correction is indicated when shoe modifications
prove ineffective; surgery should not be performed for
cosmetic reasons alone. More than 100 procedures have
been described for correction of hallux valgus; the orthopaedic surgeon treating hallux valgus need not be familiar with all of them but needs an adequate knowledge base to address all aspects of hallux valgus
deformity (Figure 3). The distal chevron osteotomy is
reserved for mild deformity. A combination of a proximal metatarsal osteotomy and distal soft-tissue procedure is indicated for moderate to severe deformity.
Many proximal first metatarsal osteotomies have been

American Academy of Orthopaedic Surgeons

Figure 1 Schematic showing intermetatarsal, hallux valgus, and hallux valgus interphalangeus angles. The angles in parentheses are angles considered physiologic in
patients without hallux valgus. (Reproduced with permission from Coughlin MJ: Hallux
valgus in men: Effect of the distal metatarsal articular angle in the hallux valgus
correction. Foot Ankle Int 1997;18:463-470.)

described. Although the crescentic, (proximal) chevron,

and proximal oblique osteotomies are most popular, the
Scarf osteotomy has gained wider acceptance. A symptomatic hallux valgus associated with hypermobility
and/or first TMT joint degeneration warrants correction
via a combination first TMT joint arthrodesis and distal
soft-tissue procedure (Lapidus procedure). A medial
closing wedge osteotomy of the proximal phalanx (Akin
procedure) is performed to correct hallux valgus interphalangeus, but may also serve as an adjunctive procedure for severe hallux valgus addressed with proximal
correction and distal soft-tissue procedure. The Keller
resection arthroplasty (base of proximal phalanx resec-

487

Foot and Ankle Reconstruction

Orthopaedic Knowledge Update 8

Figure 3 A schematic representation of a variety of osteotomies used to correct hallux valgus deformity. (1) Akin procedure (corrects hallux valgus interphalangeus). (2)
Reverdin distal medial closing wedge (corrects increased DMAA). (3) Proximal first
metatarsal osteotomy (corrects increased intermetatarsal angle). (4) Medial cuneiform
opening wedge osteotomy (corrects increased intermetatarsal angle). (Reproduced
with permission from Coughlin M, Carlson R: Treatment of hallux valgus with an increased distal metatarsal articular angle: Evaluation of double and triple first ray
osteotomies. Foot Ankle Int 1999;20:762-770.)

Figure 2 Radiograph showing hallux valgus associated with an increased DMAA. Joint
surfaces are congruent (not subluxated). (Reproduced from Coughlin MJ: Juvenile hallux valgus, in Nunley JA, Pfeffer GB, Sanders RW, Trepman E (eds): Advanced Reconstruction Foot and Ankle. Rosemont, IL, American Academy of Orthopaedic Surgeons,
2004, pp 59-66.)

tion) risks creating a cock-up toe deformity and affords

a poor potential for correction of hallux valgus; this procedure has largely been abandoned.
The DMAA must be considered in conjunction with
first ray malalignment. Typically, hallux valgus produces
an incongruent joint, and joint congruency is reestablished with surgical correction of the hallux valgus deformity. However, hallux valgus may be associated with
a congruent joint (increased DMAA) (Figure 2). Conventional correction to realign the hallux creates an incongruent MTP joint that will become stiff or lead to recurrence of deformity. To correct the DMAA in patients
with mild hallux valgus, a biplanar distal chevron osteotomy is recommended; in patients with moderate to severe disease, a proximal first metatarsal osteotomy or
Lapidus procedure should be combined with a distal
first metatarsal closing wedge osteotomy (Reverdin procedure) (Figure 3).
The most common complication of hallux valgus surgery is recurrence. Recurrence is usually observed when

488

indications are not carefully followed (for example, a

distal metatarsal osteotomy is performed when proximal
metatarsal correction is required). Loss of correction at
the osteotomy is generally avoided with adequate fixation and protected weight bearing in the immediate
postoperative period. A proximal metatarsal osteotomy
may result in a dorsiflexion malunion and transfer metatarsalgia (overload of lesser metatarsal heads). Proximal
first metatarsal osteotomies performed from the medial
aspect (oblique, chevron, Scarf) minimize the potential
for first metatarsal fixation in dorsiflexion. Hallux varus
may occur with overcorrection of the first intermetatarsal angle, excessive lateral capsular release (and overtightening of the medial capsule), overresection of the
medial aspect of the first metatarsal head, and addition
of lateral sesamoidectomy to the distal soft-tissue procedure (original McBride procedure). Recurrence may be
salvaged with revision of the original procedure, revision to a more proximal procedure that affords greater
correction, or first MTP or TMT joint arthrodesis. Continued pain and stiffness in the first MTP joint following
hallux valgus correction typically warrants first MTP
joint arthrodesis. Mild hallux varus is rarely symptomatic. With severe hallux varus, however, revision surgery
may be necessary.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Hallux Rigidus
Hallux rigidus is degenerative joint disease of the first
MTP joint. A positive family history of hallux rigidus is
more common in patients with bilateral disease, whereas
patients with unilateral disease typically recall trauma to
the great toe. The hallux MTP joint articular degeneration is associated with osteophyte formation, particularly dorsally on the first metatarsal head that creates
dorsal impingement. Patients typically have great toe
pain during push-off and forced dorsiflexion of the hallux MTP joint. The first MTP joint is stiff and has a tender dorsal prominence. Dorsiflexion is limited and osteophyte impingement is painful. Plantar flexion also
creates symptoms as the extensor tendons and sensory
nerve are stretched over the prominence. Pain at the
midrange of the motion arc is indicative of advanced
(global) arthritis. Radiographic assessment demonstrates joint space narrowing; widening or flattening of
the metatarsal head; and medial, lateral, and particularly
dorsal osteophytes. An association of hallux rigidus with
metatarsus primus elevatus has not been substantiated.
Nonsurgical treatment includes activity modification
that does not require great toe dorsiflexion (bicycle),
stretching of the shoe toe box, a carbon fiber insert or
stiff-soled shoe, and/or a rocker bottom shoe modification. The mainstay of surgical management for mild to
moderate disease remains dorsal cheilectomy to remove
dorsal osteophytes and one fourth to one third of the
dorsal articular surface from the metatarsal head. To improve the relative dorsiflexion, a dorsiflexion osteotomy
of the proximal phalanx (Moberg procedure) can be
added to decompress the joint. In patients with advanced disease, the standard of surgical care remains
first MTP joint arthrodesis. However, capsular interpositional arthroplasty has gained acceptance as a surgical
alternative for advanced hallux rigidus. Proximal phalanx base resection (Keller procedure) and dorsal cheilectomy are combined with joint soft-tissue interposition
of both the dorsal capsule and extensor hallucis brevis
tendon. To avoid destabilizing the proximal phalanx, the
Keller procedure may be performed obliquely to retain
the attachment of the flexor hallucis brevis. Alternatively, the proximal phalanx and metatarsal head can be
reamed with spherical reamers at the joint to stabilize
an anchovy soft-tissue interposition. First MTP joint
arthroplasty remains controversial. Although several
new prosthetic designs and technique modifications
have been introduced, risk of implant failure, osteolysis,
and synovitis have limited the use of first MTP joint arthroplasty in orthopaedic foot and ankle surgery. The
most reliable surgical option for symptoms related to
hallux MTP joint arthritis is arthrodesis. Successful fusion often reestablishes a more physiologic plantar pressure pattern. Gait pattern does not return to that of the

American Academy of Orthopaedic Surgeons

Chapter 40 Foot and Ankle Reconstruction

uninvolved (contralateral) side, but gait analysis suggests that patients compensate well for the fused hallux.
Multiple techniques have been described for first
MTP joint arthrodesis. Even though flat cuts provide excellent surfaces for healing, cup-and-cone techniques facilitate fine adjustments for proper alignment while
maintaining adequate bony contact. Biomechanical testing demonstrates that a combination of a compression
screw and a dorsal plate typically represents the most
stable construct. The hallux should be in slight valgus
relative to the first metatarsal without impinging on the
second toe and neutral to slight dorsiflexion relative to
the plantar aspect of the foot. Overpronation must also
be avoided to prevent painful medial hallux callus formation.

Sesamoid Disorders
The sesamoid complex includes the tibial and fibular
sesamoids, the intersesamoidal ligament, the flexor hallucis brevis tendon, and the plantar plate. Disorders specific to the sesamoids probably exist on a continuum
and comprise sesamoiditis, stress fractures, and osteonecrosis. Any of these may involve arthritis of the metatarsal head/sesamoid articulation. Females are more prone
to sesamoid problems, and the medial sesamoid is more
commonly affected than the lateral sesamoid. Sesamoid
disorders typically produce pain on the plantar first
metatarsal head with weight bearing, particularly during
push-off. Simple palpation may not clearly confirm the
diagnosis; a passive axial compression test may facilitate
a diagnosis of sesamoiditis. The sesamoids are palpated,
the hallux MTP joint is maximally dorsiflexed, compression is applied just proximal to the sesamoids, and the
toe is passively plantar flexed. This generally reproduces
the patients symptoms in sesamoiditis. Radiographs
may reveal sesamoid irregularity, although lucency in a
sesamoid may only represent bipartite sesamoid. Radiographs of the asymptomatic contralateral foot are usually helpful in distinguishing a bipartite sesamoid from
an acute fracture. Lateral and oblique radiographs to diagnose sesamoid abnormalities may be difficult to interpret, and therefore dynamic fluoroscopy, bone scan, CT,
and/or MRI are useful. Increased uptake on bone scan
and MRI signal changes may indicate acute or subacute
fracture. Sesamoid views (axial images of the foot with
the hallux dorsiflexed) may define sesamoid irregularities such as fragmentation (osteonecrosis) or metatarsal
head-sesamoid joint space narrowing (osteoarthritis).
CT provides greater detail of metatarsal head-sesamoid
degenerative changes and may confirm the diagnosis.
Nonsurgical treatment of sesamoid problems includes sesamoid pressure relief/unloading with activity
modification, temporary avoidance of dorsiflexion, and
a boot or cast. A 6- to 8-week period of unloading is followed by management with an orthotic device. Sesa-

489

Foot and Ankle Reconstruction

Orthopaedic Knowledge Update 8

Figure 4 Schematic lateral view of a Weil lesser metatarsal shortening osteotomy stabilized with screw fixation. (Reproduced from Deland JT: Angular deformities of the second
toe, in Nunley JA, Pfeffer GB, Sanders RW, Trepman E (eds): Advanced Reconstruction Foot and Ankle. Rosemont, IL, American Academy of Orthopaedic Surgeons, 2004, pp
77-84.)
moiditis is usually receptive to nonsurgical management; successful conservative management of fractures
and osteonecrosis is less predictable. Repair with bone
grafting may be possible, but generally excision of the
diseased sesamoid is necessary. The entire sesamoid
need not be routinely excised; instead, depending on
specific pathology, excision of only the proximal pole or
plantar shaving suffices.

Lesser Toes
Lesser toe deformities include hammer toes (extended
MTP joint, flexed proximal interphalangeal joint, extended distal interphalangeal joint), mallet toes (flexed
distal interphalangeal joints), and claw toes (extended
MTP joint, flexed proximal interphalangeal and distal
interphalangeal joints). Hammer toes often arise secondary to tight shoe wear and are sometimes associated
with long second and/or third rays. Mallet toes may result from tighter shoe wear but are associated with degenerative distal interphalangeal joint arthritis. Claw
toes occur with intrinsic muscle weakness, creating an
imbalance of the extrinsic and intrinsic foot musculature, and may be linked to neuromuscular disease, cavus
foot deformity, neuropathy, and sequelae of compartment syndrome. Chronic MTP joint synovitis and/or inflammatory arthritis may lead to attenuation of the
plantar plate and capsule, resulting in MTP joint extension and eventual clawing. If simultaneous collateral ligament attenuation should occur, medial or lateral deviation ensues, ultimately leading to a crossover toe
deformity. Frequently, a crossover second toe deformity
is observed with hallux valgus. With hammer and claw
toe deformities, the plantar fat pad may be displaced
distally, subjecting the metatarsal head to overload, callus formation, and potential intractable plantar keratosis. Radiographs demonstrate toe deformity and may reveal a relatively long metatarsal that contributes to the
deformity.

490

Nonsurgical treatment of flexible deformities includes pads, inserts, and extra depth to the toe box.
Hammer toe slings (Budin slings) extend the flexible
hammer or claw toes while orthotic device or metatarsal
pad can provide support to compensate for the distally
migrated fat pad. Intra-articular steroid injections can
be useful to decrease symptoms related to synovitis, but
must be used judiciously as excessive exposure to steroid may further weaken the attenuated capsular structures and worsen deformity. A flexible hammer or claw
toe can be corrected with a flexor-to-extensor tendon
transfer, whereas a fixed toe deformity requires a proximal interphalangeal joint resection arthroplasty or arthrodesis. Long-standing fixed hammer or claw toe deformity may also require extensor tendon lengthening,
dorsal MTP joint capsulotomy, and collateral ligament
release. If associated with a relatively long second metatarsal, a metatarsal shortening osteotomy can be performed with preservation of the collateral ligaments because the osteotomized metatarsal head relies solely on
its extracapsular blood supply through the collateral ligaments. Although multiple bony procedures for metatarsal shortening have been described, the Weil osteotomy
is currently favored (Figure 4). A transverse metatarsal
cut is made in line with the plantar foot through the
dorsal 5% to 10% of the articular cartilage. The metatarsal head is translated proximally several millimeters,
but not beyond the adjacent lesser metatarsal head unless it is to be shortened as well. Fixation is typically
performed with a low-profile screw. The overlapping
bone on the proximal fragment is resected to create the
new contour of the shortened metatarsal.

Bunionette Deformity
Bunionette deformities, to an extent, mirror hallux valgus deformities. Bunionette deformities can be grouped
into three types: type I, enlarged/prominent fifth metatarsal head; type II, congenital lateral fifth metatarsal
bow; and type III, increased 4-5 intermetatarsal angle.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Patients present with pain over the lateral fifth metatarsal head and difficulty with shoe wear because of fifth
metatarsal head pressure. Occasionally, a bunionette will
be associated with plantar fifth metatarsal head pain
and remain symptomatic even when shoes are removed.
Physical examination reveals lateral and/or plantar fifth
metatarsal head tenderness, a wide forefoot (in types II
and III) and medial deviation of the fifth toe. AP radiographs define the deformity. Shoe stretching over the
lateral fifth metatarsal and shoes with a wider toe box
are usually sufficient to relieve symptoms. Surgical
treatment is indicated for bunionettes not responding to
shoe modifications. All three types generally require a
lateral capsulotomy, lateral eminence excision, and lateral capsular plication. Type II can be corrected with a
distal metatarsal osteotomy that displaces the metatarsal head medially. More severe type II and type III deformities are managed with a midshaft long oblique osteotomy analogous to the proximal oblique osteotomy
of the first metatarsal head. In the fifth metatarsal, the
osteotomy is not performed proximally, given the watershed area complicating healing of Jones fractures.

Midfoot Arthritis
Midfoot arthritis may be primary, inflammatory, or posttraumatic, with primary arthritis as the most common
cause. The midfoot articulations comprise the naviculocuneiform and metatarsocuneiform/cuboid joints. Patients have pain with weight bearing in the dorsal midfoot and arch, particularly during push-off. Examination
reveals tenderness at the midfoot joints, especially with
stress at the TMT joints (dorsiflexion/plantar flexion,
twisting of midfoot, and forced forefoot abduction). Because the TMT joints may be difficult to differentiate, a
piano key test has been proposed in which the hindfoot is stabilized while a plantar force is applied to the
associated metatarsal head (like striking an individual
piano key). With the patient bearing weight, loss of the
longitudinal arch is appreciated and is often associated
with forefoot abduction. Radiographs confirm these
findings with forefoot abduction (AP radiographs) and
a break in the physiologic talar declination angle (lateral radiographs). Subtle joint space narrowing in the
TMT and naviculocuneiform joints may be detected
with standard weight-bearing foot radiographs. Nonsurgical treatment includes longitudinal arch support,
stiffer soled shoes or orthotic devices, and/or rocker bottom shoes. Steroid injections may be both diagnostic
and temporarily therapeutic. Because these are small
joints, fluoroscopic guidance is helpful to deliver the steroid directly to the involved articulation(s). Surgical
management is indicated when conservative measures
fail. The recommended surgical treatment of midfoot
degenerative arthritis is arthrodesis of the naviculocune-

American Academy of Orthopaedic Surgeons

Chapter 40 Foot and Ankle Reconstruction

iform and/or first through third TMT joints because
these are considered nonessential joints and physiologically the foot should be relatively rigid in this location.
In contrast, it is essential to maintain motion of the 4-5
metatarsal-cuboid articulations to preserve the foots accommodative function during the stance phase of gait.
Infrequently, advanced degeneration involves the 4-5
metatarsal-cuboid articulations, and use of interpositional tendon arthroplasties have proven effective in select patients. In rare instances, such as Charcot midfoot
neuroarthropathy with midfoot collapse, the lateral column of the midfoot may also be fused to ensure adequate stability.

Hindfoot and Ankle

Ankle/Subtalar Instability
Although ankle sprains are the most common sports injury, only 20% of patients develop residual symptoms.
Many of these symptoms represent associated injuries
such as peroneal tendon tears, peroneal tendon
subluxation/dislocation, fractures of the lateral process
of the calcaneus, and osteochondral lesions. These diagnoses, among others, should be considered in patients
with chronic symptoms following ankle sprain. Ankle instability that is not related to any associated pathology
is categorized into functional instability (pain that
causes the ankle to give way) and mechanical instability
(weakened static ankle restraints that are responsible
for excessive lateral ankle subluxation and resultant
pain). Furthermore, ankle (tibiotalar) instability is frequently associated with subtalar (talocalcaneal) instability. CT analysis suggests that hindfoot varus and an altered mortise position (with a more posterior fibular
position) is more prevalent in patients with chronic lateral ankle instability. Patients with mechanical instability report a history of chronic ankle sprains but are relatively pain free if the ankle is not injured. Examination
may reveal tenderness over the anterior talofibular ligament and/or the calcaneofibular ligament, and the diagnosis is confirmed with anterior drawer and inversion
testing. Although a clinical grading scheme (grades I
through III) and mechanical devices to test anterior
drawer/talar tilt exist, stress radiographs probably offer
the best objective demonstration of tibiotalar instability.
Abnormal tibiotalar tilt has been defined anywhere
from 3 to 15 relative to the contralateral side and forward talar translation on the tibia greater than 3 mm
than on the contralateral side. Subtalar instability remains a diagnostic challenge. Clinical examination typically fails to distinguish tibiotalar from subtalar instability. Brodens views have limited value. A recently
developed ultrasound method may be useful, but remains operator dependent.

491

Foot and Ankle Reconstruction

Ankle and hindfoot sprains are initially treated with
the familiar regimen of rest, ice, application, compression, and elevation, early weight bearing, and functional
rehabilitation. Generally, 15% to 30% of residual symptoms after routine initial management can be attributed
to peroneal weakness caused by inadequate rehabilitation. Treatment with more directed and supervised physical therapy is usually successful in these situations.
Surgical intervention is considered when proprioceptive and peroneal rehabilitation for chronic mechanical
instability is unsuccessful. Currently, anatomic reconstruction is favored. When the condition of residual lateral ligamentous tissues is adequate for anatomic repair,
a modified Brostrm procedure is recommended, with
imbrication of the anterior talofibular ligament and/or
the calcaneofibular ligament and augmentation with the
inferior extensor retinaculum (Gould modification).
This anatomic reconstruction has been shown to have
superior outcome when compared with a functional rehabilitation program and to a nonanatomic reconstruction using the peroneus brevis tendon (Evans procedure). Recently, advanced methods of lateral ankle
stabilization have been introduced; suture anchors (bioabsorbable and metallic) to reattach/advance the anterior talofibular ligament and calcaneofibular ligament to
the distal fibula have gained popularity. Traditionally,
with insufficient residual lateral ankle ligamentous tissue, reconstruction involved harvest of part or all of a
peroneal tendon. However, concerns over loss of the dynamic stabilizers have prompted greater interest in autologous tendon transfers. Hamstring autografts, including the semitendinosus or gracilis, have been used with
greater frequency to reestablish the deficient anterior
talofibular ligament and calcaneofibular ligament without compromising the dynamic stabilizing effect of the
intact peroneal tendons. If hindfoot varus coexists with
mechanical lateral ankle instability, then a valgusproducing calcaneal osteotomy is typically recommended. Furthermore, hindfoot varus may be driven by
a plantar flexed first metatarsal, which can be diagnosed
via the Coleman block test. In this situation, surgical
correction should include a dorsiflexion first metatarsal
osteotomy performed judiciously to avoid overload of
the lesser metatarsal heads.
Although isolated subtalar instability may exist, reconstructive procedures restricted to the subtalar joint
have not been clearly defined. Reconstructive procedures that cross both the ankle and subtalar joints (calcaneofibular ligament repair/reconstruction and augmentation with the inferior extensor retinaculum)
remain the mainstay of treatment.

Ankle Arthritis
Ankle arthritis may develop secondary to trauma
(ankle/pilon fractures), chronic ankle instability, os-

492

Orthopaedic Knowledge Update 8

teonecrosis of the talus, and inflammatory arthritides,
but may also develop as primary ankle degeneration.
The incidence of primary and inflammatory ankle arthritis is uncommon relative to the incidence of hip and
knee arthritis. Comparative studies of ankle versus hip
and knee cartilage suggest that cartilage in the ankle
possesses factors that protect it from primary degeneration; by far the most common etiology of ankle arthritis
is trauma. With isolated ankle arthritis, the patient reports pain in the anterior aspect of the ankle with
weight bearing and particularly with push-off. Whereas
diffuse ankle arthritis may also produce medial and lateral pain, early degenerative disease creates symptoms
confined to the anterior ankle. Physical examination
demonstrates tenderness about the ankle, pain with
range of motion, and restricted range of motion (particularly ankle dorsiflexion). Evaluation while the patient
is bearing weight may reveal malalignment, which is often most evident during ambulation. Weight-bearing radiographs of the ankle confirm the diagnosis, and
weight-bearing radiographs of the foot are important
to identify any associated hindfoot malalignment/
degeneration that may have a bearing on surgical management.
Nonsurgical treatment includes anti-inflammatory
agents, activity modification, bracing, rocker bottom
shoe modification, and corticosteroid injection. Failure
of conservative management is an indication for surgical
management. Mild degenerative disease with impingement during dorsiflexion can be effectively managed
with anterior ankle cheilectomy (exostectomy of anterior ankle osteophytes). Ankle arthrodesis remains the
gold standard for surgical treatment of diffuse ankle arthritis. With minimal malalignment, arthroscopic or
mini-arthrotomy techniques offer the advantage of a
minimally invasive approach and generally shorter time
to fusion when compared with traditional arthrotomy
techniques. The transfibular approach is recommended
with increased malalignment. Because long-term
follow-up of ankle arthrodesis demonstrates that many
patients develop adjacent hindfoot arthritis, some surgeons advocate that ankle arthrodesis be performed
through an anterior approach to facilitate potential future conversion of the fusion to an ankle replacement.
Improved union rates have been reported in complex
hindfoot arthrodeses with increased risk for nonunion
using implantable bone stimulation. Smokers are at 2.7
times greater risk for nonunion than nonsmokers. In patients with concomitant ankle and hindfoot arthritis, tibiotalocalcaneal or pantalar arthrodesis may be considered. Reported techniques of retrograde intramedullary
nailing and blade plate fixation (inserted via lateral or
posterior approaches) have favorable union rates and
provide acceptable pain relief.
To limit development of hindfoot arthritis following
ankle fusion or to manage patients with concomitant an-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

kle and hindfoot arthritis, several attractive surgical alternatives to ankle arthrodesis that maintain ankle motion continue to show promise. Joint distraction
arthroplasty with or without joint dbridement involves
maintaining 5 mm of ankle joint distraction with thin
wire ring fixation. Patients are allowed to bear weight
during the 3 months of joint distraction. Early to intermediate results suggest that 75% of patients experienced improvement of symptoms and did not require
further surgical management. Fresh tibiotalar shell allografts have been used in carefully selected patients.
An osteochondral allograft comprising both aspects of
the tibiotalar joint replaces the patients corresponding
resected articular surfaces. To optimize the fit of the allograft, total ankle replacement cutting guides are used.
Early results are encouraging, but the technique remains
experimental. Improved component design and surgical
technique for modern-generation total ankle arthroplasty have improved results over those for firstgeneration implants. Although several different prostheses, each with unique design features, have been
introduced worldwide, only the semi-constrained, fixedbearing Agility ankle (DePuy, Warsaw, IN) has been approved by the US Food and Drug Administration
(FDA) (Figure 5). The Agility ankle technique warrants
a syndesmotic arthrodesis to increase the surface area
and support for the tibial component. The mobile bearing Scandinavian Total Ankle Replacement (WaldemarLink, Hamburg, Germany) prosthesis affords the advantage of limited bone resection, a resurfacing talar
component, and the mobile bearing that reduces constraint and provides modularity to improve soft-tissue
balance. As of this writing, the Scandinavian Total Ankle
Replacement implant was in clinical trial, awaiting FDA
approval. Although techniques for total ankle arthroplasty have been refined, the procedure remains technically demanding. A steep learning curve has been demonstrated for total ankle arthroplasty. As with total hip
and total knee arthroplasty, proper alignment and softtissue balance are crucial for successful outcome. Current recommendations are for total ankle arthroplasty
to be performed by surgeons who have completed special training for this technique, either through fellowship
or learning center experience.

Osteochondral Lesions of the Talus

Osteochondral lesions of the ankle are commonly observed anterolaterally and posteromedially on the talar
dome. Not all of these lesions are symptomatic and may
represent incidental findings. Most osteochondral lesions of the talus are a result of ankle trauma (sprain/
fracture), but patients do not always recall a specific
traumatic event. Ankle arthroscopy performed simultaneously with open reduction and internal fixation of ankle fractures or in patients with chronic ankle sprains

American Academy of Orthopaedic Surgeons

Chapter 40 Foot and Ankle Reconstruction

reveals a high incidence of osteochondral injuries. No
clear association between focal osteochondral lesions of
the talus and development of diffuse ankle osteoarthritis has been identified. Although radiographs often suggest an osteochondral lesion of the talus, findings may
be subtle. Occasionally, a loose body (detached fragment) is evident. MRI is the favored screening tool to
identify osteochondral lesions of the talus, and both
MRI and CT provide useful information in defining specific characteristics. Although classification schemes for
osteochondral lesions of the talus have been developed
using both MRI and CT, arthroscopic evaluation is
probably most accurate.
Nonsurgical management of symptomatic osteochondral lesions of the talus consists of immobilization
and protective weight bearing to promote stabilization
of the osteochondral fragment and nonsteroidal antiinflammatory drugs and/or steroid injection for pain
relief/controlling inflammation. Adult patients rarely experience spontaneous healing of osteochondral lesions
of the talus, but some authors have recently reported
good to excellent results at intermediate-term follow-up
with nonsurgical treatment of osteochondral lesions of
the talus thought only to respond to surgical intervention. Although arthroscopic dbridement and drilling/
microfracture remains the standard of surgical care for
symptomatic osteochondral lesions of the talus (good to
excellent results of 70% to 90% at intermediate followup), the concern remains that replacing larger defects
with fibrocartilage and not hyaline cartilage may not be
a sensible solution. Although cartilage resurfacing procedures, including osteochondral autograft/allograft
transfer (mosaicplasty) and autologous chondrocyte
transplantation (Carticel procedure), continue to be
viewed as salvage procedures, some consideration is being given to applying these techniques as primary surgical treatment of larger osteochondral lesions of the talus, particularly those associated with subchondral cysts
(Figure 6). Outcome of osteochondral transfer from the
knee to the ankle (osteochondral autograft/allograft
transfer and mosaicplasty) have been promising at
short- to intermediate-term follow-up, with 88% to 94%
good to excellent results for primary and revision surgeries. Autologous chondrocyte transplantation has also
proven effective at early follow-up. Cartilage resurfacing
in the ankle often warrants medial or lateral malleolar
osteotomies to provide adequate access to the osteochondral lesions of the talus. Currently, malleolar osteotomy techniques are being refined and potential local cartilage harvest sites are being identified to
decrease the morbidity to the ankle and ipsilateral knee.
Lateral defects, particularly those associated with lateral
ligamentous instability, may be treated with release and
subsequent tightening of the anterior talofibular ligament and calcaneofibular ligament.

493

Foot and Ankle Reconstruction

Orthopaedic Knowledge Update 8

Figure 5 Several currently available total ankle prosthesis are shown. A, Agility Total Ankle System. (Courtesy of DePuy, Warsaw, IN.) B, STAR system. (Courtesy of Waldemar-Link,
Hamburg, Germany.) C, Buechel-Pappas ankle system. (Courtesy of Endotec, South Orange, NJ.) D, Hintegra ankle system. (Courtesy of New Deal, Vienne, France.)

Insertional Achilles Tendinopathy

Insertional Achilles tendinopathy is an enthesopathy of
the Achilles tendon insertion that most likely is a result
of cumulative trauma/repetitive stress. Although inflammation accompanies this disease process, the primary
pathology is tendon degeneration at the Achilles tendon
insertion. Patients report pain directly at the junction of

494

the Achilles tendon on the calcaneus with direct pressure and during push-off; swelling at the insertion limits
shoe wear with a hard heel counter. Symptoms are exacerbated by walking an incline or during activities that
cause dorsiflexion of the ankle. Physical examination
will demonstrate an intact Achilles tendon, but often
with hyperdorsiflexion of the ankle (secondary to Achil-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

les tendon attenuation). Lateral radiographs demonstrate calcification at the Achilles tendon insertion and
frequently a prominent posterior calcaneal tuberosity.
Nonsurgical treatment is successful in most instances, and includes activity modification, use of a cam
walker with a posterior heel relief, modalities to diminish associated inflammation, and a heel lift. In approximately 20% to 25% of patients, failure of nonsurgical
measures (after at least 6 months) prompts surgical
management. Current surgical treatment includes dbridement of the Achilles tendon insertion and calcaneal
exostectomy. Medial and/or lateral approaches have
been described for this procedure, but may not permit
adequate dbridement of the Achilles tendon insertion.
A central approach facilitates such dbridement, but necessitates detachment of at least 50% of the Achilles
tendon from the calcaneus, and usually requires reattachment of the residual tendon fibers with suture anchors to the residual calcaneus. When the tendon insertion is severely degenerated, necessitating resection of a
substantial portion of the insertion, then augmentation
with a flexor hallucis longus tendon transfer is recommended. A long flexor hallucis longus tendon harvest
(flexor hallucis longus division in the plantar foot) delivers approximately 3 cm more tendon than a short flexor
hallucis longus tendon harvest (flexor hallucis longus division at posteromedial ankle).

Acquired Flatfoot Deformity

Posterior tibial tendon dysfunction is the most common
etiology for the adult acquired flatfoot deformity. Posterior tibial tendon insufficiency leads to gradual loss of
the longitudinal arch, hindfoot valgus, forefoot abduction and forefoot varus/supination. Over time, the deformity may become fixed and even result in deltoid ligament attenuation with resultant valgus talar tilt. This
progressive deformity has been categorized into four
stages. Clinical evaluation is characterized by pain and
tenderness along the posterior tibial tendon and inability to perform a single limb heel rise. In the early stages
of the disease, single limb heel rise may be possible but
painful; eventually, the heel fails to turn into physiologic
varus, and ultimately unsupported single limb heel rise
is no longer possible. With advancing disease, subfibular
(calcaneofibular) impingement develops with tenderness over the compressed peroneal tendons. In fact, initial medial pain subsides and subfibular lateral foot pain
produces the greatest symptoms. Loss of the longitudinal arch, hindfoot valgus, and forefoot abduction (too
many toes sign) are evident with progressive posterior
tibial tendon attenuation. Radiographic evaluation confirms pes planus alignment. The lateral radiograph demonstrates loss of the longitudinal arch; moderate to severe talonavicular subluxation suggests spring ligament
compromise. On the AP radiograph, talar head uncover-

American Academy of Orthopaedic Surgeons

Chapter 40 Foot and Ankle Reconstruction

Figure 6 CT scan showing a subchondral cyst of the medial talar dome. Subchondral
cysts generally respond poorly to arthroscopic dbridement and drilling. Recently developed cartilage repair techniques have shown promise in the primary and secondary
surgical treatment of these lesions. (Reproduced from Scranton PE Jr: Osteochondral
lesions of the talus, in Nunley JA, Pfeffer GB, Sanders RW, Trepman E (eds): Advanced
Reconstruction Foot and Ankle. Rosemont, IL, American Academy of Orthopaedic Surgeons, 2004, pp 261-266.)

ing by the navicular demonstrates forefoot abduction.

Weight-bearing AP and mortise ankle radiographs must
be obtained to detect valgus talar tilt indicative of deltoid ligament attenuation.
Treatment of stage I flatfoot deformity (posterior
tibial tendinitis) is nonsurgical with initial immobilization and the administration of nonsteroidal antiinflammatory drugs, followed by gradual progression to
physical therapy and orthotic use. In selected patients,
surgical dbridement of the posterior tibial tendon
sheath may relieve recalcitrant stenosing tenosynovitis.
Stage II flatfoot deformity (posterior tibial tendinopathy with flexible hindfoot) can be effectively managed
with functional bracing using a University of California
at Berkely Laboratory or ankle-foot orthosis. Although
bracing supports the foot, it cannot reverse posterior
tibial tendinopathy. Surgical management is typically
joint sparing with flexor digitorum or flexor hallucis longus tendon transfer combined with either medial displacement calcaneal osteotomy or lateral column
lengthening. Medial displacement calcaneal osteotomy
is effective in protecting a posterior tibial tendon reconstruction and at least partially restoring longitudinal
arch alignment and correcting hindfoot valgus, but
proves less effective than lateral column lengthening
when deformity is associated forefoot abduction. The
medial displacement calcaneal osteotomy generally is
associated with high union and low complication rates;

495

Foot and Ankle Reconstruction

in contrast, lateral column lengthening (either through
the anterior calcaneus or calcaneocuboid joint) is associated with a higher complication rate, particularly with
regard to nonunion. With proper technique, use of structural autograft and allograft yields equal and favorable
results. Associated spring ligament attenuation can usually be corrected with soft-tissue plication; however,
larger spring ligament defects represent a greater challenge. The optimal spring ligament reconstruction has
yet to be defined. Currently, arthrodesis is usually required to manage large spring ligament tears. Subtalar
arthroereisis, a technique traditionally reserved for pediatric patients, has been introduced as an adjunctive procedure in the management of the flexible adult acquired
flatfoot deformity; preliminary results and anecdotal experience with this sinus tarsi implant that distracts the
subtalar joint show promise. In some patients, a double
calcaneal osteotomy (medial displacement calcaneal osteotomy and lateral column lengthening) has proved effective in correcting hindfoot valgus and forefoot abduction when combined with tendon reconstruction. Fixed
stage III flatfoot deformity cannot be corrected with
joint-sparing procedures and requires realignment
through arthrodesis, typically triple arthrodesis. Because
posterior tibial tendon dysfunction often is associated
with an equinus contracture, correction of deformity requires Achilles tendon lengthening. Traditionally, the
Hoke triple cut percutaneous lengthening technique has
been used, but anatomic studies demonstrate that the
triple cut technique generally fails to follow the orientation of the Achilles tendon fibers. Recently, gastrocnemius-soleus recession has gained popularity as an effective method for Achilles tendon lengthening;
although the traditional open technique is most commonly used, interest has been generated for a less invasive endoscopic technique. Stage IV flatfoot deformity
remains difficult to treat because an ideal deltoid ligament reconstruction has not been developed. When associated with ankle arthritis, pantalar arthrodesis may
be required for severe stage IV disease.

Plantar Fasciitis
The etiology of plantar fasciitis is not fully understood,
but the condition is believed to result from cumulative
trauma or repetitive stress. Symptoms usually are concentrated at the plantar fascia origin on the plantar medial heel and typically have an insidious onset. Occasionally, plantar heel pain may be associated with a
compressive neuropathy of the first branch of the lateral
plantar nerve (Baxters nerve). A heel pain triad has
been described with the coexistence of posterior tibial
tendon dysfunction, plantar fasciitis, and tarsal tunnel
syndrome. Patients typically report start-up pain (heel
pain experienced with initial weight bearing after a period of rest) and increasing heel pain after prolonged

496

Orthopaedic Knowledge Update 8

standing. The point of maximal tenderness is usually at
the plantar fascia origin on the plantar medial heel; tenderness slightly more proximal and medial is suggestive
of compressive neuropathy.
Treatment of plantar fasciitis is rarely surgical; traditional treatment with Achilles tendon stretching, night
splinting, heel cushion, and nonsteroidal antiinflammatory drugs continues to be recommended. A
prospective, randomized trial suggests that a stretching
protocol specific to the plantar fascia may provide advantages over Achilles tendon stretching. Low- and
high-energy extracorporeal shock wave therapy are
available to treat plantar fasciitis. Prospective randomized studies and meta-analyses demonstrate that both
the low- and high-energy devices are safe; however,
whether extracorporeal shock wave therapy is effective
for recalcitrant plantar fasciitis remains a controversial
issue. Surgical treatment is reserved for patients in
whom 6 months to 1 year of nonsurgical treatment is
unsuccessful. Only the medial third of the plantar fascia
should be released; complete release may result in persistent lateral foot pain. Plantar fascia release may be
performed with open or endoscopic techniques. Given
the overlap of plantar fasciitis and compression neuropathy, open plantar fascia release can be combined with
distal tarsal tunnel decompression; good to excellent results can be expected in more than 85% of patients with
recalcitrant symptoms who undergo the combined procedure.

Diabetic Foot/Ankle and Charcot Neuroarthropathy

More than 16 million people in the United States have
diabetes, and foot problems related to diabetes represent 20% of hospital admissions for diabetics. The lifetime incidence for foot ulceration in diabetic patients is
15%, and 3% to 5% of diabetic patients are amputees.
More than 50% of all amputations performed in the
United States are for diabetics. Within 3 years of lower
extremity amputation, 30% of diabetic patients lose
their contralateral leg, and 50% die. Up to half of these
amputations are considered preventable. Diabetic risk
factors for foot ulceration include neuropathy (sensory,
motor, and autonomic), vascular insufficiency, poor glycemic control, malnutrition, and impaired wound healing. Sensory neuropathy (Semmes-Weinstein monofilament < 5.07) leaves patients without protective
sensation. Motor neuropathy may lead to joint
deformities/contractures that create areas of pressure
concentration and difficulties with shoe wear. Autonomic neuropathy results in dry skin with fissures (risk
bacterial penetration) and peripheral edema (reduces
collateral blood flow/healing potential). Vascular insufficiency (ankle brachial index < 0.45, absolute toe pressures < 50 to 60 mm Hg, transcutaneous oxygen measures < 30 mm Hg) diminishes the patients ability to

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

heal. Furthermore, poor glycemic control and malnutrition contribute to impaired wound healing mechanisms.
Suspected infection requires further work-up that
can be performed by the orthopaedist or in conjunction
with an infectious disease specialist. Calluses should be
trimmed because they may mask underlying infection.
Probing to bone through an ulceration is approximately
80% reliable in diagnosing osteomyelitis. Radiographs
demonstrate cortical erosions, but it may take 7 to 14
days before these erosions become radiographically
detectable. Therefore, MRI and/or combination technetium bone scan/white blood cell-labeled indium (or gallium) scan may be required to determine if osteomyelitis is present. MRI has high sensitivity, specificity,
and accuracy in diagnosing osteomyelitis, but provides
poor localization and may in fact overrepresent the area
of involvement, thereby leading to overresection of
bone. Technetium bone scan is specific only for inflammation but combined with white blood cell-labeled indium scanning the sensitivity, specificity, and accuracy
approaches that of MRI. Limitations to technetium
bone scanning and white blood cell-labeled indium
scanning include poor circulation and prior administration of antibiotics, respectively. The most reliable means
of detecting osteomyelitis involves obtaining a deep surgical specimen of infected tissue with the patient refraining from antibiotic administration for at least 48
hours. An abscess is a surgical emergency because this
space-occupying lesion will quickly dissect more proximally through potential spaces (such as tendon sheaths)
and may create a mass effect on delicate vessels, leading
to forefoot ischemia. Typically, the only means of eradicating osteomyelitis in diabetic foot infections is to resect the infected bone. Calcaneal osteomyelitis does not
necessitate transtibial amputation; partial calcanectomy
followed by appropriate protective bracing in diabetic
patients can lead to successful limb salvage.
Most foot ulcers may be treated nonsurgically with
proper wound care and appropriate pressure relief. Following satisfactory ulcer dbridement, total contact casting remains the gold standard for forefoot/midfoot ulcer
management; hindfoot ulcers respond poorly to total
contact casting. To better monitor skin and address the
ulcer with wound management, improvements have
been made to various removable diabetic cam walker
boots. The devices have proven at least equally effective
in pressure unloading for forefoot and midfoot ulcers as
total contact casting. Forefoot ulceration associated with
equinus contractures will quickly resolve with Achilles
tendon lengthening, particularly if treated in conjunction with total contact casting. Without equinus, percutaneous flexor tenotomies and/or dorsiflexion metatarsal
osteotomies may prove beneficial for recalcitrant forefoot ulcers. Exostoses in the midfoot (frequently associated with Charcot neuroarthropathic midfoot collapse)
may be excised to effectively unload ulcerated skin.

American Academy of Orthopaedic Surgeons

Chapter 40 Foot and Ankle Reconstruction

Charcot Neuroarthropathy
Charcot neuroarthropathy remains a treatment challenge. The exact etiology of Charcot neuroarthropathy is
poorly understood, but neurotraumatic and/or neurovascular theories are currently favored. The neurotraumatic theory suggests fracture or fracturedislocation without protective sensation and a healing
response of hypertrophic bone formation in an inherently unstable fracture that has not been stabilized.
Conversely, the neurovascular theory suggests nonphysiologic vascular inflow resulting in resorption and subsequent fracture-dislocation.
Eichenholtz staging defines the clinical and radiographic progression of the neuroarthropathy. Stage I is
characterized by edema, warmth, erythema, and radiographic evidence of bony acute fracture and/or dislocation (fragmentation); the neuroarthropathy develops
during this initial stage. In stage II, the proliferative
phase, bony destruction is combined with the fracture/
dislocation. Edema and warmth are generally diminished relative to stage I. Progression to stage III is defined by coalescence and remodeling with healing in a
foot position resulting from the fracture/dislocation and
bony displacement. Typically, patients who were not immobilized and restricted from bearing weight tend to
consolidate in a less favorable, nonplantigrade foot position. As for diabetic ulceration/infection, treatment of
Charcot neuroarthropathy is improved through patient
and physician education. A heightened awareness and
appropriate differentiation from osteomyelitis generally
can lead to earlier diagnosis, immobilization, and restricted weight bearing. The advantage to identifying the
Charcot process early is that progression through the
three stages can occur while the foot is maintained in
this near anatomic position. If the Charcot process can
course through to the consolidation/remodeling phase
while avoiding deformity such as arch collapse, outcome
may be markedly improved. Unfortunately, the painless
Charcot foot often does not prompt immediate medical
care and proper diagnosis is frequently delayed. Use of
a total contact cast and no weight bearing traditionally
have been effective in the management of Charcot neuroarthropathy, but external fixation recently has gained
popularity for initial stabilization. Even with deformity
following progression to stage III, most Charcot deformities can be managed nonsurgically with total contact
inserts, extra-depth shoes, stiffer soles, rocker bottom
shoe modification, and bracing above the ankle. Severe
deformities may warrant use of a functional total contact cast (such as the Charcot restraining orthotic
walker). Some deformities are predisposed to ulceration, infection, and amputation, and brace treatment is
not possible. Salvage procedures have been effective.
Several classification schemes have been developed to
better define deformity patterns and develop treatment

497

Foot and Ankle Reconstruction

algorithms. Although complex, such classification
schemes demonstrate good interobserver and intraobserver reliability. It is predicted that such schemes may
facilitate treatment strategies. Surgical salvage procedures for ankle, hindfoot, and midfoot Charcot deformity typically require multiplanar realignment osteotomies with arthrodesis and tend to ignore traditional
anatomic landmarks. Techniques for internal fixation
have been described, but recently developed methods of
external fixation for Charcot neuroarthropathy for the
diabetic foot and ankle have shown promise. Whereas
resolution of ulceration and infection is typically necessary before internal fixation of Charcot deformity, external fixation affords the potential advantage of singlestage resection of the osteomyelitic prominence, ulcer
excision, multiplanar realignment osteotomy, and arthrodesis stabilized without implanted hardware.
Salvage procedures may be successful in limb preservation for diabetic patients with ulceration, infection,
and/or Charcot neuroarthropathy; however, amputation
may remain the only recourse in some situations. If feasible, partial foot amputations provide the advantage of
limb support without a prosthesis. However, the patient
still may be predisposed to ulceration, and transtibial
amputation may be the safest solution.

Annotated Bibliography
Forefoot: Hallux Valgus
Chi TD, Davitt J, Younger A, Holt S: Intra-and interobserver reliability of the distal metatarsal articular angle in adult hallux valgus. Foot Ankle Int 2002;23:722726.
Preoperative and postoperative radiographs of 32 patients
undergoing hallux valgus correction using a proximal bony
procedure demonstrated a reduction in the DMAA by an average of 3.9, noted by all observers. However, interobserver
reliability of preoperative and postoperative DMAA was
poor.

Coetzee JC: Scarf osteotomy for hallux valgus repair:

The dark side. Foot Ankle Int 2003;24:29-33.
The Scarf osteotomy was performed in 20 consecutive patients, resulting in a high complication rate, limited improvement in the American Orthopaedic Foot and Ankle Society
forefoot score, and only a 55% patient satisfaction rate.

Coetzee JC, Resig SG, Kuskowski M, Saleh KJ: The lapidus procedure as salvage after failed surgical treatment
of hallux valgus. J Bone Joint Surg Am 2003;85-A:60-65.
Twenty-four patients with 26 symptomatic hallux valgus
recurrences were selected to undergo the Lapidus procedure.
At an average follow-up of 24 months, the average American
Orthopaedic Foot and Ankle Society score improved from 47
to 88 points, with average improvements in the hallux valgus
angles from 37 to 17 and intermetatarsal angles from 18 to

498

Orthopaedic Knowledge Update 8

9. Complications included three nonunions, all in patients
who smoked.

Coughlin MJ, Freund E: The reliability of angular measurements in hallux valgus deformities. Foot Ankle Int
2001;22:369-379.
Interobserver and intraobserver reliability within 5 is high
for the intermetatarsal angle (97%), good for the hallux valgus
angle (86%), and poor for the distal metatarsal articular angle
(59%).

Faber FW, Kleinrensink GJ, Mulder PG, Verhaar JA:

Mobility of the first tarsometatarsal joint in hallux valgus patients: A radiographic analysis. Foot Ankle Int
2001;22:965-969.
Ninety-four feet with symptomatic hallux valgus deformity
prompting surgical correction were evaluated clinically and radiographically using the modified Coleman block test. The
mean first ray motion was 13 and proved to be significantly
greater in the group that demonstrated clinical hypermobility.

Glasoe WM, Allen MK, Saltzman CL: First ray dorsal

mobility in relation to hallux valgus deformity and first
intermetatarsal angle. Foot Ankle Int 2001;22:98-101.
A comparison of 14 hallux valgus patients and asymptomatic controls demonstrated increased mobility in the hallux
valgus group. A load-cell device was used to determine the degree of hypermobility.

Kristen KH, Berger C, Stelzig E, Thalhammer E: The

SCARF osteotomy for the correction of hallux valgus
deformities. Foot Ankle Int 2002;23:221-229.
A retrospective analysis of 111 Scarf osteotomies in 89
consecutive patients with moderate to severe hallux valgus revealed an average hallux valgus angle improvement of 19 and
intermetatarsal angle improvement of 7. The average American Orthopaedic Foot and Ankle Society forefoot score improved from 50 to 91 points. Seven patients (6%) experienced
a recurrence of deformity and 5% suffered either superficial
wound infection, displacement of the distal fragment, or hallux
MTP joint stiffness.

Nery C, Barroco R, Ressio C: Biplanar chevron osteotomy. Foot Ankle Int 2002;23:792-798.
Fifty-four biplanar distal chevron osteotomies were performed in 32 patients to correct moderate hallux valgus deformity associated with an increased distal metatarsal articular
angle. With follow-up of 2 years or longer, the average American Orthopaedic Foot and Ankle Society forefoot score improved from 50 to 90 points, the average hallux valgus angle
improved 9, the intermetatarsal angle improved 4, and the
distal metatarsal articular angle improved from 15 to 5.

Nyska M, Trnka HJ, Parks BG, Myerson MS: The Ludloff metatarsal osteotomy: Guidelines for optimal cor-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 40 Foot and Ankle Reconstruction

rection on a geometric analysis conducted on a sawbone

model. Foot Ankle Int 2003;24:34-39.

bility: Long-term results of treatment. Foot Ankle Int

2002;23:1018-1025.

Using a saw bone model, the authors describe technical aspects to achieving desired correction for the modified Ludloff
proximal oblique first metatarsal osteotomy for hallux valgus
correction.

Twenty percent (24 of 121 consecutive patients (131 feet

and 136 neuromas) were treated for concomitant excision of a
second web space neuroma and stabilization of a second MTP
joint capsular instability. At an average follow-up of 80
months, 21 feet were available for evaluation. In 15 feet, the
procedures were performed simultaneously; in 6 feet, the procedures were staged. Subjective patient satisfaction was high;
subjective and objective results were lower in patients with
persistent MTP joint instability.

Forefoot: Hallux Rigidus

Coughlin MJ, Shurnas PS: Hallux rigidus: Demographics, etiology, and radiographic assessment. Foot Ankle Int
2003;24:731-743.
This retrospective review of 114 patients treated surgically
for hallux rigidus demonstrated that hallux rigidus was associated with hallux valgus interphalangeus, female gender, unilateral disease (history of trauma), and bilateral disease (family
history). The condition was not associated with metatarsus primus elevatus, a long first metatarsal, or first ray hypermobility.

Coughlin MJ, Shurnas PS: Hallux rigidus: Grading and

long-term results of operative treatment. J Bone Joint
Surg Am 2003;85-A:2072-2088.
A retrospective review of 114 patients treated surgically
for hallux rigidus is presented. Eighty patients (93 feet) had
undergone cheilectomy; 30 patients (34 feet) were treated with
first MTP joint arthrodesis. Based on 96% of patients available at an average follow-up of 8.9 years, 97% of patients had
good or excellent results (American Orthopaedic Foot and
Ankle Society forefoot scoring system). Cheilectomy was predictable in lesser grades or first MTP arthritis; arthrodesis was
favored for greater degrees of arthritis.

DeFrino PF, Brodsky JW, Pollo FE, et al: First metatarsophalangeal arthrodesis: A clinical, pedobarographic
and gait analysis study. Foot Ankle Int 2002;23:496-502.
Clinical outcome, dynamic pedobarography (EMED) analysis, and kinematic and kinetic gait analysis were studied in
9 patients (10 feet) who underwent first MTP joint arthrodesis
for severe hallux rigidus. The mean American Orthopaedic
Foot and Ankle Society score improved from 38 to 90 points,
and the EMED analysis demonstrated restoration of the
weight-bearing function of the first ray. Kinematic data indicated a shorter step length/loss of ankle plantar flexion and
the kinetic data indicated a reduction in ankle power push-off.

Forefoot: Sesamoid Disorders

Allen MA, Casillas MM: The Passive Axial Compression (PAC) test: A new adjunctive provocative maneuver for the clinical diagnosis of hallucal sesamoiditis.
Foot Ankle 2001;22:345-346.
The authors describe a new provocative test that clinically
reproduces the symptoms of sesamoid disorders. The maneuver may be useful in initial diagnosis and monitoring response
to treatment of sesamoid problems.

Koti M, Maffulli N: Bunionette. J Bone Joint Surg Am

2001;83-A:1076-1082.
This article presents an excellent current review of evaluation and management of bunionette deformity.

Petersen WJ, Lankes JM, Paulsen F, Hassenpflug J: The

arterial supply of the lesser metatarsal heads: A vascular
injection study in human cadavers. Foot Ankle Int 2002;
23:491-495.
Epoxy resin injections performed in cadaveric foot specimens demonstrated an anastomosis of arteries about the lesser
metatarsal heads arising from both the dorsalis pedis and posterior tibial arteries. This vascular network is closely associated
with the joint capsule. The authors caution that extensive capsular stripping during metatarsal osteotomies may damage this
vascular network.

Trnka HJ, Gebhard C, Muhlbauer M, et al: The Weil osteotomy for treatment of dislocated lesser metatarsophalangeal joints: Good outcome in 21 patients with
42 osteotomies. Acta Orthop Scand 2002;73:190-194.
A retrospective review of 60 Weil metatarsal osteotomies
performed in 31 patients for dislocated lesser MTP joints at an
average follow-up of 30 months showed 42 excellent results
(21 patients). A major complication was penetrating hardware
in 10 patients.

Trnka HJ, Nyska M, Parks BG, Myerson MS: Dorsiflexion contracture after the Weil osteotomy: Results of cadaver study and three-dimensional analysis. Foot Ankle
Int 2001;22:47-50.
Using both cadaver and saw bone models, the authors
demonstrated that the Weil metatarsal osteotomy always creates plantar fragment depression that changes the center of
rotation of the MTP joint, causing the interosseous muscles to
act more as dorsiflexors than plantar flexors. These findings
are believed to be responsible for the high rate of dorsiflexion
contractures following Weil metatarsal osteotomies.

Midfoot Arthritis

Forefoot: Lesser Toes and Bunionette Deformity

Berlet GC, Anderson RB: Tendon arthroplasty for basal

fourth and fifth metatarsal arthritis. Foot Ankle Int 2002;
23:440-446.

Coughlin MJ, Schenck RC, Shurnas PJ, Bloome DM:

Concurrent interdigital neuroma and MTP joint insta-

At an average follow-up of 25 months, 12 patients undergoing tendon interpositional arthroplasty for fourth and fifth

American Academy of Orthopaedic Surgeons

499

Foot and Ankle Reconstruction

TMT joint arthritis recalcitrant to nonsurgical treatment were
evaluated. In eight patients, the diagnosis was confirmed with
preoperative differential injections; six of the eight patients
would undergo the procedure again. Outcome, based on the
American Orthopaedic Foot and Ankle Society midfoot scale,
was most favorable in patients who had a positive response to
a preoperative differential injection.

Keiserman LS, Cassandra J, Amis JA: The Piano Key

Test: A clinical sign for the identification of subtle tarsometatarsal pathology. Foot Ankle Int 2003;24:437-438.
The authors describe a simple test to identify and isolate
TMT synovitis and/or arthritis.

Hindfoot and Ankle: Ankle/Subtalar Instability

Jeys LM, Harris NJ: Ankle stabilization with hamstring
autograft: A new technique using interference screws.
Foot Ankle Int 2003;24:677-679.
An anatomic reconstruction of the lateral ankle ligaments
is described using bioabsorbable interference screw fixation of
a hamstring autograft.

Keefe DT, Haddad SL: Subtalar instability: Etiology, diagnosis, and treatment. Foot Ankle Clin 2002;7:577-609.
The authors present a comprehensive review of the current state of the art for the evaluation and treatment of subtalar joint instability.

Krips R, Brandsson S, Swensson C, et al: Anatomical reconstruction and Evans tenodesis of the lateral ligaments of the ankle: Clinical and radiological findings after followup for 15-30 years. J Bone Joint Surg Br 2002;
84:232-236.
This retrospective review compares 54 patients undergoing
an anatomic reconstruction of the lateral ankle ligaments and
45 patients treated with an Evans tenodesis for lateral ankle
instability. The study demonstrated that the functional outcome of the Evans tenodesis deteriorated more rapidly than
the anatomic reconstruction. Good to excellent results were
noted in 43 patients in the anatomic reconstruction group and
15 in the Evans tenodesis group.

Pijnenburg AC, Bogaard K, Krips R, et al: Operative

and functional treatment of rupture of the lateral ligament of the ankle: A randomized, prospective trial.
J Bone Joint Surg Br 2003;85:525-530.
The authors prospectively randomized 370 patients with a
rupture of at least one lateral ankle ligament to undergo either functional or surgical treatment. At a median follow-up of
8 years, 86% of patients were available for evaluation with the
Povacz score and anterior drawer testing. Surgical treatment
gave better long-term outcome with regard to residual pain,
recurrent sprains (22% versus 34%), and stability (anterior
drawer positive, 30% versus 54%).

500

Orthopaedic Knowledge Update 8

Sabonghy EP, Wood RM, Ambrorse CG, McGarvey
WC, Clanton TO: Tendon transfer fixation: Comparing a
tendon to tendon technique versus bioabsorbable interference. Fit screw fixation. Foot Ankle Int 2003;24:260262.
In 10 paired fresh cadaver specimens, load to failure was
greater using bioabsorbable tendon fixation when compared
with traditional fixation techniques. However, the authors supported the use of bioabsorbable screw as the mean fixation
strength provides physiologic strength at the tendon-bone interface.

Van Bergeyk AB, Younger A, Carson B: CT Analysis of

hindfoot alignment in chronic lateral ankle instability.
Foot Ankle Int 2002;23:37-42.
The authors compared simulated weight-bearing hindfoot
CT scans of 14 ankles with chronic ankle instability to 12 controls in a prospective case control format. Patients with ankle
instability had statistically significantly greater hindfoot varus
based on the radiographic parameters evaluated. The authors
suggested that although a valgus-producing calcaneal osteotomy is not routinely indicated, it may have a role in selected
patients who fail to respond to isolated lateral ankle ligament
reconstruction.

Hindfoot and Ankle: Ankle Arthritis

Anderson T, Montgomery F, Carlsson A: Uncemented
STAR total ankle prostheses: Three- to eight-year
follow-up of fifty-one consecutive ankles. J Bone Joint
Surg Am 2003;85-A:1321-1329.
Fifty-one consecutive cementless mobile-bearing Scandinavian Total Ankle Replacement ankle procedures were evaluated at intermediate follow-up. The median Kofoed score improved from 39 to 70 points at the time of follow-up. Twelve
ankles had to be revised (seven loosening, two bearing fracture, three other complications) and eight other ankles had radiographic signs of loosening. Median range of motion was the
same preoperatively and postoperatively. With revision as the
end point, the estimated 5-year survival was 70%.

Buechel FF, Buechel FF Jr, Pappas MJ: Ten-year evaluation of cementless Buechel-Pappas meniscal bearing total ankle replacement. Foot Ankle Int 2003;24:462-472.
Fifty cementless Buechel-Pappas mobile-bearing total ankle replacements in 49 patients were evaluated. Good to excellent results were observed in 88% of patients. Postoperative
range of motion was similar to preoperative motion. Revision
was required in 4% of patients. Cumulative survivorship (using revision as an end point) was 94% at 10 years.

Coester LM, Saltzman CL, Leupold J, Pontarelli W:

Long-term results following ankle arthrodesis for posttraumatic arthritis. J Bone Joint Surg Am 2001;83-A:219236.
At a mean follow-up of 22 years, 23 patients who had a
successful isolated ankle arthrodesis for posttraumatic tibiotalar arthritis were evaluated. Most patients had substantial, ac-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

celerated arthritic changes in the ipsilateral foot that often
limited function when compared with the contralateral foot.
Osteoarthritis did not develop more frequently in the ipsilateral knee or contralateral foot.

Coull R, Raffiq T, James LE, Stephens MM: Open treatment of anterior impingement of the ankle. J Bone Joint
Surg Br 2003;85:550-553.
The outcome for the open treatment of anterior ankle impingement was evaluated at a mean follow-up of 7.3 years in
23 patients. The Ogilvie-Harris score improved in all patients.
Ankle dorsiflexion did not return to normal, but symptomatic
relief allowed 79% of patients to return to athletic activity at
the same level. Two patients with preoperative joint space narrowing had a poor result.

Donley BG, Ward DM: Implantable electrical stimulation in high-risk hindfoot fusions. Foot Ankle Int 2002;
23:13-18.
The authors report a single surgeons experience with 13
implantable bone stimulators used as an adjunct for ankle/
hindfoot arthrodeses performed in patients with increased risk
for nonunion. At an average follow-up of 25 months, 92% of
patients achieved successful fusion. The subcutaneous device
was bothersome to eight patients.

Fuchs S, Sandmann C, Skwara A, et al: Quality of life 20

years after arthrodesis of the ankle: A study of adjacent
joints. J Bone Joint Surg Br 2003;85:994-998.
Retrospective, long-term follow-up of 18 successful ankle
arthrodeses (17 patients) demonstrated significant deficits of
functional outcome, limitations of activities of daily living, and
radiographic changes in adjacent foot articulations, based on
the Olerud Molander Ankle score, radiographic evaluation,
and the Medical Outcomes Study Short Form-36 outcomes instrument.

Chapter 40 Foot and Ankle Reconstruction

46 to 53 (mental). Four of seven patients reported good to excellent results; follow-up radiographs revealed joint space narrowing, osteophytes, and sclerosis even in patients with excellent clinical status. There was a 42% failure rate.
Improvements in technique may lead to a more favorable outcome.

Marijnissen AC, Van Roermund PM, Van Melkebeek J,

Lafeber FP: Clinical benefit of joint distraction in the
treatment of ankle osteoarthritis. Foot Ankle Clin 2003;
8:335-346.
The authors describe their technique and results with distraction arthroplasty for ankle osteoarthritis.

Myerson MS, Mroczek K: Perioperative complications

of total ankle arthroplasty. Foot Ankle Int 2003;24:17-21.
This retrospective review of a single surgeons experience
with 50 consecutive total ankle replacements compares the
first 25 with the second 25 procedures. The distinct improvement in component position and fewer complications in the
second group suggest that a learning curve exists in performance of total ankle arthroplasty.

Saltzman CL, Amendola A, Anderson R, et al: Surgeon

training and complications in total ankle arthroplasty.
Foot Ankle Int 2003;24:514-518.
The first 10 total ankle replacements of nine orthopaedic
foot and ankle surgeons were reviewed. No method of training
(observing inventor, hands-on surgical training course, foot
and ankle fellowship) had a statistically demonstrable positive
impact on preparing surgeons for total ankle replacement.

Thomas RH, Daniels TR: Current concepts review: Ankle arthritis. J Bone Joint Surg Am 2003;85-A:923-936.
The authors present a review of the current standards for
evaluation and treatment of ankle arthritis.

Ishikawa SN, Murphy GA, Richardson EG: The effect

of cigarette smoking on hindfoot fusions. Foot Ankle Int
2002;23:996-998.

Wood PL, Deakin S: Total ankle replacement: The results in 200 ankles. J Bone Joint Surg Br 2003;85:334341.

In a group of 160 patients who had hindfoot fusions, smokers had a significantly higher nonunion rate than nonsmokers
(19% versus 7%). The relative risk of nonunion was 2.7 times
higher for smokers than nonsmokers. No statistically significant difference was noted in the rate of infection or delayed
wound healing between the groups.

At a mean follow-up of 46 months, the authors reviewed

the results in 200 mobile-bearing cementless Scandinavian Total Ankle Replacements. The cumulative survival rate at 5
years was 93%, with decision for revision used as an end
point. Most frequent complications were delayed wound healing and malleolar fracture. A complication requiring further
surgery developed in 8 ankles, and 14 ankles were either revised or converted to arthrodesis.

Kim CW, Jamali A, Tontz W Jr, et al: Treatment of posttraumatic ankle arthrosis with bipolar tibiotalar osteochondral shell allografts. Foot Ankle Int 2002;23:10911102.
Seven patients undergoing fresh tibiotalar osteochondral
shell allografts for posttraumatic ankle arthrosis were evaluated at an average follow-up of 12 years. The ankle scores increased from 25 to 43 points; the Medical Outcomes Study 12Item Short Form scores increased from 30 to 38 (physical) and

American Academy of Orthopaedic Surgeons

Hindfoot and Ankle: Osteochondral Lesions of the

Talus
Al Shaikh RA, Chou LB, Mann JA, et al: Autologous
osteochondral grafting for talar cartilage defects. Foot
Ankle Int 2002;23:381-389.
At an average follow-up of 16 months, 19 osteochondral
lesions of the talus treated with osteochondral autograft trans-

501

Foot and Ankle Reconstruction

fer system from the ipsilateral knee trochlear border were
evaluated. All patients had failed to respond to nonsurgical
measures and 68% had failed to respond to dbridement/
excision. The average size of the lesions before the autologous
osteochondral grafting procedure was 10 mm 12 mm. Fourteen patients required malleolar osteotomy (13 medial, 1 lateral). The average postoperative American Orthopaedic Foot
and Ankle Society score at follow-up was 91; the average Lysholm knee score was 97.

Hangody L, Kish G, Modis L, et al: Mosaicplasty for the

treatment of osteochondritis dissecans of the talus: Two
to seven year results in 36 patients. Foot Ankle Int 2001;
22:552-558.
Intermediate follow-up of a single surgeons experience
with 36 osteochondral lesions of the talus treated using the
mosaicplasty technique he invented demonstrated 94% good
to excellent results (Hannover scoring system for ankle function). Osteochondral plugs were transferred from the ipsilateral knee; no long-term donor site morbidity was observed.

Hintermann B, Boss A, Schafer D: Arthroscopic findings

in patients with chronic ankle instability. Am J Sports
Med 2002;30:402-409.
Based on a cohort of 148 patients with symptomatic
chronic ankle instability undergoing ankle arthroscopy, 66% of
ankles with lateral ligament instability and 98% with deltoid
ligament injuries were noted to have associated cartilage damage, respectively.

Peterson L, Brittberg M, Lindahl A: Autologous chondrocyte transplantation of the ankle. Foot Ankle Clin
2003;8:291-303.
The inventors of autologous chondrocyte transplantation
(Carticel procedure) present their experience with this technique in the management of osteochondral lesions of the talus.

Sammarco GJ, Makwana MK: Treatment of talar osteochondral lesions using local osteochondral graft. Foot
Ankle Int 2002;23:693-698.
The authors describe a technique for local osteochondral
transfer in the management of osteochondral talar dome lesions; the graft was harvested from the medial or lateral talar
articular facet. Exposure to the defect was facilitated through
a replaceable bone block removed from the anterior tibial
plafond. In 12 patients with an average follow-up of 25
months, the average American Orthopaedic Foot and Ankle
Society score improved from 64 to 91 points. No complications
were reported.

Schafer D, Boss A, Hintermann B: Accuracy of arthroscopic assessment of anterior ankle cartilage lesions.
Foot Ankle Int 2003;24:317-320.
The authors demonstrated that iatrogenically created talar
dome defects were overestimated and underestimated when
evaluated arthroscopically in 10 cadaver specimens. These

502

Orthopaedic Knowledge Update 8

findings have implications should cartilage repair procedures
be planned based on arthroscopic sizing of osteochondral defects of the talar dome.

Shearer C, Loomer R, Clement D: Nonoperataively

managed stage V osteochondral talar lesions. Foot Ankle
Int 2002;23:651-654.
Based on nonsurgical treatment of 25 osteochondral lesions of the talus associated with subchondral cysts, the authors suggested that (1) most lesions remain radiographically
stable, (2) nonsurgical management is a viable option for stage
V osteochondral lesions with little risk of developing significant osteoarthritis, (3) the general course of stage V lesions is
benign in most patients, and (4) the development of mild radiographic changes of osteoarthritis does not correlate with
outcome. However, the authors acknowledged that lesions that
increase significantly in size correlate with poor outcome.

Schimmer RC, Dick W, Hintermann B: The role of ankle

arthroscopy in the treatment strategies of osteochondritis dissecans lesions of the talus. Foot Ankle Int 2001;22:
895-900.
The authors report that arthroscopy represents a helpful
diagnostic tool in assessing extent and, in particular, stability
and integrity of the osteochondral talar defect. They recommend that arthroscopy be performed in all patients with osteochondral lesions of the talus to define the treatment strategy.

Schuman L, Struijs PA, Van Dijk CN: Arthroscopic

treatment for osteochondral defects of the talus: Results
at followup at 2 to 11 years. J Bone Joint Surg Br 2002;
84:364-368.
Thirty-eight patients who had been treated with arthroscopic dbridement and drilling for osteochondral lesions of
the talus were evaluated at a mean follow-up of 4.8 years.
Twenty-two patients underwent a primary procedure, and 16
had failed previous surgery. Good to excellent results were
found in 86% of the primary procedures and 75% of revision
cases.

Scranton PE Jr, McDermott JE: Treatment of Type V

osteochondral lesions of the talus with ipsilateral knee
osteochondral autografts. Foot Ankle Int 2001;22:380384.
Ten consecutive patients who had osteochondral lesions of
the talus were treated with osteochondral autograft transplantation and were evaluated at short-term follow-up. All lesions
were associated with subchondral cysts. These preliminary results suggest significant improvement in all 10 patients, with an
average increase in the American Orthopaedic Foot and Ankle Society ankle/hindfoot score of 27 points.

Thordarson DB, Bains R, Shepherd LE: The role of ankle arthroscopy on the surgical management of ankle
fractures. Foot Ankle Int 2001;22:123-125.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Nineteen patients with surgical treatment of their unstable
ankle fractures were prospectively randomized to include or
not include ankle arthroscopy. Eight of nine patients randomized to the arthroscopy group had articular damage that
prompted arthroscopic treatment. At an average follow-up of
21 months, no difference was noted in the Medical Outcomes
Study Short Form-36 Health Survey or lower extremity scores
between the two groups.

Chapter 40 Foot and Ankle Reconstruction

Hindfoot and Ankle: Acquired Flatfoot Defromity

Choi K, Lee S, Otis JC, Deland JT: Anatomical reconstruction of the spring ligament using peroneus longus
tendon graft. Foot Ankle Int 2003;24:430-436.
Using a cadaver foot-ankle flatfoot model comparing
three methods of spring ligament reconstruction using the peroneus longus tendon, it was shown that a superomedial/
plantar passage of tendon through the calcaneus and navicular
was most effective.

Hindfoot and Ankle: Insertional Achilles Tendinopathy

Calder JD, Saxby TS: Surgical treatment of insertional
Achilles tendinosis. Foot Ankle Int 2003;24:119-121.
A chart review of 49 open dbridements of insertional
Achilles tendinosis in which less than 50% of the tendon was
excised suggested that early mobilization does not predispose
to postsurgical rupture. At a minimum follow-up of 6 months,
only two failures were noted: one patient with psoriatic arthritis and another who had bilateral simultaneous procedures.

Den Hartog BD: Flexor hallucis longus transfer for

chronic Achilles tendinosis. Foot Ankle Int 2003;24:233237.
Twenty-nine tendons (26 patients) undergoing reconstruction for chronic Achilles tendinosis using a flexor hallucis longus augmentation were evaluated at an average follow-up of 3
years. Time to maximum improvement was 8.2 months. The average American Orthopaedic Foot and Ankle Society anklehindfoot score improved from 42 to 90 points. No patient had
a clinically significant functional deficit of the hallux.

McGarvey WC, Palumbo RC, Baxter DE, Leibman BD:

Insertional Achilles tendinosis: Surgical treatment
through a central tendon splitting approach. Foot Ankle
Int 2002;23:19-25.
Twenty-two patients with insertional Achilles tendinosis
were treated through a central tendon splitting approach with
tendon dbridement, retrocalcaneal bursectomy, and partial
calcanectomy. This approach revealed that the disease process
was isolated to the central tendon insertion in 21 of the patients. At an average follow-up of 33 months, 82% of patients
were satisfied with the surgery, 77% would have the surgery
again, but only 59% were completely pain free and could return to unlimited activities.

Tashjian RZ, Hur J, Sullivan RJ, et al: Flexor hallucis

longus transfer for repair of chronic achilles tendinopathy. Foot Ankle Int 2003;24:673-676.
Using 14 fresh-frozen cadaver lower limbs, a short flexor
hallucis longus tendon harvest (posteromedial ankle incision)
was compared with a long (traditional) flexor hallucis longus
tendon harvest (second medial midfoot incision). The short
flexor hallucis longus tendon harvest yielded an average tendon length of 5.2 cm; the long flexor hallucis longus harvest
yielded and average tendon length of 8.1 cm.

American Academy of Orthopaedic Surgeons

Coetzee JC, Hansen ST: Surgical management of severe

deformity resulting from posterior tibial tendon dysfunction. Foot Ankle Int 2001;22:944-949.
A retrospective review of 12 feet in 11 patients undergoing
major hindfoot corrective surgery with an extended triple arthrodesis for severe acquired pes planovalgus deformity demonstrated a statistically significant improvement in the average
American Orthopaedic Foot and Ankle Society hindfoot score
(30 to 74) and radiographic parameters. Despite multiple complications (wound problems, delayed unions) requiring revision surgery, the extensive procedure provided a justifiable improvement in patients quality of life.

Guyton GP, Jeng C, Krieger LE, Mann RA: Flexor digitorum longus transfer and medial displacement calcaneal osteotomy for posterior tibial tendon dysfunction:
A middle-term clinical follow-up. Foot Ankle Int 2001;
22:627-632.
At an average follow-up of 32 months, 26 patients undergoing flexor digitorum longus tendon transfer with medial displacement calcaneal osteotomy for stage II posterior tibial
tendon dysfunction were evaluated. Only 16 patients were
evaluated by physical examination for the follow-up evaluation. The average American Orthopaedic Foot and Ankle Society hindfoot pain subscale score was 35 of 40 and the American Orthopaedic Foot and Ankle Society functional score was
27 of 28. Three failures included two early failures of fixation
of the flexor digitorum longus tendon and one failure at approximately 6 years during pregnancy.

Louden KW, Ambrose CG, Beaty SG, et al: Tendon

transfer fixation in the foot and ankle: Biomechanical
study evaluating two sizes of pilot holes for bioabsorbable screws. Foot Ankle Int 2003;24:67-72.
Bioabsorbable screw fixation for tendon transfer procedures, including flexor digitorum longus or flexor hallucis longus transfer to the navicular for correction of posterior tibial
tendon insufficiency, is being used with greater frequency in
foot and ankle surgery. This biomechanical study performed in
cadaver specimens demonstrates that initial pull-out strength
for 5- or 7-mm screws exceeds the requisite strength for tendon transfer to the navicular. The 7-mm screw diameter with
pilot holes of 5.5 mm or 6.5 mm may be preferable given the
average flexor digitorum longus or flexor hallucis longus tendon diameter approaching 5 mm.

503

Foot and Ankle Reconstruction

Malicky ES, Crary JL, Houghton MJ, Agel J, Hansen ST
Jr, Sangeorzan BJ: Talocalcaneal and subfibular impingement in symptomatic flatfoot in adults. J Bone
Joint Surg Am 2002;84-A:2005-2009.
Nineteen patients with symptomatic acquired flatfoot deformity were evaluated with simulated weight-bearing CT
analysis of the hindfoot and compared with a control group. In
the study group, sinus tarsi impingement was noted in 92%
and subfibular impingement was observed in 66% versus 0
and 5% in the control group, respectively.

Moseir-LaClair S, Pomeroy G, Manoli A: Intermediate

follow-up of the double osteotomy and tendon transfer
procedure for stage II posterior tibial tendon insufficiency. Foot Ankle Int 2001;22:283-291.
Twenty-six patients with 28 acquired pes planovalgus feet
(Johnson stage II) were managed with flexor digitorum longus
transfer, Achilles tendon lengthening, and a double calcaneal
osteotomy (medial displacement and lateral column lengthening). At a mean follow-up of 5 years, the mean American Orthopaedic Foot and Ankle Society ankle-hindfoot score was
90. All osteotomies united and average radiographic parameters remained improved at follow-up. Fourteen percent of patients demonstrated radiographic signs of calcaneocuboid arthritis.

Sammarco GJ, Hockenbury RT: Treatment of stage II

posterior tibial tendon dysfunction with flexor hallucis
longus transfer and medial displacement calcaneal osteotomy. Foot Ankle Int 2001;22: 305-312.
At an average follow-up of 18 months, 17 patients who
had undergone medial displacement calcaneal osteotomy and
flexor hallucis longus tendon reconstruction for stage II posterior tibial tendon reconstruction were evaluated. The average
American Orthopaedic Foot and Ankle Society hindfoot score
improved from 62 to 84 points. Despite radiographic assessment demonstrating that there was no statistically significant
improvement in the medial longitudinal arch in the study
group, good to excellent results were reported in most patients.

Tashjian RZ, Appel AJ, Banerjee R, DiGiovanni CW:

Endoscopic gastocnemius recession: evaluation in a cadaver model. Foot Ankle Int 2003;24: 607-613.
Endoscopic gastrocnemius recession was evaluated in 15
cadaver specimens. The sural nerve was definitively visualized
in 33% of the procedures and an average of 83% of the gastrocnemius aponeurosis was transected. Improvement in mean
ankle dorsiflexion with the knee flexed was 20.

Thomas RL, Wells BC, Garrison RL, Prada SA: Preliminary results comparing two methods of lateral column
lengthening. Foot Ankle Int 2001;22:107-119.
At a minimum follow-up of at least 1 year, 10 Evans opening wedge osteotomies were compared with 17 calcaneocuboid
distraction arthrodeses, both of which were performed with
structural iliac crest autograft and flexor digitorum longus

504

Orthopaedic Knowledge Update 8

transfer for posterior tibial tendon dysfunction. American Orthopaedic Foot and Ankle Society hindfoot scores and radiographic parameters were significantly improved for both
groups. The complication rate (reported for 34 feet) was high
for both groups; the rate of nonunion (12%) and delayed
union (18%) was considerable for the calcaneocuboid distraction arthrodesis group.

Viladot R, Pons M, Alvarez F, Omana J: Subtalar arthroereisis for posterior tibial tendon dysfunction: A preliminary report. Foot Ankle Int 2003;24:600-606.
Twenty-one patients with stage II flexible posterior tibial
tendon dysfunction were treated with flexor digitorum longus
augmentation or flexor hallucis longus tendon transfer and
subtalar arthroereisis (sinus tarsi implant). Nineteen patients
reviewed at an average 27-month follow-up had an average
improvement in the American Orthopaedic Foot and Ankle
Society score from 47 to 82 points. Two patients required removal of the implant secondary to pain.

Wacker JT, Hennessy MS, Saxby TS: Calcaneal osteotomy and transfer of the tendon of flexor digitorum longus for stage-II dysfunction of tibialis posterior:
Three-to five-year results. J Bone Joint Surg Br 2002;84:
54-58.
At mean follow-up of 51 months, 44 patients treated with
flexor digitorum longus transfer and medial displacement calcaneal osteotomy had an average improvement in the American Orthopaedic Foot and Ankle Society hindfoot score from
49 to 88 points. The outcome was good to excellent in 43 patients for pain and function, and good to excellent in 36 patients for alignment. No poor results were observed.

Hindfoot and Ankle: Plantar Fasciitis

DiGiovanni BF, Nawoczenski DA, Lintal ME, et al: Tissue specific plantar fascia-stretching exercise enhances
outcomes in patients with chronic heel pain. J Bone
Joint Surg Am 2003;85-A:1270-1277.
One hundred one patients with chronic plantar fasciitis
were randomized to undergo either plantar fascia tissue or
traditional Achilles tendonstretching regimen. Greater improvement was observed in the plantar fasciastretching program with regard to pain, activity limitation, and patient satisfaction.

Haake M, Buch M, Schoellner C, Goebel F: Extracorporeal shock wave therapy for plantar fasciitis: Randomized controlled multicentre trial. BMJ 2003;327:75.
This multicenter study of 272 patients with chronic plantar
fasciitis compared extracorporeal shock wave therapy with a
placebo group. The success rate at 12 weeks was 34% in the
shock wave therapy group and 30% in the placebo group. The
authors concluded that extracorporeal shock wave therapy is
ineffective in treating chronic plantar fasciitis.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Labib SA, Gould JS, Rodriguez-del-Rio FA, Lyman S:
Heel Pain Triad (HPT): The combination of plantar fasciitis, posterior tibial tendon dysfunction and tarsal tunnel syndrome. Foot Ankle Int 2002;23:212-220.
Fourteen patients were surgically treated for a combination of plantar fasciitis, posterior tibial tendon dysfunction,
and tarsal tunnel syndrome. At a mean follow-up of 17
months, a marked improvement was noted in 88% of patients
for pain, activity level, walking distance, walking surface, and
limp.

Rompe JD, Schoellner C, Nafe B: Evaluation of lowenergy extracorporeal shock-wave application for treatment of chronic plantar fasciitis. J Bone Joint Surg Am
2002;84-A:335-341.
This prospective, randomized, controlled trial of 112 patients with chronic plantar fasciitis compared three applications of 1,000 low-energy shock wave impulses (group I) to
three applications of 10 low-energy shock wave impulses
(group II). At 6 months, the rate of good to excellent results
was significantly better (47%) in group I than in group II. By
5 years, 13% of patients in group I and 58% in group II had
undergone surgical plantar fascia release. The authors concluded that treatment with 1,000 impulses of low-energy shock
waves may be an effective therapy for plantar fasciitis and
may help patients avoid surgery.

Speed CA, Nichols D, Wies J, et al: Extracorporeal

shock wave therapy for plantar fasciitis: A double blind
randomized controlled trial. J Orthop Res 2003;21:937940.
This double-blind randomized controlled trial compared
active therapy (moderate-dose shock wave therapy) to sham
therapy for 88 patients with plantar fasciitis of at least 3
months duration. Over a 6-month period, both groups showed
significant improvement, but no statistically significant differences were observed in any outcome measures.

Watson TS, Anderson RB, Davis WH, Kiebzak GM: Distal tarsal tunnel release with plantar fasciotomy for
chronic heel pain: An outcome analysis. Foot Ankle Int
2002;23:530-537.
Seventy-five patients (80 heels) with an average of 20
months of nonsurgical treatment underwent distal tarsal tunnel release with a partial plantar fasciotomy. Eighty-eight percent of patients had good to excellent results at final followup; 52% of patients required in excess of 6 months to reach
maximum medical improvement. In the 44 patients (46 heels)
who responded to a Medical Outcomes Study Short Form-36
and foot function index questionnaire, 91% were somewhat to
very satisfied with their outcomes.

Hindfoot and Ankle: Diabetic Foot/Ankle and Charcot

Neuroarthropathy
Bollinger M, Thordarson DB: Partial calcanectomy: An
alternative to below knee amputation. Foot Ankle Int
2002;23:927-932.

American Academy of Orthopaedic Surgeons

Chapter 40 Foot and Ankle Reconstruction

At an average follow-up of 27 months, 22 patients with
nonhealing heel wounds were treated with partial calcanectomy. Wounds healed in all patients; none required subsequent
transtibial amputation. Twelve patients had delayed wound
healing, and 11 additional procedures were performed on the
heels of 9 patients.

Cooper PS: Application of external fixators for management of Charcot deformities of the foot and ankle. Foot
Ankle Clin 2002;7:207-254.
A single surgeons experience with external fixation for
the management of infected and noninfected Charcot neuropathic deformities of the foot and ankle is presented.

Guyton GP, Saltzman CL: The diabetic foot: Basic

mechanism of disease. Instr Course Lect 2002;51:169181.
The authors present a review of the pertinent basic science, evaluation, and current state of the art in the management of diabetic foot problems.

Mueller MJ, Sinacore DR, Hastings MK, et al: Effect of

Achilles tendon lengthening on neuropathic plantar ulcers. J Bone Joint Surg Am 2003;85-A:1436-1445.
Sixty-four patients with neuropathic ulcers were randomized to receive total contact casting alone or combined with
percutaneous Achilles tendon lengthening. Eighty-eight percent of the ulcers in the total contact cast group and 100% in
the Achilles tendon lengthening group healed after a mean
duration of 41 days and 58 days, respectively. The risk for ulcer
recurrence was 75% less at 7 months and 52% less at 2 years
in the Achilles tendon lengthening group than in the isolated
total contact cast group.

Classic Bibliography
Berndt A, Hardy M: Transchondral fractures (osteochondritis dissicans) of the talus. J Bone Joint Surg Am
1959;41:988.
Brostrom L: Sprained ankles: V. Treatment and prognosis in recent ligament ruptures. Acta Chir Scand 1966;
132:537-550.
Brostrom L: Sprained ankles: VI. Surgical treatment of
chronic ligament ruptures. Acta Chir Scand 1966;132:
551-565.
Coughlin MJ, Carlson RE: Treatment of hallux valgus
with an increased distal metatarsal articular angle: Evaluation of double and triple first ray osteotomies. Foot
Ankle Int 1999;20:771-776.
Gould N, Seligson D, Gassman J: Early and late repair
of lateral ligament of the ankle. Foot Ankle 1980;1:84-89.

505

Foot and Ankle Reconstruction

Orthopaedic Knowledge Update 8

Hattrup SJ, Johnson KA: Subjective results of hallux

rigidus following treatment with cheilectomy. Clin
Orthop 1988;226:182-191.

Myerson MS, Henderson MR, Saxby T, Short KW: Management of midfoot diabetic neuroarthropathy. Foot
Ankle Int 1994;15:233-241.

Hepple S, Winson IG, Glew D: Osteochondral lesions of

the talus: A revised classification. Foot Ankle Int 1999;
20:789-793.

Papa J, Myerson M, Girard P: Salvage, with arthrodesis,

in intractable diabetic neuropathic arthropathy of the
foot and ankle. J Bone Joint Surg Am 1993;75:1056-1066.

Johnson KA, Strom DA: Tibialis posterior tendon dysfunction. Clin Orthop 1989;239:196-206.

Pell RF, Myerson MS, Schon LC: Clinical outcome after

primary triple arthrodesis. J Bone Joint Surg Am 2000;
82:47-57.

Kumai T, Takakura Y, Higashiyama I, Tamai S: Arthroscopic drilling for the treatment of osteochondral lesions of the talus. J Bone Joint Surg Am 1999;81:12291235.
Mann RA, Clanton TO: Hallux rigidus: Treatment by
cheilectomy. J Bone Joint Surg Am 1988;70:400-406.
Mann RA, Rudicel S, Graves SC: Repair of hallux valgus with a distal soft-tissue procedure and proximal
metatarsal osteotomy. A long-term follow-up. J Bone
Joint Surg Am 1992;74:124-129.

506

Sangeorzan BJ, Hansen ST Jr: Modified Lapidus procedure for hallux valgus. Foot Ankle 1989;9:262-266.
Toolan BC, Sangeorzan BJ, Hansen ST: Complex reconstruction for the treatment of dorsolateral peritalar subluxation of the foot: Early results after distraction arthrodesis of the calcaneocuboid joint in conjunction with
stabilization of, and transfer of the flexor digitorum longus tendon to, the midfoot to treat acquired pes planovalgus in adults. J Bone Joint Surg Am 1999;81:1545-1560.

American Academy of Orthopaedic Surgeons

Chapter

41

Adult Spine Trauma

David H. Kim, MD
Steven Zeiller, MD
Alan S. Hilibrand, MD

Cervical Spine Trauma

Clinical Evaluation
The initial evaluation and management of patients with
spinal injuries is usually initiated in the field by paramedical personnel. These injuries are frequently the result of high-energy trauma, and patients may require
rapid evaluation and resuscitation according to the
guidelines established by the American College of Surgeons. Treatment of potential spinal injury begins at the
accident scene with proper immobilization using a rigid
cervical collar, tape, or straps to secure the patients
neck, and transport on a firm spine board with lateral
support devices. In the setting of sports-related injuries,
helmet and shoulder pads should be left on until arrival
at the hospital where experienced personnel can perform simultaneous removal of both in controlled fashion.
During the primary survey, protection of the spine
and spinal cord is the important management principle.
It should be assumed that all trauma patients have a
cervical spine injury until proven otherwise, especially
those with altered mental status or following blunt head
or neck trauma. Inadequate initial stabilization can contribute to further neurologic deterioration in a patient
with an acute spinal cord injury and significantly worsen
eventual outcome. It has been estimated that 3% to
25% of spinal cord injuries may occur after the initial
traumatic episode during early management or transport. While securing an airway, excessive head and neck
movement should be avoided and manual in-line immobilization of the head and neck should be maintained
whenever immobilization devices are removed. Over
the past 30 years, significant improvements in the survival and outcome of patients with spinal cord injuries
have been observed, primarily because of improved initial management and rapid delivery by emergency services.
Although the initial neurologic evaluation only assesses the patients level of alertness and consciousness,
a more thorough assessment of neurologic status and
potential spinal injury is performed during the second-

American Academy of Orthopaedic Surgeons

ary survey. Absence of a neurologic deficit is not sufficient to exclude such an injury, and a hard cervical collar should be applied until the cervical spine has been
formally cleared and patients are considered with reasonable certainty to be stable and free of significant injury. Motorcyclists in particular have a higher incidence
of thoracic spinal injuries, and evidence of blunt chest
trauma should lead to further evaluation of the thoracic
spine. Abdominal ecchymoses or abrasions from lap belt
injury are associated with flexion-distraction injury of
the thoracolumbar spine. Extremity fractures may distract the emergency personnel from identification of a
spinal injury requiring treatment. Every multiple trauma
patient should undergo visual inspection of the back. A
thorough neurologic examination including sensorimotor function and level of consciousness is the final component of the secondary survey. Any neurologic deficit
suggests the possibility of an injury to the spinal axis.
Patients with ankylosing spondylitis or diffuse idiopathic skeletal hyperostosis represent a special subpopulation for which extra vigilance is required. Spinal involvement with these conditions, particularly ankylosing
spondylitis, appears to increase the risk of fracture, and
patients reporting neck or back pain after even relatively minor trauma should be considered for supplemental evaluation with CT. Nondisplaced fractures commonly occur in this setting, most frequently through an
ankylosed disk space, and carry a high rate of delayed
or missed diagnosis. These fractures are typically unstable and can lead to spinal cord injury if not stabilized
appropriately.
A variety of clinical grading systems have been developed for assessing and reporting neurologic status in
spinal cord injury patients. The Frankel scale has been
supplanted in clinical use by the American Spinal Injury
Association (ASIA) scale (Figure 1). This scale was first
introduced in 1984 and has undergone revisions in 1989,
1992, and most recently in 1996. The most recent version
includes separate motor and sensory scores as well as a
general impairment scale and incorporates the functional independence measure, a tool that assesses the
functional effect of spinal cord injury. The motor score

509

Adult Spine Trauma

Orthopaedic Knowledge Update 8

Figure 1 ASIA form for standard neurologic classification of spinal cord injury. (Reproduced from the American Spinal Injury Association.)

Figure 2 Photograph showing submandibular full-thickness skin necrosis from hard

collar immobilization in an elderly multiple trauma patient.

has been shown to correlate with potential for functional improvement and performance during rehabilitation.
The optimal algorithm for cervical spine clearance in
trauma patients remains one of the most controversial
areas in spinal trauma care. Prolonged cervical collar
immobilization in multiple trauma patients is known to

510

be associated with numerous potential complications including an increased risk of aspiration, limitation of respiratory function, development of decubitus ulcers in
the occipital and submandibular areas, and possible increases in intracranial pressure (Figure 2). Moreover,
collars limit access for devices such as endotracheal
tubes and central lines. Therefore, several strategies
have been developed to allow for rapid collar removal
in patients for whom continued immobilization is unnecessary.
Results from various studies have defined practice
standards in treating the asymptomatic trauma patient.
Cervical spine radiographs are not required in trauma
patients without neck pain or tenderness who are
awake, alert, not intoxicated, and have no distracting injuries. This standard of care is supported by class I evidence from at least nine large prospective studies involving almost 40,000 patients.
In contrast to the low incidence of spinal injury in asymptomatic patients, there is a 2% to 6% incidence of significant cervical spine injury requiring treatment in patients who present with neck pain. It is generally agreed
that symptomatic trauma patients with neck pain, tenderness, neurologic deficit, altered mental status, or distracting injuries require radiographic evaluation of the cervical spine before collar removal. Based on available class I
evidence, a practice standard has also been suggested for

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

radiographic evaluation of the cervical spine in symptomatic trauma patients. Specifically, a cervical spine series
consisting of AP and lateral views in addition to openmouth odontoid views is recommended. Supplemental CT
examination is recommended to provide more detail of inadequately visualized levels.The most common reason for
missing a significant injury appears to be inadequate visualization of the injured level, most frequently the occipitoatlantoaxial region or cervicothoracic junction. However, even with adequate plain radiographic visualization,
it has been estimated that this three-view series will miss
15% to 17% of injuries.
Following initial plain radiographs and possibly a CT
scan, multiple options for determining safe collar removal in symptomatic patients have been proposed. The
negative predictive value of a three-view series and CT
is greater than 99%, and in certain instances supplemental CT evaluation may be sufficient. However, despite
the absence of apparent osseous injury, instability can
exist from spinal soft-tissue disruption of ligaments,
facet capsules, and disk tissue. MRI is exquisitely sensitive for acute soft-tissue injury and may be an option,
but the incidence of MRI abnormalities has been shown
to be between 25% to 40%, suggesting that MRI may
be oversensitive. Moreover, MRI is only reliable for
identifying soft-tissue injury within 48 hours of the traumatic event.
Flexion-extension radiographs are frequently obtained to rule out significant instability. In awake and alert
patients, active flexion-extension radiographs are safe, and
no significant complications have been reported.The negative predictive value of plain films in conjunction with
flexion-extension views is in excess of 99%.
The most controversy surrounds cervical spine clearance in the obtunded patient. Again, MRI has been suggested as an adjunctive test but may be of limited usefulness because of the lack of correlation between MRI
findings and clinically significant injury. Passive flexionextension manipulation of the cervical spine under fluoroscopy has been advocated by several investigators.
However, there is a theoretic risk of causing iatrogenic
spinal cord injury to these patients because of unrecognized disk herniation. It has been suggested that many
obtunded patients are at low risk for any significant injury and can be cleared on the basis of plain radiographs and CT. Proposed high-risk criteria indicating
the need for further evaluation include high-velocity
motor vehicle accidents (> 35 mph), any fall from a
height of more than 10 feet, closed-head injuries, neurologic deficits referable to the cervical spine, and pelvis
or extremity fractures.

Anatomy of the Cervical Spine

The cervical spine includes seven vertebrae. The rostral
two vertebrae, the atlas (C1) and axis (C2), have distinct

American Academy of Orthopaedic Surgeons

Chapter 41 Adult Spine Trauma

morphology and biomechanical characteristics. The third
through sixth vertebrae (C3-C6) are considered typical
cervical vertebrae with small cylindrical bodies and
short bifid spinous processes. The seventh cervical vertebra (C7) is a transitional vertebra between the cervical
and thoracic spinal regions and has a large nonbifid
spinous process known as the vertebra prominens.
The atlas is a ring containing two articular lateral
masses with neither a body nor spinous process (Figure
3). Incomplete formation of the posterior ring is relatively common as a developmental variation and does
not represent a traumatic injury. The axis contains the
odontoid process or dens, which articulates with the anterior arch of the axis against which it is stabilized by
the transverse ligament (Figure 4).
Morphology of the subaxial cervical spine from C3
to C6 is relatively consistent. The articular processes are
located at the junction of the laminae and pedicles and
form pillars referred to as the lateral masses (Figure 5).
The vertebral bodies are predominantly cancellous bone
surrounded by thin cortical bone. Superior end plates
are concave in the coronal plane and convex in the sagittal plane, whereas the matching inferior end plates are
convex in the coronal plane and concave in the sagittal
plane. The uncinate processes are osseous projections
off the posterolateral surfaces of the superior end
plates. Their articulation with the convex inferolateral
surface of the more rostral vertebra comprises the joints
of Luschka.
The posterior facet joints caudal to C2 are encapsulated synovial joints with overlying hyaline cartilage and
containing small menisci. The facet joint angle approximates 45 in the sagittal plane. The transverse processes
of each cervical vertebra contain a vascular foramen. The
vertebral artery typically passes anterior to the transverse
processes of C7 before entering the spine at the C6 vascular foramen. An accessory vertebral vein typically occupies the C7 foramen. At C1, the vertebral arteries pass
through a foramen, and then turn posteromedially around
the superior articular process before entering the foramen
magnum and joining to form the basilar artery. An anomalous course for the vertebral artery is present 3% of the
time and most commonly involves medial deviation toward the vertebral body at its midpoint. The incidence of
unilateral absence or hypoplasia of the vertebral artery is
5% to 10%. Arterial supply of the anterior two thirds of
the cervical spinal cord derives from a single anterior spinal artery, which is fed by segmental arteries derived from
the vertebral arteries. Paired posterior spinal arteries derived from either the vertebral arteries or the posterior inferior cerebellar artery supply the posterior one third of
the cord. Branches of the vertebral arteries provide a robust blood supply for individual vertebra. Odontoid nonunions were previously attributed to inadequate arterial
perfusion; however, recent anatomic studies have revealed
an extensive arterial arcade surrounding the odontoid

511

Adult Spine Trauma

Orthopaedic Knowledge Update 8

Figure 3 Illustrations of the osseous anatomy of the first cervical vertebra (C1; atlas). A, Cranial view of the atlas. B, Caudal view of the atlas. (Reproduced with permission from
Heller JG, Pedlow FX Jr: Anatomy of the cervical spine, in Clark CR (ed): The Cervical Spine, ed 3. Philadelphia, PA, Lippincott-Raven, 1998, pp 3-36.)

Figure 4 Illustration of the ligamentous stabilizers of the atlantoaxial segment showing the relationship among the transverse (TR), alar (AL), and atlantodens (AD) ligaments. (Reproduced with permission from Heller JG, Pedlow FX Jr: Anatomy of the
cervical spine, in Clark CR (ed): The Cervical Spine, ed 3. Philadelphia, PA, LippincottRaven, 1998, pp 3-36.)

process supplied by anterior and posterior ascending

branches of the vertebral arteries and pharyngeal
branches from the external carotid arteries.Venous drainage from the cervical spine occurs through pairs of external veins traveling with these major supplying arteries as
well as through an internal epidural plexus of valveless sinuses, both of which drain into the superior vena cava and
azygos vein.

512

The central opening in each cervical vertebra is

known as the vertebral canal, which in continuity with
the vertebral canal of adjacent vertebrae constitutes the
spinal canal containing the spinal cord. The average
sagittal diameter of the spinal canal averages 23 mm at
C1 and decreases progressively to 15 mm at C7. Nerve
root pairs exit the canal at each level through the intervertebral foramen formed by notches on the inferior
and superior aspects of adjacent pedicles. The anterior
border of the foramen is formed by the posterolateral
uncovertebral joint and intervertebral disk, whereas the
posterior border is formed by the caudal superior articular facet. Each nerve root normally occupies one third
of the cross-sectional area of the foramen. The C3
through C8 nerve roots exit anterior to the facet joints
in contrast to the C2 nerve roots, which exit posterior to
the C1-2 facet joint. The spinal nerves pass posterior to
the vertebral artery at approximately the middle of the
corresponding lateral mass. Ventral rami of C1 to C4
make up the cervical plexus and provide innervation to
the cervical strap muscles and the diaphragm. Ventral
rami of C5 to C8 along with T1 make up the brachial
plexus. The sympathetic chain lies bilaterally in proximity to the carotid sheath between the longus capitis and
longus colli muscles.
Stability of the cervical spine is highly dependent on
the integrity of the intervertebral ligaments and disks.
The osseous anatomy of the occipitocervical junction
provides little inherent stability. Instead, the anterior
and posterior atlanto-occipital membranes and the articular capsules provide most of the stability to the craniocervical junction (Figure 6). The atlantoaxial joint provides 50% of overall cervical rotation, with stability

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 41 Adult Spine Trauma

Figure 5 Cranial (A) and lateral (B) illustrations of the osseous anatomy of the subaxial cervical spine. (Reproduced with permission from Heller JG, Pedlow FX Jr: Anatomy of
the cervical spine, in Clark CR (ed): The Cervical Spine, ed 3. Philadelphia, PA, Lippincott-Raven, 1998, pp 3-36.)
again provided largely by specific ligamentous structures. In the horizontal plane, the transverse ligament is
the primary stabilizer, whereas the apical ligament and
the paired alar ligaments constitute secondary stabilizers (Figure 4). In the subaxial spine, the anterior and
posterior longitudinal ligaments and intervertebral disk
provide significant resistance to shear forces. Posteriorly,
the ligamentum nuchae, interspinous ligaments, and ligamentum flavum comprise the posterior ligamentous
complex and provide primary resistance against flexion
distraction forces.
The cervical spinal cord is ovoid in shape. It is narrower in the sagittal plane and has an expansion between C3 and C6 to provide innervation to the upper
extremities. The white matter of the spinal cord exists in
the periphery and contains bundles of myelinated axonal tracts that are divided into three discernable columns (Figure 7). The posterior columns conduct ascending proprioceptive, vibratory, and tactile signals from the
ipsilateral side of the body. The lateral columns contain
the lateral spinothalamic tracts, which conduct ascending pain and thermal signals for the contralateral side, as
well as the lateral corticospinal tracts, which conduct
85% of descending voluntary motor signals for the ipsilateral side of the body. The anterior columns contain
the anterior spinothalamic tracts, which conduct ascending light touch signals from the contralateral side, as
well as the anterior corticospinal tracts, which conduct
descending signals underlying fine motor control. Recent evidence suggests that the previously reported
highly organized laminar structure of the white matter
tracts probably does not exist.

Cervical Spinal Instability

The concept of spinal instability is vague and controversial but remains central to clinical decision making in

American Academy of Orthopaedic Surgeons

Figure 6 Illustration of the local anatomy of the occipitocervical junction. (Reproduced with permission from Heller JG, Pedlow FX Jr: Anatomy of the cervical spine, in
Clark CR (ed): The Cervical Spine, ed 3. Philadelphia, PA, Lippincott-Raven, 1998, pp
3-36.)

spinal trauma care. The classic definition by White and

Panjabi describes instability as the loss of ability of the
spine under physiologic loads to maintain its pattern of
displacement so that there is no initial or additional
neurologic deficit, no major deformity, and no incapacitating pain. This definition is elegant, but unfortunately

513

Orthopaedic Knowledge Update 8

Adult Spine Trauma

Figure 7 Illustrations of the cross-sectional anatomy of the cervical spinal cord. A, S = sacral; L = lumbar; T = thoracic; C = cervical. B, 1 = fasciculus gracilis; 2 = fasciculus
cuneatus; 3 = dorsal spinocerebellar tract; 4 = ventral spinocerebellar tract; 5 = lateral spinothalamic tract; 6 = spinoolivary tract; 7 = ventral corticospinal tract; 8 = tectospinal
tract; 9 = vestibulospinal tract; 10 = olivospinal tract; 11 = propriospinal tract; 12 = lateral corticospinal tract. (Reproduced with permission from Heller JG, Pedlow FX Jr:
Anatomy of the cervical spine, in Clark CR (ed): The Cervical Spine, ed 3. Philadelphia, PA, Lippincott-Raven, 1998, pp 3-36.)

Table 1 | Criteria for Identifying Instability at the Occipitocervical Junction

Table 2 | Checklist for the Diagnosis of Clinical Instability

in the Middle and Lower Cervical Spine

> 8
> 1 mm
> 7 mm
> 45
> 4 mm
< 13 mm

Element

Axial rotation C0-C1 to one side

C0-C1 translation
Overhang C1-C2 (total right and left)
Axial rotation C1-C2 to one side
C1-C2 translation
Posterior body C2 to posterior ring C1
Avulsed transverse ligament

(Adapted with permission from White AA, Panjabi MM: Clinical Biomechanics of the Spine,
ed 2. Philadelphia, PA, Lippincott Williams & Wilkins, 1990.)

it is difficult to translate directly into clinical practice.

Based predominantly on data from cadaveric studies,
more specific criteria have been suggested for identifying instability at the occipitocervical junction (Table 1)
and in the subaxial cervical spine (Table 2).

Occipital Condyle Fracture

Diagnosis of occipital condyle fractures has increased in
frequency because of greater use of CT evaluation of
the cervical spine in trauma patients. Sensitivity of plain
radiography for diagnosis is as low as 3%. Occipital
condyle fractures should be considered a marker for potentially lethal trauma, with an 11% mortality rate from
associated injuries. The rate of associated cervical spine
injury at an additional level is 31%. According to the
most commonly used classification system for occipital
condyle fractures developed by Anderson and Montesano, type I fractures (3% of fractures) are comminuted
fractures resulting from axial load, type II fractures
(22%) involve extension of a basilar skull fracture into
the condyle, and type III fractures (75%) are avulsion

514

Point Value

Anterior elements destroyed or unable to function

Posterior elements destroyed or unable to function
Positive stretch test
Radiographic criteria
A. Flexion/extension radiographs
a. Sagittal plane translation > 3.5 mm or 20%
(2 points)
b. Sagittal plane rotation > 20 (2 points)
or
B. Resting radiographs
a. Sagittal plane displacement > 3.5 mm or
20% (2 points)
b. Relative sagittal plane angulation > 11
(2 points)
Abnormal disk narrowing
Developmentally narrow spinal canal
1. Sagittal diameter < 13 mm
or
2. Pavlovs ratio < 0.8 mm
Spinal cord damage
Nerve root damage
Dangerous loading anticipated
Total (5 points or more = unstable)

2
2
2
4

1
1

2
1
1

(Adapted with permission from White AA, Panjabi MM: Clinical Biomechanics of the Spine,
ed 2. Philadelphia, PA, Lippincott Williams & Wilkins, 1990.)

fractures and should raise clinical suspicion for an underlying occipitocervical dissociation. Cranial nerve palsies may develop days to weeks after injury and most
frequently affect cranial nerves IX, X, and XI or result
in visual disturbance. Treatment of occipital condyle

American Academy of Orthopaedic Surgeons

Chapter 41 Adult Spine Trauma

Orthopaedic Knowledge Update 8

fractures requires ruling out occipitocervical dissociation, particularly in patients with type III fractures, followed by external immobilization in a cervical orthosis.

Occipitocervical Dissociation
Most instances of traumatic occipitocervical dissociation
are lethal, with approximately 100 cases having been reported in the literature. Survivors may demonstrate a
wide range of neurologic injuries ranging from complete
spinal cord lesions to isolated cranial nerve palsies
(most commonly affecting cranial nerves VI, X, and
XII). Occipitocervical dissociation injuries have been
classified as anterior type I, longitudinal type II, or posterior type III.
Diagnosis of this condition can be challenging because of the poorly visualized osseous detail on plain radiographs of this region. The most frequently described
measurement is the Powers ratio, which divides the basion to posterior arch distance by the anterior arch to
opisthion distance. A ratio greater than 1 suggests possible anterior dissociation. Other measurements considered suggestive of injury include a basion to odontoid
distance greater than 10 mm, posterior mandible to anterior atlas distance greater than 13 mm, and posterior
mandible to odontoid distance greater than 20 mm. The
Harris basion-axial intervalbasion-dental interval
method measures distance from the basion to a line
drawn tangentially to the posterior border of C2 (a distance greater than 12 mm or less than 4 mm is abnormal) as well as the distance from the basion to the odontoid (greater than 12 mm is abnormal) and is
considered by some to be the most sensitive measurement. Overall, the sensitivity of plain radiographs for
occipitocervical dissociation is approximately 57%. The
sensitivity of CT and MRI has been estimated to be
84% and 86%, respectively, and one or both of these adjunctive studies is recommended for patients with suspected occipitocervical dissociation injuries.
Use of traction is associated with a 10% rate of neurologic deterioration and should be avoided in patients
with these injuries. In patients with survivable injuries,
an instrumented occipitocervical fusion is recommended. Different techniques of fixation have been described, including occipital and cervical wiring, wire
mesh and methylmethacrylate, and occipitocervical plating. More recently, modular occipital plates have been
developed that can be rigidly locked to longitudinal
rods placed across the subaxial cervical spine (Figure 8).

Atlas Fractures
Fractures of the atlas constitute approximately 7% of
cervical spine fractures. Jefferson fractures are bilateral
fractures of the anterior and posterior arches resulting
from an axial load (Figure 9). Long-term stability depends on the mechanism of injury and subsequent heal-

American Academy of Orthopaedic Surgeons

ing of the transverse ligament. Because data from cadaveric studies indicate that a combined lateral mass
displacement in excess of 7 mm strongly suggests ligament disruption, an 18% radiographic magnification
factor has been incorporated, resulting in an increase in
the measurement to 8.1 mm. However, with improvements in MRI technology, it has become a more sensitive means of detecting a ligamentous injury than plain
radiographs. Two types of transverse ligament injury
have been described. Midsubstance ruptures (type I injuries) are least likely to heal, and early surgical treatment with C1-2 fusion may be necessary. Type II injuries
involve an avulsion fracture of the ligamentous insertion. Because of higher rates of healing, an initial attempt at external immobilization using a halo vest is a
reasonable treatment option in these patients.
In general, most isolated anterior or posterior arch
fractures and lateral mass and transverse process fractures of the atlas can be treated conservatively with 6 to
12 weeks of external immobilization. Burst fractures involving both anterior and posterior arches with an intact
transverse ligament are considered stable injuries that
should also be treated with external immobilization.
Disruption of the transverse ligament introduces the option of early surgical fusion, typically involving a posterior C1-2 fusion. Multiple procedures have been described, including various wiring techniques and screwrod constructs. C1-2 transarticular screw placement is
the most stable form of fixation currently in general use
and obviates the need for postoperative halo immobilization required with C1-2 wiring techniques.

Axis Fractures
Odontoid fractures are the most common type of axis
fracture and have been classified by Anderson and
DAlonzo as type I avulsion fractures of the tip, type II
fractures through the waist of the odontoid process, or
type III fractures extending into the C2 vertebral body.
Nearly all odontoid fractures will require some form of
treatment.
Type I fractures can be treated with an external
orthosis once the possibility of an associated occipitocervical dissociation has been excluded. Type III fractures have been reported to have a sufficiently high
healing rate with rigid external immobilization in a halo
vest. Treatment of type II fractures is controversial and
depends largely on specific patient and fracture characteristics. Elderly patients tolerate halo vest immobilization poorly, demonstrate decreased healing rates, and
should be considered for early C1-2 surgical fusion. Elderly debilitated patients who are at increased risk of
medical complications from surgical treatment can be
treated with an external orthosis for 6 to 12 weeks with
the understanding that successful fusion is unlikely to
occur. In most patients, a fibrous nonunion develops

515

Adult Spine Trauma

Orthopaedic Knowledge Update 8

Figure 8 Lateral (A) and AP (B) radiographs of occipitocervical segmental fixation implants (DePuy AcroMed Summit system, Raynham, MA).
that provides sufficient stability for routine daily activities. Nevertheless, these patients should be informed
that they remain at risk for spinal cord injury in the
event of a fall or motor vehicle accident.
In younger, healthy patients with a type II fracture,
specific fracture characteristics assume increased significance. Nondisplaced fractures diagnosed early should be
treated with halo vest immobilization for 6 to 12 weeks.
Risk factors for nonunion include fracture comminution, displacement of more than 6 mm, posterior displacement, delay in diagnosis, and patient age greater
than 50 years. Early surgical treatment is an option for
patients with any of these risk factors. Surgical treatment should also be considered for fractures in which
reduction cannot be achieved or maintained.
Anterior odontoid screw osteosynthesis using a single screw placed with lag technique is an option for
treating both type II and type III noncomminuted fractures. For the best results using this technique, the fracture should be diagnosed early, reduction must be possible, and patient body habitus must allow achievement of
proper intraoperative screw trajectory (Figure 10). In
addition, odontoid fracture obliquity should run from
anterosuperior to posteroinferior. Otherwise, surgical
treatment may involve any of several methods for ac-

516

complishing a posterior C1-2 fusion. Traditional sublaminar wiring techniques (Gallie and Brooks fusions) are
being supplemented or replaced by more rigid fixation
methods such as C1-2 transarticular screws and C1 lateral mass and C2 pedicle screw-rod constructs (Figure
11). These more rigid fixation techniques provide the
surgeon with the opportunity to avoid postoperative
halo vest immobilization.

Traumatic Spondylolisthesis of the Axis

This injury is characterized by bilateral fractures of the
pars interarticularis and is also known as a hangmans
fracture, although recent efforts have been made to replace this term with traumatic spondylolisthesis of the
axis. Most injuries of this type that occur in the United
States are the result of motor vehicle accidents, and the
forces involved are likely a combination of hyperextension, compression, and rebound flexion. The currently
popular classification scheme is Levine and Edwards
modification of previous systems described by Effendi
and Francis and is based on mechanism of injury (Figure 12). Type I fractures result from axial compression
and hyperextension and demonstrate less than 3 mm
displacement and no angulation. Type II fractures result

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 41 Adult Spine Trauma

Figure 9 Illustrations of common atlas fracture patterns. A, posterior arch fracture. B,

Lateral mass fracture. C, Classic burst (Jefferson fracture). D, Unilateral anterior arch fracture. E, Transverse process fracture. F, Anterior
arch avulsion fracture. (Reproduced with permission from Klein GR, Vaccaro AR: Cervical
spine trauma: Upper and lower, in Vaccaro AR,
Betz RR, Zeidman SM (eds): Principles and
Practice of Spine Surgery. Philadelphia, PA,
Mosby, 2003, pp 441-462.)

from hyperextension and axial load followed by rebound flexion and demonstrate translation of greater
than 3 mm as well as angulation. Type IIA fractures are
characterized by angulation without significant translation and result from a flexion-distraction injury. Identification of this fracture type is important because application of traction may cause further fracture displacement
and should be avoided. Type III fractures are essentially
type I pars fractures associated with injury to the C2-3
facet joints, most commonly bilateral facet dislocation.
These fractures are thought to result from flexiondistraction followed by hyperextension. A type IA fracture pattern was recently added to this classification system to describe asymmetric fracture lines with minimal
translation and no angulation. This fracture is thought to
result from hyperextension with a component of lateral
bending forces.
Most patients with traumatic spondylolisthesis of the
axis can be successfully treated with 6 to 12 weeks of external immobilization in a cervical orthosis or halo vest.
Data from a recent cadaveric study have suggested that

American Academy of Orthopaedic Surgeons

up to 5 mm of displacement can occur without disruption

of posterior ligaments or the C2-3 disk. Suggested indications for surgical treatment include type II fractures with
severe angulation, type III fractures with disruption of the
C2-3 disk, or inability to achieve or maintain fracture reduction. All type III fractures associated with facet dislocation require open reduction and fusion. Surgical options
include anterior C2-3 interbody fusion, posterior C1-3 fusion, or bilateral C2 pars screw osteosynthesis.

Fractures and Dislocations of the Subaxial Spine

The most commonly used classification system for subaxial spine injuries is the system developed by Allen and
associates. Six distinct phylogenies were described based
on mechanism of injury, with each phylogeny being subdivided into stages of progressive severity. The three
commonly observed categories are compressive flexion,
distractive flexion, and compressive extension. Vertical
compression is observed less commonly, and the two
least common categories are distractive extension and
lateral flexion.

517

Adult Spine Trauma

Orthopaedic Knowledge Update 8

of disrupted posterior elements. In patients with this
type of injury, supplemental posterior instrumented fusion should be considered.

Figure 10 Odontoid radiograph of anterior odontoid screw fixation. (Reproduced with

permission from Klein GR, Vaccaro AR: Cervical spine trauma: Upper and lower, in
Vaccaro AR, Betz RR, Zeidman SM (eds): Principles and Practice of Spine Surgery.
Philadelphia, PA, Mosby, 2003, pp 441-462.)

Compression and Burst Fractures

Compression fractures are defined by compressive failure of the anterior half of the vertebral body (anterior
column) without disruption of the posterior body cortex
and without retropulsion into the spinal canal. Patients
are typically neurologically intact, and in the absence of
significant deformity or instability, most fractures can be
treated with external immobilization for 6 to 12 weeks.
Burst fractures are characterized by compressive
failure of the vertebral body with fracture extension
through the posterior body cortex and some degree of
bone retropulsion into the spinal canal. Burst fractures
have a significantly higher rate of instability when compared with compression fractures and are frequently associated with spinal cord injury. A burst fracture in the
presence of either complete or incomplete spinal cord
injury will typically require surgical decompression and
stabilization. In most patients with burst fractures, surgical decompression can be best achieved through an anterior approach with corpectomy of the involved level(s). Interbody fusion with anterior plate fixation may
be insufficient to stabilize these fractures in the setting

518

Facet Dislocations
Patients with cervical facet dislocations should undergo
timely reduction of their injuries upon diagnosis. Classically, plain radiographic evidence of vertebral body subluxation of 25% has been reported to suggest a unilateral facet dislocation, whereas vertebral body
subluxation of 50% has been reported to suggest a bilateral dislocation (Figure 13). Unilateral facet dislocations
are commonly associated with a monoradiculopathy
that improves following application of traction. Bilateral
facet dislocations are often associated with significant
spinal cord injuries. Awake and alert patients can safely
undergo closed reduction with progressive application
of axial traction forces. It is crucial that these patients
be closely monitored with serial neurologic examinations and plain radiographic assessment following the
placement of each additional weight. Development of
new or worsening neurologic deficits is an indication to
cease attempts at closed reduction, and an MRI scan
should be obtained to rule out herniated disk material.
Overall, as many as 26% of patients with cervical facet
dislocations will fail attempted closed reduction, with
higher failure rates observed for patients with unilateral
facet dislocations. Although 50% of patients with facet
dislocations will demonstrate signs of disk disruption on
MRI, most of these signs are of uncertain clinical significance.
Following successful closed reduction, surgical stabilization of these injuries is typically necessary. In the absence of a traumatic disk herniation, posterior instrumented fusion is recommended. However, various
treatment options are available for associated disk herniation. After successful closed reduction, an MRI scan
of the cervical spine should be obtained to rule out the
presence of associated disk herniation. Anterior diskectomy and fusion with anterior plating has been reported
to be successful, although associated with kyphotic deformity in some instances. For this reason, patients with
persistent kyphosis following closed reduction who have
anterior compression from disk material may require
anterior decompression and grafting with a concomitant
posterior stabilization.
Failure of closed reduction mandates open reduction
and instrumented cervical fusion. A preoperative MRI
scan is required to rule out the presence of a herniated
disk, which, if present, indicates the need for anterior
diskectomy and fusion. A unilateral facet dislocation
can often be reduced anteriorly using vertebral body
Caspar pins. This technique involves maintaining distraction across these pins followed by rotation of the
proximal pin toward the side of the dislocated facet
joint. Alternatively, for unilateral or bilateral disloca-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 41 Adult Spine Trauma

Figure 11 Lateral (A) and AP (B) radiographs showing C1-2 transarticular screw fixation.

Figure 12 Illustrations of types of traumatic spondylolisthesis of the axis using the Levine and Edwards modification of the Effendi classification system. A, Type I. B, Type II.
C, Type IIA. D, Type III. (Reproduced with permission from Klein GR, Vaccaro AR: Cervical spine trauma: Upper and lower, in Vaccaro AR, Betz RR, Zeidman SM (eds): Principles
and Practice of Spine Surgery. Philadelphia, PA, Mosby, 2003, pp 441-462.)
tions associated with a significant disk herniation, an
anti-kickout plate can be used to avoid the need for a
second anterior stage of the surgery. Following anterior
diskectomy and placement of an interbody graft, the
plate is fixed only to the superior body. Open reduction
can then be performed posteriorly with the plate maintaining position of the graft.
Unilateral facet fractures are the most frequently
missed significant cervical spine injuries on plain radiographs. These injuries may reduce in the supine position
during initial trauma evaluation. Subtle subluxations or
rotational malalignment visualized on plain radiographs
should prompt further study with CT. The superior facet
is more frequently fractured. Fractures without signifi-

American Academy of Orthopaedic Surgeons

cant kyphosis or subluxations can be considered for

nonsurgical treatment with 6 to 12 weeks of external immobilization in a halo vest or cervical orthosis. Kyphotic
deformity, significant subluxations, or radiculopathy
should be considered for open reduction and posterior
instrumented stabilization. Bilateral facet fractures and
lateral mass dissociations are typically unstable injuries
and may be treated with anterior or posterior instrumented fusion.

Thoracic Spine Trauma

Thoracic spine fractures represent approximately 16%
of fractures involving the thoracic and lumbar spine.
These injuries typically result from high-energy blunt

519

Orthopaedic Knowledge Update 8

Adult Spine Trauma

Figure 13 Bilateral cervical spinal facet dislocations. A, Plain radiograph demonstrates

50% subluxation of C4 on C5. B, T2-weighted
MRI scan reveals large posteriorly extruded
disk fragment at the level of injury. The patient
underwent an anterior decompression and anterior open reduction and instrumented fusion. (Reproduced with permission from Klein
GR, Vaccaro AR: Cervical spine trauma: Upper
and lower, in Vaccaro AR, Betz RR, Zeidman
SM (eds): Principles and Practice of Spine
Surgery. Philadelphia, PA, Mosby, 2003, pp
441-462.)
trauma, and patients must be evaluated thoroughly for
other injuries. Although they represent only a small percentage of thoracolumbar fractures, the narrow thoracic
spinal canal and the precarious blood supply to the thoracic spinal cord make these injuries potentially devastating.
Fractures of the thoracic spine from T2 through T10
are relatively infrequent as a result of unique anatomic
features of the thoracic spine that provide enhanced stability. Fractures of the thoracolumbar spine from T11
through L2 will be discussed separately because the
transition from the rigid thoracic spine to the more flexible lumbar spine creates a transition zone that is susceptible to injury.

Biomechanics
The thoracic spine has several unique anatomic features
that provide it with more stability than either the cervical or lumbar spine. Each thoracic vertebra has articulations with and ligamentous attachments to the adjacent
ribs at both the transverse process and the vertebral
body that increase rigidity. In addition, the ribs articulate with the sternum, which provides another point of
fixation and thereby limits thoracic motion. The facets
of the thoracic spine are oriented in the coronal plane,
with the lamina and spinous processes arranged in a
shingled fashion, which reduces the amount of extension
of the thoracic spine. The intervertebral disks of the thoracic spine are very thin, which provides increased stiffness and more rotational stability. The vertebral bodies
are asymmetric in height from anterior to posterior. The

520

vertebral bodies are typically 1 to 2 mm smaller in

height anteriorly, resulting in the kyphosis seen in the
thoracic spine. The kyphotic alignment of the thoracic
spine results in the concentration of axial load transmission in the anterior column. The apex of the kyphosis is
normally over T6 through T8, where maximal anterior
column forces are generated and fractures within the rib
cage are most likely to occur.

Diagnostic Imaging
Plain films of the thoracic spine should provide an initial assessment of any fracture as well as its impact on
overall sagittal alignment. It is of paramount importance
to evaluate the entire spine once a fracture is identified
because concomitant spinal fractures can be present in
up to 20% of patients. CT scans should be obtained in
patients with fractures on plain films, in any multiple
trauma patient with lower extremity neurologic deficits,
and for any patients with inadequate plain films. The axial images will help evaluate the vertebral body fracture
anatomy, pedicle anatomy, and presence of bony retropulsion. Coronal and sagittal reconstructions may reveal
fractures missed in the axial plane, and they provide a
better appreciation of the fracture anatomy. MRI can be
useful in evaluating the soft tissues of the spine. The anterior and posterior ligamentous complexes can be evaluated for injury as well as spinal cord encroachment by
either disk or osseous material. In addition, the spinal
cord can be evaluated for the presence of edema or
hemorrhage.

American Academy of Orthopaedic Surgeons

Chapter 41 Adult Spine Trauma

Orthopaedic Knowledge Update 8

Classification systems for spinal fractures have been
difficult to standardize because of the complex loads
that are applied to result in these injuries. A comprehensive classification system has been proposed for thoracic and lumbar fractures based on their radiographic
morphology. Three basic forcescompression, distraction, and rotationwere found to be responsible for the
injuries and are labeled A, B, and C, respectively. In addition, each group is subdivided into three subgroups.
The severity of the injury increases from group A to
group C. Because describing every variant in this system
is beyond the scope of this text, only the classic examples from each group will be discussed.

Compression Fractures
In its purest form, a compression fracture results from
an axial load applied to the spine. The anterior column
of the spine (vertebral body and disk) is involved without involvement of the middle column (posterior vertebral cortex and posterior longitudinal ligament). Radiographs will demonstrate a wedge-shaped defect in the
vertebral body that results in varying degrees of kyphosis. Neurologically intact patients with less than 30 of
kyphosis and less than 50% loss of vertebral body
height can be treated with a hyperextension orthosis.
Patients with more than 30 of kyphosis or more than
50% loss of vertebral body height can also be treated
nonsurgically, but must be watched carefully for possible
failure of the posterior ligamentous structures. These patients are more likely to develop a kyphotic deformity,
and close radiographic and clinical follow-up is warranted. In patients with fractures above T6, a cervical
extension on the thoracolumbosacral orthosis should be
used for better control.

Burst Fractures
Burst fractures also result from an axial load and are inherently unstable because of the involvement of the anterior and middle columns of the spine. Radiographs
will typically demonstrate a widening of the pedicles at
the affected vertebral body and a varying degree of
bony retropulsion into the canal. Treatment of these injuries must be based on consideration of neurologic status, sagittal alignment, and the integrity of the posterior
ligamentous structures. Patients who are neurologically
intact with less than 30 of kyphosis and less than 50%
loss of vertebral body height can be managed in a thoracolumbosacral orthosis. Patients with more than 30 of
kyphosis and more than 50% loss of vertebral body
height with failure of the posterior column should be
considered for a posterior stabilization procedure.
Patients with incomplete neurologic deficits and radiographic evidence of spinal cord compression may
benefit from an acute anterior decompression and stabilization. However, the benefit of early versus delayed

American Academy of Orthopaedic Surgeons

decompression and stabilization of these injuries has yet

to be demonstrated in the scientific literature. Patients
who have a complete neurologic injury are typically stabilized posteriorly in a nonemergent fashion to facilitate
earlier rehabilitation.

Flexion-Distraction Injuries
Flexion-distraction injuries result from a flexion moment that is combined with a fulcrum located at varying
distances from the anterior portion of the vertebral column. The resulting injury can involve bone, ligament, or
a combination of bone and ligament. When the fulcrum
is located adjacent to the vertebral body, the anterior
column will fail in compression and the middle and posterior columns will fail in tension. As the fulcrum moves
more anterior, the deforming forces become purely distractive and all three columns will fail in tension. Patients with pure ligamentous and combined bony and
ligamentous injuries should undergo a posterior stabilization procedure because of involvement of all three
columns and the poor healing properties of ligaments.
Typically, these injuries can be treated with shortsegment posterior compression constructs. Patients with
a purely bony injury can undergo reduction in extension
and can be treated in a thoracolumbosacral orthosis because of the healing properties of bone. They should be
assessed for nonunion and deformity progression.

Fracture-Dislocations
Fracture-dislocations are common injuries in the thoracic spine because of the significant forces acting on a
rigid portion of the spine. Up to 90% of these injuries
are associated with a spinal cord injury, most commonly
complete (ASIA impairment scale category A). Because
all three spinal columns are involved, these fractures are
very unstable. In patients with complete neurologic injury, a posterior stabilization procedure can be performed once their clinical condition is optimized. This
will allow for early mobilization and help minimize the
morbidity and mortality associated with these injuries.

Thoracic Pedicle Screws

The use of thoracic pedicle screws has become more
widespread based on numerous reports of safe placement techniques and success in clinical practice. Pedicle
screws offer several advantages over traditional hook
and rod constructs. The pedicle is the strongest bony attachment to the vertebra and provides excellent purchase, even in osteoporotic bone. Pedicle screws provide
fixation of all three spinal columns, which provides for
better sagittal and coronal stability. This increased stability can decrease the number of levels included in a
fusion construct.
The thoracic pedicles are the narrowest in the T4-6
region and become progressively larger toward the lum-

521

Orthopaedic Knowledge Update 8

Adult Spine Trauma

Figure 14 A, From the levels T3 through T10,

the entrance point to cannulate the pedicle
can be determined by using the intersection
point between a line drawn from the superior
border of the transverse process and a line
drawn from the midpoint of the cephalad facet. B, At the levels of T11-12, the entrance
point to cannulate the pedicle can be determined by using the intersection point between
a line drawn from the middle of the transverse
process and a line drawn from the midpoint of
the cephalad facet.
bar spine. Preoperative planning with CT is critical in
evaluating the diameter of the pedicles to ensure they
can accommodate instrumentation. Pedicle screws can
be placed with the assistance of fluoroscopy or by using
anatomic landmarks. One method that has been described for screw placement between T3 and T10 uses
the intersection of a vertical line through the middle of
the cephalad articular surface and a horizontal line at
the superior border of the transverse process (Figure 14,
A). At the T11 and T12 levels, the horizontal line should
be made from the middle of the transverse process (Figure 14, B).
The complications of thoracic pedicle screw placement can be devastating. Injuries to the spinal cord,
nerve roots, esophagus, and major blood vessels have all
been reported. Thorough preoperative planning with axial CT that documents the diameter and length of the
pedicle as well as the angle of insertion and intraoperative neurophysiologic monitoring are important for
avoiding complications.

Thoracolumbar Junction Trauma

Biomechanics
Injuries to the thoracolumbar junction (T11 through L2)
are usually the result of significant blunt trauma. These
injuries represent approximately 50% of all thoracic and
lumbar fractures. This increased incidence of fractures
to the thoracolumbar junction is the result of its location at the transition point between the rigid thoracic
rib cage and the more mobile lumbar spine. In addition,
the sagittal alignment of the spine changes from kyphosis to lordosis, which more evenly distributes stress on
the anterior and middle columns. The facet alignment
changes from a coronal orientation in the thoracic spine
to a sagittal alignment in the lumbar spine, allowing for
greater flexion and extension motion. Also, the intervertebral disks are taller than in the thoracic spine, decreasing anterior column stiffness.
The indications for surgical management of thoracolumbar fractures remain controversial. Central to defining the need for surgical management is the determi-

522

nation of fracture stability. A two-column model of the

spine was proposed in which it was hypothesized that
injuries with involvement of the posterior column are
associated with instability. Denis proposed a threecolumn concept that stressed the importance of the
newly defined middle column, which comprises the posterior 50% of the vertebral body and disk along with
the posterior longitudinal ligament (Figure 15). Based
on biomechanical data, Denis proposed that an isolated
posterior ligamentous complex injury was not sufficient
to produce instability and that concomitant involvement
of the middle column resulted in instability in flexion.
However, the integrity of the posterior ligamentous
complex (supraspinous and interspinous ligaments) is a
crucial factor in determining stability because posterior
column disruption in conjunction with involvement of
the anterior and middle columns creates instability and
increases the risk of posttraumatic kyphosis.

Treatment
The treatment of most patients with thoracolumbar
fractures is nonsurgical. Patients who are neurologically
intact with less than 25 kyphosis, less than 50% loss of
vertebral body height, and less than 50% of canal occlusion and an intact posterior longitudinal ligament are
the best candidates for nonsurgical treatment. Depending on the severity of the collapse, these patients may be
managed with hyperextension body casting and/or a
thoracolumbosacral orthosis for 3 months. Regular clinical and radiographic follow-up of these patients is important to rule out the development of progressive kyphosis or neurologic deficits.
Surgical treatment should be reserved for patients
with unstable fracture patterns and/or neurologic deficits. For patients with incomplete neurologic injuries
and spinal cord compression, anterior decompression
and stabilization is typically required. The anterior procedure may need to be performed in conjunction with a
posterior stabilization procedure in patients with posterior column involvement. Early stabilization of patients

American Academy of Orthopaedic Surgeons

Chapter 41 Adult Spine Trauma

Orthopaedic Knowledge Update 8

Figure 15 Denis three-column model of the

spine. (Reproduced with permission from Garfin SR, Blair B, Eismont FJ, Abitol JJ: Thoracic
and upper lumbar spine injuries, in Browner
BD, Jupiter JB, Levine AM, Trafton PG, Lampert
R (eds): Skeletal Trauma: Fractures, Dislocations, Ligamentous Injuries, ed 2. Philadelphia, PA, WB Saunders, 1997, pp 947-1034.)
with neurologic injuries will allow for early mobilization
and rehabilitation placement.
Patients with unstable burst fractures that include
failure of the posterior column should undergo an initial
posterior stabilization procedure. Posterior instrumentation on a lordosing frame can help restore the normal
sagittal alignment of the thoracolumbar spine and will
assist in the reduction via ligamentotaxis of the retropulsed fragment. However, patients with excessively
comminuted vertebral body fractures may be candidates
for a staged anterior procedure.
Recently, prospective studies have compared patients with stable burst fractures of the thoracolumbar
junction treated surgically and nonsurgically with regard
to pain, spinal canal compromise, and residual kyphosis.
In patients with mild to moderate initial kyphosis, much
of the sagittal correction gained surgically is lost over
time, approaching the final segmental kyphosis seen for
these fractures in the nonsurgical treatment group. In
addition, residual kyphosis has not been a reliable predictor of chronic pain. Similarly, the degree of spinal canal encroachment has also not been associated with injury to neural structures over the long term. Canal
remodeling is to be expected, and up to 50% of the canal can be restored. The degree of remodeling has been
reported to be similar in groups of patients treated surgically or nonsurgically.

American Academy of Orthopaedic Surgeons

Lower Lumbar Spine Trauma

Biomechanics
Fractures of the lower lumbar spine (L3-L5) are less
common than in the thoracolumbar region. The lower
lumbar spine is the lordotic alignment in the sagittal
plane, which results in a weight-bearing axis that falls
through the middle and posterior columns, making this
anatomic region more intrinsically stable. Such fractures
are unlikely to progress into a kyphotic deformity. In
addition, the facets of the lumbar spine are oriented in
the sagittal plane, which can accommodate greater
flexion-extension bending moments. Finally, the lumbosacral junction can withstand large forces transmitted
across it.
Because the load-bearing axis is more posterior,
compression fractures are much less common than burst
fractures in the lower lumbar spine. For the anterior column to fail, a significant flexion moment must be applied. However, these large flexion moments are typically neutralized by the increased ability of the lumbar
spine to accommodate loads in the sagittal plane. If a
compression fracture is detected, suspicion should be
high for a concomitant posterior ligamentous injury, especially in the setting of more than 50% loss of vertebral body height.
Burst fractures are the most common fracture pattern seen in the lower lumbar spine. Most of these injuries occur with the spine in a neutral position, resulting
in axial loading of both anterior and middle columns of
the spine. As with burst fractures in other parts of the
spine, a variable amount of bone may be retropulsed

523

Orthopaedic Knowledge Update 8

Adult Spine Trauma

Figure 16 A, An axial CT scan demonstrates the narrowest portion along the sacral pedicle. B, Sagittal view of the narrowest portion of the sacral ala with height and width
dimensions. C, Drawing demonstrating the adjacent bony structures and the L5 and S1 nerve roots. (Reproduced with permission from Noojin FK, Malkani AL, Haikal L, Lundquist
C: Cross-sectional geometry of the sacral ala for safe insertion of iliosacral lag screws: A computed tomography model. J Orthop Trauma 2000;14:31-35.)
into the spinal canal. However, the incidence of significant and/or permanent neurologic sequelae from these
injuries is much lower than elsewhere in the spinal column; the spinal cord ends above this level, and the
nerve roots of the cauda equina are more tolerant of
compression than the spinal cord.
Flexion-distraction injuries account for less than
10% of lumbar spine fractures. They are most commonly seen at L2, L3, or L4 and are typically the result
of the increased stability imparted at the level of L5 to
the pelvis via the iliolumbar ligaments. The large flexion
moments cause flexion of the upper lumbar segments,
whereas the lower segments are stabilized. As a result,
the posterior elements fail in tension from the distractive forces.

Treatment
Most patients with lower lumbar fractures can be
treated conservatively with external immobilization, a
short course of bed rest, and a custom-molded thoracolumbosacral orthosis. A single leg spica attachment
may be necessary for fractures of L4 and L5 to allow for
control of the pelvis and immobilization of the lumbosacral junction. Patients should wear a brace for approximately 8 to 12 weeks and undergo regular clinical
and radiographic follow-up.
Patients who have a cauda equina syndrome or significant neurologic deficit with canal compromise
should be considered for surgical treatment. Such patients will typically have near-complete canal occlusion
from bony fragments and should undergo decompression and a posterior stabilizing procedure using pedicle
screws. The decompression can usually be performed
posteriorly via a laminectomy. If a posterior decompression is going to be performed in a patient with a neurologic deficit, evaluation for the presence of a laminar
fracture should be sought. Case reports have described
herniated nerve root entrapment within the laminar
fracture site.

524

The length of the fusion in the lower lumbar spine

typically involves the fractured level and the cephalad
and caudad levels. If the fracture involves the L5 level,
fusion to the sacrum will be necessary, and attention to
the maintenance of the normal lordotic sagittal alignment is important. Isolated injuries of the posterior ligamentous complex may be addressed with a single-level
compression construct using screws or hooks. Patients
with these injuries may also require a course of postoperative immobilization in a thoracolumbosacral orthosis,
which possibly includes leg extension.

Sacral Spine Trauma

Biomechanics
Sacral fractures are usually the result of high-energy
trauma and occur in isolation in fewer than 5% of patients. As a result, a thorough evaluation of the pelvis
and spine should be undertaken for associated fractures.
Sacral fractures are generally classified according to the
direction of the fracture line. Fractures may be vertical,
transverse, or oblique, although vertical fractures are the
most common. Denis classification system divides the
sacrum into three zones. Zone 1 spans the sacral ala to
the lateral border of the neural foramen. Zone 2 represents the neural foramen. Zone 3 involves the central
sacrum and canal.
Sacral injuries commonly are associated with sacral
root deficits. Because of the relative kyphosis of the
sacrum, the sacral nerve roots are tethered and restricted within long bony tunnels and therefore not
much motion is allowed. These two anatomic factors
predispose the neural structures to injury. In addition,
the direction of the fracture line and type of fracture determine the likelihood of a neurologic injury. Neurologic
injuries of the lower sacral roots (S2-S4) are often
missed because only L5 and S1 can be evaluated by
manual muscle testing. Perianal sensory changes should
be sought in patients with this type of injury. In the
Denis classification system, zone 1 vertical injuries have
a 5.9% incidence of neural injury. If there is a neural in-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

jury, it is usually an injury to the L5 nerve root or the
sciatic nerve. Zone 2 fractures have an incidence of
28.4% of neural injury. The nerve injuries are commonly
unilateral in nature. This is important because patients
with unilateral sacral root injuries usually have normal
bowel and bladder function. Zone 3 injuries have a 50%
incidence of neurologic injuries, and most of these patients will have bilateral sacral root involvement with
bowel, bladder, and sexual dysfunction.

Treatment
Indications for surgical management of sacral fractures
include bowel and bladder dysfunction in the setting of
an unstable fracture with substantial coronal or sagittal
deformity. Vertical fractures can typically be treated
with posterior sacroiliac plating or percutaneously
placed sacroiliac screws. Placement of percutaneous sacroiliac screws can be technically demanding, and the L5
and S1 nerve roots are at risk during the procedure
(Figure 16). If this technique is used to treat a zone 2 injury, the screw should not be loaded in compression to
avoid neural injury. Patients who have displaced transverse or oblique fractures may undergo bilateral plating.
Neural decompression via laminectomy may be indicated in patients with neurologic deficits and canal compromise, and recovery of bowel and bladder function
may be seen. In patients without neurologic injury who
have minimally displaced fractures, bracing is only necessary in the setting of fractures, which extend to or
above the level of the sacroiliac joint.

Annotated Bibliography
Cervical Spine Trauma

Chapter 41 Adult Spine Trauma

restoring and maintaining sagittal alignment and minimizing
the rate of treatment failures.

Guidelines for management of acute cervical spinal injuries. Neurosurgery 2002;50(suppl 3):S1-S179.
Under the sponsorship of the American Association of
Neurologic Surgeons and the Congress of Neurologic Surgeons, a computerized English-language literature search was
performed covering the preceding 25 years. Studies involving
cervical spinal trauma were reviewed and critically evaluated.
Results were organized and presented in several topical sections designed to provide reasonable standards, guidelines, and
options for care of patients with acute cervical spinal injuries.

Peris MD, Donaldson WF, Towers J, Blanc R, Muzzonigro TS: Helmet and shoulder pad removal in suspected
cervical spine injury. Spine 2002;27:995-999.
This fluoroscopic study supports the efficacy of the current
protocol used by the National Athletic Trainers Association
that uses four individuals for the safe removal of the helmet
and shoulder pads from an injured football player with minimal cervical motion.

Vaccaro AR, Madigan L, Bauerle WB, Blescia BS, Cotler JM: Early halo immobilization of displaced traumatic spondylolisthesis of the axis. Spine 2002;27:22292233.
This retrospective study of 31 patients with traumatic
spondylolisthesis of the axis confirms that early halo immobilization after traction reduction is a safe and effective form of
treatment in this patient population. Patients with type II fractures angled greater than or equal to 12 may require more
extended periods of traction.

Blake WED, Stillman BC, Eizenberg N, Briggs C, McMeeken JM: The position of the spine in the recovery
position: An experimental comparison between the lateral recovery position and the modified HAINES position. Resuscitation 2002;53:289-297.

Thoracic Spine Trauma/Thoracolumbar Junction

Trauma/Lower Lumbar Spine Trauma

This study identifies a new patient recovery position, the

modified HAINES position, in which the patient is in the lateral position with the head lying on the fully abducted dependent arm and both legs are drawn up, with hips and knees
flexed. This position results in a more neutral position of the
spine compared with the traditional lateral recovery position
and may be preferable to the lateral recovery position during
resuscitation of trauma patients.

This prospective study evaluated 17 patients with flexiondistraction injuries of the thoracic and lumbar spine. The patients had an average follow-up of 17.6 months and were
treated with single-level posterior fixation. The average preoperative kyphosis was 10.1, which was corrected to a postoperative lordosis of 0.9. The mean Oswestry score was 11.5 and
88% of the patients reported having only minimal disabilities.

Finkelstein J, Wai E, Jackson S, Ahn H, Brighton-Knight

M: Single-level fixation of flexion distraction injuries.
J Spinal Disord Tech 2003;16:236-242.

Fisher CG, Dvorak MFS, Leith J, Wing PC: Comparison

of outcomes for unstable lower cervical flexion teardrop
fractures managed with halo thoracic vest versus anterior corpectomy and plating. Spine 2002;27:160-166.

Wood K, Butterman G, Mehbod A, Garvey T, Jhanjee

R, Sechriest V: Operative compared with nonoperative
treatment of a thoracolumbar burst fracture without
neurological deficit: A prospective, randomized study.
J Bone Joint Surg Am 2003;85:773-781.

This retrospective cohort study of 45 patients with cervical

flexion teardrop fractures indicates that anterior cervical plating is a safe and effective treatment of this patient population
and may be superior to halo vest immobilization in terms of

This prospective randomized trial evaluated 47 consecutive patients with a thoracolumbar fracture without neurologic
deficit for a minimum 2-year follow-up. The group of patients
that underwent surgical treatment had a mean preoperative

American Academy of Orthopaedic Surgeons

525

Adult Spine Trauma

and postoperative kyphosis of 10.1 and 13, respectively. The
group of patients that underwent nonsurgical treatment had a
mean preoperative and postoperative kyphosis of 11.3 and
13.8, respectively. No difference was found between groups
with respect to return to work. The average pain scores were
similar for both groups. The authors concluded that surgical
treatment provided no long-term advantage compared with
nonsurgical treatment.

Yue J, Sossan A, Selgrath C, et al: The treatment of unstable thoracic spine fractures with transpedicular screw
instrumentation: A 3-year consecutive series. Spine 2002;
27:2782-2787.
This study evaluated 32 patients in a 3-year consecutive
prospective experience of using pedicle screw fixation to treat
unstable thoracic spine injuries. Fracture healing was noted to
take place at an average of 4.8 months. Two-hundred fifty-two
pedicle screws were placed without any intraoperative complications. The Gardner segmental kyphotic deformity angle preoperative mean was 15.9 and the mean postoperative angle
was 10.6 (which was significant). All neurologically intact patients reported very good to good results with regard to pain,
activity, function, employment, and satisfaction.

Sacral Spine Trauma

Noojin F, Malkani A, Haikal L, Lundquist C, Voor M:
Cross-sectional geometry of the sacral ala for safe insertion of iliosacral lag screws: A computed tomography
model. J Orthop Trauma 2000;14:31-35.
This study evaluated 13 adult patients with intact pelves in
a trauma setting to determine the geometry of the sacral ala
for placement of iliosacral screws. Each patient had CT scans
of the pelvis with sagittal reconstructions. The narrowest portion of the sacral ala was determined, and height, width, and
slope of the geometric center were calculated. The mean width
of the sacral ala was 28.05 mm, the mean height was 27.76
mm, and the sacral slope ranged from 25 to 65. The authors
concluded that despite individual variability there is room for
two screws to be placed in the sacral ala safely.

Classic Bibliography
Allen BL Jr, Ferguson RL, Lehman TR, OBrien RP: A
mechanistic classification of closed, indirect fractures
and dislocations of the lower cervical spine. Spine 1982;
7:1-27.
Bohlmann HH: Acute fractures and dislocations of the
cervical spine: An analysis of three hundred hospitalized

526

Orthopaedic Knowledge Update 8

patients and review of the literature. J Bone Joint Surg
Am 1979;61:1119-1142.
Clark CR, White AA III: Fractures of the dens: A multicenter study. J Bone Joint Surg Am 1985;67:1340-1348.
Denis F: Instability as defined by the three-column
spine concept in acute spinal trauma. Clin Orthop 1984;
189:65-76.
Denis F, Davis S, Comfort T: Sacral fractures: An important problem. Retrospective analysis of 236 cases. Clin
Orthop 1988;227:67-81.
Hoffman JR, Mower WR, Wolfson AB, Todd KH,
Zucker MI: Validity of a set of clinical criteria to rule
out injury to the cervical spine in patients with blunt
trauma: National Emergency X-Radiography Utilization
Study Group. N Engl J Med 2000;343:94-99.
Levine AM, Edwards CC: The management of traumatic spondylolisthesis of the axis. J Bone Joint Surg
Am 1985;67:217-226.
Magerl F, Aebi M, Gertzbein SD, Harms J, Nazarian S:
A comprehensive classification of thoracic and lumbar
injuries. Eur Spine J 1994;3:184-201.
Mestdagh H, Letendart J, Sensey JJ, Duquennoy A:
Treatment of fractures of the posterior axial arch: Results of 41 cases. Rev Chir Orthop Reparatrice Appar
Mot 1984;70:21-28.
Schiff DCM, Parke WW: The arterial supply of the odontoid process. J Bone Joint Surg Am 1973;55:1450.
Spence KF, Decker S, Sell KW: Bursting atlantal fracture associated with rupture of the transverse ligament.
J Bone Joint Surg Am 1970;52:543-549.
Vaccaro AR, Rizzolo SJ, Allardyce TJ, et al: Placement
of pedicle screws in the thoracic spine: Part I. Morphometric analysis of thoracic vertebra. J Bone Joint Surg
Am 1995;8:1193-1199.
Vaccaro AR, Rizzolo SJ, Balderston RA, et al: Placement of pedicle screws in the thoracic spine: Part II. An
anatomic and radiographic assessment. J Bone Joint
Surg Am 1995;8:1200-1205.

American Academy of Orthopaedic Surgeons

Chapter

42

Cervical Disk Disease

Jonathan N. Grauer, MD
John M. Beiner, MD
Todd J. Albert, MD

Introduction
Cervical disk disease is a degenerative process that is often encountered in the aging patient. Although a patient
can occasionally identify a single inciting traumatic
event, more often no specific trauma is identified. Disk
degeneration is thought to be initiated by microtrauma,
causing a change in the proteoglycan and collagen content of the nucleus, loss of water content, and ultimately
altered biomechanics. Patients may have disk herniation
(soft disk) through an annular tear or loss of disk
height, causing bulging past the borders of the vertebral
end plates. Osteophyte (hard disk) formation can contribute to narrowing of the neural foramen, central canal stenosis, and eventually compression of the neural
elements. In most of these patients, these events are asymptomatic, even in the presence of advanced radiographic changes.
Symptoms of such disk degeneration can be insidious or acute in onset. Once foraminal or central stenosis
has developed, minor injuries have the potential to
cause local neural irritation and reaction. This can occur
at the level of the root (radiculopathy) or the spinal
cord (myelopathy). If there is compression of the nerve
roots and spinal cord, both radiculopathy and myelopathy may result (know as radiculomyelopathy).

Pathophysiology and Examination

The pathophysiology of degenerative cervical disease
involves the disk, zygapophyseal facet joints, uncovertebral joints of Luschka, posterior longitudinal ligament,
and the ligamentum flavum. As the disk loses water
content and the proteoglycan content changes, it becomes less able to support load. With progressive degeneration, there is decreased disk height, which is associated with loss of the normal cervical lordosis and the
transfer of load to the facet and uncovertebral joints.
Osteocartilaginous overgrowth may then occur in accordance with Wolffs law to stabilize the lax segment and
increased loads.
With loss of segment height, the ligamentum flavum
can become redundant, further contributing to the loss

American Academy of Orthopaedic Surgeons

of area available for the spinal cord. Radiculopathy develops as exiting nerve roots become irritated or compressed. This radiculopathy can be caused by either a
soft disk herniation, or more commonly it is secondary
to a chronic irritation from the uncovertebral or facet
joints in the patient with gradual onset of degenerative
changes.
Symptoms of radiculopathy include pain, paresthesias, or weakness with or without associated neck pain.
Objective signs of a radiculopathy include hyporeflexia
in the biceps (C5), brachioradialis (C6), or triceps (C7),
weakness or atrophy of the innervated muscle group, or
pain or paresthesias in a dermatomal fashion (Figures 1
and 2). Provocative maneuvers such as Spurlings test
can exacerbate such symptoms.
The first seven cervical nerve roots originate and
exit above their named vertebrae and the eighth cervical root originates and exits below the C7 vertebra. In
contrast to the lumbar spine, because the nerves in the
cervical spine exit the spinal canal relatively orthogonally at or below the level of the disk space, disk pathology generally affects the exiting nerve at that segment.
For example, C6-7 disk pathology will generally affect
the C7 nerve root (Figure 3). However, cervical nerve
roots exhibit a higher degree of overlap than seen in the
thoracolumbar spine, and therefore symptom patterns
may fail to localize to a specific nerve root in some patients.
The nerve roots most commonly affected by cervical
disk disease are C5, C6, and C7 because the associated
motion segments have the most flexion and extension in
the subaxial spine and are thus associated with the
greatest incidence of spondylosis. In addition, the watershed area of blood supply to the cervical spinal cord and
nerve roots makes those nerve roots most susceptible to
ischemic injury.
Cervical spondylotic myelopathy involves central
rather than foraminal stenosis. This myelopathy is most
typically caused by the combination of disk bulging and
uncovertebral hypertrophy with vertebral end plate osteophytes (the disk-osteophyte complex) in conjunction
with ligamentum flavum hypertrophy/redundancy and

527

Orthopaedic Knowledge Update 8

Cervical Disk Disease

Figure 1 Upper extremity motor testing for the cervical spine. Note there is dual innervation of some of these muscles.

Figure 2 Upper extremity sensory (A) and reflex (B) examination for the cervical
spine.

528

facet arthrosis. These changes lead to mechanical compression of the spinal cord that can be static or dynamic
(such as with neck extension). Anterior pathology may
also exert compression on the anterior spinal arteries,
which can in turn contribute to ischemia of the spinal
cord. Because such insults to the spinal cord occur very
slowly, a large degree of central canal stenosis (down to
a cross-sectional area of 17 mm2) can generally be tolerated relatively well.
Myelopathy is most commonly manifested as clumsiness (loss of fine motor skills), ataxia, and spasticity.
Specific symptoms may include dropping of objects,
changes in handwriting, and restlessness in the legs and
can progress to include loss of bowel or bladder control.
The physical examination for myelopathy should include testing for pathologic hyperreflexia below the
level of spinal cord compression, the presence of Hoffman and Babinski reflexes, and difficulty with tandem
gait. Clinicians must have a high level of suspicion for
this condition and be vigilant in their evaluation for associated signs and symptoms. Additionally, differential
diagnoses such as multiple sclerosis, anterior horn disease, and central nervous system tumors may be considered.
Because radiculopathy and myelopathy are both
caused by some of the same underlying pathology, it is
not uncommon to see patients with spondylotic radiculomyelopathy. There will often be a combination of the
signs and symptoms described for each independent pathology. Hyporeflexia is common in the upper extremities where nerve roots are compressed and hyperreflexia is usually present in the lower extremities below
the level of spinal cord compression.
Cervical spondylosis may also be associated with
neck pain. In addition to radicular pain, felt in a myotomal nerve root distribution, patients can feel pain or discomfort in a referred sclerotomal distribution corresponding to the embryologic origin of individual nerve
roots. Pain can be referred to the occiput, interscapular
region, or shoulders. The natural history of cervical disk
disease helps dictate treatment guidelines. In more than
75% of patients, symptoms of radiculopathy improve
with conservative treatment, including physical therapy,

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 42 Cervical Disk Disease

anti-inflammatory medication, and activity modification.

Soft disk herniations can resorb or shrink, and inflammation around a compressed nerve root can resolve. In
contrast, cervical myelopathy has a less predictable
course. Several studies have reported a gradual stepwise
progression of the disease, with variable intervals during
which patients symptoms remained quiescent. Most
studies, however, demonstrated a slowly worsening clinical picture.

Imaging Studies
Plain radiographs are appropriate in the evaluation of
neck pain, cervical radiculopathy, and cervical myelography. Such series include AP and lateral films to assess
overall spinal alignment and level of spondylosis, as well
as to rule out other structural lesions and deformity.
Flexion and extension radiographs can be used to assess angular or translational instability and demonstrate
whether a patient can achieve normal lordosis. Oblique
radiographs better visualize the neural foramen and facets and can facilitate visualization of the cervicothoracic
junction. A swimmers view allows for visualization of
the cervicothoracic junction, if not otherwise possible,
by limiting the obstructions imposed by the shoulders.
Other views such as the open mouth odontoid radiograph are generally not needed for degenerative conditions.
Plain radiographs are often all that is needed to initiate conservative treatment. However, if concerns are
raised, or if a patients symptoms persist beyond appropriate conservative treatments, advanced imaging may
be indicated. MRI is the axial imaging modality of
choice for the cervical spine, allowing visualization of
the soft tissues including disks, spinal cord, nerve roots,
and ligaments. Sagittal imaging provides a good overview of the levels of cord compression and central disk
pathology. However, sagittal imaging can give the false
impression of cervical kyphosis if the head of the patient if flexed during the imaging study. Parasagittal imaging can provide information about lateral disk herniations and foraminal narrowing. Axial imaging refines
information about cord or root compression and allows
for visualization of other surrounding structures such as
the vertebral artery or muscles.
CT can be used to define the bony anatomy of the
cervical spine. However, as the relationship to the neural elements must be inferred, the use of CT is best
combined with an intrathecal injection of a contrast medium for imaging of degenerative conditions (myelography). The contrast medium is injected into the thecal sac
via a C1-C2 puncture and allowed to diffuse caudally, or
via a lumbar puncture and allowed to diffuse proximally
with the patient in the Trendelenburg position. This allows precise visualization of the neural elements, with
filling voids present at sites of neural compression. Al-

American Academy of Orthopaedic Surgeons

Figure 3 The C7 nerve root exits the thecal sac above the C7 pedicle and is most
likely to be affected by pathology at the C6-7 interspace.

though much of this information can be obtained with

noninvasive MRI, myelography is particularly useful for
patients who cannot have an MRI secondary to implanted devices such as pacemakers. In other patients,
myelography can be useful if the results of MRI are inconclusive, such as with dynamic compression of the
neural elements that cannot be demonstrated with static
imaging. Flexion-extension radiographs following myelography can reveal this dynamic compression.
Diskography has been used to evaluate cervical disk
degeneration. The procedure is performed via an anterior approach in a trajectory similar to that used for the
standard anterior cervical approach. Once needles are
in place in the intervertebral disks, fluid can be injected
and the distribution and pain provocation can be studied. However, the validity and usefulness of this procedure is controversial. The risks of this procedure include
esophageal puncture and potential mediastinal or disk
space infection.

Electrodiagnostic Studies
In most patients, the history and clinical examination
can reliably identify the presence and level of nerve
root or spinal cord pathology. Combined with radiography and advanced imaging modalities, accuracy rates

529

Cervical Disk Disease

Orthopaedic Knowledge Update 8

Figure 4 Preoperative sagittal MRI (A) and postoperative lateral radiograph (B) of a patient with persistent C6 radicular symptoms who underwent anterior cervical diskectomy
and fusion with autograft and instrumentation.

are even higher. In some patients, however, the radiation of arm pain or symptoms such as paresthesias cannot be localized to a specific spinal level. In other patients, clinical examination may not correlate with
imaging studies or root symptoms may not be adequately differentiated from more distal nerve compression. In these patients, electromyogram and/or nerve
conduction velocity studies may be useful to differentiate acute or chronic radiculopathy from more distal
compressive neuropathies such as carpal tunnel syndrome or cubital tunnel compression, which may mimic
cervical root compression.

Nonsurgical Care
Management of neck pain and cervical radiculopathy
should start with conservative, nonsurgical measures, including physical therapy, traction, activity modification,
and certain medications.
With physical therapy, extremes of motion are generally not an important objective. Rather, physical therapy
should emphasize isometric exercises to build tone and
control for debilitated muscles and limit the shear forces

530

experienced by the spine. Traction may be useful for

radicular symptoms and should generally be done with
some degree of flexion to open the neural foramen. Although modalities often are useful at the time of application, they are not believed to provide lasting relief.
These therapeutic interventions may break the cycle
of irritation and splinting, allowing the reestablishment
of painless range of motion. Therapy should then focus
on neuromuscular control and active range-of-motion
and stabilization exercises. Finally, general aerobic conditioning should be added.
Medical management generally consists of nonsteroidal anti-inflammatory drugs. These drugs can help
control the inflammatory factors that are often responsible for acute symptoms. Oral steroids can be considered
for significant exacerbations of symptoms, but potential
adverse effects must be considered. The role of narcotics
in such degenerative processes is limited. Narcotics
should be reserved for acute injuries or exacerbations,
and they should only be used for a limited period. Narcotics should clearly be avoided for chronic pain.
Selective nerve root blocks and epidural steroid injections can be beneficial both diagnostically and thera-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 42 Cervical Disk Disease

peutically. Relatively high concentrations of steroids can

be achieved while limiting potential systemic adverse effects. Nevertheless, the risks associated with injections
into an already stenotic canal or neural foramen have
led to questions about their usefulness.
The role of conservative care in the management of
cervical spondylotic myelopathy is more controversial.
It appears that early myelopathy may not always follow
the pattern of progressive neurologic deterioration, and
therefore a close observational treatment may be indicated. Many surgeons believe, however, that once signs
or symptoms of myelopathy develop, surgical decompression is indicated to optimize neurologic recovery
and abate progression. Radiographic evidence of spinal
cord compression is not a surgical indication unto itself
because many patients will not have correlating symptoms or examinations.

Surgical Indications
Axial neck pain from degenerative disk disease is rarely
an indication for surgical intervention. In patients who
have been resistant to conservative measures, success
rates of surgical fusion for axial neck pain have generally only been in the 60% to 70% range. This may be related to an incomplete understanding of the associated
pain generators and potential painful foci at other cervical levels, posterior facets, or nonspinal sources.
In recent years, clinical trials of cervical disk replacements have begun in the United States. Goals of this
technology are to allow decompression or removal of a
degenerative disk while preserving motion. However,
the role and long-term outcome of such implants have
not yet been defined.
In contrast to axial neck pain, radiculopathy responds well to a variety of surgical treatments. When an
appropriate course of nonsurgical management has
failed, and radiculopathy persists, the surgeon can offer
greater than 90% success rates with surgical intervention. Clinically significant myelopathy is generally believed to be an indication for surgical intervention.

Surgical Treatment Options for Radiculopathy

and Myelopathy
The type of surgical procedure advocated for cervical
arthrosis is dictated by the location and extent of the
pathology. There are also situations in which similar pathology can be addressed in several ways, with roughly
similar results.
Anterior decompression can be performed, as initially described in the classic study by Smith and Robinson. When there is a central disk herniation, significant
uncovertebral spurring, or central canal stenosis, anterior decompression is the most direct means of decompressing the neural elements. Transverse incisions are
made for one, two, or arguably three levels, and oblique

American Academy of Orthopaedic Surgeons

Figure 5 When a patient has cord compression behind the vertebral bodies as well
as at the disk spaces, corpectomy should be considered as shown in this radiograph,
with a C4 and C5 corpectomy, allograft strut graft, and anterior instrumentation.
incisions may be necessary for longer exposures. The
level of incision can be guided by anatomic landmarks.
For example, the carotid tubercle, which can be palpated
percutaneously, is the lateral process of C6. The cricoid
cartilage is approximately at the C6 level, and the thyroid cartilage at the C4-C5 level. The anterior approach
is carried down medial to the sternocleidomastoid and
carotid sheath and lateral to the trachea and esophagus.
This provides exposure to the anterior aspect of the cervical spine. The longus coli are then elevated to allow
access to the entire disk space.
Once a diskectomy is carried back to the posterior
disk space, the posterior osteophytes can be taken down
with or without the associated posterior longitudinal ligament. This scenario may be associated with more immediate and complete relief of symptoms, more complete foraminal decompression, and better identification
of extruded disk fragments. Some authors have found
that taking down the posterior osteophytes and posterior longitudinal ligament is not always necessary and
that disk height restoration and fusion (with its associated elimination of motion) may be adequate and associated with lesser bleeding and surgical times.

531

Cervical Disk Disease

Orthopaedic Knowledge Update 8

Figure 6 Preoperative sagittal MRI (A) and postoperative lateral (B) and AP (C) radiographs of a patient with multilevel cervical stenosis and myelopathy who was treated with
posterior decompression and fusion.

Once the diskectomy and decompression have been

achieved, interbody fusion is performed by decorticating
the end plates and inserting a bone graft. This has traditionally been iliac crest autograft. However, according
to a recent study, allograft has been shown to afford almost identical fusion rates for single-level procedures
without the donor site morbidity associated with autograft. This phenomenon does not appear to be as true
for multilevel procedures or in the presence of factors
that inhibit fusion such as smoking. Diskectomy without
fusion has largely been abandoned because of its association with kyphotic collapse, neck pain, and potential
for recurrent foraminal narrowing.
Plate fixation can be added to potentially increase
the rate of union, limit kyphotic collapse, and minimize
the need for external cervical orthoses (Figure 4). Many
types of plates exist. All current plate systems lock the
screw to the plate to prevent screw backout. Some systems allow for variable angle screw insertions to facilitate implantation. Others systems allow for settling of
the screws within the plate (dynamic plates) to allow the
graft to be better loaded. Although plates are now routinely used by many surgeons, the justification for this is
uncertain given the high fusion rate in single-level noninstrumented fusions. The potential benefits of plating
must be contrasted with the expense and the potential,
but limited, risks of implantation.
With larger procedures such as multilevel diskectomy, corpectomy, or partial corpectomy, recent studies
have indicated that strut grafting is a good surgical alternative. With this technique, central compression behind

532

the vertebral bodies can be addressed. It also offers

fewer surfaces to fuse compared with multiple interbody
grafts, potentially increasing fusion rates (Figure 5).
Bone graft can be autograft, structural allograft (potentially filled with local corpectomy bone), or synthetic
cages. As constructs grow in length to corpectomies of
two or more vertebral bodies, consideration should be
given to supplementation with posterior stabilization.
However, it is generally believed that lordosis can be restored and maintained better with multiple interbody
grafts.
Alternatively, if there is purely foraminal stenosis
and radiculopathy from a soft disk or hypertrophied
facet or uncovertebral joint, posterior laminoforaminotomy may be considered. This procedure allows decompression without fusion, and destabilization is avoided
as long as the surgeon avoids total facetectomy unilaterally or partial facetectomy bilaterally. Because the disk,
which is often thought to be the pain generator for axial
pain, is not directly addressed, posterior laminaforaminotomy is not a good treatment option if axial pain is a
large component of a patients symptoms. This approach
avoids the dangers of anterior dissection, such as recurrent laryngeal nerve palsy, and preserves the motion
segment. However, central pathology cannot be addressed because the spinal cord cannot be retracted to
allow access to the region. Furthermore, disadvantages
lie in the muscle dissection and facet capsule disruption
inherent in the procedure.
Myelopathy can be treated from posterior as well as
anterior approaches. The choice of approach depends on

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 42 Cervical Disk Disease

Figure 7 A and B, Imaging studies of a

patient with multilevel cervical stenosis
and ossification of the posterior longitudinal ligament who presented with myelopathy. Treatment was laminaplasty at C3-C6
with plate fixation (C and D). Note the increase in canal diameter between preoperative (B) and postoperative (C) CT reconstructions.

factors including the site of compression (anterior spurs/

ossification of the posterior longitudinal ligament versus
posterior ligamentum flavum hypertrophy/redundancy),
overall spinal alignment, and number of levels involved.
The anterior approach allows removal of anterior cord
compression, but this must often involve removal of the
vertebral body if compression is in this region.
Posterior decompression in the form of laminectomy
has previously been the standard procedure, but was associated with postlaminectomy kyphosis in a large percentage of patients. Posterior cervical fusion is thus now
generally performed with a laminectomy to restore and
maintain cervical lordosis and allow the spinal cord to
float away from any anterior compressive pathology (Figure 6). Posterior fixation can be achieved with lateral mass

American Academy of Orthopaedic Surgeons

screws at C3-C6 and pedicle screws at C2, C7, and below.

The disadvantages of this technique include the morbidity of muscle dissection from a posterior cervical approach
and the associated loss of motion.
As an alternative to fusion, laminaplasty was developed in Japan for the treatment of cervical myelopathy associated with ossification of the posterior longitudinal ligament. Laminaplasty involves hinging open the lamina at
multiple levels, affecting decompression. Such openings
can be held with sutures, instrumentation, and/or structural bone grafts. The goal of the procedure is to decompress without adding the complications inherent to fusion.
Outcome studies support these goals, but potential adverse effects include increased axial neck pain and loss of
motion (Figure 7).

533

Cervical Disk Disease

Fusion procedures may be associated with accelerated degenerative changes adjacent to the fused segments. This phenomenon may be alleviated with laminaplasty as opposed to decompression and fusion. In
addition, disk arthroplasty may help limit the incidence
of adjacent level disease in the cervical spine, but this
has not yet been conclusively demonstrated.

Annotated Bibliography
Bryan VE: Cervical motion segment replacement. Eur
Spine J 2002;11:S92-S97.
Cervical disk arthroplasty is being developed as an alternative to fusion procedures. The possibility of limiting adjacent
level degeneration is one of the potential benefits.

Edwards CC, Heller JG, Murakami H: Corpectomy versus laminoplasty for multilevel cervical myelopathy: An
independent matched-cohort analysis. Spine 2002;27:
1168-1175.
In this study, corpectomy and laminaplasty were found to
arrest myelopathic progression and offer the potential for neurologic recovery. However, this study suggested that the laminaplasty group had less pain at follow-up than the multilevel
corpectomy group.

Fouyas IP, Statham PFX, Sandercock PA: Cochrane review of the role of surgery in cervical spondylotic radiculomyelopathy. Spine 2002;27:736-747.
In this critical review of the literature, it was difficult to
draw reliable conclusions about the risk/benefit balance for
cervical spine surgery for spondylotic radiculopathy or myelopathy.

Hilibrand AS, Fye MA, Emery SE, Palumbo MA, Bohlman HH: Increased rate of arthrodesis with strut grafting after multilevel anterior cervical decompression.
Spine 2002;27:146-151.
A higher rate of fusion was seen after corpectomy and
strut grafting than after multilevel diskectomy and interbody
grafting. This therefore suggests that strut grafting should be
considered in patients requiring surgical attention at multiple

534

Orthopaedic Knowledge Update 8

disk spaces even if there is not compression behind the vertebral body.

Malloy KM, Hilibrand AS: Autograft versus allograft in

degenerative cervical disease. Clin Orthop 2002;394:2738.
Although allograft is the gold standard bone grafting material, high fusion rates have been observed when allograft is
used, especially for single-level cases in nonsmokers. This offers the potential of eliminating the morbidity associated with
the harvest of autograft.

Patel CK, Fischgrund J: Complications of anterior cervical spine surgery. Instr Course Lect 2003;52:465-469.
This article presents a review of the potential complications associated with anterior cervical spine surgery for which
the incidence is relatively low.

Sampath P, Bendebba M, Davis JD, Ducker TB: Outcome of patients treated for cervical myelopathy: A prospective, multicenter study with independent clinical review. Spine 2000;25:670-676.
Authors of this multicenter study of patients with cervical
myelopathy concluded that patients treated with surgery appear to do better than those treated nonsurgically.

Classic Bibliography
Bohlman HH, Emery SE, Goodfellow DB, Jones PK:
Robinson anterior cervical discectomy and arthrodesis
for cervical radiculopathy. J Bone Joint Surg Am 1993;
75:1298-1307.
Herkowitz HN: A comparison of anterior cervical fusion, cervical laminectomy, and cervical laminoplasty for
the surgical management of multiple level spondylotic
radiculoparhy. Spine 1988;13:774-780.
Smith GW, Robinson RA: The treatment of certain
cervical-spine disorders by anterior removal of the intervertebral disc and interbody fusion. J Bone Joint Surg
Am 1958;40:607-624.

American Academy of Orthopaedic Surgeons

Chapter

43

Thoracic Disk Disease

Chetan K. Patel, MD

Introduction
Symptomatic thoracic disk disease is relatively uncommon compared with disk disease in the cervical and
lumbar regions, with an estimated incidence of between
1 in 10,000 and 1 in 1 million persons. The occurrence of
symptomatic thoracic disk herniation is greatest between the fourth and sixth decade of life, with a peak incidence in the fifth decade. Diagnosis can be difficult
because of a variety of clinical presentations. Widespread availability and use of MRI has aided in the diagnosis of symptomatic thoracic disk herniations (Figure 1). However, one MRI study documented thoracic
degenerative changes in 73% of asymptomatic individuals, with 37% showing a disk herniation.

Anatomy and Biomechanics

The thoracic spine, a relatively rigid structure, is stabilized by the rib cage. Thoracic facets are generally oriented vertically, which allows lateral bending and rotation while limiting flexion and extension. These
anatomic and biomechanical properties would be expected to decrease the potential of injury to the thoracic
intervertebral disk. This is confirmed clinically by the
observation that less than 1% of symptomatic disk herniations occur in the thoracic spine.
When a thoracic disk herniation occurs, biomechanical studies have shown the mechanism to be a combination of torsion and bending load. Several anatomic factors are suspected in the pathogenesis of neurologic
compromise in thoracic disk herniations. Kyphosis of the
thoracic spine places the spinal cord directly on the posterior longitudinal ligament and the posterior aspect of
the vertebrae and disks. Additionally, the vascular supply to the thoracic spine is much more tenuous than that
of the cervical and lumbar spine, especially from T4
through T9. Although the size of the spinal cord is
smaller in the thoracic spine than in the cervical spine,
the cord to canal ratio is actually higher because of the
smaller size of the thoracic spinal canal. These factors
together explain how the thoracic spinal cord is vulnerable to injury.

American Academy of Orthopaedic Surgeons

Figure 1 Axial MRI scan demonstrating a thoracic herniated nucleus pulposus.

Etiology
Between 33% and 50% of patients report a history of
trauma or significant physical exertion before the onset
of symptoms. However, the role of trauma as the cause
of thoracic disk herniations is controversial. Most authors favor degenerative processes as the major cause of
disk herniations. This theory is supported by the common findings of disk degeneration at the level of herniation and the higher incidence of herniations in the lower
thoracic spine where greater degenerative changes have
been reported. End plate changes consistent with
Scheuermanns disease are seen more often in symptomatic patients than in those who are asymptomatic,
suggesting an association between Scheuermanns disease and symptomatic herniated thoracic disks.

Clinical Presentation
The clinical presentation of patients with thoracic disk
disease is variable, and the differential diagnosis for tho-

535

Thoracic Disk Disease

racic back pain and myelopathy is extensive. The differential diagnosis includes thoracic herniated nucleus pulposus, thoracic degenerative disk disease, cervical and
lumbar herniated nucleus pulposus, intercostal neuralgia, multiple sclerosis, amyotrophic lateral sclerosis,
ankylosing spondylitis, fractures, metabolic bone disorders, tumors, infection, herpes zoster, and diseases of the
thoracic and lumbar viscera including aneurysm. A thorough history and physical examination are the best tools
for the expeditious diagnosis and treatment of patients
with thoracic disk disease.
Symptoms of thoracic disk herniations occur in one
of three distinct patternsaxial, radicular, or myelopathic. Axial pain usually presents in the mid- to lower
thoracic regions, worsens with activity, and improves
with rest. Radicular symptoms typically arise from a lateral disk herniation impinging on a nerve root and producing symptoms that follow the course of that nerve
root. Most commonly, pain and paresthesias are reported starting in the back and radiating anteriorly in a
band-like pattern in the distribution of the more caudal
nerve root. Thoracic dermatomes can be roughly identified by the following landmarks: T4 at the nipple line,
T7 at the xiphoid process, T10 at the umbilicus, and T12
at the inguinal crease. If any skin lesions are noted in a
dermatomal pattern, herpes zoster may be the underlying cause. Although motor testing is difficult in this region, the patient can do a partial sit-up to help identify
an asymmetric contraction of the segmentally innervated rectus abdominis.
Myelopathy may also be the presentation of a thoracic disk herniation when there is significant spinal
cord compression. The presentation can vary from subtle sensory changes to obvious paraparesis and bowel
and bladder dysfunction. In addition to the motor and
sensory examination, attention should focus on upper
motor neuron signs such as lower extremity hyperreflexia, clonus, Babinski sign, wide-based gait, and superficial abdominal reflexes. The Romberg sign can help
detect subtle changes in proprioception. The results of
an upper extremity neurologic examination should be
normal in addition to the patient exhibiting a negative
Hoffmann sign.

Diagnostic Imaging
Plain radiographs should be obtained first. In addition
to being used to assess overall alignment, the plain radiographs should be scrutinized for degenerative
changes, calcification in the disk space or in the canal,
fractures, and tumors. Intradiskal calcification is noted
in 45% to 71% of patients with symptomatic thoracic
herniated disks compared with 10% of asymptomatic individuals.
MRI is the diagnostic modality of choice in further
evaluation of these patients because it is noninvasive

536

Orthopaedic Knowledge Update 8

and highly sensitive. However, as mentioned previously,
the findings should be interpreted in the context of clinical findings because of the high percentage of abnormal
MRI findings in asymptomatic individuals. A CT myelogram is also highly sensitive, but this modality is invasive and should be reserved for patients who cannot undergo MRI. Diskography is controversial but can be
considered in patients with axial pain, multilevel disease, and no neurologic findings. At this time, no outcome studies exist that demonstrate a correlation between diskography results and surgical outcomes.

Nonsurgical Treatment
To decide on the best treatment course for thoracic disk
disease, its natural history must first be understood.
Asymptomatic patients with abnormal MRI findings
were noted to remain asymptomatic at a follow-up of
more than 26 months. Children with painful calcified
thoracic disks improve spontaneously with resorption of
calcification; however, these children should be closely
followed because a few instances of neurologic deficit
requiring surgical intervention have been reported in
this population. In adults, the natural history of acute
thoracic disk herniations without neurologic deficit is
benign. Most patients can be treated with activity modification, anti-inflammatory medications, exercise, and
with bracing in rare occasions. Most patients are expected to return to their normal activities, including vigorous sports and work. In patients with radicular symptoms, corticosteroid injections of intercostal nerves
should be considered when other modalities do not provide adequate pain relief.

Surgical Treatment
Indications for surgery include progressive neurologic
deficit, myelopathy, and pain refractory to conservative
treatment. The herniated thoracic disk can be accessed
using the posterior, posterolateral, lateral, or anterior
approach (Figure 2).
Laminectomy and the transpedicular approach are
posterior approaches. The surgical approach of choice
was once straight posterior laminectomy and disk excision. However, because this approach was associated
with significant risk of neurologic deterioration, it has
been largely abandoned. The transpedicular approach
was developed to allow less retraction of the spinal cord
by removing the pedicle and facet joint. Although this
approach is well suited for the treatment of a lateral
disk herniation, a central or paramedial disk herniation
is difficult to excise using this approach because of poor
visualization. Removal of the pedicle and facet joint
complex may lead to instability and postoperative pain.
The posterolateral approach is also known as a costotransversectomy. In this approach, the posteromedial portion of the ipsilateral rib along with the transverse process,

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 43 Thoracic Disk Disease

Figure 2 Illustration of four surgical approaches to the thoracic disk pathology: transpedicular (A), extracavitary (B), costotransversectomy (C), and transthoracic (D). (Reproduced from Wood KB, Mehbad A: Thoracic disk herniation, in Koval KJ (ed): Orthopaedic Knowledge Update 7. Rosemont, IL, American Academy of Orthopaedic Surgeons, pp
621-625.)

a portion of the pedicle, and the posterosuperior vertebral

body are removed. This allows gentle removal of the disk
fragment without manipulating the cord. This approach
can be used for multiple levels and allows resection of
paramedian as well as lateral disks. Visualization of central disks is poor with this approach, and it does involve a
generous amount of bone resection and significant disruption of the paraspinal musculature.
The lateral extracavitary approach, another extrapleural approach, is a modification of the lateral thoracotomy approach that involves resection of the posteromedial rib in addition to the transverse process, pedicle,
facet, and posterosuperior portion of the vertebral bodies. Midline access is less than optimal, and a large
amount of bone resection and disruption of paraspinal
musculature can lead to postoperative pain.
The anterior approaches include thoracotomy and
video-assisted thoracoscopic surgery (VATS). Anterior
approach via a thoracotomy was first reported in 1969
and since then has gained wide popularity. This approach allows excellent visualization of midline and lateral pathology from T5 through T12. Multilevel access is
also readily available. The base of the rib articulating
with the body and a portion of the pedicle are removed,
followed by a partial diskectomy. Using a burr, a trough
is created into which disk material can be pulled forward to decompress the canal and remove the herniated
fragment. Fusion can easily be added to the diskectomy

American Academy of Orthopaedic Surgeons

with the available exposure. Many authors have reported good visualization and excellent results with this
technique. The drawbacks of this technique include the
possibility of pneumothorax, pulmonary contusion,
atelectasis, pneumonia, and the need for a postoperative
chest tube.
VATS was first reported in 1993 as a minimally invasive procedure that can minimize the morbidity of an
open thoracotomy and allow excellent visualization of
the anterior approach. One of the main disadvantages
of this technique is the steep learning curve to attain the
high level of technical skills required for the procedure.
Calcified protrusions can be adherent to or penetrate
through the dura and can be difficult to treat via VATS.
Two-year follow-up data are now available and suggest
that VATS is effective in a select group of patients.
The role of fusion in thoracic disk surgery is controversial. Relative indications for fusion include multilevel
diskectomy, Scheuermanns disease, and bony resection
that removes a large portion of the vertebral body or
the pedicle facet complex. The resected rib often provides sufficient autologous bone graft material without
the need for added morbidity from harvesting a graft. A
small fibular allograft augmented with local autograft
can also be used. Instrumentation is generally not used
because the rib cage provides a protective splinting effect. If deformity correction is desired in multilevel
cases, instrumentation should be considered.

537

Thoracic Disk Disease

Most patients with thoracic disk herniation and axial
pain alone can be treated successfully with nonsurgical
treatment. In patients who do not respond well to nonsurgical treatment, thoracic diskectomy and fusion can
be considered; however, the role of surgical intervention
is controversial.

Annotated Bibliography
Anand N, Regan JJ: Video-assisted thoracoscopic surgery for thoracic disc disease: Classification and outcome study of 100 consecutive cases with a 2-year minimum follow-up period. Spine 2002;27:871-879.
The authors present their experience of 100 patients who
underwent VATS for the treatment of thoracic disk herniations with at least a 2-year follow up. This prospective, nonrandomized study supports the conclusions that VATS is effective
in a select group of patients, it is a reasonably safe procedure,
and satisfactory outcomes are achieved for most patients.

Oskouian RJ, Johnsin JP, Regan JJ: Thoracoscopic microdiscectomy. Neurosurgery 2002;50:103-109.
The authors of this article present a detailed description of
a technique for thoracic diskectomy via thoracoscopy.

Classic Bibliography

Orthopaedic Knowledge Update 8

Dickman CA, Rosenthal D, Regan JJ: Reoperation for
herniated thoracic discs. J Neurosurg 1999;91(suppl 2):
157-162.
Fessler RG, Sturgill M: Review: complications of surgery
for thoracic disc disease. Surg Neurol 1998;49:609-618.
Fidler MW, Goedhart ZD: Excision of prolapse of thoracic intervertebral disc: A transthoracic technique.
J Bone Joint Surg Br 1984;66:518-522.
Otani K, Yoshida M, Fujii E, Nakai S, Shibasaki K: Thoracic disc herniation: Surgical treatment in 23 patients.
Spine 1988;13:1262-1267.
Rogers MA, Crockard HA: Surgical treatment of the
symptomatic herniated thoracic disk. Clin Orthop 1994;
300:70-78.
Schellas KP, Pollei SR, Dorwart RH: Thoracic discography: A safe and reliable technique. Spine 1994;19:21032109.
Simpson JM, Silveri CP, Simeone FA, Balderston RA,
An HS: Thoracic disc herniation: Re-evaluation of the
posterior approach using a modified costotransversectomy. Spine 1993;18:1872-1877.

Arce CA, Dohrmann GJ: Herniated thoracic disks.

Neurol Clin 1985;3:383-392.

Vanichkachorn JS, Vaccaro AR: Thoracic disk disease:

Diagnosis and treatment. J Am Acad Orthop Surg 2000;
8:159-169.

Awwad EE, Martin DS, Smith KR Jr, Baker BK: Asymptomatic versus symptomatic herniated thoracic
discs: Their frequency and characteristics as detected by
computed tomography after myelography. Neurosurgery
1991;28:180-186.

Wood KB, Blair JM, Aepple DM, et al: The natural history of asymptomatic thoracic disc herniations. Spine
1997;22:525-530.

Bohlmann HH, Zdeblick TA: Anterior excision of herniated thoracic discs. J Bone Joint Surg Am 1988;70:10381047.
Brown CW, Deffer PA Jr, Akmakjian J, Donalson DH,
Brugman JL: The natural history of thoracic disc herniation. Spine 1992;17(suppl 6):S97-S102.
Currier BL, Eismont FJ, Green BA: Transthoracic disc
excision and fusion for herniated thoracic discs. Spine
1994;19:323-328.

538

Wood KB, Garvey TA, Gundry C, Heithoff KB: Magnetic resonance imaging of the thoracic spine: Evaluation of asymptomatic individuals. J Bone Joint Surg Am
1995;77:1631-1638.
Wood KB, Schellhas KP, Garvey TA, Aeppli D: Thoracic
discography in healthy individuals: A controlled prospective study of magnetic resonance imaging and discography in asymptomatic and symptomatic individuals.
Spine 1999;24:1548-1555.

American Academy of Orthopaedic Surgeons

Chapter

44

Lumbar Degenerative Disorders

Raj D. Rao, MD
Kenny S. David, MD

Introduction
Low back pain accounts for more than 15 million patient visits to the physicians office per year in the
United States, second only to the number of patient visits for respiratory infections. Complaints of back pain
begin around age 35 years and increase in prevalence up
to age 50 years in men and age 60 years in women. The
overall point prevalence of back pain in the United
States is estimated to be 18%. The annual cost for managing back pain is approximately $50 billion, the bulk of
which is spent on an estimated 1% of the patients. The
three most common lumbar degenerative disorders are
lumbar spinal stenosis, lumbar disk herniation, and discogenic low back pain.

Lumbar Spinal Stenosis

Lumbar spinal stenosis is a reduction in the dimensions
of the central or lateral lumbar spinal canal that occurs
most frequently as a result of chronic degenerative
changes at the lumbar motion segment (Table 1). Patients with lumbar spinal stenosis have back pain associated with neurogenic claudication and/or radicular pain.

Pathoanatomy
Absolute stenosis is defined as a decrease in the midsagittal lumbar canal diameter of less than 10 mm, whereas
10 to 13 mm represents relative stenosis. The normal
cross-sectional area of the lumbar canal is 150 to 200
mm2, and a decrease to less than 100 mm2 is a more reliable indicator of the combined effects of central and
lateral lumbar stenosis. Central stenosis results from
congenitally short pedicles, diffuse posterior protrusion
of the degenerative disk, and infolding of the ligamentum flavum.
The lateral portion of the lumbar canal is divided
into three zones: the lateral recess, foraminal zone, and
extraforaminal zone (Figure 1). The most common pathology in the lateral recess is bony overgrowth of the
superior articular process caused by degenerative facet
joint arthrosis. The foraminal zone lies distal to the pedicle and ventral to the pars interarticularis and contains

American Academy of Orthopaedic Surgeons

portions of the ventral nerve root, dorsal root ganglion,

and nerve root. Stenosis in this region is commonly the
result of cartilaginous overgrowth or osteophytes arising
from the anterior-inferior aspect of a pars defect, foraminal disk or end plate protrusion, or superior-inferior
compression between two pedicles as a result of loss of
disk height. Foraminal height normally ranges from 20
to 23 mm, and anterior-posterior depth ranges from 8 to
10 mm in the upper foramen. A foraminal height of less
than 15 mm and a posterior disk height of less than 4
mm are associated with nerve root compression in 80%
of patients. The lumbar nerve roots lie in the upper foramen and occupy approximately 30% of the available foraminal area. Evidence of obliteration of perineural fat
on radiographic images is an early indicator of foraminal stenosis. The extraforaminal zone is lateral to the intervertebral foramen and contains the exiting root.
Alteration of lumbar canal dimensions with trunk
posture results in dynamic stenosis. Extension of the
lumbar spine leads to reduced interlaminar space and
buckling of the ligamentum flavum, which can result in
spinal stenosis. Foraminal height and width decrease by
14% to 18% in extension. Foraminal area decreases by
20% during extension. CT studies show that the crosssectional area of the foramen increases by 12% in flexion, and that nerve root compression is least in flexion
and highest in extension.

Pathophysiology
The development of symptoms in a subset of these
patients can be explained by the pathophysiologic
changes that occur concurrently with the morphologic
changes of stenosis. In animals, 50% constriction of the
cauda equina results in major changes in cortical evoked
potentials and mild motor weakness. These findings generally resolve by 2 months despite persistent compression. With constriction to 75%, motor and sensory deficits are more profound and show only slight recovery at
2 months. Claudication and neurologic symptoms may
initially result from venous distension in the nerve roots
and dorsal root ganglion. Obstruction of microcircula-

539

Orthopaedic Knowledge Update 8

Lumbar Degenerative Disorders

Table 1 | Etiologic Classification of Lumbar Spinal Stenosis

Congenital/Developmental
Chondrodystrophy-achondroplasia
Congenital small spinal canal with short vertebral pedicles
Congenital cysts from the dura/arachnoid narrowing an otherwise normal
canal
Osteoporosis
Acquired
Degenerative
Central canal/lateral recess/foraminal zone
Degenerative spondylolisthesis
Degenerative scoliosis
Inflammatory arthritis
Ankylosing spondylitis
Rheumatoid arthritis
Pseudogout
Spondylolytic
Iatrogenic
Postdiskectomy/laminectomy/fusion
Postchemonucleolysis
Vertebral fractures
Traumatic
Pathologic
Spinal infections with abscess bone collapse
Miscellaneous
Pagets disease
Fluorosis
Acromegaly
Diffuse idiopathic skeletal hyperostosis
Pseudogout
Oxalosis
Combined
Degenerative changes superimposed on a developmentally narrow canal

tion at the site of constriction, nerve root fibrosis, and

wallerian degeneration of the motor root and posterior
spinal tracts contribute to subsequent neurologic manifestations. The same general rules appear to apply to the
human lumbar canal, with significant symptoms appearing when the stenosis is greater than 75%. Claudication
symptoms from lumbar spinal stenosis occur in 90% of
patients who have a cross-sectional spinal canal area
less than 100 mm2.
In animal studies, the rate of onset of compression
plays a role in the pathophysiology of stenosis. Rapid
onset (0.05 to 0.10 s) compression causes more damage
than insidious onset (20 s) pressure. The compressive lesion results in greater mechanical deformation at the
edge of nerve root, a phenomenon referred to as the
edge effect. Histologic examination of edge segments
has revealed significant microvascular and neural tissue

540

Figure 1 Illustration of lateral recess stenosis in which the superior articular process
impinges on the traversing lumbar nerve root. The coronal section shows the relative
positions of the central canal, lateral recess, foraminal zone, and extraforaminal zone.

damage. Compression at multiple sites may explain development of symptoms in some patients. Animal studies have shown that single-level compression of 10 mm
Hg had marginal effect on nerve function, whereas twolevel compression caused significant reduction in blood
flow of the cauda equina.
At the cellular level, bone morphogenetic proteins
are related to chondrogenesis within the aging disk. The
migration of these factors from the region of the vertebral end plates to fibrous cells within the anulus fibrosus
may contribute to osteophyte formation and the bone
overgrowth seen in the degenerative spine.

Clinical Features
Classically, patients have back, buttock, or posterior leg
pain that gets worse with standing and walking. Leg
symptoms may also be described as cramps, burning
pain, or weakness. Flexion of the trunk alleviates symptoms, and extension aggravates the symptoms. Leaning
on a cart helps, primarily because of the flexed posture
of the trunk in this position. Sleeping is comfortable in
the fetal position. The limitations in activity and stooped
forward posture are commonly attributed to age, and
many patients learn to work within their limitations. Autonomic sphincter dysfunction manifests as recurrent
urinary tract infection associated with an atonic bladder,
incontinence, or retention, and occurs in up to 10% of
patients with advanced degrees of stenosis.
Claudication manifests as diffuse buttock and/or leg
pain, nonspecific paresthesias, or radicular symptoms

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

that commence with walking for varying distances. Unlike vascular disease, the claudication distance with
symptomatic spinal stenosis may vary considerably and
is relieved by sitting and not by standing. Activities performed with the trunk flexed, such as cycling or walking
uphill, are performed more easily. The exertion-induced
symptoms of spinal stenosis can be accurately reproduced using a treadmill, providing a method for preoperative and postoperative quantification of symptoms.
The manifestations of central and lateral canal
stenosis are not easily distinguished. Unilateral radicular
pain from foraminal stenosis is worsened by extension
to the painful side (Kemp sign). Paresthetic and dysesthetic symptoms may be an indicator of dorsal root ganglion entrapment in the foramen. Unlike disk herniations, radicular pain from stenosis is typically not
aggravated by Valsalva maneuvers or accompanied by a
positive straight leg raising sign.

Management
Nonsurgical therapy with anti-inflammatory agents, analgesics, activity modification, exercises, and soft braces
may help patients with exacerbations of pain but seldom
achieve sustained improvement. Epidural steroid injections or selective nerve root injections result in substantial relief of radicular pain and may obviate the need for
surgery over the short term in a subset of patients. Ideally, the injections are performed at the level of the
symptomatic nerve root through a fluoroscopically directed transforaminal technique. One to three injections
are typically performed, with 1-to 2-week intervals between injections.
Patients with persistent lower extremity symptoms
from lumbar spinal stenosis are generally offered surgical decompression. The physician should make clear
that the aim of surgery is to relieve current disability
rather than prevent future complications. There is no
conclusive evidence that disability from spinal stenosis
worsens over time. Less frequent indications for surgery
include progressive neurologic deficit or cauda equina
syndrome.
The optimal surgical approach combines maximal
thecal sac and nerve root decompression and preserves
stability. The facet joints, capsule, intervertebral disk,
and interspinous ligaments are important lumbar spine
stabilizers. In typical degenerative lumbar spinal stenosis, maximal compression of the thecal sac occurs at the
level of the disk. Decompression of a single level is
achieved by resection of approximately 50% of the
cephalad and caudad laminae and the intervening ligamentum flavum. After creation of a central trough, the
decompression is extended laterally to the medial wall
of the pedicle bilaterally to ensure that the traversing
nerve root is free of pressure. Greater than 50% excision of the bilateral facets or unilateral complete face-

American Academy of Orthopaedic Surgeons

Chapter 44 Lumbar Degenerative Disorders

tectomy can result in iatrogenic instability and may require fusion.
There are no guidelines on whether adjacent levels
with lesser degrees of stenosis should be included in the
decompression. The benefits of a greater decompression
of the neural elements need to be weighed against the
potential morbidity of a longer operation. In general, a
level with mild stenosis may be excluded if radicular
symptoms corresponding to that level are absent. Decompressing too few levels may be a source of persistent symptoms postoperatively. Somatosensory-evoked
potential techniques may serve as a guide to the extent
of decompression in the future.
The anulus fibrosus is left intact when possible, resecting only free fragments or extruded segments of the
disk that result in neural impingement. Diffuse bulging
of the disk is unlikely to result in continued neural compression in patients who have undergone a wide laminectomy with bilateral partial facetectomy.
Based on the existing literature, a fusion should be
considered following decompressive laminectomy in (1)
patients who have stenosis with degenerative spondylolisthesis; (2) when radiographic criteria for instability
exist preoperatively; (3) when more than a combined
50% of the bilateral facet joints are resected as part of
the decompression; and (4) when decompression is performed in a patient with a flexible or progressive degenerative scoliosis.
The precise role of instrumentation in lumbar spinal
stenosis is currently unclear. The aims of pedicle screw
instrumentation in spinal stenosis surgery include deformity correction, immediate stabilization of an unstable
spine, improvement in fusion rates, and faster mobilization of patients. The advantages of instrumentation need
to be weighed against the potential morbidity of extending the surgical procedure. A randomized prospective
study showed that the addition of instrumentation to
lumbar decompression and posterolateral arthrodesis
for degenerative spondylolisthesis with stenosis resulted
in improved fusion rates, but it did not improve the
overall clinical outcome of patients.

Alternative Surgical Techniques

Several limited approaches have been proposed as alternatives to the traditional decompressive laminectomy
and partial facetectomy. The rationale for these approaches involves decreased patient morbidity and
faster rehabilitation with a more limited operation, limiting the surgery to the pathologic area (the bony and
soft-tissue pathology in most patients with degenerative
stenosis is at the level of the interlaminar window and
can be adequately addressed through an interlaminar
fenestration or laminotomy), decreased postoperative
radiographic listhesis, reduced scarring posterior to the
thecal sac and root, awareness that paraspinal muscle

541

Orthopaedic Knowledge Update 8

Lumbar Degenerative Disorders

atrophy with wide exposure may be a source of postoperative back pain, and limited surgical destabilization,
which potentially obviates the need for fusion. The benefits of these approaches need to be weighed against
their limitations, which include a higher incidence of
nerve root injury and dural tears, bone regrowth potentially compromising the result, the need for technically
demanding procedures and longer surgical times, the
fact that most of these procedures do not address central stenosis, and the lack of long-term clinical efficacy
studies.
A unilateral hemilaminectomy or laminotomy may
be indicated in a patient with one-sided symptoms from
lateral stenosis and no significant central stenosis. Simultaneous diskectomy should be performed if there is residual bulging of the disk, especially when the patient
has good disk height. Bilateral or multilevel laminotomy
may be considered in patients with degenerative stenosis at multiple levels. Myelography in these patients usually shows an hourglass compression of the thecal sac at
several levels, with spacious intervening areas. The procedure differs from the traditional multilevel laminectomy by leaving the spinous processes and interspinous
ligaments intact. In unilateral laminotomy with bilateral
decompression, an ipsilateral laminotomy and decompression is followed by decompression of the contralateral side by angling the microscope or endoscope across
the top of the dural sac. Resection of the contralateral
ligamentum flavum allows good exposure of the contralateral side.

preoperatively may be a negative prognostic indicator.

The results in general deteriorate over a 5- to 10-year
period, with increasing back pain, claudication, and rates
of revision surgery. Mortality in these elderly patients is
0.6% to 1.0% following surgery.
The rates of lumbar fusion vary considerably across
geographic regions and specialties, suggesting uncertainty among surgeons on the indications for fusion following decompressive laminectomy. In a multicenter
study of patient selection and outcomes following decompressive laminectomy, instrumented and noninstrumented fusion in patients with degenerative spinal
stenosis, the major predictor of the decision to perform
arthrodesis was found to be the individual surgeon.
Other factors associated with this decision were the
presence of more than 5 mm of spondylolisthesis,
younger age, female patients, scoliosis greater than 15,
patients with greater severity of back pain, and fewer
levels decompressed. The authors reported significantly
greater relief of back pain after 6 and 24 months in the
noninstrumented arthrodesis group. The decreased satisfaction in the instrumented arthrodesis group at these
earlier stages cannot be well explained, but it may be related to pain associated with the instrumentation itself.
The increased likelihood of a successful arthrodesis with
instrumentation may result in improved results in this
group over a longer period. Hospital costs were highest
in patients who underwent instrumented arthrodesis.

Lumbar Disk Herniation

Pathoanatomy

Prognosis and Outcomes

There is little published information regarding the true
natural history of lumbar spinal stenosis. Studies that include nonsurgical treatment show that about 15% to
20% of patients improve, 15% to 20% get worse, and
60% to 70% remain unchanged. Surgery appears to improve short-term outcomes in patients with severe
stenosis. In one study, 52% of nonsurgically treated patients showed improvement at 4-year follow-up compared with 70% of surgically treated patients. Patients
undergoing surgical treatment generally had more severe degrees of stenosis and were more symptomatic
than the nonsurgically managed group.
Appropriate patient selection and thorough decompression result in good long-term results following decompressive lumbar laminectomy with or without fusion. A meta-analysis of the surgical treatment of spinal
stenosis found that the average proportion of good to
excellent results in 74 studies was 64%. There were no
consistent predictors of clinical outcome, but younger
age, greater severity and duration of symptoms, previous
back surgery, multilevel involvement, coexisting medical
morbidity, and litigation issues were associated with
poorer outcome. The presence of bladder symptoms

542

A disk herniation is a localized displacement of disk

material beyond the limits of the intervertebral space.
The disk material can be nucleus pulposus, cartilage, anulus fibrosus, fragments of apophyseal bone, or any
combination of these materials. A herniated disk should
be distinguished from a disk bulge, which is a diffuse
symmetric outpouching of the anulus fibrosus associated
with varying degrees of disk degeneration. A true herniation may either be a protrusion, extrusion, or sequestration (Figure 2). The base of a protrusion is wider than
any diameter of the displaced material. In an extrusion,
the diameter of the displaced material in at least one
plane is greater than the width at its base. If a displaced
fragment loses all continuity with the parent disk, it is
termed a sequestration, and is prone to migrate within
the spinal canal. The relationship of the herniated nuclear material to the surrounding anulus fibrosus and
the overlying posterior longitudinal ligament defines its
containment. A herniation completely enveloped by anulus fibrosus, posterior longitudinal ligament, or both is
termed a contained herniation. Nuclear material that escapes through the annular periphery and rests beneath
the posterior longitudinal ligament has been referred to
as subligamentous.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 44 Lumbar Degenerative Disorders

to induce mild local discomfort. Stimulation of a root
exposed to nucleus pulposus reproduced sciatic pain in
the leg. Epidural application of extracts of nucleus pulposus in animal models when combined with mild deformation of the nerve root and dorsal root ganglion resulted in increased tissue edema, fibrosis in and around
nerve root, demyelination of axons, and Schwann cell
hypertrophy.
Inflammatory cytokines such as interleukin-1,
interleukin-6, prostaglandin-E2, and phospholipase A2
are found in significant concentrations in the nerve root
and dorsal root ganglion, suggesting they play a role in
the inflammatory process. The presence of these agents
incites vascular changes around the nerve root and also
has a direct effect on the blood-nerve barrier, promoting
intraneural edema and reducing neuronal perfusion. Interactions between nerve growth factors at the nerve
root and dorsal root ganglion may play a role in the
central modulation of spinal pain.

Figure 2 Illustration of four morphologic types of disk herniations: contained disk

protrusion (A), noncontained subligamentous disk extrusion (B), noncontained transligamentous disk extrusion (C), and noncontained disk sequestration (D).

The geographic relationship of the herniated nucleus

to the circumference of the anulus fibrosus and the intervertebral foramen is used to classify herniations as
central, paracentral, foraminal, or extraforaminal. Paracentral herniations are the most common pattern encountered in clinical practice. They are caused by the
weaker posterolateral portion of the posterior longitudinal ligament overlying this segment of the anulus fibrosus. Posterolateral herniations compress the lower exiting root (for example, the L5 root at L4-5 level),
whereas the more lateral and extraforaminal herniations
are more likely to compress the upper root (for example, the L4 root at L4-5 level). Axillary and shoulder descriptions refer to the position of the herniated disk
with respect to the origin of the nerve root from the
thecal sac.

Pathophysiology
Several pathophysiologic events occur in the nucleus
pulposus and adjacent areas that act in concert to produce radicular symptoms. Tumor necrosis factor-
(TNF-) may be a key component of this process in that
it exerts its effect by sensitizing the nerve root to produce pain in the presence of a mechanically deforming
force. Local accumulation of sodium ion channels and
spontaneous axonal activity may be pathways through
which TNF- acts.
The effects of mechanical deformation are compounded by chemical sensitization of the nerve root.
When diskectomy is performed using local anesthetic in
human subjects, light mechanical stimulation of nerve
roots not exposed to nucleus pulposus have been noted

American Academy of Orthopaedic Surgeons

Clinical Features
Lumbar disk herniation is the most common cause of
radicular pain in the adult working population, with an
estimated 2.8 million herniations (1% of the general
population) occurring annually. Ninety-five percent of
these herniations involve the L4-5 or L5-S1 lumbar disk
spaces, and most patients are between the ages of 20
and 50 years. Patients typically present with back pain
and sharp, stabbing leg pain accompanied by a feeling of
numbness or tingling in a specific dermatomal distribution. One study reports a dermatomal sensitivity and
specificity of 74% and 18% for paresthesias from lumbar disk herniation. Referred (sclerotomal) pain in the
buttock or posterior thigh arises from stimulation of
muscles, ligaments, periosteum, and other structures of
mesodermal origin and does not go beyond the knee.
Symptoms are aggravated by activities and maneuvers
that raise the intra-abdominal and intradiskal pressure,
such as coughing, sneezing, and sitting.
Motor, sensory, and reflex evaluation corresponding
to the lumbar roots should be specifically evaluated
(Figure 3). The straight leg raising test is a clinical maneuver that demonstrates limited excursion of inflamed
lumbosacral nerve roots. In lumbar disk herniation, the
test is sensitive (true positive in 72% to 97% of patients) but not specific (false positive in 11% to 66% of
patients). In contrast, the crossed straight leg raising test
has a lower sensitivity (true positive in 23% to 42% of
patients) but much higher specificity (false positive in
85% to 100% of patients). There is minimal movement
in the sciatic nerve or roots during the first 20 to 30 of
straight leg raising; most tension in the roots develops at
35 to 70 of elevation.
Large lumbar disk herniations may result in a cauda
equina syndrome, characterized by bilateral leg pain,

543

Orthopaedic Knowledge Update 8

Lumbar Degenerative Disorders

Figure 3 Illustration of the motor, reflex, and sensory radicular findings in various types of lumbar disk herniation.

544

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

lower limb weakness, saddle anesthesia, and bowel/
bladder changes. One percent to 10% of all patients undergoing surgery for lumbar disk herniation have symptoms of cauda equina syndrome. The incidence may be
higher when patients with subtle urinary symptoms are
included. Pain varies substantially in patients, and it is
modified and influenced by several factors including individual cultural and psychologic factors, sleep deprivation, and secondary gain. Evaluation of the patient
should include a skillful appraisal of these issues.

Management
Nonsurgical measures including activity modification,
anti-inflammatory agents, physical therapy modalities,
exercises, spinal manipulation, corsets, epidural injections, and nerve root blocks result in good resolution of
symptoms in most patients. It is unclear whether any of
these treatment options actually alters the natural history of disease.
Foraminal epidural steroid injections accurately administered under fluoroscopic control may help in combating the chemical mediators of pain and inflammation
associated with disk herniations. A positive response is
usually indicated by a reduction of leg pain by more
than 50%. The steroid load generally precludes more
than three to four injections over a 1-year period. The
most important application of these injections may be in
shortening the pain-control phase of treatment, which
allows early reconditioning to begin.
Persistent intractable pain following nonsurgical
treatment during a minimum 6-week period is the most
frequent indication for surgery. There is some evidence
that results of surgery deteriorate when nonsurgical care
exceeds 12 months. Imaging studies must correlate with
the symptoms and neurologic findings. The presence of
a nerve tension sign improves the likelihood of a good
postoperative result.
Other factors that influence the decision to proceed
with surgery include disk herniation into a stenotic canal, which may lead to recurrent or persistent symptoms; inability of patients to comply with the dictates of
a conservative therapy regimen; and the number of sciatica episodes experienced by a patient. Among patients
experiencing a second episode of sciatica, 90% will improve but 50% will have a recurrence of symptoms. The
incidence of future episodes of sciatica rises to almost
100% in patients who have experienced three prior episodes. Absolute indications for surgery in lumbar disk
herniation are bladder and bowel involvement and progressive neurologic deficit.
A laminotomy and diskectomy (microdiskectomy) is
the gold standard for surgical treatment of a posterolateral lumbar disk herniation. This treatment is frequently
performed as an outpatient or short-stay procedure using an operating microscope or surgical loupes with a

American Academy of Orthopaedic Surgeons

Chapter 44 Lumbar Degenerative Disorders

headlight and a small posterior midline incision centered on the disk space. Surgeons must be aware of the
anatomic peculiarities of the upper lumbar canal when
treating patients with high lumbar disk herniations. Anatomic peculiarities include smaller interlaminar spaces,
greater coverage of the disk space by the lamina of the
superior vertebra, proximity to the conus medullaris, a
smaller spinal canal with a higher volume of nerve tissue, and a more horizontal orientation of the upper lumbar nerve roots. An extraforaminal disk herniation is
ideally accessed through the Wiltse paraspinal approach,
which preserves motion segment stability by avoiding
injury to the lamina and facet joints. Injury to the dorsal
root ganglion with postoperative dysesthesias is a potential complication when treating patients with this form
of disk herniation. Endoscopic and other percutaneous
techniques for disk decompression are discussed in
chapter 50.

Prognosis and Outcome Following Intervention

Lumbar disk herniation is a self-limiting disease in most
patients. Eighty percent to 90% of patients obtain satisfactory resolution of symptoms with nonsurgical treatment. Imaging studies that monitor the changes in size
of nonsurgically treated disk herniations have shown a
progressive decrease in the size of herniations over
time. Large extrusions have a greater likelihood of total
resorption; they are phagocytosed by cells attracted by
the disk-incited inflammatory response or from desiccation and atrophy secondary to loss of nutrient supply.
Nonsurgical treatments will likely result in favorable
outcomes when the duration of sciatica is less than 6
months, the patient is younger, and there is no litigation
involved. The results of surgical treatment are better in
patients with large anterior-posterior herniated disk
length and in patients who have a greater compromise
of the canal area by the disk herniation. Patients who
have an extruded or sequestered fragment with a minimal fissure in the anulus fibrosus or those who have a
detached fragment lying beneath an intact anulus fibrosus are likely to do better following surgery than those
who have an extruded disk with a large or massive annular defect or those who have an intact anulus fibrosus
and no detached fragment beneath the anulus fibrosus.
The likelihood of recovery of neurologic deficit is independent of surgical intervention, and patients with
neurologic deficit but no pain do not require surgery.
Surgery is considered in some patients with painless motor deficits when there is functional weakness in a major
muscle group or in those with significant motor deficits
and no return of function after 6 weeks. Recovery of
motor deficit is the general rule, although in some patients complete recovery may take a long time. Longterm studies have shown that 30% of patients will continue to have sensory deficits despite resolution of pain.

545

Lumbar Degenerative Disorders

This finding is consistent with the fact that sensory fibers are the most vulnerable to compression (they are
affected first and recover last).
One prospective controlled study compared outcomes of surgical and nonsurgical treatment in a group
of patients with lumbar radicular pain and concordant
myelograms. Surgery was superior to nonsurgical treatment at 1-year follow-up and continued to be slightly
better at 4-year follow-up. Outcomes were similar in
both groups at 10-year follow-up. Approximately 60%
of both groups of patients were symptom free at 10-year
follow-up, although clinical recovery occurred earlier in
the surgically treated group. In another 5-year follow-up
study, 56% of nonsurgically managed patients reported
symptomatic improvement compared with 70% of surgically treated patients.
Long-term success rates following open diskectomy
are between 76% and 93%. In a study of 63 patients 10
years after interlaminar fenestration and diskectomy,
75% of patients reported some back pain, but only 13%
reported severe back pain. Younger patients with preoperative degenerative changes at the disk had a higher incidence of back pain. Sensory disturbances were present
in 81% of patients preoperatively, but in only 31% of
patients at final follow-up. Motor deficits were present
in 76% of patients preoperatively, and in 14% of patients at 10-year follow-up. A complete foot drop was
present in four patients preoperatively and this symptom did not improve after surgery. The straight leg raising test was negative in 97% of patients at the 10-year
follow-up. Mild preoperative urinary disturbances in
four patients resolved completely.

Orthopaedic Knowledge Update 8

herniations in patients in this age group are more frequent in the upper lumbar spine. Nerve root tension
signs such as the straight leg raising test are commonly
negative in elderly patients. This may be secondary to
reduced resting tension in the nerve roots as a result of
reduced disk height from the degenerative process or
reduced limb length from degenerative hip and knee
changes. Chronic fibrosis of the roots from compromised microcirculation may contribute to the severity of
the radicular pain. Histologically, the herniated disk
fragment primarily contains sections of anulus fibrosus.
Spontaneous resorption and improvement is less likely
to occur.
Disk Herniation in Young Patients
Pediatric disk herniations constitute 1% to 3% of all instances of lumbar disk herniation. Previous trauma and
vertebral column abnormalities such as congenital
stenosis or transitional vertebrae may increase the likelihood of herniation. The clinical presentation is similar
to that for adults, with back pain and nerve tension signs
being frequently reported; neurologic deficits in these
patients is typically uncommon. The herniation is frequently an avulsed fragment of the ring apophysis of the
vertebral body. In children and adolescents who have
persistent symptoms despite nonsurgical treatment, surgery usually results in good relief of symptoms. A unilateral hemilaminectomy with partial diskectomy is
sufficient treatment for most patients. A bilateral laminectomy is occasionally necessary when the avulsed
apophysis fragment is large.

Discogenic Low Back Pain

Special Situations
Recurrent Herniation and Reoperation
Recurrent herniations at the same level occur in approximately 5% of surgically treated patients at 5-year
follow-up and can be ipsilateral or, less frequently, contralateral. Survival analysis or rates of revision are a
better measure of surgical success. A recent large
population-based study from Finland analyzed the risk
of reoperations after lumbar diskectomy. Fourteen percent of a total of 35,309 patients in the study group underwent a second procedure during the 10-year period
of the study. The second procedure was diskectomy in
63% of patients, decompression in 23%, and fusion in
14%. Among the patients who underwent a second procedure, there was a 25% cumulative risk of needing a
third spinal surgical procedure within the next 10 years.
A pain-free interval of more than 1 year was associated
with a lower risk of revision.
Disk Herniation in the Elderly
Disk herniation in elderly patients commonly occurs in
the setting of spinal stenosis or spondylolisthesis. Disk

546

Discogenic pain refers to pain originating from a degenerative lumbar disk, which is characterized by axial low
back pain without associated radicular findings, spinal
deformity, or instability. The controversy surrounding
discogenic low back pain primarily exists because degenerative changes at the disks are ubiquitous, yet
symptoms arise in only a few patients; multiple additional anatomic sources of low back pain exist (Table 2);
the diagnosis of discogenic low back pain is made primarily with provocative diskography, which is controversial in itself; it is unclear whether the treatment options
for discogenic low back pain are superior to the natural
history of the disorder over the long term. Notwithstanding these issues, the degenerative lumbar disk is
being increasingly recognized as a valid source of axial
low back pain.

Pathoanatomy and Pathophysiology

As the disk ages, several chemical and mechanical
changes occur within it, which result in a continuum of
changes that can be observed using radiographic imaging. Type II collagen in the nucleus pulposus and anulus

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 44 Lumbar Degenerative Disorders

Table 2 | Differential Diagnosis of Low Back Pain

Spinal Causes
Trauma
Fractures
Musculoligamentous injury
Fracture in pathologic bone
Infections
Diskitis, osteomyelitis
Epidural abscess
Inflammatory
Seronegative spondyloarthropathy
Rheumatoid arthritis
Tumors
Primary arising from bone or marrow elements
Metastatic
Degenerative conditions
Spinal stenosis
Spondylolisthesis
Scoliosis
Discogenic low back pain
Miscellaneous
Pagets disease
Arachnoiditis
Sickle cell disease
Extraspinal Causes
Visceral origin
Urinary system-kidney stones, pyelonephritis
Reproductive system (endometriosis, retroverted uterus, ectopic pregnancy)
Gastrointestinal tract (duodenal ulcers, pancreatitis, biliary colic)
Abdominal aortic aneurysm
Retroperitoneal tumors
Musculoskeletal origin
Myofascial pain
Hip arthrosis
Sacroiliac joint pathology
Miscellaneous causes
Psychogenic
Central pain syndromes

fibrosus is increasingly replaced by type I collagen,

chondroitin sulfate is replaced by keratan sulfate, and
there is increasing dissociation between the collagen
and proteoglycans within the disk. The resulting loss in
hydrostatic properties of the disk leads to altered load
transmission, with asymmetric and abnormal stresses
being transmitted from the vertebral end plate to the
periphery of the anulus fibrosus. Radial tears develop
within the anulus fibrosus, with subsequent fissuring and
ingrowth of granulation tissue. As the process continues,
the disk loses height, and the vertebral bodies come

American Academy of Orthopaedic Surgeons

Figure 4 Illustration of the axial lumbar vertebra showing the origin, course, and
structures innervated by the sinuvertebral nerve.
closer to each other. Laxity in the peripheral attachments of the anulus fibrosus and the facet joint capsules
may allow motion or displacement between the vertebral bodies. Increased load transfer at the facet joints
and vertebral end plates results in degenerative changes
at both these sites.
Nerve fibers and nerve endings found in the peripheral portions of the disk offer a possible mechanism by
which lumbar disks act as pain generators. The disk is
innervated by the sinuvertebral nerve, which is formed
by branches from the ventral nerve root and sympathetic plexus (Figure 4). Once formed, the nerve turns
back into the intervertebral foramen along the posterior
aspect of the disk, supplying portions of the anulus fibrosus, posterior longitudinal ligament, periosteum of
the vertebral body and pedicle, and adjacent epidural
veins. The free nerve endings in the peripheral anulus fibrosus are immunoreactive for several pain-related neuropeptidessubstance P, calcitonin gene-related peptide, and vasoactive intestinal peptide. In degenerative
disks, nerve endings penetrate deep into the anulus fibrosus and even into nucleus pulposus. The free nerve
endings that penetrate deep into the disk are also immunoreactive for substance P.
Chemical factors may help explain why pain develops in a subset of patients with degenerative changes.
Mechanical deformation of the anulus fibrosus stimulates both mechanoreceptors as well as nociceptors by
lowering their firing thresholds. Inflammatory mediators
eluted by the disk (phospholipase A2, interleukin-1, and
matrix metalloproteinases) may play a role in the sensitization of pain receptors. Phospholipase A2 also stimulates the dorsal root ganglion, which can serve as another pathway for axial pain generation.
Mechanoreceptors and nociceptors in the facet joint
capsules and synovium may play an accessory role in
the symptoms of discogenic pain. The number of these
receptors in the facet joints falls during the weeks fol-

547

Orthopaedic Knowledge Update 8

Lumbar Degenerative Disorders

Table 3 | Radiographic Criteria for Lumbar Spine Instability

Resting radiographs
1. Sagittal displacement > 4.5 mm or 15%
2. Relative sagittal plane angulation > 22
or
Flexion-extension radiographs
1. Sagittal displacement > 4.5 mm or 15%
2. Relative sagittal plane rotation
> 15 at L1-2, L2-3, L3-4
> 20 at L4-5
> 25 at L5/S1
(Reproduced with permission from Posner I, White AA III, Edwards WT, Hayes WC: A biomechanical analysis of the clinical stability of the lumbar and lumbosacral spine. Spine
1982;7:374-389.)

lowing interbody fusion in animals, suggesting that appropriate stabilization of the disk space reduces nociception from the facets. Vertebral end plates and the
underlying cancellous bone have an increased density of
sensory nerves in patients with degenerative disk disease, thereby providing another pathway for pain generation.

Clinical Features
Clinically, discogenic pain is characterized by axial low
back pain without associated radicular pain, nerve tension signs, spinal deformity, or instability. The pain is
generally deep, aching, and exacerbated by sitting, bending, and axial loading. Symptoms are predominantly mechanical, and rest may provide relief. There may be a
history of prior injury to the spine such as a fall, lifting
with outstretched arms, or sudden twisting that resulted
in back pain. Instead of getting better, the pain gradually gets worse. Referred pain may radiate in a sclerotomal fashion to the sacroiliac joints, buttocks, or posterior
thighs, and occasionally into the inguinal region with involvement of the L5-S1 disk. Greater degrees of back
pain appear to be associated with a more distal pattern
of pain referral.
Traditional instability of a motion segment is a recognized cause of low back pain. This instability is diagnosed radiographically by intervertebral translation or
angulation, using criteria put forth by White and Panjabi
(Table 3). Patients frequently have mechanical back
pain related to posture or motion but no apparent radiographic instability. These patients may have a painful
arc of motion during forward bending or extension or
report a sense of shifting of their trunk with certain postures. Micromotion at the intervertebral segment may
play a role in the pathogenesis of pain in these patients.
There are currently no reliable methods of verifying

548

small degrees of instability, but intraoperative measurements of motion may provide more information in the
future.

Diagnostic Testing
Patients with discogenic low back pain typically have
MRI scans showing a degenerative disk without significant stenosis or herniation and concordant provocative
diskography. Patients who have normal MRI scans correlate highly with negative lumbar diskograms, and such
patients should be presumed to have a nondiscogenic
cause of low back pain. Degenerative changes are interpreted with caution in older patients because of the
ubiquitous nature of these changes in asymptomatic
older patients. Patients who have obvious radiographic
instability or multilevel severe degenerative disk and
facet joint changes cannot be categorized as having diskogenic back pain. It is helpful to categorize patients
into two groups: those who have radiographic evidence
of loss of disk height and those who do not.
Patients With Loss of Disk Height
In patients with loss of disk height, plain radiographs
show a decrease in disk height, which is often associated
with vertebral end plate sclerosis or mild facet joint arthrosis. MRI will usually show a greater degree of degenerative changes at the disk. There is loss of disk
height often associated with bulging of the posterior anulus fibrosus into the canal, some infolding of the ligamentum flavum, minor degrees of canal and foraminal
stenosis, and abnormal signals at the vertebral end plate.
Provocative diskography is used to confirm the diagnosis of discogenic low back pain when surgery is being
considered. The procedure is performed using fluoroscopic control with the patient awake. Several criteria
must be met for a diskogram to be considered positive.
There must be a concordant pain response from the patient, evidence of abnormal disk morphology on fluoroscopy and postdiskography CT examination, and negative control levels in the lumbar spine. Some authors
report that low-pressure pain responses suggest a chemical pathway for pain generation and have better outcomes following fusion.
Although complications from the procedure are uncommon, diskography continues to be controversial.
Positive diskograms have been found in up to 25% of
patients who were only mildly symptomatic, which
raises the risk of overdiagnosing discogenic disease.
When using strict criteria, including low-pressure injection and a normal control disk, positive results may be
as low as 30% in patients with chronic low back pain.
Normal psychologic profiles have been reported in
about 20% of patients with chronic low back pain who
are candidates for diskography.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Patients Without Loss of Disk Height
Plain radiographs in patients without evidence of loss of
disk height generally do not show any significant
changes. Disk height is maintained, and there are no
facet joint nor end plate changes. MRI typically shows
mild changes such as desiccated disks with low signal on
T2-weighted imaging or annular tears.

Prognosis
The natural history of acute low back pain is generally
excellent. Ninety percent of patients will go on to complete pain relief within 2 to 6 weeks. In patients who
have chronic pain, one study found that 40% of patients
could be diagnosed as having discogenic back pain with
positive diskograms at L4-L5 or L5-S1. There is limited
information on the natural history of confirmed discogenic low back pain. One study found 68% of patients
improved, 24% worsened, and 8% remained unchanged
over a 5-year period.

Management
Anti-inflammatory, analgesic, and antispasmolytic medications are effective in the management of acute back
pain, although their usefulness in managing chronic
back pain is unclear. Gastrointestinal toxicity and renal
impairment are concerns with the use of these drugs, especially in the elderly population. Flexible or rigid spinal supports may reduce lumbar mobility and decrease
intradiskal pressure in certain positions of lumbar flexion. A program of physical therapy should be directed
at ergonomic instruction, stretching maneuvers, and isometric stabilization exercises. Nonimpact type exercise
such as swimming or cycling is recommended. Other
measures used with varying degrees of success include
acupuncture, hydrotherapy, ultrasound, biofeedback,
electrical stimulation, manipulation, massage, and psychotherapy. Intradiskal electrothermal therapy has been
proposed as an option in the management of discogenic
back pain (see chapter 50).
Lumbar fusion is the surgical procedure of choice
for the treatment of discogenic low back pain in patients
who have intractable pain after an aggressive nonsurgical management program, MRI findings of disk degeneration, and concordant diskography at one or two
levels. Disk excision or other disk decompression procedures are not recommended for these patients.
The fusion approach may be posterior, anterior, or
both. Instrumentation is generally used with posterior
fusion, and implants or bone graft are used with interbody fusion. Successful fusion is obtained in 60% to
90% of patients who undergo a posterior procedure, but
clinical outcomes are satisfactory in only 40% to 70%.
Even with posterior instrumentation, there is some motion across the disk space that may account for persistent symptoms in some patients. Residual symptoms in

American Academy of Orthopaedic Surgeons

Chapter 44 Lumbar Degenerative Disorders

other patients may be associated with approach-related
muscle denervation and necrosis or disruption of the adjacent facet joints. Posterior lumbar interbody fusion or
transforaminal interbody fusion can allow posterior decompression and concurrent anterior column stabilization, while avoiding the morbidity of a separate anterior
operation. The drawbacks include limitations on the size
of the anterior graft or implant, nerve root injury, epidural fibrosis and chronic radiculitis, and potential instability from resection of the facet joints needed for exposure.
Anterior lumbar interbody fusion (ALIF) has several theoretic benefits in the treatment of discogenic
low back pain. It can remove pain fibers and receptors
from the anulus fibrosus and nucleus pulposus, eliminate motion across disk, restore disk height and indirectly decompress foramen, and enable surgeons to
avoid posterior muscle disruption during the approach.
Fusion rates are high, varying from 80% to 96% of patients. The disadvantages of the anterior approach are
the risks of major vessel injury and, in males, retrograde
ejaculation and impotence. Laparoscopic ALIF techniques require a skilled general surgeon for the approach and are more likely to be successful at the L5-S1
interspace. The type of anterior structural support used
in ALIF surgery has received much attention. Autologous cancellous bone shows high fusion rates, but migration and collapse have been drawbacks. Threaded cages
or machined allografts are being increasingly used with
interbody fusion, with the aim of reducing the morbidity
associated with bone graft harvest. Bone morphogenetic
protein used within threaded interbody cages in ALIF
shows high fusion rates and relief of back pain similar to
autograft over the short term.

Annotated Bibliography
Lumbar Spinal Stenosis
Atlas SJ, Keller RB, Robson D, Deyo RA, Singer DE:
Surgical and nonsurgical management of lumbar spinal
stenosis: 4-year outcomes from the Maine Lumbar
Spine Study. Spine 2000;25:556-562.
A cohort of 119 patients underwent either surgical or nonsurgical treatment and were followed for 4 years. Seventy percent of surgically treated patients and 52% of nonsurgically
treated patients reported improvement in symptoms. The authors report that the relative benefits of surgery diminished
over time but remained superior to those of nonsurgically
treated patients.

Fischgrund JS, Mackay M, Herkowitz HN, Brower R,

Montgomery D, Kurz LT: Degenerative lumbar spondylolisthesis with spinal stenosis: A prospective, randomized study comparing decompressive laminectomy and
arthrodesis with and without spinal instrumentation.
Spine 1997;22:2807-2812.

549

Lumbar Degenerative Disorders

This randomized study showed that fusion rates were superior in instrumented patients undergoing single level surgery, but overall clinical improvement was identical in instrumented and noninstrumented groups.

Iguchi T, Kurihara A, Nakayama J, Sato K, Kurosaka M,

Yamasaki K: Minimum 10-year outcome of decompressive laminectomy for degenerative lumbar spinal stenosis. Spine 2000;25:1754-1759.
In this study, decompressive laminectomies without fusion
in patients older than 60 years were shown to produce longterm satisfactory results in more than half of the patients. The
authors concluded that the presence of sagittal rotation of
more than 10 in patients needing multiple laminectomies may
be related to deterioration of outcome.

Katz JN, Lipson SJ, Lew RA, et al: Lumbar laminectomy

alone or with instrumented or noninstrumented arthrodesis in degenerative lumbar spinal stenosis: Patient selection, costs, and surgical outcomes. Spine 1997;22:11231131.
The authors of this study report that adding arthrodesis to
decompression for spinal stenosis depends primarily on surgeon preference and that noninstrumented fusions produce
superior relief of back pain, whereas instrumentation adds
substantial costs to the treatment.

Rao RD, Wang M, Singhal P, McGrady LM, Rao S: Intradiscal pressure and kinematic behavior of lumbar
spine after bilateral laminotomy and laminectomy. Spine
J 2002;2:320-326.
This article presents a biomechanical analysis of bilateral
laminotomy as an alternative to wide laminectomy in the decompression of patients with lumbar spinal stenosis.

Lumbar Disk Herniation

Atlas SJ, Keller RB, Chang Y, Deyo RA, Singer DE:
Surgical and nonsurgical management of sciatica secondary to a lumbar disc herniation: Five-year outcomes
from the Maine Lumbar Spine Study. Spine 2001;26:
1179-1187.

Orthopaedic Knowledge Update 8

Osterman H, Sund R, Seitsalo S, Keskimaki I: Risk of
multiple reoperations after lumbar discectomy: A
population-based study. Spine 2003;28:621-627.
This 11-year review of 35,309 patients who underwent diskectomy showed that 14% had at least one more operation,
whereas 2.3% had two or more operations. Sixty-three percent
of the second operations were diskectomies, 14% were fusions,
and 23% were decompressions.

Yorimitsu E, Chiba K, Toyama Y, Hirabayashi K: Longterm outcomes of standard discectomy for lumbar disc
herniation. Spine 2001;26:652-657.
In this study, 74% of patients who underwent lumbar diskectomy reported some back pain at 10 years, although disabling pain was reported in only 12.7%. Decreased disk height
was associated with more disability, whereas recurrent herniations were more common in those with preserved disk height.

Discogenic Low Back Pain

Fritzell P, Hagg O, Wessberg P: Nordwall Anders, Swedish Lumbar Spine Study Group: Lumbar fusion versus
nonsurgical treatment for chronic low back pain. Spine
2001;26:2521-2534.
This randomized controlled multicenter study showed that
lumbar fusion in a carefully selected group of patients with severe chronic low back pain decreases pain and disability more
than nonsurgical treatment.

Riew KD, Yin Y, Gilula L, Bridwell KH, Lenke LG, Lauryssen CC, Goette K: The effect of nerve-root injections
on the need for operative treatment of lumbar radicular
pain: A prospective, randomized, controlled, doubleblind study. J Bone Joint Surg Am 2000;82:1589-1593.
This prospective study demonstrated the efficacy of selective nerve root injections of corticosteroid in patients with
lumbar radiculopathy. Twenty of 28 patients who received injection of bupivacaine and betamethasone declined surgery
over a 13- to 28-month period after receiving one to four injections, whereas the remainder of the patients eventually underwent surgery.

A 5-year follow-up of 402 patients showed that 70% of

surgically treated patients and 56% of nonsurgically treated
patients reported improvement in their predominant symptoms. Patients with moderate or severe symptoms did better
after surgery than after nonsurgical care.

Classic Bibliography

Carragee EJ, Han MY, Suen PW, Kim D: Clinical outcomes after lumbar discectomy for sciatica: The effects
of fragment type and annular competence. J Bone Joint
Surg Am 2003;85:102-108.

Bogduk N, Tynan W, Wilson AS: The nerve supply to the

human lumbar intervertebral discs. J Anat 1981;132:3956.

This prospective study correlated intraoperative morphologic patterns of disk herniation with outcomes following surgical intervention, with particular reference to reherniation, reoperation rates, and the incidence of persistent symptoms.

Freemont AJ, Peacock TE, Goupille P, Hoyland JA,

OBrien J, Jayson MI: Nerve ingrowth into diseased intervertebral disc in chronic back pain. Lancet 1997;350:
178-181.

550

Abumi K, Panjabi MM, Kramer KM, Duranceau J, Oxland T, Crisco JJ: Biomechanical evaluation of lumbar
spinal stability after graded facetectomies. Spine 1990;
15:1142-1147.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Herkowitz HN, Kurz LT: Degenerative lumbar spondylolisthesis with spinal stenosis: A prospective study comparing decompression with decompression and intertransverse process arthrodesis. J Bone Joint Surg Am
1991;73:802-808.
Kuslich SD, Ulstrom CL, Michael CJ: The tissue origin
of low back pain and sciatica: A report of pain response
to tissue stimulation during operations on the lumbar
spine using local anesthesia. Orthop Clin North Am
1991;22:181-187.

American Academy of Orthopaedic Surgeons

Chapter 44 Lumbar Degenerative Disorders

Mixter WJ, Barr JS: Rupture of the intervertebral disc
with involvement of the spinal canal. N Engl J Med
1934;211:210-215.
Turner JA, Ersek M: Herron L, Deyo R: Surgery for
lumbar spinal stenosis: Attempted meta-analysis of the
literature. Spine 1992;17:1-8.
Weber H: Lumbar disc herniation: A controlled, prospective study with 10 years of observation. Spine
1983;8:131-140.

551

Chapter

45

Spondylolysis-Spondylolisthesis
Thomas J. Puschak, MD
Rick C. Sasso, MD

Introduction
Spondylolysis refers to a bony defect in the pars interarticularis, which is the isthmus or bony bridge that connects the superior and inferior articular facets of the
posterior neural arch. Spondylolysis can occur unilaterally or bilaterally and is an acquired condition because
it has never been reported at birth. Spondylolysis is
more common in males and occurs in approximately 6%
of the general population. A higher prevalence (up to
53%) is seen in Eskimo populations. Also, athletes who
participate in sports that repeatedly cause the spine to
be hyperextended, such as gymnastics, football, and
wrestling, may have an increased predisposition to developing spondylolysis. Spondylolysis occurs most often
at L5, with decreasing incidence at the more cranial
lumbar levels. Several factors have been implicated in
the etiology of spondylolysis; however, the primary lesion is believed to be a stress fracture of the pars interarticularis that remains unhealed. This theory is supported by the fact that there are no reported instances
of spondylolysis in patients who have never walked.
There also seems to be a genetic predisposition for
spondylolysis because relatives of index cases have a
greater than fourfold increased incidence.
Spondylolisthesis refers to the translation of a vertebral body on the caudal vertebra. This translation can be
anterior, lateral, or posterior. The term comes from the
Latin roots spondy, which means the spine, and olisthesis, which means a slipping. Spondylolisthesis most
often occurs in the lower lumbar spine. Several different
forms of spondylolisthesis exist and can be classified by
severity of slip, etiology, and potential for progression.

Classification Systems
Meyerding
The Meyerding classification system is a radiographic
system based on the severity of vertebral slippage. Slips
are classified as grade 1 through grade 4 based on the
percentage of translation of the cranial vertebra on the
caudal vertebra. The superior end plate of the caudal
vertebra is divided into quarters, and the percentage of

American Academy of Orthopaedic Surgeons

slippage is recorded. Grade 1 is 0 to 25% slippage (Figure 1), grade 2 is 26% to 50%, grade 3 is 51% to 75%
(Figure 2), and grade 4 is 76% to 100% (Figure 3). Slippage of 100% or more is referred to as spondyloptosis.
This classification system is frequently used because it is
simple and reliable.

Wiltse and Associates

The Wiltse and associates classification system is based
on anatomic etiology and classically consists of five
types of spondylolisthesis, although iatrogenic postoperative instability is sometimes included as a sixth type.
Type 1 slips are dysplastic or congenital. These occur at
the lumbosacral junction as a result of incomplete or inadequate development of the facet joints and superior
sacral end plate. Facets are often sagittally oriented and
the superior end plate of S1 is rounded or domed, which
predisposes patients to slippage. The intact pars interarticularis usually limits slippage to approximately 30%.
Spondyloptosis can occur, and patients with highergrade slips of this type can develop significant neurologic symptoms from severe stenosis resulting from an
intact pars interarticularis. Type 2 slips are referred to as
lytic or isthmic because they are secondary to defects of
the pars interarticularis. There are three subtypes in this
category: a pars defect, an elongated pars (possibly
caused by repeated fracture and healing of the pars),
and an acute pars fracture. Type 3 slips are degenerative
and occur as a result of the incompetence of arthritic
facet joints in the degenerative lumbar spine. Degenerative spondylolisthesis can occur with or without spinal
stenosis. Type 4 slips are caused by traumatic injuries to
the posterior neural arch in areas other than the pars interarticularis and lead to a destabilization of the facet
joints. Type 5 slips are caused by pathologic destabilization of the spine from tumors, infection, and other systemic diseases.

Marchetti and Bartolozzi

The system described by Marchetti and Bartolozzi attempts to classify spondylolisthesis based on anatomic

553

Orthopaedic Knowledge Update 8

Spondylolysis-Spondylolisthesis

Figure 2 Lateral radiograph showing a Meyerding grade 3 slip. Dysplastic features

such as a trapezoidal L5 body and rounded sacral dome are often seen in slips that
progress beyond Meyerding grade 2.

Figure 1 Lateral radiograph showing a Meyerding grade 1 slip.

slips, and therefore, they have less potential for slip progression.
Acquired spondylolisthesis occurs as a result of acquired pathology, such as pars defects, trauma, degenerative facets, pathologic instability, and iatrogenic injury.
The main difference between the developmental and acquired categories is that the basic spinal architecture is
developed normally in the acquired group and therefore
has more potential stability or less potential for progression of slippage.

Diagnostic Imaging
Plain Radiographs

Figure 3 Lateral radiograph showing a Meyerding grade 4 slip. As in Figure 2, significant dysplastic features are present.
etiology and prognostic factors. Slips are divided into
two main categories: developmental and acquired. Developmental spondylolisthesis is categorized as either
high dysplastic or low dysplastic. Dysplastic features of
the anterior and posterior elements, which lead to instability, characterize developmental slips. Posteriorly, the
pars interarticularis, laminae, or facets may be incompetent; anteriorly, the L5 body tends to be trapezoidal and
oriented toward the floor and the S1 superior end plate
tends to be rounded. Low dysplastic slips tend to have
less profound dysplastic features than high dysplastic

554

Initial assessment should at least consist of standing AP

and lateral radiographs. The standing position may accentuate existing translational deformity. Spondylolisthesis can be missed on supine lateral films up to 20%
of the time because of reduction of the deformity when
the patient is in the supine position. Standing films may
also accentuate angular deformities better than those
obtained with the patient supine. Pars interarticularis
defects and other defects of the posterior arch such as
spina bifida occulta may be identified on AP views. In
patients with high-grade slips with significant angulation
of the cephalad vertebra, a Napoleons hat sign may be
seen on AP views (Figure 4). A standard AP view may
not be helpful for visualization of the pedicles and other
surgical anatomy in patients with high-grade slips. In
these patients, a Ferguson view (AP radiograph angled
30 cephalad) will yield a true AP view of the lumbosacral junction.
Although pars interarticularis defects are usually
identifiable on lateral films, right and left oblique radiographs may help in the diagnosis of subtle pars defects
not readily seen on AP or lateral views.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 45 Spondylolysis-Spondylolisthesis

Dynamic flexion-extension lateral radiographs can

be helpful in assessing translational stability. Patients
with bilateral pars interarticularis defects as well as
those with arthritic degenerative facets (especially sagittally oriented facets) may elicit significant translational
motion on flexion-extension radiographs that is not evident on standing lateral films. The difficulty in interpreting dynamic radiographs is that there is no clear consensus on the definition of normal translational motion.
Traditionally, the definition of instability is greater than
3 mm of translation on flexion-extension views. Several
clinical studies have reported large variations in translational motion in asymptomatic individuals. Also, the reliability of dynamic studies is questionable because the
quality of the study depends on the effort of the individual. Dynamic radiographs can be valuable diagnostic
tools when used in conjunction with other data, but they
should not be used as the sole criteria for treatment.

Computed Tomography
CT scans are useful in the diagnosis of occult pars interarticularis defects. Thin section cuts (1 to 2 mm) should
be obtained because larger axial sections may miss the
defects. Sagittal reconstruction technology is also helpful because pars interarticularis defects may be difficult
to see on axial images. These defects often lie in a similar plane to the axial cuts. Pars interarticularis defects
tend to lie more dorsal and posterior to the facet joints
on the axial images and are often associated with significant bony and cartilaginous overgrowth. In the degenerative spine, axial CT images allow assessment of the
orientation of facet joints (coronal or sagittal) to help
determine the relative stability of a segment after decompression. CT scans are also helpful in assessing surgical bony anatomy, such as pedicle size and orientation.
Myelograms in conjunction with CT scans are helpful in assessment of the neuroanatomy. Although central
and lateral recess stenosis is easily identified, foraminal
stenosis may be missed because the compression occurs
lateral to the root sleeve beyond the extent of the myelogram dye.

Magnetic Resonance Imaging

MRI scans are not often helpful in identifying spondylitic defects. They are most helpful in the evaluation of
the neuroanatomy and disks. Central and lateral recess
stenosis caused by facet hypertrophy, overgrown pars interarticularis defects, and translational offset is easily
seen. Also, sagittal images allow accurate assessment of
the nerve root in the foramina.
MRI scans can evaluate the status of the intervertebral disks at the affected and adjacent levels. Normal
disks appear bright on T2-weighted images because of
good hydration, and they appear dark when degeneration (relative dehydration) is present. Reactive bony

American Academy of Orthopaedic Surgeons

Figure 4 AP radiograph of a high-grade slip. The Napoleons hat sign is outlined.

Severe angulation of the L5 body allows an end-on axial projection on AP views.
end plate changes are also signs of disk degeneration.
Often disks adjacent to slips may be radiographically
normal, but they elicit significant degenerative changes
on MRI, which must be considered during surgical planning. Although knowing the status of the adjacent disk
is important, there are no conclusive findings that warrant routinely incorporating the adjacent degenerative
level in patients with spondylolisthesis.

Nuclear Medicine Studies

Technetium bone scans have been historically popular
in the diagnosis of spondylolysis. These scans are used to
determine the acuteness of a pars injury, with positive
scans suggesting a more acute injury and greater potential for healing with bracing or conservative care. Negative scans indicate an old or chronic defect with little
healing potential. Recently, single photon emission CT
has been shown to be more specific and sensitive in detecting radiographically occult pars defects than technetium scans.

555

Spondylolysis-Spondylolisthesis

Orthopaedic Knowledge Update 8

Degenerative Spondylolisthesis

Clinical Presentation

Pathogenesis/Pathoanatomy

Patients with degenerative spondylolisthesis initially report low back pain secondary to the degenerative
changes in the spine. These are usually mechanical complaints worsened with activity and improved with rest.
With time, neurogenic back pain may develop in which
pain is exacerbated with prolonged standing or walking
and improved with sitting and flexion. The back pain
tends to be located in the lower lumbar and buttock region and is often described as an aching, burning, or
pulling sensation.
As nerve compression progresses, lower extremity
symptoms increase. Leg pain may be unilateral or bilateral and neuroclaudicatory or radicular in nature. Symptoms tend to be worsened in an upright or extended
posture and improved with sitting and flexion. The L5
root is most commonly affected; however, in patients
with instability and foraminal stenosis, the L4 roots are
also involved. Reflexes are often diminished or absent,
and sensation may be anywhere from normal to severely impaired. Bladder dysfunction only occurs in 3%
to 4% of patients.
Peripheral neuropathy must be considered in patients with a history of diabetes and stocking glove pattern dysesthesias. Other diagnoses that can mimic spinal
stenosis symptoms are cervical myelopathy and primary
hip disease with anterior thigh pain.

Degenerative spondylolisthesis tends to occur in patients older than 60 years. It is more prevalent in women
and more common in African Americans than Caucasians. It occurs five to six times more frequently at L4-5
than at L3-4 or L5-S1. The average amount of anterior
slippage is 15% to 33%. Hormonal influence may contribute to the development of degenerative spondylolisthesis. The increased incidence in women with a history
of pregnancy may be the result of increased ligamentous
and joint laxity in conjunction with a large flexion moment on the lumbar spine.
The role of facet joint orientation has also been investigated as a potential cause of degenerative spondylolisthesis. Several authors have shown a positive correlation between sagittally oriented facet joints and a
predisposition for spondylolisthesis. A bilateral facet angle greater than 45 at L4-5 has been shown to result in
a 25-fold increased incidence of degenerative spondylolisthesis. The L5-S1 facet joint tends to be oriented
more in the coronal plane, offering greater resistance to
anterior translational forces. This orientation may explain the greater incidence of degenerative spondylolisthesis at L4-5 than at L5-S1.
Degenerative changes in the intervertebral disk combined with hormonal factors and facet orientation can
lead to intersegmental instability. Disk space collapse
leads to buckling of the ligamentum flavum and altered
stress loading of the facet joints. Ligamentous laxity and
remodeling of the facet joints create a hypermobile segment, which leads to spondylolisthesis. The hypertrophy
and buckling of ligamentum flavum combined with bony
translational offset lead to central canal stenosis. Hypertrophy of the facet joints creates lateral recess and foraminal stenosis. Facet cysts, often associated with advanced
degenerative facet arthrosis, can add to the degree of
nerve root compromise in the lateral recess.

Natural History
Few data are available regarding the natural history of
degenerative spondylolisthesis. Severe disk space narrowing has been shown to be associated with a lower
likelihood of progression of slippage. In a study following patients with degenerative spondylolisthesis who
were treated nonsurgically over 10 years, 76% of patients without neurologic symptoms remained symptom
free. Eighty-six percent of patients with lower extremity
complaints saw an initial improvement in symptoms;
however, 37% had redevelopment of symptoms. Eightythree percent of patients presenting with neurologic
symptoms who refused surgical treatment eventually experienced a deterioration of symptoms. Although degenerative spondylolisthesis symptoms tend to be intermittent, complete resolution of symptoms is not likely.

556

Nonsurgical Management
Most patients with degenerative spondylolisthesis will
respond to nonsurgical treatment. Initial treatment is focused on the mechanical back pain and consists of a
short rest period, administration of nonsteroidal antiinflammatory drugs, a short course of oral analgesics,
and passive physical therapy modalities. Once the acute
pain phase is controlled, active physical therapy can be
instituted for trunk stabilization and aerobic conditioning. Avoidance of hyperextension activities is recommended to limit recurrence of acute episodes.
Similarly, leg symptoms from spinal stenosis can be
managed with rest, nonsteroidal anti-inflammatory
drugs, and oral analgesics. Oral steroids and injectable
steroids in the form of epidural steroid injections or selective nerve root blocks may also be used. Despite being somewhat controversial and lacking significant prospective data proving their efficacy, epidural steroid
injections are widely used and are often effective (if
only in temporizing lower extremity symptoms).

Surgical Management
The main goals of surgery are pain reduction, restoration of function, and preservation of neurologic function. The most common indication for surgery is persistent incapacitating claudication and radicular leg pain,
which significantly compromises function, and the fail-

American Academy of Orthopaedic Surgeons

Chapter 45 Spondylolysis-Spondylolisthesis

Orthopaedic Knowledge Update 8

ure of 6 to 12 weeks of nonsurgical therapy to relieve
symptoms. Absolute indications for surgery include progressive and evolving motor deficits and cauda equina
syndrome. Surgical intervention should address the two
components of degenerative spondylolisthesisspinal
stenosis and instability.
The management of spinal stenosis includes laminectomy or laminotomy with partial or total facetectomy.
Depending on symptoms, decompression may be unilateral or bilateral. All regions of compression must be addressed because incomplete decompression of the lateral recess and stenosis can lead to persistent symptoms
and poor outcome. Outcomes of decompression without
fusion in the face of degenerative spondylolisthesis are
mixed. Several studies show significant improvement in
leg symptoms initially; however, up to 33% of patients
had severe worsening of back and leg symptoms by 5
years, with nearly 25% requiring reoperation for the
treatment of recurrence of stenosis and/or progression
of instability. In one study, more than half of the patients with decompression alone were unable to walk
more than two blocks after 3 to 6 years, and 25% of patients were very dissatisfied with the treatment outcome.
Several studies have addressed the role of arthrodesis in addition to decompression for the treatment of degenerative spondylolisthesis. In a prospective randomized study comparing decompression alone to
decompression with posterolateral noninstrumented fusion, at 3-year follow-up 96% of the patients who underwent decompression alone had radiographic slip progression compared with 28% of those in the arthrodesis
group. Despite a pseudarthrosis rate of 36%, 96% of patients in the arthrodesis group had good or excellent results compared with 44% of those who underwent decompression alone.
Numerous studies have also assessed the effects of
using instrumentation in conjunction with arthrodesis. A
randomized prospective study comparing single-level
laminectomy and posterolateral fusion with and without
transpedicular fixation showed that the addition of instrumentation significantly improves fusion rates. Despite the significant difference in fusion rates, in 83% of
patients in the instrumented group versus 45% of those
in the noninstrumented group, there was no significant
difference in clinical outcomes, with 78% to 85% of
those in both groups reporting good and excellent results. Several other authors have reported an improvement in both fusion rates and clinical outcomes when
comparing instrumented and noninstrumented fusions
for the treatment of patients with degenerative spondylolisthesis.
When surgical treatment is indicated, a thorough decompression is mandatory, and the addition of arthrodesis improves intermediate- and long-term outcomes. The
addition of instrumentation increases fusion rates, but it

American Academy of Orthopaedic Surgeons

has not been shown to consistently improve outcome

over noninstrumented arthrodesis. No known randomized prospective data have been published to date regarding the role of interbody fusion in the treatment of
degenerative spondylolisthesis.

Acquired Isthmic Spondylolisthesis in Adults

Pathogenesis/Pathoanatomy
The primary lesion responsible for isthmic or lytic
spondylolisthesis is a defect of the pars interarticularis.
Several etiologies for this defect have been proposed,
including acute or chronic traumatic injuries. Congenital
abnormalities can contribute to instability as well. There
are several types of pars defects: chronic stress fractures,
elongated pars from healing of multiple stress fractures
over time, and acute pars fractures. These lesions are acquired, not congenital, and result from repeated microtrauma of the pars as it is loaded in extension. Isthmic spondylolisthesis occurs at L5-S1 in 82% of patients
compared with 11% at L4-5 and 0.5% at L3-4 and L2-3.
Forces in the lumbar spine are greatest at the lumbosacral junction and likely are the reason for the high prevalence of pars defects at L5-S1.

Natural History
Relatively few patients with spondylolysis will acquire
spondylolisthesis. Those who do typically have some associated dysplastic features such that the true incidence
of acquired isthmic spondylolisthesis may be much less.
Many patients who eventually acquire spondylolisthesis
are asymptomatic. True acquired slips are almost always
grade I or II and rarely progress beyond grade I or II
severity unless there is dysplasia present. Isthmic
spondylolisthesis that progresses beyond Meyerding
grade II almost always is associated with dysplasia as
well.
Progression of isthmic spondylolisthesis is most common in the adolescent population. Significant increase
in slippage in adulthood is uncommon. In contrast to
the L5-S1 isthmic slip, lesions at L4-5 or other more cranial levels may remain unstable into the third and
fourth decades of life, with progression of translation
and increase in back and leg symptoms. An isthmic lesion at L4-5 is less stiff in sagittal rotation and shear
translation than a lesion at L5-S1. Overloading a lesion
at L4-5 may lead to premature translational wear of the
disk, contribute to the inherent instability of the segment, and may be the reason why previously asymptomatic isthmic slips at L4-5 progress later in adulthood.

Clinical Presentation
The most common presenting symptom is low back
pain. Patients often have a long history of periodic selflimited low back pain episodes that vary in intensity
and/or duration. Neurologic deficits are infrequent be-

557

Spondylolysis-Spondylolisthesis
cause slips rarely progress beyond grade II and the relative detachment of the posterior arch prevents significant central canal stenosis. Patients may present with
unilateral or bilateral radiculopathy. Radicular symptoms may be caused by nerve root irritation from the
reactive tissue around the pars defect combined with
the micromotion of the unstable posterior arch. Additionally, significant foraminal stenosis may occur as a result of loss of sagittal foraminal height from the translation and degeneration of the intervertebral disk.
Typically, the exiting nerve root (L5) is most affected for
an L5-S1 isthmic spondylolisthesis.

Nonsurgical Treatment
Most patients with acquired isthmic spondylolisthesis
will respond to nonsurgical management. Medical treatment should follow similar guidelines for nonspecific
low back pain. Oral anti-inflammatory drugs may reduce acute pain and improve function. Long-term use of
these drugs should be avoided if possible because of potential renal and gastrointestinal adverse effects. Narcotic pain medication, muscle relaxers, and other controlled substances should be used with extreme caution
and definitely should be avoided in long-term treatment.
There are no known studies that address the role of
injections in the facet joints or pars defects in patients
with isthmic spondylolisthesis. Injection of local anesthetic and corticosteroid into the facets or pars defects
may have therapeutic effects. However, diagnostic information from these injections is at best difficult to interpret because the pars defects often communicate with
the facet joints, causing an uncontrolled extravasation of
steroid and local anesthetic. In patients with radicular
symptoms, selective nerve blocks or epidural steroid injections may be used for diagnostic and therapeutic purposes. No prospective studies have yet been conducted
that address the effectiveness of these injections in isthmic spondylolisthesis.
The use of external bracing in the treatment of patients with acquired isthmic spondylolisthesis has been
reported. In addition, patients may benefit from initial
rest in the acute phase followed by physical therapy that
is focused on strengthening of the abdominal and lumbosacral muscles. Physical therapy has been shown to
decrease pain and functional disability. Once the acute
pain is resolved, the focus of physical therapy is on hamstring stretching, pelvic tilts, and abdominal and pelvic
stabilizer strengthening for approximately 6 months.

Surgical Treatment
Failure of conservative management (persistence of
pain, progression of neurologic symptoms, or progression of slippage) is an indication for surgical treatment.
Numerous approaches to surgical reconstruction have

558

Orthopaedic Knowledge Update 8

been used. These approaches include anterior interbody
fusion, posterolateral fusion with and without instrumentation, anterior-posterior procedures, and posterior
interbody procedures.
Posterolateral fusion is commonly used to treat patients with isthmic spondylolisthesis. Midline dissection
is performed laterally over the facet joints and transverse processes and sacral alae. Abundant lateral bone
graft leads to stabilization of the segment by intertransverse process fusion. A Gill laminectomy may be done
in conjunction with the posterior fusion. Removal of the
loose lamina and inflammatory pars defect allows decompression of the nerve roots, direct visualization of
the pedicles for hardware placement, and potentially removes part of the inflammatory back pain generator. In
addition, the laminectomy bone may be processed for
bone graft. Although the roles of instrumentation and
decompression with posterolateral fusion remain controversial, Gill decompression without fusion is not recommended. Potential complications of the posterior approach include infection, dural tear, nerve root injury,
hematoma, and pedicle fracture.
Anterior lumbar interbody fusion (ALIF) can be
successfully used to treat isthmic spondylolisthesis even
when radicular symptoms are present. Although the direct decompression of nerves is difficult, indirect decompression can be obtained by foraminal distraction.
A transabdominal retroperitoneal approach is usually
used to minimize the risk to abdominal viscera. Caudal
to the iliac vessels, cautery should be used sparingly to
decrease the risk of injuring the presacral nerves, which
may result in retrograde ejaculation and sexual dysfunction. The anterior approach allows for an aggressive diskectomy with good removal of cartilage from the vertebral end plates and increase of the fusion surface area.
Additionally, anterior grafts provide excellent anterior
column support and may aid in partial reduction of the
slip as well as restoration of disk space height. The
choice of graft material is generally based on surgeon
preference and includes autologous iliac crest, structural
allograft, and various metallic cages packed with autogenous cancellous graft. This method of anterior diskectomy and interbody fusion cannot be done in patients
with high-grade spondylolisthesis because of the presence of significant translational and angular deformity.
Risks of the anterior approach include wound infection,
vascular injury, deep venous thrombosis, pulmonary embolism, sympathetic chain injury, retrograde ejaculation,
and ventral hernia.
Stand-alone ALIF reconstruction for patients with
isthmic spondylolisthesis works best for those with a
Meyerding grade 1 slip. Several clinical studies have
shown adequate results with good and excellent results
ranging from 87% to 94% at 2 years. One study reported that 75% of patients had excellent and 20% had
satisfactory results at 10-year follow-up. The fusion rates

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

in these studies did not correlate with clinical outcome.
Some patients had good clinical improvement despite
eventually having a pseudarthrosis. Persistent leg pain
may occur as a result of inadequate indirect foraminal
decompression. Patients may also have persistent low
back pain because of the pain fibers present in the inflammatory tissue of the pars interarticularis defect,
which is not removed with an isolated anterior approach. Nociceptive pain fibers are histologically
present in the pars interarticularis defect.
Circumferential (360) fusion can be accomplished
by a combined anterior-posterior fusion or by a posterolateral fusion combined with a posterior lumbar interbody fusion (PLIF) or transforaminal lumbar interbody
fusion. Both techniques provide anterior column support, foraminal distraction, posterior stabilization, arthrodesis, and the ability to perform a Gill laminectomy
decompression if indicated. Three series have shown excellent success with circumferential procedures when
compared with other approaches alone. In one study,
posterolateral fusion alone was compared with posterolateral fusion combined with posterior lumbar interbody
fusion. Outcomes were comparable when good and excellent results were grouped together (95% and 97%,
respectively). However, the circumferential group had
75% excellent results compared with 45% in the posterolateral fusion group. Several other studies have shown
higher fusion rates for circumferential fusions when
compared with anterior or posterior fusions alone. Specific risks related to circumferential fusions include increased surgical time, more blood loss, and potentially
longer hospital stays.

Developmental Spondylolisthesis
Developmental spondylolisthesis was formerly referred
to as congenital spondylolisthesis. Current thought is
that the defects and slip are not present at birth but develop over time. Unlike acquired isthmic spondylolisthesis, the pars fractures are thought to develop as a result
of the slippage rather than causing the slip. The dysplastic nature of this type of spondylolisthesis often leads to
the development of high-grade slips. Developmental
spondylolisthesis is the most common type of slip seen
in children, and as such will be covered in greater detail
in chapter 66. The focus of this section will be on the
treatment of high-grade spondylolisthesis in adults.

Pathomechanics/Pathogenesis
Developmental spondylolisthesis is categorized as either
high dysplastic or low dysplastic slips based on radiographic findings. Radiographic findings include deficiencies of the posterior arch, a trapezoidal L5 body,
rounded sacral dome, incompetent L5-S1 disk, and
poorly formed facet articulations. Severe lumbosacral
kyphosis with verticalization of the sacrum and hyper-

American Academy of Orthopaedic Surgeons

Chapter 45 Spondylolysis-Spondylolisthesis

Figure 5 Preoperative photograph of a patient with a high-grade slip. Note the

crouching posture caused by severe hamstring tightness. Upper lumbar hyperlordosis
is also seen.
lordosis of the upper lumbar spine usually develops.
This type of slip has the highest risk for malignant progression. Patients with this type of slip and an intact
pars interarticularis must be observed closely because
they may develop cauda equina syndrome as the slip
progresses.
Low dysplastic slips also tend to occur at the lumbosacral junction. This type of slip is characterized by
dysplastic changes in the posterior elements with relatively normal anatomy anteriorly. These slips are less
likely to progress given the increased stability of the relatively normal anterior structures. A vertical sacrum and
hyperlordosis are not traditionally seen in patients with
low dysplastic slips.

Clinical Presentation
Most developmental spondylolisthesis presents in adolescence during the growth spurt. The disease course of
patients with low dysplastic slips may progress slowly;
consequently, these patients may not present until early
adulthood. Traditionally, patients present with severe
acute low back pain without neurologic findings. Some
patients present in a listhetic crisis (severe back pain,
hamstring spasm, and various neurologic deficits). Patients walk with a crouched gait because of severe hamstring tightness (Figure 5). Compensatory upper lumbar
hyperlordosis often creates a significant abdominal
crease (Figure 6). Neurologic symptoms range from isolated radiculopathy resulting from the stretch of exiting
nerve roots to cauda equina syndrome in patients with
high-grade slips and intact posterior arches. Because
these slips usually occur at the lumbosacral junction, the

559

Orthopaedic Knowledge Update 8

Spondylolysis-Spondylolisthesis

Figure 6 Photograph of a patient with the

combination of severe lumbosacral kyphosis
with compensatory upper lumbar hyperlordosis, which typically creates a prominent
abdominal crease in patients with highgrade spondylolisthesis.

L5 nerve roots are most commonly affected. Some

adults with high-grade slips may have subtle bowel and
bladder neurologic symptoms resulting from the slow
progression of slip over time and may warrant urologic
evaluation.

Treatment Algorithm
Treatment recommendations are based on the type of
slip (high dysplastic slips versus low dysplastic slips), patient age, neurologic status, slip severity, and the patients symptoms.

In Situ Fusion
Posterior in situ fusion has been a widely recommended
surgical treatment for patients with spondylolisthesis because of the low rates of neurologic injury and high
rates of clinical success. Several techniques have been
used for in situ fusion, including posterior, posterolateral, anterior interbody, and posterior interbody fusionall of which can be done with or without instrumentation. In situ fusion with decompression has been
shown to provide good results when neurologic symptoms are present preoperatively. Other studies show
that in situ fusion without decompression can be effective in treating back pain and radicular symptoms. In
one study in which eight high-grade slips were treated
with in situ fusion without decompression, all patients
achieved fusion and had good resolution of their back
and radicular symptoms. In the presence of progressing
neurologic deficits or cauda equina symptoms, decompression should be performed in addition to fusion. In
the absence of these neurologic symptoms, in situ fusion

560

without decompression may be considered.

The indication for instrumentation is not well defined. Good clinical results with noninstrumented in situ
fusion have been reported in several studies; however,
postoperative slip progression has been reported even
after successful arthrodesis. The addition of instrumentation may help increase fusion rates and decrease the
risk for postoperative slip progression; however, this has
not been definitively proven. One randomized prospective study comparing in situ fusions with and without instrumentation found that at 2-year follow-up there was
no significant difference in fusion rates and that pain
and functional disability were similar in the two groups.
The addition of interbody fusion may help increase
fusion rates and minimize progression of slippage postoperatively. The traditional ALIF and PLIF techniques
described earlier are not possible in patients with highgrade slips when in situ fusion is performed because of
the presence of significant translational and angular deformities. Interbody fusion using fibular struts placed
across the L5-S1 disk space through the bodies of L5
and S1 has been described (Figure 7). This technique
was first reported with the grafts placed from posterior
to anterior after a wide posterior decompression and
then augmented with a posterolateral fusion. Patients
were followed for 2 to 12 years, and all patients
achieved fusion and had significant or complete resolution of preoperative neurologic deficits. This technique
has also been described with placement of the grafts
from an anterior approach with similar results. No difference in outcomes has been reported when comparing
allograft and autogenous fibula.

Reduction
The indications for reduction of spondylolisthesis are
extremely controversial, and there are currently no
widely accepted reduction guidelines. Reduction of the
translational displacement and slip angle may occur independently, either partially or fully. The argument for
reduction is to restore sagittal alignment, reduce lumbosacral kyphosis and translation, and restore the normal lumbosacral biomechanics.
The main detraction of using reduction in the treatment of patients with high-grade slips is the risk of significant associated complications. Neurologic complications include L5 and S1 nerve root injuries, paresis,
paralysis, cauda equina syndrome, sensory deficits,
bowel/bladder dysfunction, and sexual dysfunction.
Other complications include nonunion, hardware failure, sacral fracture, dural tear, graft resorption, loss of
fixation, and prolonged immobilization or bed rest secondary to unstable fixation. Reported rates of neurologic deficits range from 8% to 30% after reduction of
high-grade slips. Although most of these deficits are
transitory, permanent deficits have been reported. High

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 45 Spondylolysis-Spondylolisthesis

Figure 7 A, Intraoperative photograph of the placement of a transsacral fibula posterior interbody graft. The cauda equina is protected with nerve root retractors as the
graft is placed across the L5-S1 disk space through the sacrum and L5 body from a
posterior approach. B, Postoperative sagittal reconstruction CT scan of a transsacral
fibular graft. C, Axial CT scan of transsacral fibular grafts in which the grafts are placed
bilaterally and purchase in both the S1 and L5 bodies is seen on the same CT cut
because of the severity of the slip.

rates of pseudarthrosis, progression of slippage, and

hardware failure have also been reported after reduction and posterolateral instrumented fusion. One study
reported implant failure in five of six patients who had
an isolated posterolateral instrumented fusion following
reduction. The addition of anterior interbody fusion after reduction and posterolateral instrumented fusion has
been shown to decrease these complications.
Recent improvements in spinal instrumentation have
increased the interest in the use of open reduction to
treat patients with spondylolisthesis. Several techniques
have been described. One technique involves pedicle
screw instrumentation from L4 to S1 or S2. A wide decompression is performed so that the nerve roots are directly visualized during reduction. Distraction and poste-

American Academy of Orthopaedic Surgeons

rior translation forces are used to gradually reduce the

slip. In the past, repeated wake-up tests were used
during the reduction. Electrophysiologic monitoring
(somatosensory-evoked potentials, electromyograms,
and motor-evoked potentials) should be used with this
procedure.
High rates of implant failure and loss of reduction
have been reported when reduction is performed and
only posterior instrumented fusion is used for stabilization. The reduction of the spondylolisthesis normalizes
the shear forces at the lumbosacral junction; however,
an anterior column defect is also created by the reduction. Posterior instrumentation may fatigue and fail, resulting in loss of reduction. Posterior distraction forces
do not adequately resist shear, and compression forces

561

Spondylolysis-Spondylolisthesis
applied through posterior instrumentation may result in
foraminal stenosis. The best way to biomechanically resist these shear forces is by providing anterior column
support in the form of interbody fusion to augment the
posterior instrumentation. If reduction is near anatomic,
traditional ALIF or PLIF procedures can be used for
anterior column support, depending on surgeon preference. Several studies have reported decreased nonunion
rates when either ALIF or PLIF procedures were combined with posterior instrumented open reduction. In
the treatment of patients with high-grade spondylolisthesis the benefits of reduction must be weighed against
the significant potential complications and these issues
should be discussed with the patient and family in great
detail.

Annotated Bibliography
Bartolozzi P, Sandri A, Cassini M, Ricci M: One-stage
posterior decompression-stabilization and transsacral interbody fusion after partial reduction for severe L5-S1
spondylolisthesis. Spine 2003;28:1135-1141.
This retrospective study suggests that posterior decompression with partial reduction and stabilization with pedicle
screw fixation and titanium cage transsacral interbody fusion
is a safe and effective treatment for patients with high-grade
spondylolisthesis.

Beutler WJ, Frederickson MD, Murtland A, Sweeney

MA, Grant WD, Baker D: The natural history of
spondylolysis and spondylolisthesis: 45-year follow-up
evaluation. Spine 2003;28:1027-1035.
This is a prospective study of the natural history of
spondylolysis and spondylolisthesis over 45 years. The authors
report that patients with pars defects follow a clinical course
similar to the general population and that slip progression
markedly slows with each decade.

Hanson DS, Bridwell KH, Rhee JM, Lenke LG: Correlation of pelvic incidence with low- and high-grade isthmic spondylolisthesis. Spine 2002;27:2026-2029.
This study shows that pelvic incidence is significantly
higher in low- and high-grade slips compared with control
groups. Pelvic incidence also has a significant correlation with
Meyerding-Newman scores.

Hanson DS, Bridwell KH, Rhee JM, Lenke LG: Dowel

fibular strut grafts for high-grade dysplastic isthmic
spondylolisthesis. Spine 2002;27:1982-1988.
This study shows that fibular strut grafting is a safe and effective method for treating patients with partially reduced
high-grade slips. The authors found no significant difference
between the use of allograft and autograft.

La Rosa G, Conti A, Cacciola F, et al: Pedicle screw fixation for isthmic spondylolisthesis: Does posterior lum-

562

Orthopaedic Knowledge Update 8

bar interbody fusion improve outcome over posterolateral fusion? J Neurosurg 2003;99(suppl 2):143-150.
Addition of a PLIF procedure to posterolateral instrumented fusion in patients with isthmic spondylolisthesis improves slip correction and maintenance of reduction over posterolateral instrumented fusion alone. The authors report no
difference in fusion rates or functional and neurologic outcome.

Natarajan RN, Garretson RB III, Biyani A, Lim TH,

Andersson GBJ, An HS: Effects of slip severity and
loading directions on the stability of isthmic spondylolisthesis: A finite element model study. Spine 2003;28:11031112.
This study shows that the stiffness of a spondylolisthetic
motion segment decreases with increasing slip progression.
The authors also report that lateral bending and torsion cause
the most resultant motion.

Classic Bibliography
Boos N, Marchesi D, Zuber K, Aebi M: Treatment of severe spondylolisthesis by reduction and pedicular fixation: A 4-6 year follow up study. Spine 1993;18:16551661.
Bradford DS, Boachie-Adjei O: Treatment of severe
spondylolisthesis by anterior and posterior reduction
and stabilization: A long-term follow-up study. J Bone
Joint Surg Am 1990;72:1060-1066.
Carragee EJ: Single-level posterolateral fusion, with or
without posterior decompression, for the treatment of
isthmic spondylolisthesis in adults. J Bone Joint Surg Am
1997;79:1175-1180.
Fischgrund JS, Mackay M, Herkowitz HN, Brower R,
Montgomery DM, Kurz LT: Degenerative lumbar
spondylolisthesis with spinal stenosis: A prospective,
randomized study comparing decompressive laminectomy and arthrodesis with and without spinal instrumentation. Spine 1997;22:2807-2812.
Herkowitz HN, Kurz LT: Degenerative spondylolisthesis
with spinal stenosis: A prospective study comparing decompression with decompression and intertransverse arthrodesis. J Bone Joint Surg Am 1991;73:802-808.
Katz JN, Lipson SJ, Chang LC, Levine SA, Fossel AH,
Liang MH: Seven- to 10-year outcome of decompressive
surgery for degenerative lumbar spinal stenosis. Spine
1996;21:92-98.
Marchetti PG, Bartolozzi P: Classification of spondylolisthesis as a guideline for treatment, in Bridwell KH,
DeWald RL (eds): Textbook of Spinal Surgery. Philadelphia, PA, Lippincott-Raven, 1997, pp 1211-1254.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Smith MD, Bohlman HH: Spondylolisthesis treated by a
single-stage operation combining decompression with in
situ posterolateral and anterior fusion: An analysis of
eleven patients who had long-term follow-up. J Bone
Joint Surg Am 1990;72:415-421.
Suk S, Lee C, Kim WJ, Lee JH, Cho KJ, Kim HG: Adding posterior lumbar interbody fusion to pedicle screw

American Academy of Orthopaedic Surgeons

Chapter 45 Spondylolysis-Spondylolisthesis
fixation and posterolateral fusion after decompression
in spondylolotic spondylolisthesis. Spine 1997;22:210219.
Wiltse LL, Newman PH, Macnab I: Classification of
spondylolysis and spondylolisthesis. Clin Orthop
1976;117:23-29.

563

Chapter

46

Adult Spinal Deformity

Bobby K-B Tay, MD

Introduction
The evaluation and management of adult patients with
spinal deformity has undergone a rapid evolution in the
past decade. Increasing knowledge about the natural
history of adult spinal deformity and the biologic events
that mediate the processes of spinal fusion have provided the spinal surgeon with a greater ability to evaluate and treat these complex disorders. Despite these advances, evaluation and optimal treatment of adult spinal
deformity remains a significant challenge. In addition,
the increasing number of technical options available to
the spinal surgeon as well as the many spinal instrumentation systems available today have not made management of adult spinal deformity any less challenging.
It is now well established that untreated scoliosis in
the adult is not a benign condition. Common associations include painful, degenerative, spinal osteoarthritis;
progressive deformity; spinal stenosis with radiculopathy; muscle fatigue from coronal and sagittal plane imbalance; and poor cosmesis. Accurate determination of
pain, impairment in the quality of life, and cosmetic effects of deformity are difficult to measure and compare
among groups of patients. These factors, when weighed
against the increased complication rate of surgical treatment, makes the decision to operate as important or
more important than the technique and expertise with
which the procedure is performed.

Prevalence and Natural History

Adult spinal deformity can arise as a sequela of untreated adolescent idiopathic scoliosis, failed surgical or
nonsurgical treatment, or de novo spinal deformity developing in the adult. The latter can be caused by degenerative changes of the spinal column leading to instability and deformity, by iatrogenic instability after
decompressive procedures for spinal stenosis and radiculopathy, and possibly by metabolic bone diseases such
as osteoporosis. The prevalence of scoliosis in adults has
been reported to range from 1.4% to 20%.
Clearly, scoliosis can continue to progress after skeletal maturity and into late adulthood. Large thoracic

American Academy of Orthopaedic Surgeons

curves (> 60) show the greatest risk for progression. In

a large cohort of patients with adolescent idiopathic
scoliosis that was followed for more than 40 years, 68%
of patients had progression of their curvature after skeletal maturity. Thoracic curves greater than 50 progressed an average of 1 per year. Thoracolumbar
curves progressed about 0.5 per year and lumbar
curves progressed 0.24 per year. Thoracic curves less
than 30 tended not to progress.
Patients with degenerative scoliosis exhibited a
higher rate of curve progression, which on average was
about 3.3 per year. The prevalence of lumbar scoliosis
in adult patients with low back pain was estimated to be
about 7.5%. The prevalence increased with age and included 15% of patients with low back pain who were
older than 60 years. Risk factors for curve progression
include the presence of a curvature greater than 30,
apical rotation greater than 33%, more than 6 mm of
lateral listhesis, and poor seating of L5 on S1.
In contrast to adolescent patients with scoliosis who
are usually asymptomatic, adult patients with scoliosis
often report back pain. In a recent European study involving 16,394 adolescents and their parents, the lifetime
prevalence of low back pain was 50.9% for boys and
69.3% for girls with scoliosis as an independent risk factor. Other smaller but better designed studies have demonstrated that patients with scoliosis have a higher incidence of back pain than age- and sex-matched controls.
In general, the overall incidence of pain ranges from
40% to 90%, and the overall prevalence of painful deformity ranges from 60% to 80%. The etiology of the
pain is multifactorial. It can arise from muscle fatigue
on the convexity of the curvature, trunk imbalance, facet
arthropathy on the concavity of the curvature, and degenerative disk disease. Back pain is more common in
patients with lumbar curves and in patients with thoracolumbar and lumbar curves exceeding 45 with apical
rotation and coronal imbalance. Although the incidence
of back pain in patients with adult scoliosis is similar to
that in the general population, many studies suggest that
back pain in patients with scoliosis is greater and more
persistent. In addition to back pain, adult patients are

565

Adult Spinal Deformity

more likely to experience symptoms of spinal stenosis
and radiculopathy from osteoarthritis of the spine, rotatory subluxations, anterior listhesis, and lateral listhesis
of the vertebral bodies that result in stretching and/or
impingement of nerve roots. Again, neurogenic symptoms are more common in patients with lumbar curvatures. However, paresis or paraplegia from untreated
scoliosis has not been reported.
Decreased vital capacity and other parameters of pulmonary function may occur in patients with severe thoracic curves (> 60), especially in the presence of thoracic
lordosis that effectively decreases the anterior-posterior
chest diameter. In one study, up to 35% of patients with
thoracic curves reported cardiopulmonary symptoms.
However, significant alterations in vital capacity as a result of restrictive lung disease have not been observed until the curve magnitude exceeds 90 to 100. In addition,
there is no evidence to suggest that an adult patient with
previously normal pulmonary function after skeletal maturity will experience deterioration in pulmonary function
with progression of the curvature without a preexisting
history of smoking or some other pulmonary disease.
Finally, the psychologic impact of chronic pain and
deformity must be considered. In a world where appearance and health become increasingly important for social acceptance and well-being, the inability of individuals to function at the level of their peers because of pain
or deformity can create a huge psychologic burden. The
influence of spinal deformity on psychologic well-being
has not been determined in the adult population. Small
series of patients with adult scoliosis have been reported
to show a significant negative impact of the spinal deformity on patients perception of health.

Patient Assessment
History and Physical Examination
A complete history followed by a careful physical examination (including a complete neurologic examination)
is essential in a patient with adult spinal deformity. The
importance of social and family history and occupational history cannot be overstated. Depression, substance abuse, and chronic smoking in patients with adult
spinal deformity can result in a less than ideal outcome
after major spinal reconstructive surgery. Overall, the
evaluation of the adult with scoliosis is much more difficult than that of adolescents because the usual criteria
for surgical treatment are more difficult to interpret. In
addition, the surgical treatment of scoliosis in adults carries a higher rate of complications and involves a longer
recovery period than in adolescents. Proper patient selection is critical for achieving a successful result.

Imaging Studies
Standard PA and lateral full-length spine radiographs
are an essential part of the evaluation process. Long cas-

566

Orthopaedic Knowledge Update 8

sette 14- 36-inch weight-bearing scoliosis radiographs
are necessary to fully assess the extent of the primary
curvature as well as any compensatory curves that may
exist. The weight-bearing radiographs also allow physicians to evaluate both coronal and sagittal plane balance. If surgery is contemplated, lateral supine bending
films are obtained to assess the flexibility of secondary
curvatures. Occasionally, flexion and extension views are
helpful to determine lumbar spine flexibility and the
presence of sagittal plane instability. These views have
been sufficient in most patients for preoperative planning. Traction views may be helpful for assessing severe
deformities, but in deformities less than 60, bending
films provide a better assessment of curve flexibility. A
fulcrum bending view (in which the PA radiograph is
taken with the patient in the lateral decubitus position
with the apex of the curvature over a large bolster) or a
push-prone radiograph (in which the pelvis of the prone
patient is stabilized while the surgeon applies lateral
pressure to the major curve) can help access thoracic
curve flexibility.
In general, curve magnitude and patient age are the
main predictors of curve flexibility. Every 10 increase
in curve magnitude over 40 results in a 10% decrease
in curve flexibility. Every 10-year increase in age decreases the flexibility of the structural curve by 5% and
the lumbosacral fractional curve by 10%.
MRI is reserved to evaluate rapidly progressive
curves and to determine the cause of any neurologic abnormalities on physical examination. Myelography is
helpful for evaluating the neural elements in patients
with severe rotational deformity and in those with metal
implants that may obscure the clarity of MRI scans. Diskography has been recommended by some authors as a
means of evaluating discogenic pain. This is especially
relevant in the evaluation of patients with low back pain
and lumbar scoliosis to help determine the distal extent
of fusion, especially in patients with disk degeneration
and axial low back pain. A positive diskogram in a patient with a degenerated disk below the anticipated fusion levels would be a relative indication to extend the
fusion to encompass the involved degenerated disk.
However, there are no prospective studies to confirm or
refute the efficacy of this rationale because the results
of provocative diskograms seem to depend to a great
extent on the interpretation and the technique of the
examiner.
Preoperative pulmonary function testing is useful for
patients with thoracic curves exceeding 70, patients
with a history of pulmonary disease (chronic smokers),
and patients who will require thoracoplasty. Adults who
undergo thoracoplasty experience a 27% decline in pulmonary function by 3 months postoperatively, which, in
contrast to adolescents, does not improve appreciably
after 2-year follow-up. Thus, an adult patient with bor-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

derline pulmonary function may not tolerate thoracoplasty or thoracotomy.

Treatment
Nonsurgical Treatment
The initial treatment of patients with back pain and
scoliosis should not differ from the treatment of patients
with mechanical back pain in the absence of deformity.
A relative indication for nonsurgical treatment includes
patients who physically cannot tolerate the amount of
surgery necessary to properly address their pain, deformity, and neurologic dysfunction. A physical therapy
program should be instituted to improve aerobic capacity, strengthen muscles, and improve flexibility and joint
motion. Although local heat application, analgesics, and
bracing all may aid in the amelioration of symptoms,
they do not prevent curve progression. Corticosteroid
injections in the form of nerve root blocks, facet injections, and epidural steroid injections may be of considerable value in the arsenal of conservative management of
adult spinal deformity. In patients who are not surgical
candidates, spinal bracing during ambulation may provide some symptomatic relief to improve functionality.

Surgical Treatment
The indications for surgery in the adult patient with
scoliosis include thoracic curve greater than 50 to 60,
with chronic pain that is unrelieved by conservative
management; significant loss of pulmonary function not
attributable to underlying pulmonary disease; documented curve progression with coronal or sagittal plane
imbalance; symptomatic deformity that is unacceptable
to the patient; and lumbar curvature with associated
back or radicular pain or symptoms of spinal stenosis.
Adult patients have a greater risk of experiencing
surgical complications than adolescents. Major complications include pseudarthrosis in 5% to 27% of patients,
residual pain in 5% to 15%, neurologic injury in 1% to
5%, infection in 0.5% to 5%, and thromboembolism in
1% to 20%.
To avoid the detrimental effects of prolonged immobilization, surgical procedures should be designed to
provide maximum stability and thus allow early mobilization with minimal external support. Combined procedures are preferable to staged procedures if this is technically and physiologically feasible. Combined anteriorposterior spinal reconstructive surgery has a lower
infection rate than staged procedures as a result of patient malnutrition at the time of the posterior procedure.
Normalization of nutritional status does not occur
until 6 to 12 weeks after the index procedure. If surgical
procedures need to be staged, the use of hyperalimentation or enteral nutritional supplementation between
stages is recommended to help decrease complications,

American Academy of Orthopaedic Surgeons

Chapter 46 Adult Spinal Deformity

which can result from malnutrition. Use of sequential
compression devices or foot pumps to minimize the risk
of deep venous thrombosis, mobilization with the assistance of physical therapy, and aggressive respiratory
therapy should be routine. Other useful adjuncts to surgery include an autologous blood exchange program,
the use of the Cell Saver autologous blood recovery system (Haemonetics, Braintree, MA) at the time of surgery, carefully monitored intraoperative hypotension,
routine neurologic monitoring, and an experienced anesthesia staff that can manage the blood loss and fluid
shifts that occur during complex spinal reconstruction.
Finally, redosing of antibiotics every 4 hours or after
1,500 mL of blood loss to maintain the antibiotic concentration above the minimum inhibitory concentration
is recommended.
It is important to remember that the amount of correction obtained in the adult patient is secondary to the
achievement of coronal and sagittal plane balance. Thoracic rib prominence is best corrected with thoracoplasty rather than with overcorrection of the thoracic
curve. The choice of fixation used to achieve correction
and balance is extremely important because osteoporosis is often present in the adult patient. Segmental fixation provides improved purchase in weakened bone and
creates a large area for force transmission in the correction of deformity.
The use of thoracic pedicle screws in the treatment
of traumatic conditions of the spinal column has become more popular as experience with the technique increases. The primary benefit of pedicle screw fixation is
the ability to shorten the length of fusion across the injured level while still having significant threedimensional control over the spinal deformity to be able
to reduce it. Pedicle screw fixation may also provide a
more secure anchor at the cephalad extent of the fusion
compared with that provided by hooks. The use of transpedicular fixation in the thoracic spine is still controversial because the presence of deformity makes insertion
of screws more difficult and hazardous. In addition, in
the presence of osteoporosis, the spinal lamina is often
the most secure point of fixation. In such instances, excellent correction of thoracic curves can be obtained
with a combination of laminar hooks and sublaminar
wires. In contrast to adolescent scoliosis, adult curvatures tend to be less flexible and adult patients tend to
have osteoporosis and less healing potential. Thus, pedicle screw fixation alone does not substitute for appropriate anterior column releases and fusion.
Intraoperative somatosensory-evoked potential and
motor-evoked potential monitoring are especially helpful when there is a possibility of extensive blood loss,
hypotension, significant deformity correction, and stiff
curvatures. When this type of monitoring is not available, a wake-up test should be performed after instrumentation and correction of the curvatures. Combined

567

Adult Spinal Deformity

somatosensory-evoked potential and motor-evoked potential monitoring is superior to single modality techniques in the assessment of real-time root and spinal
cord function. With the development of intercostal muscle and rectus abdominis intraoperative electromyographic monitoring, thoracic pedicle screws can be
tested to minimize the risk of perforation into the spinal
canal or the neural foramen.
Before any surgical procedure, both surgeon and patient must have realistic expectations for the outcome of
the operation. Most studies report a 69% to 95% reduction in the severity of pain and a 30% to 40% correction
of deformity.
In general, the approach to the surgical treatment of
adult spinal deformity is based on accepted methods of
choosing fusion levels, placement of instrumentation,
and curve reduction techniques that were derived from
experience in the treatment of adolescent spinal deformities. These techniques are then modified to take into
consideration the various challenges presented by adult
patients. These surgical challenges include the presence
of spinal stenosis, stiffer deformities, sagittal plane imbalance, less potential for healing, osteoporosis, the need
to extend the fusion to the sacrum because of stiff fractional curvatures in the lower lumbar spine or the presence of spinal stenosis, and discogenic pain. In addition,
an anatomic correction of the spinal deformity is not often achieved or warranted in the adult patient. The goal
of surgical correction is to obtain a balanced spine in
the sagittal and coronal planes.
Fusion levels are often determined from the weightbearing PA radiographs, with modifications determined
by lateral bending and weight-bearing lateral radiographs. Some general guidelines exist. All of the structural curves should be included in the fusion. The end
vertebra of the fusion should be within the Harrington
stable zone and should be neutrally rotated when the
deformity corrects on lateral bending. The first unfused
disk space at the caudal end of the fusion should be
flexible on lateral bending. The fusion should not be
stopped within the apex of the thoracic kyphosis or adjacent to a kyphotic motion segment.
Curves less than 60 to 70 on the Cobb measurement without significant coronal or sagittal plane decompensation can be treated effectively using a single
posterior approach. In patients with this type of spinal
deformity, radical facetectomies and osteotomies of
ankylosed segments will allow sufficient mobility to
achieve curve correction and balance. Curvatures
greater than 70 are often associated with coronal and
sagittal plane decompensation. These deformities are
best treated with a combined anterior-posterior approach. Anterior releases and fusions of the most structural portions of the curvature followed by posterior releases and osteotomies with or without rib resections
will, in most patients, allow sufficient mobility of the de-

568

Orthopaedic Knowledge Update 8

formity to obtain a balanced correction of the curvature.
Curvatures that exceed 90 to 100 are often associated
with rigid ankylosis of the spine and severe coronal imbalance. In patients with this degree of deformity, a
three-stage posterior-anterior-posterior procedure or a
vertebral resection procedure is required to gain sufficient spinal mobility and to gain sufficient curve correction to obtain spinal balance.
Patients with spinal stenosis should have decompression of the stenotic levels before instrumentation and
correction of the curvatures. Patients with severe lumbar spinal stenosis who also have decompensation in the
sagittal plane can be treated with staged posterior and
anterior surgery. In these patients, decompression of the
stenotic levels should be performed before restoration
of sagittal balance. Restoration of lordosis in these patients before decompression of the stenotic levels may
result in a neurologic deficit.
Once the spinal deformity is corrected with the appropriate release and instrumentation, a meticulous spinal fusion with bone grafting should be performed.Autogenous
graft from the local fusion bed and the iliac crest are
placed over the decorticated posterior elements.

Degenerative Lumbar Scoliosis

Degenerative lumbar scoliosis can result from untreated
idiopathic scoliosis, but it more frequently occurs independently as a sequela of the aging process in combination with osteoporosis. Many patients with degenerative
lumbar scoliosis have a combination of both mechanical
back pain and spinal stenosis. Patients may also present
with radiculopathy because the neural foramen are narrowed at the concavity of the spinal deformity. Fortunately, most of these patients are well balanced in the
sagittal and coronal planes despite the magnitude of the
curvature, and their main symptoms are the result of
neural compression. Decompression of the stenotic spinal canal and neural foramen often results in significant
improvement in pain and function in these patients. Unless the decompressed segments are severely ankylosed,
these levels should be fused to prevent progression of
the scoliosis (Figure 1). The entire deformity does not
need to be included in the fusion as long as the patient
is balanced in the sagittal and coronal planes. However,
fusion levels should be chosen so as not to end the fusion at a listhetic level, the apex of the curvature, or a
kyphotic segment. If instrumentation is used and the patient still has lumbar lordosis, mild distraction between
the pedicle screws at the concavity of the curvature
helps to enlarge and stabilize the neuroforamen.
Patients with long fusions to the sacrum should also
have selective anterior fusion at the L4-5 and L5-S1 segments, with structural grafting to maximize fusion rates
and decrease the possibility of failure of the sacral
screws.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 46 Adult Spinal Deformity

Figure 1 PA (A) and lateral (B) weight-bearing radiographs of the lumbar spine of a 50-year-old woman with degenerative scoliosis and spinal stenosis who was treated with
lumbar laminectomy radical facetectomies, instrumentation with correction of scoliosis, and spinal fusion. Postoperative PA (C) and lateral (D) weight-bearing radiographs of the
same patient after decompression and fusion.
In general, arthrodesis to the sacrum for idiopathic
scoliosis should be avoided if possible. Long fusions to
the sacrum are associated with a higher rate of pseudarthrosis, fixed sagittal deformity, instrumentation failure,
and limited function. Indications to extend the fusion to
the sacrum include lumbosacral pain secondary to degenerative disk disease below a lumbar curvature when
a decision has been made to correct the lumbar curve,
an unbalanced lumbosacral curvature with lumbar scoliosis for which balance in the lumbosacral curve is not
achieved (as assessed using appropriate side bending radiographs), and the presence of substantial degeneration of the motion segments of L4-5 and L5-S1 anteriorly and posteriorly. In selected patients with a fixed
lumbosacral fractional curve, fusion to the sacrum can
be avoided by an end plate osteotomy at L4 or L5 (with
concave osteophyte excision) to make the end vertebra
horizontal, reduce the fractional curve, and create a stable end vertebra above the pelvis.
When performing long fusions to the sacrum in
adults, a combined approach is recommended to maximize the fusion rate, reestablish lumbar lordosis, and
prevent implant failure across the lumbosacral junction.
Two-stage surgery is preferable with the anterior diskectomies and fusion, with structural grafts or cages performed first followed by posterior fusion and instrumentation.
Many techniques are available to secure fixation
across the lumbosacral junction. These include sacral
screws placed in a bicortical fashion, sacral screws with
intrasacral rods, iliac wing screws, Galveston technique,
and convergent and divergent sacral screws. Bicortical
sacral screw fixation with structural anterior fusion at
L5-S1 appears to be an adequate anchor for most long
fusions to the sacrum. In patients with poor bone stock

American Academy of Orthopaedic Surgeons

or in those in whom previously placed sacral screws

have failed, iliac fixation is usually necessary. In the
adult patient, Luque-Galveston fixation is associated
with a high rate of pseudarthrosis (30% to 40%). Although iliac fixation with iliac screws is biomechanically
strong, iliac screws can be prominent in very thin individuals (Figure 2).

Salvage Procedures in Adult Patients

The indications for revision scoliosis surgery in the adult
include painful pseudarthrosis, progressive deformity
from pseudarthrosis or inadequate fusion levels, flatback syndrome from distraction implants across the
lumbar spine, symptomatic adjacent segment degeneration or vertebral body fracture, and unacceptable residual deformity.
The goals of surgery are the same as in primary surgeryto obtain a solid arthrodesis, achieve threedimensional spinal balance, and provide rigid internal
fixation. The management of pseudarthrosis in the adult
patient with scoliosis depends on the level of the pseudarthrosis, the presence or absence of deformity, the security of preexisting fixation, and especially the presence or absence of pain. A painful pseudarthrosis
associated with progressive deformity and loss of fixation is a clear indication for spinal revision surgery via a
combined approach. However, thoracic pseudarthrosis
that is not associated with deformity or loss of fixation
can be adequately managed by single-stage anterior and
posterior repair with copious autogenous bone grafting.
Adult patients may present with curve progression
above or below a previous fusion that was done in childhood or adolescence. This condition may be secondary
to crankshaft phenomena or the primary fusion being
too short. For minor degrees of deformity in young

569

Adult Spinal Deformity

Orthopaedic Knowledge Update 8

Figure 2 Preoperative PA (A) and lateral (B) weight-bearing radiographs of an adult patient with scoliosis who was treated with staged anterior and posterior fusion with
posterior fixation to the pelvis using iliac screw fixation. Postoperative PA (C) and lateral (D) weight-bearing radiographs of the same patient.

adults with good bone stock, extension of the fusion to

include the involved segments and reinstrumentation
are sufficient treatment. However, in patients with a
fixed deformity with coronal or sagittal imbalance, especially in the presence of osteopenia, a combined approach is preferable.
Disk degeneration may develop below a fusion and
require salvage reconstruction. If there is isolated spinal
stenosis, decompression alone is usually sufficient. If the
patient has a vertebral insufficiency fracture below the
fusion, extension of the fusion and instrumentation is
necessary.
Loss of lumbar lordosis occurs gradually with aging.
Iatrogenic loss of lumbar lordosis is typically caused by
the use of distraction implants across the lumbar spine.
Some patients who undergo salvage procedures experience loss of lumbar lordosis secondary to sagittal plane
imbalance. In these patients, the primary goal of the index procedure was coronal plane correction. These patients are best treated with a combined surgical procedure. For patients with a solid arthrodesis, a pedicle
subtraction osteotomy with vertebral body decancellation using the posterior approach and rigid internal fixation can provide up to 30 to 40 of sagittal plane correction. This procedure is particularly useful in patients
with combined anterior and posterior fusion who still
have flat-back deformity (Figure 3). A poorer clinical

570

result is associated with increasing patient comorbidities, thoracic pseudarthrosis, and adjacent segment
breakdown caudad to the fusion. To restore lumbar lordosis in patients with multiple pseudarthrosis or fusions
that are not intact to the sacrum, a combined approach
is preferable. This may be performed as a first-stage posterior procedure with osteotomies and instrumentation
to the sacrum followed by anterior interbody fusion
with structural allograft or vice versa (Figure 4). For severe, rigid, unbalanced deformity, a spinal shortening
procedure such as a vertebral body resection is necessary.
Postoperatively, adult patients are mobilized within
24 to 48 hours depending on pain tolerance and overall
general condition. Then the patient is given intravenous
narcotic medication via a patient-controlled pump until
oral pain medication is tolerated. The use of postoperative anti-inflammatory agents such as ketorolac is contraindicated because the fusion rate is adversely affected. Perioperative antibiotics can be continued for 36
to 48 hours. Antiembolic stockings and sequential compression devices are used to minimize the incidence of
venous thrombosis until the patients are ambulatory.
Anticoagulation is not performed routinely, but it may
be considered in high-risk patients and in patients with
a preexisting history of thromboembolic disease. In patients with a history of pulmonary embolism, a vena

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 46 Adult Spinal Deformity

Figure 3 PA (A) and lateral (B) full-length weight-bearing radiographs of a 56-year-old woman with lumbar flat-back syndrome following lumbar decompression and instrumented fusion. The patient had severe sagittal plane decompensation and recurrent spinal stenosis with back pain and bilateral buttock and leg radiculopathy. Note that the
lateral radiograph demonstrates evidence of severe sagittal plane decompensation. Postoperative PA (C) and lateral (D) full-length weight-bearing radiographs of the same
patient after revision laminectomy and transpedicular wedge resection at L3 with restoration of sagittal balance.
cava filter may be considered before surgical intervention. The patients are fitted with a lightweight plastic
orthosis within 5 to 7 days of surgery and are instructed
to wear the brace full time, except when in bed and for
hygiene.

Controversial Issues in Managing Adult

Patients With Scoliosis
The absolute indications for combined surgery versus
posterior fusion and instrumentation alone are unclear.
Combined surgery allows for better correction and reestablishment of physiologic lumbar lordosis through the
use of anterior structural grafting, and it increases the
probability of successful fusion. In patients with complex deformities involving the lumbar spine in which fusion across the lumbosacral junction is necessary to
achieve spinal balance, combined procedures are preferable because 15% to 20% of patients who undergo posterior instrumentation and fusion have pseudarthrosis,
even when the stiffer third-generation instrumentation
systems are used.
Thoracoplasty is a useful adjunct in treating patients
with spinal deformity. Thoracoplasty reduces rib prominence by 71% compared with only a 17% reduction in
control subjects. Rib resection also provides a good

American Academy of Orthopaedic Surgeons

source of bone graft. It is well known that pulmonary

function can deteriorate after thoracotomy and thoracoplasty. These procedures should be used with extreme
caution in patients with severely compromised pulmonary function. Unlike adolescents, adult patients with
scoliosis do not recover lost pulmonary function, even at
2-year follow-up.

Results and Complications

The results of surgical correction of spinal deformity in
adults have improved significantly over the past decade.
The rate of complications ranges from 12% to 23% of
patients. Pseudarthrosis rates have varied between 5%
to 27% of patients, with higher rates of pseudarthrosis
occurring after revision surgery and after using nonsegmental posterior implants in distraction mode. Pain in
these patients is seldom totally alleviated by surgical
correction. However, reports of residual pain after complex spinal reconstructive surgery vary between 5% and
25% of patients. Approximately 61% of patients report
improved sleep patterns after surgical treatment, and
57% are able to return to work. Mortality remains low
in patients who undergo complex spinal reconstructive
surgery, but it is not insignificant (< 1% to 5% of patients). Overall, patient outcomes as assessed by the

571

Adult Spinal Deformity

Orthopaedic Knowledge Update 8

Figure 4 Preoperative PA (A) and lateral (B) full-length weight-bearing radiographs of a 50-year-old woman with flat-back syndrome and pseudarthrosis after posterior lumbar
interbody fusion with threaded cages at L4-5 and instrumented posterolateral fusion at L5-S1. Note that the lateral radiograph shows evidence of significant sagittal plane
decompensation. Postoperative PA (C) and lateral (D) radiographs of the same patient after multiple level Smith-Peterson osteotomies and instrumented fusion and repair of
pseudarthrosis. Note that the postoperative lateral radiograph shows evidence of restoration of lumbar lordosis.
Medical Outcomes Study Short Form 36-Item Health
Survey (SF-36) and the American Academy of Orthopaedic Surgeons Modems Instrument are significantly
associated with radiographic correction of the lumbar
lordosis to greater than 25 and coronal plumb alignment to within 2.5 cm. Patients with primary degenerative lumbar scoliosis as well as those who have had long
fusions to L4, L5, or the sacrum have improved gait parameters (both speed and endurance) after spinal decompression and fusion, with restoration of coronal and
sagittal balance. Outcomes assessment in this challenging group of patients is necessary to determine the impact of deformity correction on the patients quality of
life and daily function. The Modified Scoliosis Research
Society Outcomes Instrument for Adult Deformity, the
Oswestry Disability Index, and the SF-36 are the most
commonly used validated outcomes questionnaires for
adult patients undergoing spinal deformity surgery.
Neurologic injury occurs in fewer than 1% to 5% of
patients who undergo surgical treatment for spinal deformities. Significant risk factors for major intraoperative neurologic deficits include combined anterior and
posterior surgery and hyperkyphosis. Delayed postoperative paraplegia is another devastating complication after extensive spinal reconstructive surgery, and it can occur several hours after the completion of the procedure.
This phenomenon has been attributed to ischemia of
the spinal cord from postoperative hypovolemia, mechanical tension of spinal blood vessels on the concavity

572

of the curve, and preexisting atherosclerosis. Thus,

proper postoperative volume support and vigilance are
essential. Loss of vision as a complication of spinal reconstructive surgery is a rare but often permanent event
that can occur as the result of ischemic optic neuropathy, retinal artery occlusion, or cerebral ischemia. The
incidence of pulmonary embolism varies from 1% to
20% of patients, depending on the series. Air embolism
can occur in patients with patent foramen ovale. However, only a few isolated instances of this complication
have been reported. Flail chest can occur with overaggressive thoracoplasty, especially in revision situations.
Infection is a relatively rare event in patients who
undergo surgery for spinal deformities; it occurs in between 1% and 8% of patients. In addition, infection after anterior surgery alone occurs in about 1% of patients. However, the sequelae of deep postoperative
infection are substantial, with up to 38% to 64% of patients having pseudarthrosis at 37-month follow-up. Delayed infections can occur more than 2 years postoperatively, typically with low virulent organisms.
If coronal and sagittal balance is achieved and maintained with a solid fusion, the outcomes are generally
excellent. Pulmonary improvement after correction of
spinal deformity is unlikely to occur in the adult and the
use of thoracoplasty may actually reduce pulmonary
function. The correction of curvature varies from 30%
to 60% and depends on the nature and flexibility of the
patients preoperative curve and the technique used for

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

correction. Despite these relatively modest gains, satisfaction with surgical correction is generally high and can
reach up to 90% of patients.

Annotated Bibliography
Prevalence and Natural History
Kovacs FM, Gestoso M, Gil del Real MT, Lopez J, Mufraggi N, Mendez JI: Risks factors for non-specific low
back pain in school children and their parents: A population based study. Pain 2003;103:259-268.
In this European study involving 16,394 adolescents and
their parents, the lifetime prevalence of low back pain was
50.9% for males and 69.3% for females. Scoliosis was an independent risk factor.

Schwab F, Dubey A, Pagala M, et al: Adult scoliosis: A

health assessment analysis by SF-36. Spine 2003;28:602606.
This prospective self-assessment of a consecutive series of
adult patients with scoliosis showed that scoliosis has a significant impact on patients perception of health.

Weinstein SL, Dolan LA, Spratt KF, Peterson KK,

Spoonamore MJ, Ponseti IV: Health and function of patients with untreated idiopathic scoliosis: A 50-year natural history study. JAMA 2003;289:559-567.
This prospective cohort analysis of 117 untreated patients
with idiopathic scoliosis followed over a 50-year span showed
a greater incidence for back pain and cosmetic deformity in
patients with scoliosis compared with 62 control patients. Patients with thoracic curves greater than 80 were more likely
to report shortness of breath, suggesting a deterioration of
pulmonary function in this subgroup.

Patient Assessment
Davis BJ, Gadgil A, Trivedi J: Ahmed el NB: Traction radiography performed under general anesthetic: A new
technique for assessing idiopathic scoliosis curves. Spine
2004;29:2466-2470.
Traction radiographs taken with the patient under general
anesthesia allow a better assessment of curve flexibility than
standard supine bending radiographs. This finding may obviate
the need for anterior release and fusion in select patients.
Comparison to fulcrum bending radiographs was not performed in this study.

Treatment
Ali R, Boachie-Adjei O, Rawlins BA: Functional and radiographic outcomes after surgery for adult scoliosis using third-generation instrumentation techniques. Spine
2003;28:1163-1169.
This retrospective radiographic and chart review of 28 patients with adult scoliosis treated with primary corrective surgery showed significant clinical and radiographic improvements using third-generation spinal implants.

American Academy of Orthopaedic Surgeons

Chapter 46 Adult Spinal Deformity

Suk S, Kim WJ, Lee SM, et al: Thoracic pedicle screw
fixation in spinal deformities: Are they really safe?
Spine 2001;26:2049-2057.
In this study, 4,604 thoracic pedicle screws were placed in
the treatment of adult patients with spinal deformity; the authors report an overall rate of screw malposition of 1.5%.
Most thoracic pedicle screws were placed inferior or lateral to
the pedicle. Screw-related complications occurred in 0.8% of
patients. Deformity correction was 69.9% for patients with
coronal plane deformities and 47 for those with kyphosis.

Voos K, Boachie-Adeji O, Rawlins BA: Multiple vertebral osteotomies in the treatment of rigid adult spinal
deformities. Spine 2001;26:526-533.
This retrospective chart and radiographic review of 27
consecutive adult patients with spinal deformity demonstrated
the efficacy of multiple vertebral osteotomies in the management of rigid adult spinal deformities. The authors reported
that the average scoliosis correction for these patients was
40% and the average correction in sagittal balance was 6.5 cm.

Degenerative Lumbar Scoliosis

Berven S, Hu SS, Deverin V: Lumbar end plate osteotomy in adult patients with scoliosis. Clin Orthop 2003;
411:70-76.
The authors of this study discuss a technique of end plate
osteotomy and concave osteophyte excision in adults with
fixed lumbosacral fractional curves that avoids extension of
fusion to the sacrum.

Bridwell KH, Edwards CC II, Lenke LG: The pros and

cons to saving the L5-S1 motion segment in a long scoliosis fusion construct. Spine 2003;28:S234-S242.
This article presents a review of the literature and the authors personal experiences concerning the advantages and
disadvantages of extending a long scoliosis fusion to the
sacrum. The subject is controversial and no good studies exist
that definitively answer the question.

Tribus CB: Degenerative lumbar scoliosis: Evaluation

and management. J Am Acad Orthop Surg 2003;11:174183.
This article presents a review of evaluation and treatment
of degenerative scoliosis.

Salvage Procedures in Adult Patients

Berven SH, Deviren V, Smith JA, Hu SH, Bradford DS:
Management of fixed sagittal plane deformity: Outcome
of combined anterior and posterior surgery. Spine 2003;
28:1710-1715.
Twenty-five patients with fixed sagittal imbalances were
treated with combined anterior-posterior fusion. This study
showed that combined surgery is an effective method to treat
fixed sagittal deformity. Patients with preoperative hypolordosis (relative flat-back syndrome) who had good restoration of
lumbar lordosis and sagittal plane balance had the best outcomes.

573

Adult Spinal Deformity

Bridwell K, Lewis S, Lenke LG, et al: Pedicle subtraction osteotomy for the treatment of fixed sagittal imbalance. J Bone Joint Surg Am 2003;85-A:454-463.
In this study, 27 consecutive patients with fixed sagittal imbalance and treated with pedicle subtraction osteotomy were
evaluated using Oswestry and Modified Scoliosis Research Society Outcomes Instrument for Adult Deformity questionnaires. The authors report that pedicle subtraction osteotomy
is an effective procedure in this group of patients, with significant improvements in overall Oswestry scores and decreased
pain scores. A poorer clinical result was associated with increasing patient comorbidities, pseudarthrosis in the thoracic
spine, and subsequent breakdown caudad to the fusion.

Results and Complications

Eck KR, Bridwell KH, Ungacta FF, et al: Complications
and results of long adult deformity fusions down to L4,
L5, and the sacrum. Spine 2001;26:E182-E192.
The authors of this study report the outcomes and complication rates of patients treated surgically for adult lumbar
scoliosis at an average 5-year follow-up (minimum 2-year
follow-up). They report that patients with fusions short of the
sacrum who developed distal spinal degeneration had worse
outcomes that those patients with fusion to the sacrum.

Emami A, Deverin V, Berven S, et al: Outcome and

complications of long fusions to the sacrum in adult spinal deformity. Spine 2002;27:776-786.
This is a retrospective study of 54 adult patients with spinal deformities who were treated with different techniques to
achieve long fusion to the sacrum. All patients had combined
anterior and posterior reconstruction with a minimum 2-year
follow-up. The authors report a 34% pseudarthrosis rate in patients who had Luque-Galveston fixation compared with bicortical sacral screw and iliac screw fixation.

Lapp MA, Bridwell KH, Lenke LG, et al: Long-term

complications in adult deformity patients having combined surgery: A comparison of primary to revision patients. Spine 2001;26:973-983.
This outcomes analysis compares the complication rates of
primary surgery and revision surgery in 44 consecutive patients who underwent combined procedures for correction of
adult spinal deformity with a minimum 2-year follow-up. The
results demonstrated similar rates of minor complications
(22% to 23% of patients). The primary surgery group had a
higher rate of major complications and a significantly higher
rate of pseudarthrosis compared with the revision surgery
group.

Parsch D, Gaertner V, Brocai DR, Carstens C: The effect

of spinal fusion on the long-term outcome of idiopathic
scoliosis: A case control study. J Bone Joint Surg Br
2001;83:1133-1136.
In this study, 68 patients with scoliosis (34 treated and 34
untreated) compared with age- and sex-matched controls were

574

Orthopaedic Knowledge Update 8

evaluated for severity of back pain and overall function. Functional scores for patients with scoliosis were lower than those
for the control group. Frequency and severity of back pain
scores were lower in patients treated with fusion than in the
patients with untreated scoliosis.

Classic Bibliography
Albert TJ, Purtill J, Mesa J, McIntosh T, Balderston RA:
Health outcome assessment before and after adult deformity surgery: A prospective study. Spine 1995;20:
2002-2005.
Ascani E, Bartolozzi P, Logroscino CA, et al: Natural
history of untreated idiopathic scoliosis after skeletal
maturity. Spine 1986;11:784-789.
Boachie-Adjei O, Bradford DS: Vertebral column resection and arthrodesis for complex spinal deformities.
J Spinal Disord 1991;4:193-202.
Bradford DS, Tribus CB: Vertebral column resection for
the treatment of rigid coronal decompensation. Spine
1997;22:1590-1599.
Collis DK, Ponseti IV: Long-term follow-up of patients
with idiopathic scoliosis not treated surgically. J Bone
Joint Surg Am 1969;51:425-445.
Dickson JH, Mirkovic S, Noble PC, Nalty T, Erwin WD:
Results of operative treatment of idiopathic scoliosis in
adults. J Bone Joint Surg Am 1995;77:513-523.
Goldberg MS, Mayo NE, Poitras B, Scott S, Hanley J:
The Ste-Justine Adolescent Idiopathic Scoliosis Cohort
Study: Part II. Perception of health, self and body image, and participation in physical activities. Spine 1994;
19:1562-1572.
Hu SS, Fontaine F, Kelly B, Bradford DS: Nutritional
depletion in staged spinal reconstructive surgery: The effect of total parenteral nutrition. Spine 1998;23:14011405.
Hu SS, Holly EA, Lele C, et al: Patient outcomes after
spinal reconstructive surgery in patients > or = 40 years
of age. J Spinal Disord 1996;9:460-469.
Jackson RP, Simmons EH, Stripinis D: Incidence and severity of back pain in adult idiopathic scoliosis. Spine
1983;8:749-756.
Lenke LG, Bridwell KH, Blanke K, Baldus C: Analysis
of pulmonary function and chest cage dimension
changes after thoracoplasty in idiopathic scoliosis. Spine
1995;20:1343-1350.
Nachemson A: Adult scoliosis and back pain. Spine
1979;4:513-517.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Thomasen E: Vertebral osteotomy for correction of kyphosis in ankylosing spondylitis. Clin Orthop 1985;194:
142-152.

Chapter 46 Adult Spinal Deformity

Weinstein SL, Ponseti IV: Curve progression in idiopathic scoliosis. J Bone Joint Surg Am 1983;65:447-455.

Vanderpool DW, James JI, Wynne-Davies R: Scoliosis in

the elderly. J Bone Joint Surg Am 1969;51:446-455.

American Academy of Orthopaedic Surgeons

575

Chapter

47

Spinal Infections
Eric S. Wieser, MD
Jeffrey C. Wang, MD

Introduction
Historically, patients with spinal infections experienced
poor results with high morbidity and mortality rates.
Medical innovations over the past several decades, including the improvement of antimicrobial chemotherapy, powerful laboratory and imaging techniques, and
advancements in surgical techniques have significantly
improved the outcomes of patients with spinal infections. Earlier detection of vertebral osteomyelitis and
diskitis through increased clinician awareness and the
use of advanced imaging modalities are crucial in limiting or avoiding complications associated with progressive spinal infection such as epidural abscess, structural
deformities, chronic osteomyelitis, paralysis, sepsis, and
death.

Pyogenic Vertebral Osteomyelitis

Epidemiology and Pathogenesis
The demographic features of patients with vertebral osteomyelitis have changed over the past years because of
medical advances and the aging of the population. Vertebral osteomyelitis affects approximately 2% to 7% of
all patients with pyogenic osteomyelitis in developed
countries. Despite an overall decrease in the incidence
of tuberculous spondylitis, the incidence of pyogenic
vertebral osteomyelitis appears to have increased. This
increase is probably related to the growth in the elderly
and immunocompromised populations, increases in the
number of invasive medical procedures, and an increase
in intravenous drug use. Risk factors for pyogenic vertebral osteomyelitis that have recently been described in
the literature are listed in Table 1.
Several theories exist regarding the mechanism of
bacterial seeding of the spine. Pyogenic vertebral osteomyelitis is a bacterial infection with several possible
sources including direct inoculation, contiguous spread
from local infection, or hematogenous seeding. Direct
inoculation can occur after penetrating trauma, open
spinal fracture, and following percutaneous or open spinal procedures (such as diskography, diskectomy,
chemonucleolysis, or any other diagnostic or therapeutic

American Academy of Orthopaedic Surgeons

injection). Contiguous spread of infection to the spine is

most commonly associated with retropharyngeal and
retroperitoneal abscesses. Any condition that causes a
transient bacteremia may ultimately lead to hematogenous vertebral osteomyelitis.
The pathogenesis of spinal infection in children differs significantly from that of adults because of the differences in the vascular anatomy of the vertebrae and
intervertebral disks. Isolated diskitis primarily occurs in
children and vertebral osteomyelitis is common in adulthood. In children, vascularity extends through the cartilaginous growth plate into the nucleus pulposus, allowing direct deposition of bacterial emboli into the center
of the disk. However, in adults, blood vessels reach only
to the anulus fibrosus, limiting deposition of bacteria to
the vertebral body metaphysis and end plate.
Hematogenous seeding of the vertebrae is described
by two major theories. First, Batson described retrograde flow through the valveless vertebral venous
plexus within the vertebral metaphyseal region where
bacterial seeding could occur under increased abdominal pressure. This theory was refuted with another theory that proposed that bacteria can become wedged in
the end metaphyseal arteriole loops. Infection spreads
after the microorganisms are lodged in the low-flow vascular loops in the metaphysis and at the end plates. After the infection is established near the end plate of one
vertebral body, it can penetrate through the end plate
into the adjoining disk. The avascular disk material is
destroyed by bacterial enzymes, allowing the infection
to spread to the adjacent vertebral body. Posterior spinal arteries have abundant anastomoses about the disk,
which probably contributes to the spread of the infection from one level to the next.
Staphylococcus aureus remains the most common organism cultured from pyogenic vertebral osteomyelitis
and is found in 50% to 65% of culture positive cases.
S aureus accounts for more than 80% of pediatric spinal
infections. There also has been an increase in the number of gram-negative bacillus spinal infections over the
past decade. The genitourinary tract, respiratory tract,
and soft-tissue infections are often the source of these

577

Spinal Infections

Table 1 | Risk Factors for Pyogenic Vertebral Osteomyelitis

Male sex
Intravenous drug abuse
Diabetes mellitus
Urinary tract infection
Respiratory tract infection
Recent genitourinary procedure
Previous spinal procedure
Morbid obesity
Alcohol abuse
Human immunodeficiency virus/acquired immunodeficiency syndrome
Malignancy
Corticosteroid use
Penetrating trauma

gram-negative infections. Escherichia coli, Pseudomonas, and Proteus infections often occur after genitourinary infections or procedures. Pseudomonas infections
are also often seen in intravenous drug users. From intestinal flora, Salmonella can cause vertebral osteomyelitis in children with sickle cell disease. Enterococcus,
Propionobacterium acnes, Streptococcus viridans, Staphylococcus epidermidis and diphtheroids have all been
causative organisms of pyogenic vertebral osteomyelitis.

Clinical Characteristics
The clinical presentation of vertebral osteomyelitis is
highly variable depending on the location of infection,
the virulence of the organism, and the immunocompetency of the host. Pyogenic infections of the spine occur
50% of the time in the lumbar spine followed by approximately 40% in the thoracic spine, and only about
10% of the time in the cervical spine. Fever and constitutional symptoms are present in approximately 50% of
the affected population. Weight loss is common but
rarely recognized by the patient.
Approximately 90% of patients with pyogenic infections will have back or neck pain that is often quite severe and insidious in onset, which accounts for the frequent delay in diagnosis. Because the pain is often
present at rest and at night, there is concern for the differential diagnosis of potential malignancy. Muscle
spasms often are associated with neck or back pain. Torticollis and dysphagia often accompany fever as the only
symptoms of cervical infection. Lumbar infection can
lead to loss of lumbar lordosis, hamstring tightness, hip
flexion contracture, or a positive straight leg raising test.
Neurologic deficits are present in about 10% of patients secondary to nerve root or spinal cord compression, especially with cervical or thoracic level disease.
Patients may report radicular pain, motor nerve root
paresis, and paralysis. Risk factors predisposing patients

578

Orthopaedic Knowledge Update 8

Table 2 | Risk Factors for Neurologic Deterioration With

Vertebral Osteomyelitis
Diabetes mellitus
Rheumatoid arthritis
Systemic corticosteroid use
Advanced age
Cephalad level of infection (high thoracic or cervical)
Staphylococcal infection

to paralysis include diabetes, rheumatoid arthritis, systemic steroid use, increasing age, Staphylococcus infection, and a more cephalad level of infection. The risk
factors for neurologic deterioration with vertebral osteomyelitis are summarized in Table 2.

Laboratory Evaluation
Laboratory studies often support the diagnosis of infection but remain nonspecific. The white blood cell count
is elevated in about 50% of patients. The erythrocyte
sedimentation rate (ESR) is a much more sensitive test
and is elevated in more than 90% of patients; however,
its specificity for infection is poor. All patients should
have a C-reactive protein (CRP) test, which is slightly
more sensitive and specific than the ESR. The CRP is
also elevated sooner than the ESR. Both tests are helpful in following the course of treatment of the infection.
However, each of these markers of inflammation will be
elevated following an invasive procedure without any
infection present. A substantial decrease in the ESR and
CRP suggests an adequate response to treatment.
The definitive diagnosis of spinal pyogenic osteomyelitis requires identification of the organism through either a positive blood culture with confirmatory clinical
and imaging features or from a biopsy and culture of
the infected site. Blood and urine cultures should be
done on all patients before the administration of any
antibiotics. Blood cultures have been reported to be
positive in 25% to 60% of patients. A positive urine culture does not necessarily confirm the diagnosis because
a different organism may be identified at the time of
vertebral biopsy. In general, it is appropriate to delay
antibiotics until all cultures have been obtained; however, if the patient is septic or critically ill, antibiotics
should be initiated immediately.
Biopsy of the infected site is often necessary to identify the infecting organism and exclude other potential
etiologies. Spinal biopsies may be performed percutaneously using fluoroscopy or more accurately with CT. A
second closed biopsy is recommended if the diagnosis is
not confirmed after the first attempt. CT-guided biopsy
provides the best results, with positive cultures in 68%
to 86% of patients. Open biopsy is indicated when needle biopsy fails to identify the organism. Minimally inva-

American Academy of Orthopaedic Surgeons

Chapter 47 Spinal Infections

Orthopaedic Knowledge Update 8

sive techniques such as endoscopic or thoracoscopic approaches may be considered when appropriate. Open
biopsies yield positive results in more than 80% of patients. All biopsy specimens should be sent for Gram
stain, acid-fast and potassium hydroxide staining, and
cultures (including aerobic, anaerobic, mycobacterium,
fungi, and atypical mycobacterium). All cultures should
be retained to allow for the growth of mycobacterium
and low virulence organisms. Especially with lowvirulence and indolent bacteria, polymerase chain reaction has facilitated earlier identification of the infecting
organism. Caution is needed to avoid cross contamination with the use of polymerase chain reaction. Specimens also should routinely be sent for histologic analysis to rule out malignancy.
Malnutrition is often coincident with spinal infection; therefore, an evaluation of the patients nutritional
status is crucial in determining appropriate nutritional
therapy as a part of treatment. Laboratory measures
such as serum albumin level less than 3 g/dL, a serum
transferrin level of less than 150 ug/dL, and an absolute
lymphocyte count less than 800/mL suggest severe malnutrition, which should be addressed as part of the
treatment of the vertebral osteomyelitis.

Radiographic Evaluation
Imaging studies lag behind the clinical course of pyogenic vertebral osteomyelitis but are vital in localizing
and determining the extent of involvement of the infection and for assessing the response to treatment. Plain
radiographs can show subtle paravertebral soft-tissue
swelling in the first few days of infection. After 7 to 10
days, disk space narrowing can be observed. After several weeks, radiographs show frank erosion and destruction of the vertebral end plates and anterior vertebrae
with extension into the central portion of the vertebral
body. The disk space continues to collapse and vertebral
compression and paraspinal mass are noted. CT scans
show paravertebral soft-tissue masses and, most importantly, define the extent of bony involvement of the infection. CT scans show the anatomy in detail and can be
used to guide percutaneous drainage or biopsy and for
preoperative planning.
Radionuclide studies also are useful in evaluating
spinal infections and can be positive before the development of radiographic changes. Technetium Tc 99m bone
scintigraphy is more than 90% sensitive, but it lacks
specificity for infection. Scans that combine technetium
Tc 99m with gallium 67 increase both the sensitivity and
specificity for identifying infection. Indium 111-labeled
leukocyte scans are not recommended for vertebral osteomyelitis because of poor sensitivity (17%). Gallium
scans may be used to follow treatment response because
they begin to normalize during the recovery phase. MRI
with gadolinium contrast has become the imaging mo-

American Academy of Orthopaedic Surgeons

dality of choice for the diagnosis of spinal infections.

MRI has a sensitivity of 96% and specificity of 93% for
the diagnosis of vertebral osteomyelitis. On T1-weighted
images, the affected vertebral body and disk both show
low signal intensity with loss of distinction at their margins, secondary to the marrow edema and decreased fat
content. In contrast, on T2-weighted images, the vertebral body has increased signal intensity secondary to the
increase in water content associated with the inflammation and edema. On gadolinium-enhanced T1-weighted
images, abscess collections show abnormal rim enhancement, whereas areas of active inflammation in the vertebral body and disk are enhanced.

Treatment
The goals for treatment of spinal infections include establishing a diagnosis and identifying the organism,
eliminating the infection, preventing or improving neurologic involvement, and maintaining spinal stability. As
with other illnesses, nutritional repletion and optimization of medical comorbidities are crucial to eradication
of the infection.
Treatment of vertebral osteomyelitis usually entails a
trial of nonsurgical treatment with spinal immobilization, early ambulation, proper nutritional support, and
intravenous antibiotics followed by oral antibiotics (specific for the organism cultured). If the offending organism cannot be identified even after biopsy, empiric
parenteral antibiotics should be administered.
Parenteral antibiotics are generally recommended for 4
to 6 weeks to prevent high failure rates of nonsurgical
treatment in patients with pyogenic infections. Patients
are converted to oral antibiotics after signs of clinical
improvement, normalization of the ESR and CRP levels, or resolution of the infection on imaging studies.
Immobilization of the affected area helps prevent
deformity and aids in pain relief. The application of a
rigid contact brace is effective in the lumbar region. A
rigid cervicothoracic orthosis or halo is often required
for cervical osteomyelitis. Serial laboratory tests (CRP
and ESR) should be followed to monitor response to
treatment. Approximately 75% of patients respond to
nonsurgical treatment with resolution of pain and often
spontaneous fusion.
Surgical intervention is warranted to obtain a tissue
diagnosis after failed percutaneous needle biopsies, to
address neurologic deficit secondary to compression, to
treat spinal instability or significant deformity, to drain
infectious foci causing sepsis, or for failure of nonsurgical medical treatment alone. Nonsurgical treatment is
often unsuccessful in elderly and immunocompromised
patients, who then require surgical management. The location of the infection and the goals of the surgery dictate the intervention performed. If the surgery is intended to obtain a specimen for diagnosis, then an

579

Spinal Infections

Orthopaedic Knowledge Update 8

Figure 1 Imaging studies of an 80-year-old man with a 3-month history of increasing neck pain and recent onset of fever. A, A lateral radiograph shows collapse of the C5
vertebral body and end plate erosion of C4 and C6 with segmental kyphosis secondary to pyogenic vertebral osteomyelitis. The patient was treated with anterior dbridement and
corpectomy of C5, with partial corpectomy of C4 and C6, and with allograft fibula strut graft. This treatment was followed by posterior stabilization with C3-7 lateral mass fixation
and iliac crest bone graft. Postoperative lateral (B) and AP (C) radiographs.

anterior or posterior transpedicular biopsy (possibly using minimally invasive techniques) is appropriate.
The anterior approach is preferred for the treatment
of vertebral osteomyelitis because it permits dbridement of the infected bone and tissue, decompression of
the neural elements, drainage of an epidural abscess,
and stabilization of adjacent spinal segments. Posterior
infections are exceedingly rare but are amenable to a
posterior approach for dbridement. Cultures should always be obtained intraoperatively and thorough irrigation and dbridement of all infected and necrotic tissue
is required. The anterior dbridement inevitably leaves
a bony void, which often requires stabilization of the anterior column with interbody arthrodesis and posterior
stabilization (Figure 1). If a kyphotic deformity is
present, it can be reduced and maintained with appropriate interbody graft placement. Autogenous tricortical
iliac crest, rib, or fibular strut grafting (vascularized or
nonvascularized) has proven safe and effective in the
presence of acute infection. Freeze-dried allografts are
being used with successful results, but autogenous
sources are preferred because of better incorporation.
In the presence of a severe kyphotic deformity or when
a multilevel anterior construct is required, the addition
of posterior fusion with instrumentation is recommended to adequately stabilize the spine (Figure 2). The
procedure can be performed concomitantly or in a
staged manner. A recent study showed superior deformity correction with anterior titanium mesh cages filled
with autograft followed by posterior instrumentation.

580

Spinal Epidural Abscess

Epidemiology and Pathogenesis
There has been an increased incidence of spinal epidural abscess over the past decade. The reported incidence
of the disease, however, is only 1.2 per 10,000 hospital
admissions. Patients are most often age 60 years or older; men and women are equally affected. Factors that
contribute to the relative increased incidence of epidural abscess include an increase in intravenous drug use,
an increase in the number of immunocompromised
hosts secondary to malignancy or human immunodeficiency virus and/or acquired immunodeficiency syndrome, and an increase in the number of invasive spinal
operations and procedures (such as epidural anesthesia,
spinal injections, and diskography).
A spinal epidural abscess is a collection of pus or inflammatory granulation tissue between the dura mater
and the surrounding epidural adipose tissue.The presence
of a spinal epidural abscess is typically associated with a
coexistent vertebral osteomyelitis or diskitis. Contiguous
spread of pathogens from adjacent disk or bone is the
most common route of infection. However, hematogenous
seeding of bacteria into the epidural space occurs rarely.
Recently, studies have documented spinal epidural abscess after direct inoculation during spinal procedures.
The causative organism remains S aureus in more than
60% of patients. Gram-negative rods account for approximately 18% of infections. Pseudomonas is commonly
found in intravenous drug users.
Epidural abscesses are found in the cervical spine in
approximately 14% of patients, and most are found an-

American Academy of Orthopaedic Surgeons

Chapter 47 Spinal Infections

Orthopaedic Knowledge Update 8

Figure 2 Imaging studies of a 60-year-old man who had worsening back pain for 2 months. A, Lateral radiograph of the lumbar spine shows complete loss of disk space at L3-4
and end plate erosions of the L3 and L4 vertebrae. B, T2-fat suppressed MRI scan of the lumbar spine shows L3-4 diskitis and associated pyogenic vertebral osteomyelitis. C, The
patient was treated with anterior dbridement with interbody fusion using allograft strut followed by posterior fusion with instrumentation.

teriorly. The majority of spinal epidural abscesses are

found in the thoracic (51%) and lumbar (35%) spine.
Most of these infections are posterior unless contiguous
with vertebral osteomyelitis. Three to four spinal segments are usually involved, but the entire spinal column
is at risk.
Initially, there is an inflammatory reaction in the epidural fat that progresses to suppuration, necrosis, and fibrosis in the epidural space. The pathogenesis of the
neurologic manifestations is related either to direct
compression of the neural elements or to compromise
of the intrinsic circulation of the spinal cord.

Clinical Characteristics and Diagnosis

Intractable back or neck pain is the most common
symptom in patients with a spinal epidural abscess.
However, the clinical presentation of a patient with an
epidural abscess can be highly variable; this variability
leads to frequent misdiagnosis on initial presentation.
Most patients with an acute spinal epidural abscess have
fever, back pain, and spinal tenderness; however, these
outward signs may be absent if the disease is chronic. If
treatment is not initiated early, the back pain is followed
by radicular pain, weakness, paralysis, and even frank
sepsis. The immune status of the host and the virulence
of the offending pathogen dictate the timing and severity of the progression of the infection. If the abscess
penetrates the dura, a subdural abscess or meningitis
may occur.

American Academy of Orthopaedic Surgeons

Patients with an acute epidural abscess usually have

more systemic illness than those patients with vertebral
osteomyelitis. Laboratory evaluation shows leukocytosis
(mean, 22,000 cells/mm3), elevated ESR (mean, 86.3 mm
in 1 hour), and an elevated CRP. Plain radiographs are
usually normal unless contiguous vertebral osteomyelitis
or diskitis has been present long enough for the radiographic findings to be positive. CT scans have a poor
sensitivity for epidural abscess. MRI with gadolinium
contrast is the imaging modality of choice for the diagnosis and evaluation of spinal epidural abscess. The exact location and extent of the abscess, the amount of
neurologic compression, and the presence and severity
of contiguous infection can all be evaluated with MRI.
Pus in the epidural space will enhance with gadolinium,
whereas the cerebrospinal fluid will show low signal intensity on T1-weighted images. Biopsy of the adjacent
disk, vertebral body, or surrounding tissue provides the
definitive diagnosis of spinal epidural abscess and identification of the offending organism.

Treatment and Prognosis

Historically, the presence of spinal epidural abscess has
been regarded as a medical and surgical emergency. The
objectives of treatment of epidural abscesses are eradication of infection, prevention or improvement of neurologic sequelae, and preservation of spinal stability.
Nonsurgical treatment of a spinal epidural abscess is
only indicated in patients who are neurologically stable
or are unsuitable surgical candidates because of medical

581

Spinal Infections

Orthopaedic Knowledge Update 8

Figure 3 Imaging studies fof a 65-yearold man who developed cervical pyogenic
osteomyelitis with associated epidural abscess and neurologic deficits with concomitant lumbar diskitis. A, Sagittal T2weighted MRI scan of the cervical spine
shows increased signal intensity throughout the C6 and C7 vertebral bodies with
associated diskitis and epidural abscess
causing anterior mass effect on the spinal
cord. B, Sagittal T2-weighted image of
lumbar spine shows signal enhancement
of L2-3 disk and end plates consistent
with diskitis and early pyogenic vertebral
osteomyelitis.

comorbidities. A recent study showed that nonsurgical

treatment of spinal epidural abscess is possible if no
neurologic compromise has occurred. A new or progressive neurologic deterioration warrants immediate
change from nonsurgical to surgical treatment.
Surgical intervention is the current treatment for
most patients with spinal epidural abscess. The surgical
approach and technique depend on the location and extent of the infection. Because the location is usually posterior, a decompressive laminectomy over the involved
levels generally is the treatment of choice. Wide decompression with facetectomy is not usually required for adequate dbridement and should be avoided because iatrogenic spinal instability may result. If anterior
vertebral osteomyelitis or diskitis is present, combined
anterior and posterior dbridement may be necessary
(Figure 3). Reconstruction of the anterior column with
structural graft or instrumentation may be required if
significant bony destruction has occurred. Posterior fusion with instrumentation often is preferred when acute
infection is present after anterior dbridement.
Parenteral antibiotics are chosen based on the culture
and sensitivity results, and are continued for 2 to 4
weeks after adequate surgical dbridement. If vertebral
osteomyelitis is present, parenteral antibiotics should be
administered for at least 6 weeks.
Spinal epidural abscesses continue to have significant morbidity and mortality as a result of diagnostic
delay. Recent literature reports that early diagnosis remains the most essential factor in preventing devastating outcomes. A recent retrospective study found that
factors associated with improved outcome of patients

582

with spinal epidural abscess included younger age, less

than 50% thecal sac compression, lumbosacral location,
the presence of pustular abscesses as opposed to granulation tissue, and shorter duration of symptoms.

Nonpyogenic Vertebral Osteomyelitis

Nonpyogenic or granulomatous infections of the spine
may be caused by fungi, atypical bacteria, or spirochetes. These infections are grouped together because,
although the lesions resulting from these causes are not
common, when they occur they have a similar clinical
and histologic presentation. Most spinal infections in patients in the United States are pyogenic; however, with
the increased number of immunocompromised hosts,
more nonpyogenic infections have emerged in the past
decade. The pathophysiology of granulomatous spinal
infection differs from that of pyogenic infections. By far,
the most common granulomatous disease of the spine is
caused by Mycobacterium tuberculosis. Tuberculous
spondylitis (Potts disease) is the most extensively researched disease in this group of infections.

Epidemiology and Pathogenesis

Although tuberculosis (TB) is endemic in many developing countries, it was nearly eradicated in the United
States. Nevertheless, there has been a resurgence of TB
infection in the past decade, with the disease often occurring in immunocompromised patients and with resistant strains of the organism. The human immunodeficiency virus and acquired immunodeficiency syndrome
epidemic is a major cause of the increase in TB infection in the United States. Ten percent of patients in-

American Academy of Orthopaedic Surgeons

Chapter 47 Spinal Infections

Orthopaedic Knowledge Update 8

fected with TB will develop musculoskeletal infection,
but 50% of these patients will have spinal involvement.
Hematogenous seeding from the respiratory or genitourinary tract is the usual mode of spread of infection
to the spine. Local spread by direct extension is also
possible. Most active cases of spinal TB in adults are actually reactivations of quiescent lesions from previous
infection.
The three major types of spinal involvement of vertebral TB describe the pattern of bony involvement. The
most common form is the peridiskal type, which begins
with infection in the metaphyseal area that spreads under the anterior longitudinal ligament and later involves
the adjacent vertebral bodies. In contrast to pyogenic infection, the disk is relatively resistant to infection and is
often preserved until late in the infection process despite extensive bony destruction. With the anterior type,
infection spreads beneath the anterior longitudinal ligament and may involve several vertebral levels, causing
anterior erosions on the involved vertebral bodies. Central type infections often are mistaken for tumors because they remain isolated to one vertebra and often
cause collapse and deformity. Isolated posterior element
involvement has been described but is exceedingly rare.
The pathologic findings in tuberculous spondylitis
differ in several respects from pyogenic vertebral osteomyelitis. The disk is relatively resistant to infection by
TB. In addition, the development of the infection takes
a longer period of time, and more deformity is typically
observed at the time of presentation. Large paraspinal
abscesses are more common with tuberculous infections.
The thoracic spine is the most common location for spinal TB followed by lumbar involvement. The cervical
spine is rarely involved.
Other etiologies of granulomatous infection are also
increasing in incidence with the increasing number of
immunocompromised hosts. Atypical bacteria such as
Actinomyces israelii, Nocardia asteroids, and Brucella
species also cause chronic granulomatous infection. Treponema pallidum, the organism responsible for syphilis,
causes gummatous lesions in the spine, which are syphilitic granulomas representing the local reaction of the
tissues to the organism and its products. Fungal infections of the spine are uncommon and often have a significant delay in diagnosis. Coccidioides immitis, Blastomyces dermatitidis, Cryptococcus neoformans, Candida
species, and Aspergillosis are all responsible for fungal
infections of the spine. In a recent study of patients with
fungal spinal infections, the Candida species were the
most common fungal pathogens.

Clinical Presentation and Evaluation

The onset of symptoms with tuberculous spondylitis is
typically more insidious than with a pyogenic infection.
Patients usually report pain in the spine and constitu-

American Academy of Orthopaedic Surgeons

tional symptoms of chronic illness including intermittent

fever, malaise, night sweats, and weight loss. Neurologic
deficit is present in 10% to 47% of patients with Potts
disease. Deformity at the time of diagnosis is more common in TB infection than in pyogenic spinal infection.
Laboratory studies are helpful but nonspecific. The
white blood cell count often is not elevated. The ESR is
usually elevated but may be normal in up to 25% of patients. Purified protein derivative skin tests typically are
positive, which suggests either active or previous disease. False negative results are possible because of anergy, which could result from any condition that compromises host immunity. A chest radiograph should be
obtained in all patients to rule out active lung involvement.
Plain radiographs are useful in diagnosis, but they
will vary in appearance depending on the pathologic
type and chronicity of the infection. After several weeks
of infection, radiographs show vertebral end plate lucency and loss of cortical margins. Soft-tissue swelling
and expansion across two to three spinal segments follow. Eventually, the classic pattern of disk space destruction with lucency and compression of adjacent vertebral
bodies is seen, followed by further collapse into a kyphotic deformity.
Nuclear medicine studies with a combination of
technetium and gallium nuclear scans have been shown
to have the highest sensitivity for detecting infection.
CT scans are beneficial in evaluating the extent of bony
destruction and in surgical planning. However, the preferred imaging study for the diagnosis of tuberculous
and fungal infections remains MRI with gadolinium
contrast. MRI helps to differentiate metastatic disease
from vertebral osteomyelitis; the lack of disk space involvement is seen with MRI only with metastatic disease.
Definitive diagnosis requires a positive biopsy with
culture of the organism. Studies have shown that CTguided biopsy with cultures and staining allows identification of the organism. Mycobacteria are acid-fast bacilli and may take up to 10 weeks to grow in culture.
Polymerase chain reaction has been used for fast identification of mycobacteria with a sensitivity of 95% and
accuracy of 93%.

Treatment
Nonpyogenic vertebral osteomyelitis is usually treated
with chemotherapy directed at the offending pathogen.
Additionally, an external immobilization device may be
used for pain control and prevention of deformity. Tuberculous spondylitis is treated with isoniazid, rifampin,
and pyrazinamide for 9 to 18 months depending on response to treatment. Ethambutol or streptomycin is usually added to the regimen for at least part of the treatment. It is recommended that an infectious disease

583

Spinal Infections

Orthopaedic Knowledge Update 8

Figure 4 Imaging studies of a 48-year-old woman who presented with a 6-month history of severe low back pain. A, Plain radiograph shows evidence of chronic pyogenic
vertebral osteomyelitis of L4-5 with sclerosis and apparent autofusion consistent with the duration of symptoms. B, Sagittal T2-weighted MRI shows continued enhancement of L4
and L5 vertebral bodies as well as diskitis at L5-S1. AP (C) and lateral (D) postoperative radiographs of the patient who was treated with anterior dbridement and interbody
fusion of L4-L5 and L5-S1, with femoral ring allografts followed by posterior fusion with instrumentation.

consultation be done in conjunction with the chemotherapy regimen because of varying regional resistance
patterns. Amphotericin B and ketoconazole provide the
mainstay for the treatment of most fungal infections.
Brucellosis is typically treated with tetracycline and
streptomycin. Nocardia infections are treated with sulfonamides, whereas actinomycosis is still treated with
penicillin.
The surgical indications for treatment of nonpyogenic spinal infections are neurologic deficits, failure of
response to nonsurgical treatment after 3 to 6 months,
the need for tissue for diagnosis, spinal instability, progressive kyphotic deformity, and/or recurrence of the
disease. Surgical options include anterior dbridement
and strut graft alone or with posterior instrumentation
and fusion (Figure 4), or posterior dbridement without
anterior surgery done only for isolated posterior disease. If a laminectomy were performed for isolated posterior arch disease or posterior epidural abscess, then a
supplemental fusion would be recommended. Anterior
dbridement and reconstruction at the site of pathology
has shown the best long-term neurologic and structural
results. Graft choice is dependent on surgeon preference. Autogenous and allograft strut grafts are acceptable with good results. A recent study showed that anterior spinal instrumentation with structural allograft
fibula could be used after proper anterior dbridement
of tuberculous spondylitis with a 96% fusion rate and
no recurrence of infection. Patients did not require external support in the postoperative period. Another
study, using fresh frozen femoral allografts and stabilization with a single-rod construct after anterior dbridement, showed excellent results with incorporation of the
allografts between 12 and 18 months. These studies sug-

584

gest that anterior instrumentation reduces kyphotic deformity without increasing the risk of disease recurrence. In patients with neurologic deficit, earlier
dbridement led to a faster and better neurologic recovery. There is a direct correlation between duration of
preoperative symptoms and neurologic recovery.

Postoperative Spinal Infections

An increase in the incidence of postoperative infections
has been documented in the past decade. Postoperative
infections result from inoculation during the index procedure or by hematogenous seeding of the surgical site.
Lumbar diskectomy performed with the use of prophylactic antibiotics carries a 0.7% incidence of infection.
The use of the operating microscope for diskectomy increases that rate to 1.4%. Studies have shown that lumbar fusion with instrumentation has an estimated incidence of postoperative infection of 6%. A recent
retrospective review showed a wound infection rate of
10% in trauma patients who underwent fusion for thoracolumbar fractures. The authors concluded that the
overall risk of infection is higher in the trauma patient
than in the elective surgery population and that those
patients with complete neurologic deficit are at a
greater risk of infection. Another recent study showed
that late pain at the surgical site appearing 12 to
20 months after scoliosis surgery with posterior instrumentation is most often attributable to a subacute lowgrade implant infection.
Prophylactic antibiotics have been shown in several
studies to significantly reduce the incidence of postoperative infections. Prophylactic antibiotics should reach
peak concentrations quickly and cover S aureus and
S epidermidis. Cefazolin, a first-generation cephalospo-

American Academy of Orthopaedic Surgeons

Chapter 47 Spinal Infections

Orthopaedic Knowledge Update 8

Table 3 | Risk Factors for Developing Postoperative

Infections

tibiotics taken for 6 weeks are required for patients with

deep infection, bony involvement, or retained hardware.

Advanced age
Obesity
Diabetes mellitus
Smoking
Immunocompromised host
Length of preoperative hospitalization
Myelodysplasia
Revision surgery
Increased surgical time
Spinal instrumentation
Bone graft
Methylmethacrylate
Arthrodesis
Trauma

Annotated Bibliography

rin, has good Staphylococcus coverage and should be administered at least 20 minutes before surgery. To prevent
resistance, vancomycin should be used for prophylaxis
only in patients at high risk for methicillin-resistant S aureus. Risk factors for postoperative spinal infections are
listed in Table 3.
Clinical presentation of postoperative infections depends on timing and the depth of the infection. In the
immediate postoperative period, patients with superficial wound infections may have pain, fever, tenderness,
erythema, and drainage from the incision site. Diagnosis
of deep wound infections is more difficult because complete onset may be delayed, with only constitutional
symptoms and a well healed surgical incision. Laboratory values including leukocyte count, ESR, and CRP
are often elevated. The acute phase reactants are normally elevated in the immediate postoperative period.
The ESR remains elevated for up to 6 weeks, whereas
the CRP normalizes in approximately 2 weeks. S aureus
is cultured in about 60% of wound infections.
Aggressive surgical intervention is generally recommended for postoperative infections. Administration of
antibiotics should be delayed until intraoperative superficial and deep cultures are obtained. Aggressive dbridement followed by copious irrigation is recommended. Wound closure over closed suction drains is
required unless the wound is packed open for repeat
dbridements. Recent studies advocate the use of antibiotic beads, especially in the presence of hardware. Unless the fusion is solid, most surgeons retain the instrumentation and bone graft. If significant soft-tissue
necrosis or dead space is present, plastic surgical techniques including musculocutaneous flaps may be necessary. For patients with soft-tissue and wound infections,
10 to 14 days of antibiotics are sufficient. Parenteral an-

American Academy of Orthopaedic Surgeons

Pyogenic Vertebral Osteomyelitis

Hee HT, Majd ME, Holt RT, Pienkowski D: Better
treatment of vertebral osteomyelitis using posterior stabilization and titanium mesh cages. J Spinal Disord Tech
2002;15:149-156.
The authors evaluated the efficacy of using titanium mesh
cages anteriorly and posterior instrumentation after anterior
dbridement in the surgical treatment of vertebral osteomyelitis. Patients treated with cages had better sagittal and coronal
correction.

Schimmer RC, Jeanneret C, Nunley PD, Jeanneret B:

Osteomyelitis of the cervical spine: A potentially dramatic disease. J Spinal Disord Tech 2002;15:110-117.
A retrospective review of 15 patients treated for osteomyelitis of the cervical spine is presented. Good results were
achieved by early and aggressive surgical intervention.

Spinal Epidural Abscess

Tang HJ, Lin HJ, Liu YC, Li CM: Spinal epidural abscess: Experience with 46 patients and evaluation of
prognostic factors. J Infect 2002;45:76-81.
This article reviewed 46 patients treated for spinal epidural abscess with emphasis on defining clinical characteristics,
treatment options, and evaluation of prognostic factors.

Nonpyogenic Vertebral Osteomyelitis

Frazier DD, Campbell DR, Garvey TA, Wiesel S, Bohlman HH, Eismont FJ: Fungal infections of the spine:
Report of eleven patients with long-term follow-up.
J Bone Joint Surg Am 2001;83:560-565.
This article presents a retrospective review of 11 patients
treated for spinal osteomyelitis caused by a fungus. The authors describe delay in treatment and diagnosis as important
factors in poor outcomes. They recommend performing fungal
cultures whenever a spinal infection is suspected.

Govender S: The outcome of allografts and anterior instrumentation in spinal tuberculosis. Clin Orthop 2002;
398:50-59.
This article presents a review of 41 patients with neurologic deficits caused by spinal TB who were treated with radical anterior decompression with reconstruction of the anterior
column with fresh-frozen femoral ring allograft and stabilized
with a single-rod screw instrumentation construct.

Ozdemir HM, Us AK, Ogun T: The role of anterior spinal instrumentation and allograft fibula for the treatment of pott disease. Spine 2003;28:474-479.
The authors retrospectively reviewed 28 patients with multilevel spinal TB who had anterior dbridement, decompression, and fusion with anterior spinal instrumentation and fibu-

585

Spinal Infections

Orthopaedic Knowledge Update 8

lar allograft. A 96% fusion rate with no graft complications

was recorded.

tumors and infections with magnetic resonance imaging.

Spine 1991;16(suppl 8):S334-S338.

Postoperative Spinal Infections

Carragee EJ: Pyogenic vertebral osteomyelitis. J Bone

Joint Surg Am 1997;79:874-880.

Kothari NA, Pelchovitz DJ, Meyer JS: Imaging of musculoskeletal infections. Radiol Clin North Am 2001;39:
653-671.
This article reviews the epidemiology, pathophysiology,
and the clinical and imaging presentations of musculosketetal
infections of all types. Discussion is presented on the imaging
characteristics of plain radiographs, CT scans, MRI, and nuclear studies for various spinal infections.

Rechtine GR, Bono PL, Cahill D, Bolesta MJ, Chrin

AM: Postoperative wound infection after instrumentation of thoracic and lumbar fractures. J Orthop Trauma
2001;15:566-569.
Twelve of 117 patients (10%) with thoracolumbar fractures who had surgical intervention developed postoperative
wound infections. The authors observed that the overall risk of
infection is higher in trauma patients especially those with
neurologic deficit.

Currier BL, Eismont FJ: Infections of the spine, in

Herkowitz HN, Garfin GR, Balderston RA, Eismont FJ,
Ball GR, Wiesel SW (eds): The Spine. Philadelphia, PA,
WB Saunders, 1999, pp 1207-1258.
Eismont FJ, Bohlman HH, Soni PL, Goldberg VM, Freehafer AA: Pyogenic and fungal vertebral osteomyelitis
with paralysis. J Bone Joint Surg Am 1983;65:19-29.
Emery SE, Chan DP, Woodward HR: Treatment of hematogenous pyogenic vertebral osteomyelitis with anterior debridement and primary bone grafting. Spine 1989;
14:284-291.
Heggeness MH, Esses SI, Errico T, et al: Late infection
of spinal instrumentation by hematogenous seeding.
Spine 1993;18:492-496.

Tay BK, Deckey J, Hu SS: Spinal Infections. J Am Acad

Orthop Surg 2002;10:188-197.

Moon MS: Tuberculosis of the spine: Controversies and

a new challenge. Spine 1997;22:1791-1797.

A thorough review of the literature and a concise description of current methods of diagnosis, laboratory assessment,
imaging, and treatment of spinal infections in both children
and adults are presented.

Rezai AR, Woo HH, Errico TJ, Cooper PR: Contemporary management of spinal osteomyelitis. Neurosurgery
1999;44:1018-1026.

Vaccaro AR, Harris BM, Madigan L: Spinal infections,

pyogenic osteomyelitis, and epidural abscess, in Vaccaro
A, Betz R, Zeidman S (eds): Principles and Practice of
Spine Surgery. Philadelphia, PA, Mosby, 2003, pp 165174.
This chapter presents a current comprehensive review of
spinal infections.

Classic Bibliography
An HS, Vaccaro AR, Dolinskas CA, Colter JM, Balderston RA, Bauerle WB: Differentiation between spinal

586

Thalgott JS, Cotler HB, Sasso RC, LaRocca H, Gardner

V: Postoperative infections in spinal implants classification and analysis: A multicenter study. Spine 1991;16:
981-984.
Wheeler D, Keiser P, Rigamont D, Keay S: Medical
management of spinal epidural abscesses: Case report
and review. Clin Infect Dis 1992;15:22-27.
Yilmaz C, Selek HY, Gurkan I, Erdemli B, Korkusu Z:
Anterior instrumentation for the treatment of spinal tuberculosis. J Bone Joint Surg Am 1999;81:1261-1267.

American Academy of Orthopaedic Surgeons

Chapter

48

Tumors of the Spine

Gurvinder S. Deol, MD
Rex Haydon, MD, PhD
Frank M. Phillips, MD

Introduction
Primary and metastatic tumors of the spine encompass a
wide spectrum of disease processes requiring many different treatment algorithms. The treatment of spinal tumors has evolved over the course of the past decade
with the advent of improved diagnosis, staging, and nonsurgical and surgical treatment.

Primary Tumors
Primary tumors of the spine account for 2% to 5% of
all spinal neoplasms, with metastatic tumors accounting
for most spinal tumors. Within the group of primary tumors, benign tumors are far more common than malignant tumors.

Benign Tumors
Osteoid Osteoma
Osteoid osteomas are probably the most common primary benign vertebral tumors, and they are usually diagnosed during the first three decades of life, with a
peak incidence at age 15 years. Ten percent to 25% of
all osteoid osteomas occur in the spine, and nearly 70%
of painful juvenile scoliotic deformities are associated
with osteoid osteomas that typically occur at the apex of
the concavity of the curve as a cortically based nidus of
osteoid-producing cells surrounded by a dense halo of
sclerosis, which may be the only radiographic sign at diagnosis. Histologically, the lesion manifests as a nidus of
highly vascular osteoid-producing spindle cells surrounded by dense sclerotic bone. Pain is the most common presenting symptom and is characteristically worse
at night and relieved by treatment with nonsteroidal
anti-inflammatory drugs (NSAIDs). On plain radiographs, the overlying bony structures often obscure the
appearance of osteoid osteoma, making additional imaging studies necessary. The most sensitive study for osteoid osteoma is the bone scan, which targets the rapid
bone turnover, a hallmark of this lesion. Increased uptake of technetium Tc 99m occurs in the area of the lesion, often surrounded by a zone of diminished uptake,
creating a distinctive target sign. Although bone scans

American Academy of Orthopaedic Surgeons

represent perhaps the most sensitive test for osteoid osteoma, a CT scan is the most specific. Treatment of this
disorder includes both medical and surgical options.
Pain associated with osteoid osteoma, as a rule, responds to treatment with NSAIDs. Given the usually
self-limited nature of osteoid osteoma, NSAIDs and observation are the initial treatment. In patients in whom
NSAIDs are either not tolerated or are contraindicated
or in patients whose osteoid osteoma is associated with
progressive scoliotic deformities, more aggressive therapies can be considered. Excision of the lesions results in
reliable pain relief, and most associated scoliotic deformities improve. Good short-term results with percutaneous radiofrequency ablation of osteoid osteoma have
been reported.
Osteoblastoma
Histologically, osteoblastomas are often indistinguishable
from osteoid osteomas except for their size, but the clinical features and natural history of these two disorders
have notable differences. Spine involvement is even more
commonly associated with osteoblastoma, accounting for
approximately 40% of instances; lesions typically localize
to the posterior elements in 55% of patients. The most
common presenting symptom in patients with osteoblastoma is focal pain, which is less responsive to NSAIDs
than the pain associated with osteoid osteoma.The pain is
more typically activity-related. Cortical expansion can result in impingement of neural elements. Painful scoliotic
deformities can also occur in the setting of osteoblastoma; however, this is much less common than with osteoid
osteoma. Osteoblastomas are more readily detected on
plain radiographs because of their larger size (> 2 cm), and
their propensity to cause cortical expansion. The internal
characteristics of osteoblastomas can be variable, but ossification is the predominant pattern, which is consistent
with its osteoblastic origin. Osteoblastoma is a slowly progressive lesion that does not normally respond to conservative management. Surgical resection of the lesion is
therefore indicated; however, local recurrences occur in
10% to 15% of patients, and the recurrence rate can be as
high as 50% in patients with high-grade subtypes of os-

587

Tumors of the Spine

teoblastoma. Debate exists regarding the adequacy of
curettage in the treatment of osteoblastoma, and it has yet
to be determined whether marginal resection results in a
lower risk of recurrence. Surgical treatment (simple curettage or resection) should be planned based on the location of the lesion, concomitant symptoms, and risk of morbidity. Malignant transformation of osteoblastomas has
been documented.
Giant Cell Tumor
Giant cell tumors can range in behavior from slowly
growing, relatively innocuous tumors to locally aggressive tumors that metastasize. The spine is a relatively
common site for giant cell tumors, comprising between
5% and 10% of all instances of giant cell tumor. Unlike
their appendicular counterparts, however, there is often
a significant delay between the onset of symptoms in
the spine and the diagnosis. Pain and radiculopathy are
the most common presenting symptoms, and they have
often been present for several months before the patients initial contact with a physician. Spinal giant cell
tumors are most commonly diagnosed during the third
and fourth decades of life, and they occur at a slightly
more frequent rate in women. Within the axial skeleton,
the sacrum is the most common region affected, and lesions are typically found in the vertebral body. Plain
films generally demonstrate a well-demarcated, radiolucent lesion with a variable amount of cortical expansion
and local remodeling. Even in the most aggressive instances of giant cell tumor, a thin shell of cortical bone
will usually remain at the periphery of the lesion, which
helps distinguish it from a malignant bone tumor. Giant
cell tumors involving the spine are typically treated by
curettage. En bloc surgical resection is considered to be
the optimal treatment of this disorder and appears to
reduce the rate of local recurrence. In the spine, the
proximity of vital structures to the lesion, as well as the
considerable morbidity associated with this approach
has limited the use of en bloc resections. Spinal giant
cell tumors have a considerably poorer prognosis than
those in the appendicular skeleton, with recurrence
rates reaching almost 80% in grade III giant cell tumors.
Furthermore, metastasis occurs in just under 10% of patients.
Chondroblastoma
Chondroblastoma of the spine is extremely rare, and
published articles regarding spinal chondroblastoma are
mostly limited to case reports. Nonetheless, most patients with this disorder have been diagnosed during the
second or third decade of life. Radiographic evaluation
of spinal chondroblastomas generally reveals a welldemarcated radiolucent lesion. Evidence of internal matrix calcification may be apparent on plain radiographs
or CT scans; however, this is not a universal finding in
all patients with chondroblastoma. Because of the lim-

588

Orthopaedic Knowledge Update 8

ited number of patients with this disorder, it is difficult
to define a clear predilection for either the anterior or
posterior spinal elements; however, it has been reported
that most spinal chondroblastomas occur in the posterior elements. Chondroblastomas usually contain numerous osteoclast-like giant cells; matrix calcification,
when present, often has a distinctive chicken-wire appearance, which is a histologic hallmark of this lesion.
Cellular atypia can vary from moderate to high, which is
believed to reflect the spectrum of behavior of chondroblastoma from slowly progressive local growth to aggressive local growth and metastatic spread. Treatment
of these lesions consists predominantly of curettage or
excision; however, it is not clear whether this treatment
is effective in the spine, given the limited number of patients with spinal chondroblastomas. The recurrence rate
for chondroblastomas of the spine is likewise unclear.
Osteochondroma
Spinal osteochondromas account for less than 10% of
all osteochondromas. They usually arise from the posterior elements. Typically, they are slow-growing masses
and rarely cause mechanical or compressive symptoms.
They are therefore largely observed unless they begin to
cause problems such as persistent pain or radiculopathy.
If this occurs, these tumors can be resected. The rate of
recurrence after resection in adults is low, and in skeletally immature patients it is imperative to resect the cartilage cap to prevent recurrence. Malignant transformation to chondrosarcomas has been reported and occurs
more commonly in patients with multiple hereditary exostoses. These lesions tend to be low grade and respond
well to local wide resection.
Langerhans Cell Histiocytosis
This condition, more commonly known as eosinophilic
granuloma, represents a usually benign, self-limiting
process that usually causes focal areas of welldemarcated bone resorption, which in the spine gives
rise to the classic vertebra plana sign. The underlying
etiology of this disorder is unknown, and it usually affects individuals during the first and second decades of
life, with a 2:1 predilection for males. The spine is involved in 10% to 15% of patients, but the condition
more commonly affects the skull and flat bones of the
pelvis, rib cage, and shoulder. Although most instances
of this disorder are self-limiting and resolve on their
own, surgical intervention may be indicated in patients
with progressive kyphosis or neurologic symptoms. Lowdose radiation therapy has also shown to be effective in
patients who are not amenable to resection. It is important to be aware of two systemic disease processes associated with eosinophilic granuloma: Hand-SchllerChristian disease and Letterer-Siwe disease. Both of
these disease processes involve multifocal lesions along
with other systemic manifestations.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Aneurysmal Bone Cyst
Aneurysmal bone cysts involve the spine in 10% to 30%
of patients, and they are most commonly found in the
posterior elements of the thoracolumbar spine. They can
occur at any age but are most common during the first
two decades of life. Unique to aneurysmal bone cysts is
their ability to involve multilevel adjacent spinal segments. Radiographically, aneurysmal bone cysts cause
cortical expansion and thinning, and have a characteristic bubbly appearance created by multiple cavernous
chambers. MRI is needed to detect smaller aneurysmal
bone cysts, and fluid/fluid levels, which represent old areas of hemorrhage, are almost pathognomonic for aneurysmal bone cysts. Treatment includes curettage, wide local excision, embolization, and sometimes radiation. The
rate of local recurrence has been estimated to be between 15% and 30% of patients; recurrence is treated
by repeat curettage.
Hemangioma
Hemangiomas of the spine represent the most common
tumor of the spine and are usually identified as an incidental finding. At autopsy, 11% of individuals are reported to have hemangiomas. These occur as singular lesions in approximately two thirds of individuals, and the
lesions are characterized radiographically by vertical
trabecular striations resembling a honeycomb that most
commonly involve the vertebral body, with the thoracic
spine being the most commonly affected. Plain radiographs can be sufficient to diagnose those lesions that
involve greater than 30% to 40% of the vertebral body;
however, CT or MRI may be more helpful in detecting
the presence of more subtle lesions. Neurologic symptoms may occur as a result of neural compression
caused by cortical expansion or soft-tissue extension beyond the vertebral body. Hemangiomas are radiosensitive, and low-dose radiation has been shown to be effective as a treatment of symptomatic lesions, as has
embolization via angiography. In those instances where
pathologic fracture or deformity results in instability or
neurologic compromise, surgical resection and stabilization may be indicated. Vertebral cement augmentation
procedures (vertebroplasty and kyphoplasty) have also
been successfully used to treat hemangiomas.

Malignant Tumors
Osteosarcoma
Osteosarcoma of the spine carries with it an especially
bleak prognosis. Osteosarcoma of the spine accounts for
approximately 2% of all osteosarcomas throughout the
body, and 3% to 14% of malignant tumors involving the
spine. Most tumors arise in the lumbosacral region and
involve the vertebral body in up to 90% of patients. As
an entity, osteosarcoma includes any malignant spindle
tumor that produces osteoid; however, this encompasses

American Academy of Orthopaedic Surgeons

Chapter 48 Tumors of the Spine

a variety of histologic subtypes. Treatment and prognosis are, therefore, dependent on appropriate histologic
diagnosis. There is a bimodal distribution in the age of
presentation of patients with osteosarcomas involving
the spine; a young patient population (age 10 to 25
years) typically has the more classic type of osteosarcoma, whereas patients older than 50 years typically
have secondary osteosarcomas. When viewed collectively, osteosarcoma of the spine has a generally poorer
prognosis and occurs in older age groups when compared with appendicular osteosarcoma. The primary lesion can be either radiolucent or radiodense, with prominent periosteal reaction and usually soft-tissue
extension. The internal characteristics of the lesion demonstrate ossification consistent with the degree of osteoid production. Numerous histologic subtypes of osteosarcoma exist that are designated by location
(central, parosteal, and periosteal), grade (low versus
high), predominant cell type (osteoblastic, chondroblastic, and fibroblastic), or etiology (radiation-induced, and
Pagets sarcoma). With the exception of low-grade lesions such as parosteal osteosarcomas, patients with osteosarcoma of the spine receive preoperative chemotherapy followed by surgical resection and usually
adjuvant therapy. As with other malignant lesions of the
spine, wide resections or even marginal resections are
often not possible, making radiation therapy and adjuvant chemotherapy necessary to treat the residual disease. Estimation of the outcome in patients with osteosarcoma of the spine has been hampered by its
relative rarity. Metastasis at diagnosis, large size, sacral
location, and intralesional resections are associated with
adverse outcomes.
Chondrosarcoma
After chordomas, chondrosarcoma is the most common
primary malignant tumor of bone in the spine, accounting for approximately 7% to 12% of all spine tumors.
Chondrosarcomas occur more commonly in men than
women and later in life, with the average age at diagnosis being 45 years. Chondrosarcoma can vary considerably in its behavior, and it is generally described according to grades I through III, with each grade
corresponding to an increasing tendency for metastasis
and, therefore, a poorer prognosis. Plain films typically
demonstrate a centrally based destructive lesion with
calcification. In patients with low-grade chondrosarcoma, the lesion can cause scalloping of the bony cortex
or cortical expansion. In patients with high-grade chondrosarcoma, the tumor can erode through the cortex
and form a large extraosseous mass, also containing diffuse areas of calcification. Identification of the grade of
the tumor is essential to determine prognosis, and it is
based primarily on the cellularity of the tumor and pleomorphism of the tumor cells. Such information is also
useful to determine the usefulness of adjuvant treat-

589

Tumors of the Spine

ments such as radiation therapy. The treatment of chondrosarcoma is complicated by its lack of response to
conventional chemotherapy and/or radiation therapy.
No known clinical trials have demonstrated any survival
benefit among patients receiving chemotherapy, and the
use of radiation in this patient population is controversial. Surgical excision, therefore, is the mainstay of treatment of chondrosarcoma. As a result, the survival rate
of patients with chondrosarcoma is closely associated
with adequate lesion excision and clean margins. Median survival in patients with chondrosarcoma of the
spine is approximately 6 years.
Ewings Sarcoma
Ewings sarcoma of the spine is a relatively rare entity,
accounting for only 8% of a large reported series of patients with Ewings sarcoma. It is actually more common
for this tumor to metastasize to the spine from other locations than it is for it to originate in the spine as a primary tumor. Its clinical features are similar to those for
all patients with Ewings sarcoma; it most commonly
arises during the second decade of life, and it affects
males more frequently than females. The most common
presenting symptom among patients with Ewings sarcoma is pain, often in the sacrococcygeal area. Although
instances of Ewings sarcoma involving the cervical
spine have been reported, they are extremely rare. Osseous findings in patients with Ewings sarcoma can be
extremely subtle. Plain radiographs can, therefore, appear normal, often belying a large soft-tissue mass. In
this respect, CT or MRI may be far more informative in
defining tumor extent. In patients with Ewings sarcoma, these imaging modalities can demonstrate evidence of a large mass originating in the vertebral body,
with variable amounts of internal mineralization.
C-reactive protein levels and erythrocyte sedimentation
rates are often elevated, and can therefore be useful adjuncts for the diagnosis of Ewings sarcoma. Histologically, Ewings sarcomas are composed of sheets of small
blue cells, occasionally forming pseudorosettes around
areas of necrosis. Nearly all Ewings sarcomas possess a
characteristic t11:22 chromosomal translocation. Although the treatment of patients with Ewings sarcoma
has varied considerably, the condition currently is
treated using a combination of chemotherapy, surgical
resection, and radiation therapy, each of which is effective against this tumor individually. The use of adjuvant
treatments is especially critical given the difficulty of
performing wide resection in the spine.
Chordoma
Chordomas are the most common nonhematogenous
primary malignant tumors of the spine, and, unlike
other malignant tumors discussed, they do not normally
occur outside of the spine. They account for 1% to 4%
of all primary bone tumors, with 20% of those occurring

590

Orthopaedic Knowledge Update 8

in the spine. Originating from remnants of the notochord, chordomas typically involve either the sacrococcygeal or sphenooccipital regions of the axial skeleton.
The average age at diagnosis is 56 years; however, these
tumors can occur in almost any age group. Clinical presentation is often subtle, with a gradual onset of neurologic symptoms including pain, numbness, motor weakness, incontinence, and constipation. Chordomas are
slowly growing lesions and are often quite large when
initially discovered. When located in the sacrum, the
mass usually protrudes anteriorly, thereby preventing
the lesion from causing a noticeable external mass. On
radiographic evaluation, the bone from which the tumor
arises may demonstrate noticeable changes, but the
most impressive feature of chordomas is the large softtissue mass. Unless a significant amount of internal calcification is present, the soft-tissue component can be
missed or underestimated on the basis of plain radiographs; therefore, either CT or MRI is required for definitive evaluation. On MRI chordomas are lobulated
masses with a distinctly myxoid (mucoid) consistency.
Because they are generally slow-growing tumors, they
are associated with a pseudocapsule. The histologic appearance of chordomas can vary from relatively cellular
masses to fluid-filled cysts. Classically, they are composed of chords of physaliphorous cells that are organized into lobules. Similar to chondrosarcoma, chordomas demonstrate a poor response to both radiotherapy
and chemotherapy. Surgical excision with wide margins,
therefore, offers the only significant possibility of cure
in these patients. The average 10-year survival rate
among patients with sacral chordomas is 20% to 40%,
usually because of recurrence and direct spread of the
tumor. The reported rate of metastasis varies widely in
different series, ranging from 10% to 27% for sacral lesions; however, this rarely represents the cause of death
in patients with chordomas.
Multiple Myeloma
Multiple myeloma and its solitary counterpart, plasmacytoma, are both B-cell lymphoproliferative diseases.
Solitary plasmacytoma often progresses to multiple myeloma. Multiple myeloma is the most common primary
malignancy of bone and the spine. Patients often
present with pain from spontaneous vertebral compression fractures. Multiple punched-out lesions on a lateral
skull radiograph are a classic finding, and this warrants a
skeletal survey which may reveal other lesions. Bone
scans are classically negative because of the lack of local
bone reaction, and the diagnosis is confirmed with the
presence of a monoclonal spike with serum electrophoresis. Chemotherapy and radiation therapy have
long been the mainstays of treatment of this condition.
With advances in chemotherapeutic treatments, patients with multiple myeloma often live longer than patients with bone metastases from other malignancies.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

This has prompted a more aggressive approach in the
treatment of myeloma affecting the spine to improve
quality of life. Radiation may be quite effective in treating spinal cord compression from epidural tumor. In patients with vertebral collapse with neurologic symptoms
or advanced kyphosis, surgery may be indicated. Even
with more advanced disease, surgical decompression
and reconstruction may be warranted, and the outcomes
from a recent study seem to favor this approach in carefully selected patients. Similarly, in patients with myeloma and pathologic vertebral compression fractures,
kyphoplasty has been shown to result in improved quality of life as well as correction of vertebral deformity.

Metastatic Tumors
Because the spine is the most common site of skeletal
involvement of metastatic disease, metastases account
for most tumors surgically encountered in the spine. Approximately 50% of these tumors arise from carcinoma
(lung or breast), lymphoma, or myeloma. In regard to
the solid tumor primaries, breast, lung, and prostate are
the most common, followed by renal, thyroid, gastrointestinal, and rarely primary soft-tissue sarcomas.
Breast, prostate, and renal metastases are more likely to
be seen by the spine surgeon than either pulmonary or
gastrointestinal tumors because of longer patient survival rates.
The most common and usually the first symptom of
spinal metastases is that of localized pain, which may
occur at night and awaken patients from sleep. Patients
may also present with radicular or myelopathic symptoms. If a suspected metastatic spinal lesion is found, a
search for the primary tumor as well as for other sites of
metastatic disease must be made. This workup may
identify other lesions that are more accessible for biopsy
than the spine. The type of tumor often dictates the
treatment; however, the goals of treatment include decreasing the tumor burden, relief of pain, prevention
and reversal of neurologic deficits, and preserving or restoring spinal stability. The modes of treatment will be
discussed later in the chapter.

Spinal Cord Tumors

Tumors of the spinal cord are divided into three categories: extradural, intradural-extramedullary, and intramedullary.
Extradural tumors are very uncommon. The most
common malignancy in the epidural space is lymphoma,
with most being spread from the vertebral body or
paraspinous nodes; however, a small number do arise
primarily from the epidural space. Excision and posterior neural decompression is usually the treatment of
choice for extradural tumors. Ten percent of neurofibromas are extradural. They usually arise from a spinal
nerve root, and because malignant degeneration is pos-

American Academy of Orthopaedic Surgeons

Chapter 48 Tumors of the Spine

sible, excision is indicated. Although epidural lipomas
are uncommon, they are a well-documented entity and
can cause cord compression. Lipomatosis, especially
when steroid-induced, is another possible diagnosis in
patients with extradural tumors. Extradural tumors can
initially be observed using MRI, but excision is recommended in patients with any progressive neurologic deficits. Epidural hemangioma and meningioma are both
rare.
Intradural-extramedullary tumors make up approximately 85% of intradural tumors and largely consist of
neurofibromas, meningiomas, and ependymomas. The
two predominant intramedullary tumors are astrocytoma (high and low grade) and ependymomas, with hemangiomas and lipomas being exceedingly rare. As with
extradural neurofibromas, the intradural counterpart
also arises from the nerve roots, usually the dorsal sensory roots. The so-called dumbbell tumor is an intradural neurofibroma that is following the nerve root
through the foramen. Again, complete excision is the
treatment of choice. Meningiomas are usually found in
the anterolateral canal and should be excised when
identified. Ependymomas usually arise from within the
cord and should be excised; however, extramedullary
ependymomas can arise from the tip of the conus at its
junction with the filum terminale. Astrocytomas are intramedullary and should be excised. High-grade astrocytomas have no visible margin and are usually debulked
followed by neoadjuvant chemotherapy. Patients with
high-grade astrocytomas have a bleak prognosis.

Diagnosis and Evaluation

The most common presenting symptom among patients
with tumors of the spine is pain, which is reported by
more than 80% of patients at initial visit. Typically, tumors with a more rapid onset of pain reflect more aggressive tumors, whereas benign or more slowly growing
tumors may be characterized by gradually increasing
pain spanning several months to years, with sudden increases in pain suggesting pathologic fracture. Neurologic compromise can also be a presenting symptom because of compression of either nerve roots or the spinal
cord. Depending on the precise location and size of the
tumor, neurologic symptoms can range from subtle motor or sensory deficits to paraplegia. Other presenting
symptoms such as spinal deformity are considerably less
common. A careful history should also include constitutional symptoms such as weight loss, fevers and chills,
and lethargy as well as a complete medical history, including any personal history or family history of malignancy.
Radiographic evaluation generally begins with plain
films, with notation of the location of the lesion, any
soft-tissue extension, the zone of transition, and internal
characteristics. For tumors involving the spine, certain

591

Tumors of the Spine

Orthopaedic Knowledge Update 8

Figure 2 Ennekings oncologic stages (IA, IIA, IB, and IIB) of malignant tumors to the
spine. 1 = tumor capsule, 2 = pseudocapsule, 3 = tumor with pseudocapsule, 4 = skip
metastases. (Reproduced with permission from Boriani S, Biagini R, DeJure F: Bone
tumors of the spine and epidural cord compression: Treatment options. Semin Spine
Surg 1995;7:317-322.)

Figure 1 AP radiograph demonstrating the winking owl sign in which the right-sided
pedicle has been destroyed by tumor, so that the pedicle ring is absent (arrow).

classic radiographic signs have been described, such as

the winking owl sign, in which the pedicle has been
eroded by an expanding tumor (Figure 1), or vertebra
plana, in which there is vertebral body collapse. Although these signs are considered to be classic radiographic signs for spinal lesions, they are typically seen
only after considerable bony destruction has occurred.
Advanced imaging studies are mandatory in any evaluation of spinal tumors. MRI provides not only unparalleled soft-tissue detail, but it also evaluates the neural
elements. MRI can reveal important tissue characteristics of the tumor, such as density, vascular perfusion and

592

necrosis, and extent of marrow involvement. When the

lesion is primarily bony, CT provides excellent detail
and is a valuable tool for preoperative planning. For patients with malignant tumors or metastatic disease, staging studies should also include a chest radiograph, chest,
abdominal, and pelvic CT, and bone scan. Finally, before
beginning any definitive treatment, a pathologic diagnosis should be obtained by performing a biopsy, either
CT-guided or open, depending on the location and size
of the lesion.
Staging systems attempt to predict outcome among
patients with benign and malignant tumors by using
clinical variables such as the histologic grade of the tumor, local behavior, and metastatic spread. In bone and
soft-tissue sarcomas, staging systems have also been designed as preoperative aides to define the surgical planning for optimal local control. For benign tumors of
bone, the Enneking system is commonly used and divides tumors into one of three categories: inactive, active, and aggressive. To address malignant bone lesions,
Enneking developed a surgical staging system based on
the grade of the tumor, intracompartmental or extracompartmental status, and metastatic disease (Figure 2).
Although this staging system has been adopted for the
staging of pelvic and extremity tumors, it has proved to
be less useful in the treatment of spinal tumors. A stag-

American Academy of Orthopaedic Surgeons

Chapter 48 Tumors of the Spine

Orthopaedic Knowledge Update 8

ing system for spinal tumors was developed based on
the lesions location and local extension (Figure 3). The
vertebral body is divided into twelve sectors similar to a
clockface, and the tumor location is plotted according to
this grid. Furthermore, five layers are described, beginning with the paraspinal soft tissues peripherally to the
intradural space centrally. The system is primarily descriptive, and it can therefore be for the staging of both
benign and malignant tumors. The Harrington staging
system has been applied to metastatic spinal tumors
(Figure 4).

Nonsurgical Treatment
The management of patients with benign primary tumors includes close clinical and radiographic observation. Bracing has a limited role in treatment, but it may
provide some symptomatic relief of pain. Specific strategies for certain tumors have been discussed individually.
Surgical excision should be considered when the patient
has evidence of neurologic deficit, tumor progression,
progressive deformity, or pain.
In patients with malignancies, chemotherapy plays
an important role in treating tumors such as osteosarcoma, Ewings sarcoma, and lymphoma for which chemotherapeutic regimens have proved to be systemically
effective. Chemotherapy can be used both preoperatively and postoperatively as an adjuvant agent, and in
some patients, it can even be used as a primary treatment to reduce tumor burden. Stem cell and bone marrow salvage is allowing higher doses of chemotherapy to
be administered for longer periods, which in turn is
more lethal to tumors. Radiation therapy, as with chemotherapy, can be used either as an adjuvant or primary
treatment. Patients with radiosensitive tumors (lung,
prostate, breast) who present early with metastatic disease without spinal instability or dense neurologic compromise can be managed with radiation therapy. In patients with higher tumor load requiring surgical
intervention, external beam irradiation can be used as
an adjuvant therapy either preoperatively or postoperatively. Recent advances in radiosurgery allow the delivery of a single, large dose of radiation (proton beam) to
a localized tumor using a stereotactic approach, resulting in precise delivery to the target. This may be useful
in the treatment of chordomas, recurrent tumors, or sarcomas without clear resection margins.

Surgical Treatment
The primary goals of surgery in patients with spinal tumors are to reduce pain, to preserve or restore neurologic function, and to establish spinal column structural
integrity. Less commonly, excision of the tumor in an attempt at a cure is performed. In patients with tumors
that do not respond to radiotherapy and chemotherapy,
earlier surgery to prevent impending vertebral collapse

American Academy of Orthopaedic Surgeons

Figure 3 Proposed system of surgical staging of spine tumors, defining the extent of
the tumor. A = extraosseous soft tissues, B = intraosseous (superficial), C = intraosseous (deep), D = extraosseous (extradural), E = extraosseous (intradural), and
M = metastasis. Numbers 1 through 12 represent location of tumor. (Reproduced with
permission from Boriani S, Weinstein JN, Biagini R: Primary bone tumors of the spine.
Spine 1997;22:1036-1044.)
and/or neurologic deterioration may be considered. Specific treatment considerations for the various types of
spinal tumors have been discussed. In patients with
highly vascular tumors such as renal cell carcinoma, preoperative tumor embolization may help reduce intraoperative bleeding. In general, the location and extent of
the tumor dictate the approach (anterior, posterior, or
combined), and the extent of destabilization caused by
the decompression or removal of tumor then determines the method of restabilization and reconstruction
of the spine. The immediate proximity of vital structures
(particularly the neural elements) to the tumor often
precludes wide surgical excision, and spinal tumors are
invariably treated with intralesional surgery. Such surgeries are therefore palliative, and are performed in an
attempt to improve the quality of the patients life.

Spinal Stability
The definition of spinal instability in the setting of vertebral destruction by tumor remains elusive. Unlike long
bones, the spine may continue to exhibit a degree of
load bearing without catastrophic failure after fracture,
so that the concept of patients being at risk for pathologic fracture is not as intuitive for those with tumorrelated spinal instability. Although various criteria for
spinal stability in the face of spinal tumors have been
reported, they have not been particularly useful in predicting which patients may benefit from spinal surgery
before the development of profound instability or neurologic deficit.
Varied grading systems and parameters have been
devised to determine when surgical intervention is re-

593

Tumors of the Spine

Orthopaedic Knowledge Update 8

Figure 4 Illustration of Harringtons classification of spinal metastases. Numbers

1 through 4 represent the anatomic location of tumor. A = tumor confined to the
vertebra; B = tumor extends beyond the
vertebra; C = distant metastases. (Reproduced from Beaty JH (ed): Orthopaedic
Knowledge Update 6. Rosemont, IL, American Academy of Orthopaedic Surgeons,
1999, pp 723-736.)
quired to treat impending instability as a result of
pathologic vertebral replacement. In the thoracic spine,
50% to 60% of vertebral body involvement or 25% to
30% of body involvement with costovertebral involvement have been suggested as criteria for impending collapse. In the lumbar spine the suggested criteria for impending collapse are 35% to 40% of body involvement
alone or 25% of body involvement with pedicle or posterior body involvement. Even when adhering to these
guidelines, each patient should be assessed individually,
taking tumor load, type, location, and prognosis of the
patient into account. More recently, finite element models have shown that tumor size, magnitude of spinal
loading, and bone density are predictive of the initiation
of burst fractures in metastatically affected vertebrae
and that pedicle involvement and disk degeneration are
less important.

Radiation Treatment
In patients with spinal tumors causing neural compression, treating physicians are required to weigh the relative advantages of initiating treatment with radiation
therapy against primary surgical decompression. If the
tumor is radiosensitive and neural progression is gradual, radiotherapy may be the initial treatment of choice.
Patients with limited ambulatory function as a result of
tumor causing neural compression have a 60% chance
of improvement after undergoing radiation treatment.
Patients who have lost sphincter function have less than
a 40% chance of regaining function after radiation
treatment. When spinal radiation is ineffective in improving neurologic deficits, subsequent surgical decompression is fraught with complications. Operating
through a radiated field will increase the risk of wound
nonhealing and surgical site infections. In some patients,
radiation injury to the skin in the surgical field may prevent or delay surgical intervention. In addition, the potential advantages of initiating radiation treatments after surgical resection (when the tumor volume has been
significantly reduced) must be considered.

Surgical Approach
Performing an anterior and posterior surgery when either approach alone would suffice exposes the patient

594

to unnecessary risks, morbidity, and a prolonged recovery. In patients for whom posterior stability is maintained, reconstructing the anterior spinal column with
anterior surgery only after tumor removal may be all
that is necessary. The corollary to this is that performing
too little surgery does the patient a similar disservice.
In patients with significant comorbidities and reduced
life expectancy, it is tempting to perform a lesser procedure. If however the patient does not leave the operating room with effective neural decompression and spinal stability, the likelihood of early or late failure is
great. Having to return a patient to surgery after a
failed initial surgical procedure is an extremely undesirable and potentially avoidable situation.
In general, the site of the lesion dictates the surgical
approach. When the vertebral body is involved and neural compression is anterior, an anterior approach will allow for tumor resection and direct decompression of the
neural elements. Anterior reconstruction, usually involving a strut spanning the excised level or levels is required to restore spinal stability. The involved segments
are typically further stabilized with rigid anterior instrumentation. If the morbidity of an anterior approach is
prohibitive, vertebral body excision and anterior neural
decompression may be accomplished through a posterolateral approach involving removal of the pedicles so
that the vertebral body can be accessed lateral to the
dura. A posterior approach is also preferred when the
tumor involves the posterior elements (Figure 5). The
posterior approach will also allow for multilevel posterior segmental fixation to stabilize the spine, whereas
anterior approaches tend to allow for stabilization of
fewer segments. If both the vertebral body and posterior
elements of the spine are involved, posterior instrumentation in addition to anterior decompression-reconstruction will be necessary to stabilize the involved
spine (Figure 6). Isolated laminectomy has a small role
in the treatment of spinal tumors.
Optimal timing of treatments such as chemotherapy
and radiation therapy after spinal surgery remains
poorly defined. These treatments will interfere with
wound healing as well as with bone graft incorporation
and fusion. Animal research suggests that radiation therapy should be delayed for 6 weeks after spinal recon-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 48 Tumors of the Spine

Figure 5 AP (A) and lateral (B) radiographs

of a 29-year-old man with a 2-year history of
slowly progressive low back pain; the pain occasionally radiated to the right thigh, but no
paresthesias nor weakness was reported. The
AP radiograph is significant for a classic winking owl sign at the right pedicle of L4 (arrow).
C, A subsequent CT scan reveals an expansile
lesion involving the right pedicle and transverse process of L4 that contains a limited
amount of intralesional mineralization. D, MRI
further demonstrates the extent of the lesion
and suggests that the lesion may be compressing the L3 nerve root. No central compression was present. Clinical history and imaging suggested a benign lesion, although a
precise diagnosis was not possible. Osteoblastoma was considered to be the most likely diagnosis. E, Histologic analysis of the pathology specimen contained polygonal
chondroblasts with numerous giant cells, indicating that the lesion was, in reality, a chondroblastoma. F, Postoperative AP and lateral
radiographs show that an excision was performed with posterior instrumented L3-L5 fusion.
struction involving arthrodesis to permit the critical
early phases of bone graft revascularization. Obviously,
decisions regarding the timing of adjuvant treatments
need to be made by taking into account the likely
course of the patients underlying malignancy.

En Bloc Tumor Resection

Surgery for spinal tumors typically involves intralesional
excision with debulking followed by adjuvant therapy.

American Academy of Orthopaedic Surgeons

This has been reported to provide palliation in many series; however, the recurrence rates have been quite high.
This has prompted interest in en bloc spondylectomy
for the treatment of patients with solitary metastases or
intracompartmental primary malignant tumors. In case
reports, en bloc resection or total spondylectomy has
been suggested to decrease recurrence and improve survival. En bloc spondylectomy usually involves removal
of the posterior elements and osteotomy of the pedicles,
followed by posterior stabilization with pedicle fixation

595

Tumors of the Spine

Orthopaedic Knowledge Update 8

Figure 6 AP (A) and lateral (B) radiographs of a 44-year-old man with a

several-month history of progressive neck
pain, no radicular symptoms, and no upper motor neuron signs at presentation
demonstrate a lesion (arrow) involving the
body and posterior elements of the C3
vertebra with soft-tissue extension. C, Axial CT demonstrates the mass and suggests that it contains mineralized matrix.
D, MRI further delineates the extent of the
tumor and reveals that there is no spinal
cord compression. The patient underwent
a CT-directed biopsy of the C3 mass, and
the results of histologic analysis (E) were
most consistent with the presence of a
low-grade chondrosarcoma. Because wide
resection was not possible because of the
proximity of the mass to both vertebral arteries and the spinal cord, the patient underwent preoperative radiotherapy, followed by a front-back resection of C3 with
an anterior cage/C2-C4 plate and C2-C5
posterior instrumentation. Intraoperative
radiotherapy was administered to the anterior dura, and the patient underwent
proton radiotherapy postoperatively.
F, Postoperative radiograph.

one to two levels above and below the resected level.

This is followed by the en bloc removal of the vertebral
body and reconstruction of the anterior column, which
can be accomplished either through the same posterior
surgical approach or through a separate anterior approach.

vere pain and preventing neurologic impairment, including retention of bowel and bladder control. Recent studies have demonstrated that in appropriately selected
patients, surgical management is able to positively affect
the overall quality of life in those with spinal metastases.

Metastatic Disease

Vertebral Augmentation

Diffuse metastatic disease still poses a significant challenge for the spinal surgeon, and the combination of debulking, intralesional excision, and spinal reconstruction
is the mainstay of treatment. The surgical approach is either anterior, posterior, or both, depending on the location of the tumor. Even though surgery is palliative, it
can enhance a patients quality of life by relieving se-

Vertebroplasty and kyphoplasty have shown promise in

the treatment of patients with multiple myeloma or spinal metastases. These percutaneous modalities have
been widely used for the treatment of benign osteoporotic fractures and are increasingly being used to
treat pathologic compression fractures caused by metastatic disease and vertebral body pain secondary to tu-

596

American Academy of Orthopaedic Surgeons

Chapter 48 Tumors of the Spine

Orthopaedic Knowledge Update 8

mor infiltration. The indications for these techniques include refractory pain without neurologic compromise.
The posterior vertebral body wall must be intact to prevent extravasation of cement into the spinal canal.
Short-term results appear promising, with good and
rapid relief of pain, improved function, and even restoration of vertebral body height with kyphoplasty in
some patients. Most reported complications in patients
who have undergone vertebroplasty have been the result of extravertebral cement extravasation. This risk is
higher when treating patients with metastatic as opposed to osteoporotic vertebral fractures. Kyphoplasty,
which involves the creation of an intravertebral cavity
with a balloon tamp, has a lower risk of extravertebral
cement extravasation than vertebroplasty. Although
these minimally invasive treatments may not impact the
prognosis or the disease process, they can provide substantial and immediate pain relief with a reported low
risk of complications.

Annotated Bibliography
Primary Tumors
Abe E, Koboyashi T, Murai H, et al: Total spondylectomy for primary malignant, aggressive benign and solitary metastatic bone tumors of the thoracolumbar spine.
J Spinal Disord 2001;14:237-246.
Fourteen patients with malignant or aggressive benign vertebral tumors of the thoracolumbar spine underwent total
spondylectomy. Pain relief was achieved in all 14 patients, and
no serious complications occurred. Local recurrence was
found in three patients at a mean follow-up of 3.2 years. Total
spondylectomy appears to be an effective method for controlling local recurrence without major complications.

Durr HR, Wegener B, Krodel A, Muller PE, Jansson V,

Refior HJ: Multiple myeloma: Surgery of the spine. Retrospective analysis of 27 patients. Spine 2002;27:320-324.
This is a report on the clinical course of 27 consecutive patients who were surgically treated for solitary or multiple myeloma of the spine. Life quality, reported to be 48% before
surgery, improved to 59% 1 month after surgery and to 73%
in 24 survivors after 1 year. The authors concluded that surgical treatment of myeloma of the spine seemed to be an effective method of treatment with respect to neurologic function
and life quality in selected patients.

Nonsurgical Treatment
Ryu S, Fang YF, Rock J, et al: Image-guided and
intensity-modulated radiosurgery for patients with spinal metastasis. Cancer 2003;97:2013-2018.
In this study, 10 patients with spinal metastasis were
treated with image-guided and intensity-modulated radiosurgery. The authors reported that the most patients had pain relief within 2 to 4 weeks of treatment.

Surgical Treatment
Dudeney S, Lieberman IH, Reinhardt MK, Hussein M:
Kyphoplasty in the treatment of osteolytic vertebral
compression fractures as a result of multiple myeloma.
J Clin Oncol 2002;20:2382-2387.
Fifty-five consecutive kyphoplasties were prospectively
evaluated in 18 patients. Mean follow-up was 7.4 months, and
there was significant improvement in Short Form-36 scores for
bodily pain, physical function, vitality, and social function.

Fourney DR, Schomer DF, Nader R, et al: Percutaneous

vertebroplasty and kyphoplasty for painful vertebral
body fractures in cancer patients. J Neurosurg 2003;98:
21-30.
In this study, 97 procedures (65 vertebroplasty and 32 kyphoplasty) were performed in 56 patients with myeloma and
primary malignancies. The authors report that 84% of patients
experienced complete pain relief at a median follow-up of 4.5
months.

Krepler P, Windhager R, Bretschneider W, et al: Total

vertebrectomy for primary malignant tumors of the
spine. J Bone Joint Surg Br 2002;84:712-715.
Vertebrectomy was performed on seven patients, with a
mean follow-up of 52 months. In five patients, a wide resection
was achieved. The authors conclude that vertebrectomy seems
to be an appropriate procedure for the treatment of primary
lesions of the spine.

Sundaresan N, Rothman A, Manhart K, Kelliher K: Surgery for solitary metastases of the spine, rationale and
results of treatment. Spine 2002;27:1802-1806.
This is a retrospective review of 80 consecutive patients
with solitary sites of spine involvement from solid tumors. The
overall median length of survival after surgery was 30 months.
Complete surgical excision before irradiation was recommended to increase the prospects of palliation and possible
cure.

Ghanem I, Collet LM, Kharrat K, et al: Percutaneous

radiofrequency coagulation of osteoid osteoma in children and adolescents. J Pediatr Orthop B 2003;12:244252.

Wai EK, Finkelstein JA, Tangente RP, et al: Quality of

life in surgical treatment of metastatic spine disease.
Spine 2003;28:508-512.

In this study, 23 patients who underwent percutaneous radiofrequency coagulation for osteoid osteoma were retrospectively reviewed. Pain disappeared immediately after the procedure in 21 patients. At an average of 3.5-year follow-up, all
patients were pain free.

In this study, 25 consecutive patients undergoing surgery

for spinal metastases were prospectively evaluated. After surgery, the largest improvement was with pain, but there were
also improvements with tiredness, nausea, anxiety, drowsiness,
appetite, and well-being.

American Academy of Orthopaedic Surgeons

597

Tumors of the Spine

Whyne CM, Hu SS, Lotz JC: Burst fracture in the metastatically involved spine: Development, validation, and
parametric analysis of a three-dimensional poroelastic
finite-element model. Spine 2003;28:652-660.
A finite-element study and in vitro experimental validation was performed for an investigation of features that contribute to burst fracture risk. The authors report that the principal factors affecting the initiation of fracture were tumor
size, magnitude of spinal loading, and bone density.

Classic Bibliography
Bohlman HH, Sachs BL, Carter JR, Riley L, Robinson
RA: Primary neoplasms of the cervical spine. J Bone
Joint Surg Am 1986;68:483-494.
Boriani S, Biagini R, De Iure F, et al: En bloc resections
of bone tumors of the thoracolumbar spine: A preliminary report of 29 patients. Spine 1996;21:1927-1931.
Galasko CS, Norris HE, Crank S: Spinal instability secondary to metastatic cancer. J Bone Joint Surg Am 2000;
82:570-594.
Hart RA, Boriani S, Biagini R, Currier B, Weinstein JN:
A system for surgical staging and management of spine
tumors: A clinical outcome study of giant cell tumors of
the spine. Spine 1997;22:1773-1783.
Harrington KD: The use of methylmethacrylate for
vertebral-body replacement and anterior stabilization of
pathologic fracture-dislocations of the spine due to metastatic malignant disease. J Bone Joint Surg Am 1981;63:
36-46.
Harrington KD: Anterior decompression and stabilization of the spine as a treatment for vertebral collapse
and spinal cord compression from metastatic malignancy. Clin Orthop 1988;233:177-197.
Hekster RE, Luyendijk W, Tan TI: Spinal cord compression caused by vertebral haemangioma relieved by per-

598

Orthopaedic Knowledge Update 8

cutaneous catheter embolisation. Neuroradiology 1972;
3:160-164.
Kostuik JP, Errico TJ, Gleason TF, Errico CC: Spinal stabilization of vertebral column tumors. Spine 1988;13:
250-256.
Marsh BW, Bonfiglio M, Brady LP, Enneking WF: Benign osteoblastoma: Range of manifestations. J Bone
Joint Surg Am 1975;57:1-9.
Pettine KA, Klassen RA: Osteoid osteoma and osteoblastoma of the spine. J Bone Joint Surg Am 1986;68:
354-361.
Samson IR, Springfield DS, Suit HD, Mankin HJ: Operative treatment of sacrococcygeal chordoma: A review
of twenty-one cases. J Bone Joint Surg Am 1993;75:14761484.
Shives TC, Dahlin DC, Sim FH, Pritchard DJ, Earle JD:
Osteosarcoma of the spine. J Bone Joint Surg Am 1986;
68:660-668.
Stener B: Complete removal of vertebra for extirpation
of tumors: A 20 year experience. Clin Orthop 1989;245:
72-82.
Sundaresan N, Galicich JH, Lane JM, et al: Treatment of
neoplastic epidural cord compression by vertebral body
resection and stabilization. J Neurosurg 1985;63:676-684.
Sundaresan N, Steinberger AA, Moore F, et al: Indications and results of combined anterior-posterior approaches for spine tumor surgery. J Neurosurg 1996;85:
438-446.
Yuh WT, Quets JP, Lee HJ, et al: Anatomic distribution
of metastases in the vertebral body and modes of hematogenous spread. Spine 1996;21:2243-2250.

American Academy of Orthopaedic Surgeons

Chapter

49

Spondyloarthropathy
Tushar Patel, MD
Mark J. Romness, MD

Introduction
Many inflammatory diseases are known to occur in the
musculoskeletal system, but the primary inflammatory
conditions that affect the spine, in descending prevalence, are rheumatoid arthritis, ankylosing spondylitis,
and juvenile rheumatoid arthritis. These three inflammatory conditions alone are estimated to affect over 3 million patients in the United States. Juvenile rheumatoid
arthritis is discussed in chapter 21.

Rheumatoid Arthritis
Definition
Rheumatoid arthritis is a systemic autoimmune disease
of unknown etiology that causes progressive joint swelling, pain, and stiffness secondary to synovitis. The onset
of symptoms usually occurs between 20 and 45 years of
age. Seventy percent of patients with rheumatoid arthritis are female. A genetic component continues to be
supported but is not well defined. A positive rheumatoid factor is present in approximately 85% of patients,
but is not specific for rheumatoid arthritis and can be
present in normal individuals and in other medical conditions. A positive rheumatoid factor is clinically associated with more severe symptoms, including extra-

articular manifestations and a more aggressive disease

course. Patients with rheumatoid arthritis who have a
negative rheumatoid factor may eventually convert to a
positive rheumatoid factor and follow a similar clinical
progression as patients who are initially positive. Severe
spinal involvement is primarily cervical in patients with
rheumatoid arthritis, and up to 85% of patients with
rheumatoid arthritis develop cervical radiographic
changes within 10 years of disease onset. Lumbar problems are less significant, but low back pain was present
in 40% and radiographic evidence of lumbar pathology
was found in 57% of 106 patients with rheumatoid arthritis. The primary spinal deformities in rheumatoid arthritis are atlantoaxial subluxation, superior migration
of the odontoid, and subaxial subluxation (Figure 1). All
of these conditions can lead to neurologic impairment.
Superior migration of the odontoid is also known as
basilar invagination, cranial settling, and atlantoaxial impaction.

Natural History
The progressive nature of rheumatoid arthritis is well
known, as is the variable expression of the disease in
different patients. Several studies have attempted to cor-

Figure 1 Lateral radiograph (A) and corresponding sagittal MRI (B) showing the
combined manifestations of cranial settling, atlantoaxial subluxation, and subaxial subluxation.

American Academy of Orthopaedic Surgeons

599

Orthopaedic Knowledge Update 8

Spondyloarthropathy

myelopathy secondary to irreducible atlantoaxial subluxation, all were confined to bed within 3 years and
died within 8 years.

Imaging
Plain Radiographs
Plain radiographs remain the standard modality to diagnose, classify, and monitor spinal arthropathy. Plain images provide cost-effective screening for skeletal
changes. Early radiographic changes including the presence of osteophytes and disk space narrowing may not
be symptomatic. Flexion and extension lateral views are
used to define stability between the spinal segments and
are the primary method to assess the cervical spine for
the primary spinal deformities that occur in rheumatoid
arthritis. The use of select digital imaging and software
allows electronic measurements such as the atlanto-dens
interval.
Magnetic Resonance Imaging
MRI provides information on soft tissues and structural
bone details not seen with conventional radiography. Tomographic representation clarifies anatomic structures,
and the use of gadolinium enhancement has been shown
to increase the sensitivity of diagnosis and to help in determining the extent of the disease. Dynamic studies are
possible with MRI (Figure 2), but when dynamic MRI
and plain flexion-extension films were compared in 23
patients with rheumatoid arthritis, the magnitude of atlantoaxial subluxation was less with MRI in all patients;
MRI was not able to detect atlantoaxial subluxation
that was seen in four patients using radiography. MRI is
required for evaluation of neurologic structures and is
the standard method for evaluation of superior migration of the odontoid. Correlation of cord compression
or impingement at the atlantoaxial level on initial MRI
and development of subarachnoid space encroachment
on sequential MRI scans have both been shown to be
predictive of neurologic deterioration.

Figure 2 Dynamic flexion-extension MRI panels showing evidence of dynamic cord

compression at the occipitocervical junction.

relate serologic tests and radiographic changes with cervical progression. Rheumatoid factor-positive serology,
more extensive peripheral joint involvement, male gender, and corticosteroid use are factors that have been
linked to greater cervical involvement. Cervical myelopathy has been associated with progression to bed confinement or death within an average of 35 months if untreated in patients with rheumatoid arthritis and
extensive mutilating-type changes of the peripheral
joints. In 21 untreated patients with rheumatoid arthritis

600

CT Myelogram and Bone Scan

The use of CT myelograms is beneficial for unique situations such as for patients who had previous surgery
and who have ferromagnetic implants in the area to be
imaged, or for those who are not candidates for MRI
because of cochlear implants, pacemakers, or neurostimulators.Nuclear imaging is a sensitive screening tool for
identifying occult fractures in patients with rheumatoid
arthritis. The advantage over MRI is the ability to image
the whole body efficiently and cost-effectively.
Bone Density Studies
Osteoporosis is common in most forms of spinal arthropathy. Contributing factors include the disease process itself, genetics, limitations in activity, nutrition, and

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 49 Spondyloarthropathy

treatments such as the use of corticosteroids. However,

the presence of osteoporosis is not limited to patients
with severe disease or those treated with corticosteroids.
Bone density studies quantitate the extent of osteoporosis and help determine whether treatment is indicated.

Indications for Surgery

Neurologic impairment is the most urgent indication for
surgery in patients with spinal arthropathy. Impairment
can be determined by history, clinical examination, electrodiagnostic studies, or MRI. Because progression of
myelopathy is the natural history in rheumatoid arthritis, any myelopathic findings should be considered a surgical indication.
Pain and instability often are associated with neurologic involvement, except in patients with gradual displacement over time. Pain alone is a subjective indication; instability without neurologic deficit has specific
parameters to warrant surgery. These parameters include posterior atlanto-dens interval of 14 mm or less in
flexion on plain radiographs for atlantoaxial subluxation
and MRI findings of cord diameter of less than 6 mm in
flexion, cervicomedullary angle of less than 135, and 13
mm or less of space available for the spinal cord. Superior migration of the odontoid alone is not an indication
for surgery and documentation of neurologic impingement or myelopathy on flexion MRI or by examination
should be documented before surgery is recommended.
Spinal fractures in patients with rheumatoid arthritis
usually require fixation, and fixation needs to compensate for the often-associated osteoporosis and accompanying deformity. This may require fusion and instrumentation of more levels than would ordinarily be
considered for patients with normal bone quality.

Surgical Treatment
Decompression
Myelopathy may require decompression if the deformity is fixed, but reduction of the deformity may preserve bony areas for fusion healing. Solid fusion prevents spinal cord irritation and has been shown to
relieve myelopathy. Anterior transoral decompression
was previously recommended for spinal cord compression, but the only current indication for transoral decompression is the presence of cranial nerve deficits.
Arthrodesis
Arthrodesis is the treatment of choice for either primary instability resulting from inflammatory process or
secondary instability caused by late manifestations of
pseudarthrosis or sterile spondylodiskitis. Fusion to the
occiput is indicated when adequate fixation at C1 is not
possible or when there is instability at the occiputC1
junction.

American Academy of Orthopaedic Surgeons

Figure 3 An example of posterior occiputocervical fusion construct.

Instrumentation
Wire fixation is extensively used for stabilization in the
cervical spine and has been described in the literature.
Additional methods for instrumentation of the cervical
spine have been recently described. At the occipital
area, cranial bolts placed using the inside-outside technique and connected to plates and lateral mass screws
were used for 21 patients with rheumatoid arthritis; results showed no instrument failures and significantly reduced pain scores. Cadaver studies have shown increased axial rotation stiffness with rod and screw
fixation (Figure 3) compared with plate and screw or
loop and cable constructs, and better restoration of sagittal plane balance and maintenance of correction.
The use of screws throughout the spine has increased treatment options. Transarticular C1-2 fixation,
and more recently, segmental fixation of C1 to C2 using
polyaxial screws inserted into the lateral masses of C1,
has replaced the requirement of fusion from the occiput
to C2 (or lower) in most instances when C1 laminectomy is performed. The technique of placing screws into
the lateral mass of C1 increases the risks of bleeding
from the venous plexus accompanying the dorsal ramus
of the C2 nerve root as well as direct injury to the root,
but affords additional safety compared with the use of

601

Orthopaedic Knowledge Update 8

Spondyloarthropathy
transarticular screws because the path of the vertebral
artery is avoided.

Surgical Outcomes
The benefits of surgical intervention have been well
documented and recent outcomes studies have supported this treatment option. For irreducible atlantoaxial instability with myelopathy in patients with rheumatoid arthritis, C1 laminectomy and fusion with
rectangular rod fixation from the occiput to C2 (16 patients) to C3 (2 patients) or to C4 (1 patient) led to 68%
improvement of at least one Ranawat class and 5- and
10-year survival rates of 84% and 37%, respectively.
Without surgery, 21 matched patients were all bedridden
within 3 years and died within 8 years. Similar neurologic improvements were noted in 67% of patients (37
of 55) with occipital neuralgia, myelopathy, or both
treated with various procedures. Mortality at 2 years
was 27% and resulted mainly from nonsurgery-related
causes.
Patients with inflammatory spondylitis often have
multiple comorbid medical conditions that contribute
significantly to management and potential complications. Involvement of the temporomandibular joint may
interfere with intubation, and peripheral joint contractures can make surgical positioning and rehabilitation
more challenging. Osteoporosis impairs fixation potential, and many authors recommend the use of orthotic
devices after surgery if osteoporosis is present. Poor
skin quality, impaired healing, and dental problems increase the risk of infection. Despite all the potential
risks, careful patient preparation and planning combined with up-to-date skilled care can improve function
and survival.

Ankylosing Spondylitis
Definition
Ankylosing spondylitis is the prototype disease of seronegative spondyloarthropathiesa family of arthritic
conditions not associated with positive rheumatoid factor serology. Other related conditions are reactive arthritis (Reiters syndrome), spondylitic forms of psoriatic arthritis and inflammatory bowel disease (Crohns
disease), juvenile spondyloarthropathy, and undifferentiated spondyloarthropathy. Disease onset typically occurs between the ages of 20 and 30 years. Males are
more commonly affected, and symptoms in females are
usually milder and harder to diagnose. Early symptoms
of ankylosing spondylitis are commonly overlooked in
young adults as simply back pain, but awareness of
early findings such as lumbar stiffness and decreased
lordosis may lead to a correct diagnosis. Positive HLAB27 assay is present in 80% to 98% of patients with
ankylosing spondylitis versus 8% of the general population. Spinal involvement is primarily lumbar at onset

602

and spreads cephalad. Neck pain caused by spontaneous

anterior atlantoaxial subluxation has been reported as
an initial symptom. One study showed that 49% of patients (89 of 181) with ankylosing spondylitis had radiographic changes, but atlantoaxial subluxation was
present in 2% to 20% of patients with ankylosing
spondylitis compared with 16% to 25% of adult patients
with rheumatoid arthritis; superior migration of the odontoid has not been reported with ankylosing spondylitis alone.

Natural History
The natural history of ankylosing spondylitis is progressive loss of posture and mobility. The rate of progression
is variable. Mild spinal changes in ankylosing spondylitis
occur first, with squaring of the lumbar bodies and syndesmophytes. More severe involvement leads to changes
characteristic of so-called bamboo spine. Spinal deformity in ankylosing spondylitis develops secondary to
limitation of motion at ankylosed areas or increased
motion at nonankylosed areas. Pain can develop at remaining motion segments but resolves once that segment is ankylosed. Ankylosis in a relatively kyphotic position occurs at the cervical, thoracic, and lumbar spine,
but etiology for kyphosis has not been defined.
With long rigid spinal segments and focused areas of
osteoporosis, fractures are common even with minimal
trauma and can cause significant morbidity and mortality. Mortality can occur from initial injury, fracture treatment, or epidural hematomas that can develop up to
weeks after injury. Recent articles on trauma have emphasized the importance of strict neck immobilization
for patients with ankylosing spondylitis, even in those
who have experienced mild low-energy injuries.

Imaging
Plain Radiographs
Plain radiographs may provide clues for the initial diagnosis of ankylosing spondylitis. Classic changes at the
sacroiliac joints may not be identified initially, but initial
presentation with thoracic spine changes in a female patient has been reported. Spondylodiskitis (also known as
Andersson lesions) is a destructive diskovertebral lesion
occurring in ankylosing spondylitis that usually occurs at
the thoracolumbar junction. Both inflammatory and
noninflammatory types of destructive lesions occur. The
inflammatory type is defined radiologically as a reduced
disk space with a defect of vertebral bodies and dense
cancellous bone sclerosis. The noninflammatory type is
associated with a fracture through an ankylosed disk or
a pseudarthrosis. Lateral flexion and extension views of
the cervical, thoracic, and lumbar spine are essential to
identifying instability and ankylosis in patients with
ankylosing spondylitis.

American Academy of Orthopaedic Surgeons

Chapter 49 Spondyloarthropathy

Orthopaedic Knowledge Update 8

Computed Tomography and Magnetic Resonance
Imaging
Excellent evaluation of the sacroiliac joints is possible
with CT. Sacroiliac joint space narrowing, erosions, and
sclerosis are all classically seen on the axial views without reconstruction or contrast. Canal stenosis secondary
to the combined effects of kyphosis and ankylosis is well
defined with MRI. Contrast gives better definition of
which structures are involved and if spinal cord pathology is present.
Bone Scan
The presence of activity in the sacroiliac joints and in
the spinal column on bone scans may help with the diagnosis of ankylosing spondylitis. Nuclear imaging is an
effective screening tool to identify occult fractures in
ankylosing spondylitis and should be considered early in
the assessment of patients with this disease who sustain
even minor trauma. There have been reports of devastating complications from undiagnosed fractures.
Bone Density Studies
A decrease in bone density has been shown to be related to persistent systemic inflammation in patients
with ankylosing spondylitis. Spinal density measured by
dual-energy x-ray absorptiometry (DEXA) increases as
ankylosis progresses; therefore, DEXA hip measurements or quantitative CT should be used.

Surgical Indications
Neurologic compromise may be less obvious in patients
with ankylosing spondylitis than in those with rheumatoid arthritis. A recent meta-analysis of 52 articles discussing 86 patients with ankylosing spondylitis and
cauda equina syndrome showed that nearly all patients
had sensory, motor, or reflex deficits on examination
and 30% had previous prostate surgery without improvement for incontinence. Compensations in gait
caused by spinal ankylosis may mask the true myelopathic contribution to gait. As myelopathy is associated
with rheumatoid arthritis, any myelopathic findings
should be considered a surgical indication.
There are no specific indications for correction of
chin on chest deformity from cervical kyphosis; however, commonly accepted indications for surgical intervention include the inability to perform the activities of
daily living (including those related to personal hygiene). Correction of thoracolumbar kyphosis was recommended if the global kyphosis was greater than 50,
or in patients with less severe kyphosis in whom nonsurgical management of symptomatic spondylodiskitis was
unsuccessful.
Fractures associated with ankylosing spondylitis often are associated with neurologic deficit that rarely re-

American Academy of Orthopaedic Surgeons

solves but can improve with surgery. In 11 patients with

ankylosing spondylitis and fractures, 6 had neurologic
deficits; 3 of 6 patients had improved neurologic function with surgery. Most fractures are unstable because
the ankylosis forms a rigid column that does not follow
the same three-column criteria of Denis for stability as
seen in normal spines. The rigid columns act as long
level arms concentrating forces (such as motion) at the
fracture site. Epidural hematomas commonly produce
neurologic impairment because of the confined rigid canal. Surgery for unstable fractures and epidural hematomas is required and combined anterior and posterior
approaches and extended fixation must be considered.
Only minor nonstructural fractures should be considered stable fractures that can be treated nonsurgically in
patients with ankylosing spondylitis.

Surgical Treatment
Decompression
Emergency decompression is indicated for epidural hematomas. Laminectomy is required for cauda equina
syndrome in patients with ankylosing spondylitis. A
study found that 55 patients who had nonsurgical treatment showed no improvement of sensory, bowel, or
bladder deficit; only 2 of these patients had motor improvement.
Osteotomy
Extension osteotomy to correct kyphosis at both the
cervicothoracic junction and lumbar spine is well described in patients with ankylosing spondylitis. Cervical
deformity is best corrected with osteotomy between C7
and T1, which is the widest area of the cervical canal
and is caudal to the vertebral artery entry at C6. Cervical osteotomies above C7 are rarely required and have a
higher risk of complications.
Thoracic kyphosis is best treated with extension osteotomy at or below L2. Thoracic osteotomies are rarely
indicated. Patients who have osteotomy at L4 may have
difficulty sitting on the floor; however, the procedure
may need to be considered if two osteotomies are being
performed. Two levels of osteotomy are recommended
for deformity greater than 70 and should be separated
by one or two levels to distribute the amount of anterior
angulation and distraction.
Arthrodesis and Instrumentation
Arthrodesis is required for fractures and after kyphectomy. Fusion rates are similar to those seen in the normal spine. Modifications to the standard techniques of
instrumentation may be necessary because of the obliteration of bony landmarks that ordinarily serve as reference points.

603

Spondyloarthropathy

Surgical Outcomes
Correction of kyphosis by posterior subtraction osteotomy has previously been described and reported. In a
recent large series, 92 osteotomies were done in 78 patients with mean correction of 34.5% per osteotomy level; the maximum correction attained was 100. Neither
deaths nor vascular complications were noted compared
with previously reported mortality of up to 10% with
anterior opening wedge osteotomy. Loss of correction
occurred in only two patients. Patients reported outcome as excellent (83%) or good (15%); one patient
outcome was not reported.

Annotated Bibliography
Ahn NU, Ahn UM, Nallamshetty L, et al: Cauda equina
syndrome in ankylosing spondylitis (the CES-AS syndrome): Meta-analysis of outcomes after medical and
surgical treatments. J Spinal Disord 2001;14:427-433.
A case report of acute onset of cauda equina syndrome associated with ankylosing spondylitis is presented and a review
of 52 articles with 86 patients is evaluated regarding treatment
outcome. Onset of symptoms was found to be gradual. Only
22% of the patients had radicular symptoms, yet nearly all patients had some neurologic deficit on physical examination.
Thirty percent of the male patients had undergone prostatectomy for misdiagnosed prostatic hypertrophy. Improvement of
sensory, bowel, and bladder dysfunction was not noted with
nonsurgical treatment, but in 6 of the 15 patients who had surgery, some form of improvement was shown. Neurologic deficit progressed much less in the surgically treated group.

Chen IH, Chien JT, Yu TC: Transpedicular wedge osteotomy for correction of thoracolumbar kyphosis in
ankylosing spondylitis: Experience with 78 patients.
Spine 2001;26:E354-E360.
A retrospective study of a single surgeons experience is
evaluated. Modifications to Thomasens original description
for closing wedge osteotomy are included. With a mean
follow-up of 3.8 years, excellent results were noted in 83% of
patients and good results in 15%. Only two patients had a loss
of correction within the osteotomy segment. Average correction of the osteotomy was 34.5.

Chou LW, Lo SF, Kao MJ, Jim YF, Cho DY: Ankylosing
spondylitis manifested by spontaneous anterior atlantoaxial subluxation. Am J Phys Med Rehabil 2002;81:952955.
A case report of spontaneous anterior atlantoaxial subluxation is presented. There was no history of trauma or back
pain and no alteration of bowel or bladder function. Examination revealed severe cervical but only mild lumbar spine mobility limitations. The patient underwent C1 laminectomy and
occiput to C2 fusion and fixation and had almost complete
resolution of his neurologic symptoms.

604

Orthopaedic Knowledge Update 8

Kawaguchi Y, Matsuno H, Kanamori M, Ishihara H,
Ohmori K, Kimura T: Radiologic findings of the lumbar
spine in patients with rheumatoid arthritis, and a review
of pathologic mechanisms. J Spinal Disord Tech 2003;16:
38-43.
In this study, 106 patients were evaluated by questionnaire
and lumbar spine radiographs. A scoring system was used for
the radiographs and consensus of significant changes by three
observers was required. Low back pain was noted by 40% of
the patients and was believed to be severe by 3%. Leg pain
and leg numbness were noted in 18% and 14%, respectively.
Significant radiologic findings were noted in 57% of patients.
In order of frequency, these findings were disk space narrowing, postural anomaly, olisthesis, end of plate erosion, and facet
erosion. End plate erosion correlated closest to clinical symptoms.

Kurugoglu S, Mihmanli I, Kanberoglu K, Kanberoglu A:

Destructive diskovertebral lesions in ankylosing
spondylitis: Appearance on magnetic resonance imaging. South Med J 2001;94:837-841.
A case report of an 18-year-old woman with initial complaints of back and chest pain is presented. The patient had no
restriction in motion or neurologic symptoms. Diskovertebral
lesions were found diffusely throughout the spine. Involvement of the sacroiliac joint was found 6 months later by nuclear scintigraphy and CT scan. HLA-B27 was positive.

Laiho K, Soini I, Kautiainen H, Kauppi M: Can we rely

on magnetic resonance imaging when evaluating unstable atlantoaxial subluxation? Ann Rheum Dis 2003;62:
254-256.
Twenty-two patients with rheumatoid arthritis and one patient with juvenile idiopathic arthritis were evaluated for neck
pain. Evaluation included flexion and extension lateral cervical spine radiographs and functional cervical spine MRI. The
magnitude of anterior atlantoaxial subluxation in flexion was
greater in all patients measured by radiography than by MRI.
Seventeen percent of the flexion MRI scans did not show subluxation that was seen by radiography.

Matsunaga S, Sakou T, Onishi T, et al: Prognosis of patients with upper cervical lesions caused by rheumatoid
arthritis: Comparison of occipitocervical fusion between
c1 laminectomy and nonsurgical management. Spine
2003;28:1581-1587.
Treatment of myelopathy secondary to irreducible atlantoaxial dislocation was compared between 19 patients undergoing C1 laminectomy with occipital cervical fusion and instrumentation and 21 matched patients who were treated
nonsurgically. All patients were followed until their deaths,
which averaged 9.7 years in the surgically treated patients and
4.2 years in the nonsurgically treated patients. Pain improved
with all patients following surgery and neural improvement of
one or more Ranawat levels was found in 68% with worsening
in only 5%. The 10-year survival rate with surgery was 37%
and 0% without surgery.

American Academy of Orthopaedic Surgeons

Chapter 49 Spondyloarthropathy

Orthopaedic Knowledge Update 8

Reijnierse M, Dijkmans BA, Hansen B, et al: Neurologic
dysfunction in patients with rheumatoid arthritis of the
cervical spine: Predictive value of clinical, radiographic
and MR imaging parameters. Eur Radiol 2001;11:467473.
Forty-six patients with rheumatoid arthritis for at least
5 years were followed annually with examination, radiographs,
and MRI to correlate neurologic dysfunction with diagnostic
changes. All patients had a second examination and eight had
five sequential examinations. Subjective muscle weakness was
the only symptom associated with neurologic dysfunction. No
significant radiographic abnormalities could be statistically
correlated to neurologic dysfunction. The only MRI abnormality that was statistically significant was decreased subarachnoid space along the entire cervical spine, which was a subjective interpretation.

Sandhu FA, Pait TG, Benzel E, Henderson FC: Occipitocervical fusion for rheumatoid arthritis using the
inside-outside stabilization technique. Spine 2003;28:414419.
The technique using cranial bolts for occipital cervical stabilization is described, and the results in 21 patients who underwent a stabilization and fusion are presented. No implant
complications or failures occurred. Ranawat neurologic level
improved in 62% of the patients. There was no decline in neurologic level immediately following surgery or during the
follow-up period that averaged 25.5 months.

van Asselt KM, Lems WF, Bongartz EB, et al: Outcome

of cervical spine surgery in patients with rheumatoid arthritis. Ann Rheum Dis 2001;60:448-452.
Various types of cervical fusion in patients with rheumatoid arthritis and cervical myelopathy and/or occipital neuralgia were followed for 2 years. Mortality at 2 years was 27%
because of a variety of reasons. Of the surviving patients, 73%
had neurologic improvement at 3 months and 67% had an improved neurologic level at 2 years.

American Academy of Orthopaedic Surgeons

Classic Bibliography
Boden SD, Dodge LD, Bohlman HH, Rechtine GD:
Rhematoid arthritis of the cervical spine: A long term
analysis with predictors of paralysis and recovery.
J Bone Joint Surg Am 1993;75:1282-1297.
Conaty JP, Mongan ES: Cervical fusion in rheumatoid
arthritis. J Bone Joint Surg Am 1981;63:1218-1227.
Crockard HA: Surgical management of cervical rheumatoid problems. Spine 1995;20:2584-2590.
Lipson SJ: Rheumatoid arthritis in the cervical spine.
Clin Orthop 1989;239:121-127.
Morizono Y, Sakou T, Kawaida H: Upper cervical involvement in rheumatoid arthritis. Spine 1987;12:721725.
Neva MH, Kauppi MJ, Kautiainen H, et al: Combination
drug therapy retards the development of rheumatoid atlantoaxial subluxations. Arthritis Rheum 2000;43:23972401.
Rana NA: Natural history of atlanto-axial subluxation in
rheumatoid arthritis. Spine 1989;14:1054-1056.
Ranawat CS, OLeary P, Pellicci P, Tsairis P, Marchisello
P, Dorr L: Cervical spine fusion in rheumatoid arthritis.
J Bone Joint Surg Am 1979;61:1003-1010.
Santavirta S, Slatis P, Kankaanpaa U, Sandelin J, Laasonen E: Treatment of the cervical spine in rheumatoid
arthritis. J Bone Joint Surg Am 1988;70:658-667.
Thomasen E: Vertebral osteotomy for correction of kyphosis in ankylosing spondylitis. Clin Orthop
1985;194:142-152.

605

Chapter

50

Endoscopic and Minimally Invasive

Spine Surgery
Eeric Truumees, MD

Introduction
Minimally invasive and endoscopic spine surgery does
not refer to a single technique, but rather to a set of
tools in a continuum of less morbid approaches to the
treatment of spine problems. These tools continue to
emerge and will ultimately include injection techniques
(epidural injections, diskography, and facet and nerve
root blocks), percutaneous therapeutic modalities (intradiskal electrothermal therapy [IDET], vertebroplasty,
and kyphoplasty), true endoscopic procedures (endoscopic diskectomies, endoscopic lumbar fusions, and
endoscopic transthoracic procedures), image-guided surgery, bone substitutes and enhancers, nuclear replacement, and injection of growth factors.
There is no clear delineation between traditional
and minimally invasive spine surgery. Minimally invasive approaches reflect a trend in orthopaedic surgery to
closely target the pathology during a given therapeutic
intervention while minimizing damage to the surrounding tissues. Typically, these techniques represent new
ways to perform traditional surgical procedures, such as
instrumentation and fusion. Occasionally, newer surgical
interventions are introduced, such as percutaneous vertebral body polymethylmethacrylate augmentation.
Surgical indications are not changed by the way in
which the procedure is done. In considering the role of
minimally invasive spine surgery as part of a continuum
of spine care, it is useful to remember that the most
minimally invasive modality remains nonsurgical care.
Nonsurgical treatment is appropriate and effective for
most patients with degenerative conditions of the spine,
especially those with axial pain in the absence of neurologic dysfunction. For newer technologies, surgical indications are evolving.
These newer surgical technologies can be categorized by the spectrum of invasiveness each requires
from truly percutaneous, to endoscopic, to mini-open.
Spinal endoscopy refers to the use of an endoscope and
light source for visualization and magnification through
small percutaneous portals. Conceptually, endoscopy is
an attractive treatment option because the spinal col-

American Academy of Orthopaedic Surgeons

umn lies centrally in the body. Open surgical approaches

involve significant soft-tissue dissection, which affects
risk, recovery, and long-term function. Endoscopically
assisted and mini-open approaches rely on advances in
surgical lighting, corridor retractors, and image guidance
during standard operations with standard instruments to
perform spine surgery with less fascial plane violation
and dead space creation.
Most thoracic and lumbar spine procedures are performed posteriorly. Posterior approaches offer relatively
direct access to the bony elements and the spinal canal.
However, canal exposure may result in symptomatic
epidural fibrosis. The dissection and retraction of the
paraspinal muscles may lead to dead space formation
and extensor muscle disruption. Such disruption has
been referred to as fusion disease, which may be associated with early fatigability and other long-term symptoms. Less invasive posterior techniques may allow for
less disruption of the posterior musculature and a
smaller laminotomy.
Anterior fusion procedures have become more common in the treatment of symptomatic lumbar disk degeneration, spinal deformity correction, and the ablation
of tumors and infections. Traditional anterior approaches may require large incisions, rib resections, and
division of major muscle groups. Several endoscopic approaches, including transthoracic, transperitoneal, and
retroperitoneal approaches, have been described in an
effort to limit the morbidity associated with anterior
surgical approaches.

Percutaneous Disk Procedures

Percutaneous Diskectomy Techniques
Over time, various radiographically guided procedures
have been developed as both diagnostic and therapeutic
interventions that can significantly limit perioperative
risk and morbidity. Chymopapain, a proteolytic enzyme
obtained from a papaya extract, has been used as a percutaneous treatment modality for disk herniations. The
enzyme hydrolyzes the nuclear bulge and thereby decreases nerve root pressure. However, it does not affect

607

Endoscopic and Minimally Invasive Spine Surgery

Table 1 | Fractures Less Likely to Improve With Standard

Medical Management
Fractures of thoracolumbar junction (T11-L2)
Bursting fracture patterns
Wedge compression fractures with > 30 of sagittal angulation
Vacuum shadow in fractured body (ischemic necrosis of bone)
Progressive collapse seen in office follow-up

extruded or sequestered fragments. Although good results continue to be published in the world literature,
chymopapain injection has fallen out of favor in North
America because of complications, including anaphylaxis and transverse myelitis.
Several percutaneous techniques of microdiskectomy have been developed as well. In the 1970s, instrumentation was developed to access the disk space percutaneously from a posterolateral approach. These
procedures were believed to debulk the central disk,
thereby indirectly reducing nerve root irritation and nociceptor stimulation of the anulus fibrosus. Modifications of the procedure allowed direct visualization
through an arthroscope. Like chymopapain injections,
these techniques, which continue to be performed at
some institutions, are reserved for patients with contained herniations and bulges. Studies comparing automated percutaneous diskectomy with open microdiskectomy usually demonstrate better results with open
microdiskectomy.
Endoscopic diskectomy is performed in a manner
nearly identical to open diskectomy, with the exception
that the surgical instruments are passed through a tubular retractor. With these systems, a transmuscular rather
than subperiosteal approach is undertaken because it is
theorized that the smaller incisions will result in reduced postoperative pain and improved mobilization.
Potential disadvantages with these techniques include limited visualization, a long learning curve during
which complications are more frequent, the risk of inadequate exposure or incomplete decompression, the risk
of vessel or nerve root damage, limits in the ability to
treat lateral recess and foraminal stenosis, and difficulty
accessing the L5-S1 disk space from a far lateral approach in some patients (especially male patients with a
narrow pelvis).

Intradiskal Electrothermal Therapy

Given the mixed outcomes of fusion surgery when treating patients with discogenic back pain, newer modalities
such as thermal energy have been applied to the disk.
IDET is conceptually based on the conversion of radiofrequency energy into thermal energy. With fluoroscopic
guidance, a catheter probe is steered along the posterior
annular wall. Once in place, the tissues are heated to 65

608

Orthopaedic Knowledge Update 8

for 17 minutes. Possible goals include collagen fibril
shrinkage to stabilize the motion segment or destruction
of nociceptive fibers.
The published indications for IDET include more
than 6 months of discogenic pain, more than 3 months
of failed medical treatment and physical therapy, concordant diskography, maintenance of at least 50% residual disk height, and the absence of neurocompressive
pathology.
Initial outcomes reports have been highly varied. A
designer series of 25 consecutive patients reported a
2-point difference in pain scores at an average follow-up
of 7 months. Improvements in the Medical Outcomes
Study Short Form 36-Item (SF-36) scores for physical
function and body pain subscales were noted in 72% of
patients. No complications were reported. A recent update, also by the products designers, found that improvements increased from 1- to 2-year follow-up. These
and other IDET articles have been criticized for the relatively low hurdle (for example, 2 points on the visual
analog scale) that was used to determine significance.
Others argue that in the absence of significant complications, IDET remains a viable modality in a small number of patients. Yet, complications have been reported,
including diskitis and cauda equina syndrome.
Recent basic studies questioned the ability of the
IDET probe to affect collagen denaturation or nociceptive fibers. One biomechanical study showed an increase
in motion at the treated segment, raising the possibility
of long-term destabilization. A recent MRI study found
no change in the appearance of the high-intensity zone
in degenerated disks after IDET.

Vertebral Body Augmentation (Vertebroplasty and

Kyphoplasty)
In the United States, approximately 700,000 osteoporotic vertebral compression fractures occur each
year. Traditionally, these fractures were thought to heal
uneventfully with few, if any, sequelae. However, compression fractures represent a serious and growing
health care problem worldwide. Long-term sequelae of
vertebral compression fractures are common and include pain, deformity, pulmonary decline, gastrointestinal disturbance, gait and functional decline, and increased mortality. Population studies report that
significant functional impairment can occur among people with vertebral compression fractures who do not
seek medical attention.
The pain associated with a given vertebral compression fracture is highly variable. Of those patients presenting to physicians with pain after a vertebral compression fracture, many respond to treatment with
narcotic pain medications and bracing. In some patients,
bracing is not well tolerated or it does not immobilize
the fracture adequately (Table 1). It has been estimated

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 50 Endoscopic and Minimally Invasive Spine Surgery

Figure 1 A, Preoperative T2-weighted MRI

study demonstrating several osteoporotic fractures. The healed fracture has a normal marrow signal, whereas the acute fracture demonstrates increased signal (marrow edema). B, A
lateral fluoroscopic image of the same patient
undergoing kyphoplasty with balloon reduction
of the fracture. C, Postoperative axial CT scans
of the treated level show evidence of a wellcontained cement mantle.
that each year approximately 150,000 patients with compression fractures require hospitalization with protracted periods of bed rest and administration of intravenous narcotics. In elderly patients, osteoporosis with
bed rest is associated with an additional 4% loss of bone
mineral density.
On plain radiographs, vertebral compression fractures have different configurations. In the thoracic
spine, fractures often have a wedge appearance caused
by kyphotic loading. In the lumbar spine, central end
plate cupping of lordotic lumbar bodies is seen. These
levels are also more likely to demonstrate a bursting
pattern. It may be difficult to distinguish fracture age
using plain radiographs alone. A sclerotic band below
the end plate may represent healing or merely compressed trabeculae. Comparison with older films (even a
recent chest radiograph) may help. MRI has a high sensitivity for fracture acuity, which is reflected by peri
end-plate edema and is typically seen as an increased
signal on T2-weighted or short T1 inversion recovery
images (Figure 1). T1-weighted signal intensity is usually
diminished. MRI is also used to assess fracture architec-

American Academy of Orthopaedic Surgeons

ture and possible instrument trajectories. In patients

who are unable to undergo MRI, a bone scan may estimate fracture acuity. A CT scan will provide similar
morphologic information, particularly when sagittal reconstructions are ordered.
Vertebroplasty involves a unilateral or bilateral percutaneous approach to the vertebral body. Various needle and applicator systems are available through which
liquid polymethylmethacrylate bone cement is introduced into the vertebral body. The early clinical results
of vertebroplasty suggest that it is effective in decreasing pain in most patients. However, the mechanism of
pain relief is controversial. Patient groups responding
well to these techniques typically have mechanical incompetence of their vertebral body. Polymethylmethacrylate bone cement restores the load-bearing capacity of the vertebral body. This restoration of stiffness
is thought to provide pain relief. A degree of postural
reduction is also possible in some fractures treated with
vertebroplasty.
Kyphoplasty extends the vertebroplasty concept.
During this procedure, a bilateral approach to the verte-

609

Orthopaedic Knowledge Update 8

Endoscopic and Minimally Invasive Spine Surgery

Table 2 | Precautions and Contraindications to Kyphoplasty

and Vertebroplasty

Table 3 | Applications for Video-Assisted Thoracic Spinal

Surgery

Neurologic symptoms
Young patients*
Pregnancy
High-velocity fractures
Fractures pedicles or facets
Burst fracture with retropulsed bone
Medical issues
Allergy to devices
Allergy to contrast medium
Bleeding disorders
Severe cardiopulmonary difficulties
Technically not feasible
Vertebra plana
Multiple painful vertebral bodies
Level above T5
Neoplasm
Osteoblastic metastasis
Patients with significant long-term survivability
Primary spinal neoplasm
Severe cortical destruction
Local spinal infection

Infection
Biopsy
Dbridement
Drainage of abscess
Tumor
Biopsy
Tumor excision
Corpectomy and grafting
Degenerative disease
Excision of herniated thoracic disk
Trauma
Corpectomy
Cancellous bone grafting
Deformity
Anterior releases for scoliosis or kyphosis
Anterior fusion and instrumentation
Hemivertebra excision
Internal thoracoplasty

*Age range varies; caution in patients younger than 65 years

Severe pulmonary problems precluding prone positioning because of vertebral size or
imaging problems

bral body is made through 1-cm incisions. Cannulas allow introduction of balloon tamponades. Sequential inflation of these tamponades creates a void in the
cancellous bone of the vertebral body and attempts to
reduce the deformity. The balloon tamponades are then
removed. Another theoretic advantage is that viscous
polymethylmethacrylate bone cement can be introduced
into the void for a more controlled fill.
Vertebral body augmentation should be considered
in patients who are bedridden because of pain or in
those whose pain does not begin to decline after several
weeks of nonsurgical care. Results are best for patients
with focal, intense, deep pain in the midline. Usually,
pain worsens with activity and is relieved when recumbent. The contraindications of vertebral body augmentation are listed in Table 2.
Vertebral body augmentation procedures are usually
well tolerated. Good to excellent short-term pain relief
has been reported in more than 80% of patients. Complications typically occur in association with polymethylmethacrylate bone cement extravasation and can lead to
canal compromise or pulmonary embolism. These complications appear to be more common among patients
who undergo vertebroplasty. Additional reduction is

610

Table 4 | Contraindications to Thoracoscopic Spine Surgery

Inability to tolerate one-lung ventilation
Severe or acute respiratory insufficiency
High airway pressures with positive pressure ventilation
Pleural symphysis
Bullous lung pathology
Empyema (relative)
Previous thoracotomy (relative)
Previous tube thoracostomy (relative)
Narrow anterior posterior chest diameter (relative)

possible with kyphoplasty, but it may be variable as a

function of fracture age.

Endoscopic and Mini-Open Procedures

Thoracic Spine
Most endoscopic spine surgery is directed anteriorly
where larger incisions and postthoracotomy pain may
be avoided. Video-assisted thoracic spinal surgery has
the same indications and goals as open thoracotomy
(Tables 3 and 4).
Many of these techniques are technically demanding
and are only performed in select centers worldwide. Before attempting thoracoscopic approaches to the spine,
the surgeon must be thoroughly familiar with open anterior spinal anatomy and have considerable animal laboratory and proctored surgical experience. The largest
early experience with video-assisted thoracic surgery
has been in patients undergoing treatment for spinal de-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

formities. A thoracoscope can be used to assist anterior
release surgery in patients with large kyphotic or scoliotic deformities. The advantages and disadvantages of
thoracoscopic surgery are listed in Table 5.
Thoracoscopic spinal surgery requires a fluoroscopic
table and selective double lumen endotracheal intubation. Surgery proceeds in stages, beginning with the selection of access sites. The sixth intercostal space at the
midaxillary line gives an unobstructed view of the entire
hemithorax. The rib cage and chest wall form a rigid
open space in which to work. Unlike laparoscopy, carbon dioxide insufflation is not required during thoracoscopic spinal surgery.
Instruments are centered at the level of the pathology. The lung may require retraction with an endoscopic
fan retractor or strategically placed sponges. Once the
level of the pathology has been identified, the parietal
pleura is incised. The parietal pleura is bluntly dissected
proximally, distally, and anteriorly to expose as much of
the vertebral margins as necessary.
After the anulus fibrosus is incised with cautery, the
disk and cartilage end plates are removed to bleeding
bone with long pituitaries, Cobb elevators, and curets.
For patients undergoing kyphosis correction, the anterior longitudinal ligament and opposite side of the anulus fibrosus is incised. In patients undergoing releases
for the treatment of scoliosis, it is mandatory to incise
the posterolateral corner of the disk spaces in the concavity of the curve. Morcellized rib or iliac crest graft
can be delivered by a funnel or a structural graft (femoral ring allograft or cage) delivered through an enlarged
portal.
In certain patients, a thoracoscopic release, with the
patient in the prone position, may be done concurrently
with open posterior stabilization. Similarly, internal thoracoplasty may also be performed concurrently or as an
independent procedure. Rib resections are planned as
an ellipse to ensure smooth chest contours. The periosteum of the rib portion to be resected is stripped, and a
high-speed burr creates the osteotomy.
Patients with symptomatic thoracic disk herniations
may be amenable to treatment with endoscopic transthoracic decompressions. Patient selection is a critical issue because patients rarely have true radicular or myelopathic symptoms. Patients with diffuse midthoracic
back pain and diagnostic imaging showing evidence of
multiple degenerative segments with or without herniations are not good candidates for either open or endoscopic diskectomy.
Comparable endoscopic techniques have been described for vertebral corpectomies or osteotomies. The
spine is exposed by bluntly dissecting the pleura and ligating only those segmental vessels in the surgical field.
A combination of pituitary rongeurs, curets, Cobb elevators, and high-speed burrs are used as they would be
with the chest open. The corpectomy reconstruction can

American Academy of Orthopaedic Surgeons

Chapter 50 Endoscopic and Minimally Invasive Spine Surgery

Table 5 | Advantages and Disadvantages of Endoscopy

Advantages
Less postoperative pain
Alleviates the physiologic concerns of healing associated with large
incisions into the thoracic, peritoneal, or retroperitoneal cavities
Improved visualization of the target tissues
Improved cosmetic result with smaller incisions
Reduced recovery time and hospital stay
Disadvantages
Steep learning curve
Cost
Limitations in tactile feedback
Increased working distance from 4 to 30 cm
Two-dimensional video visualization
Triangulation
Limited depth perception
Necessity for specialized instruments
Adverse effect on ability to deal with certain complications (such as
brisk arterial bleeding)
As experience is gained, disadvantages much less limiting

be completed by negotiating allograft struts or mesh

cages into the defect after inserting them into the chest
through an enlarged portal.
Techniques for endoscopic transthoracic spinal instrumentation continue to evolve with the development
of new implants capable of stabilization and correction.
Threaded cylindrical interbody fusion cages, in widespread use in the treatment of deformities of the lumbar
spine, are now being used to treat deformities of the
thoracic spine in the coronal or oblique plane. Such thoracic cage application may be technically less demanding, but the consequences of malpositioning are significant. The fusion rates for patients with single thoracic
cages have yet to be evaluated. Existing posterior or anterior rod and screw implant systems have been modified for application in an endoscopic fashion.
Increasingly, open and endoscopic approaches are
being combined for instrumentation. In such procedures, one of the portals is enlarged to a 5-cm incision,
which allows the use of more standard surgical instruments and significantly simplifies implant introduction.
In the postoperative period, most patients are extubated immediately. A chest tube is maintained at water
seal and removed 1 to 2 days postoperatively. An intensive care unit stay is not usually needed. Aggressive respiratory care is required to prevent down lung
atelectasis and pneumonia.
Complications of thoracoscopic surgery are essentially the same as with an open approach. As in any spinal cord level procedure, spinal cord injury or ischemia
is possible. Dural laceration may be noted. Intercostal

611

Endoscopic and Minimally Invasive Spine Surgery

neuralgia is common (21% of patients), but it is usually
transient. Endoscopic releases may be less complete
than open releases. Also, trocar or instrument injury to
the lung, diaphragm, heart, great vessels, thoracic duct,
azygos vein, esophagus, segmental arteries, sympathetic
chain, and splanchnic nerves is possible.
One series compared video-assisted thoracoscopic
surgery to open thoracotomy in a sheep model. In this
series, histologic, biomechanical, and radiographic outcomes were comparable to those of open surgery. However, endoscopic procedures were associated with protracted learning periods, long surgical times, increased
blood loss, and increased animal morbidity. Others report that, with increased surgeon experience with these
procedures, tissue damage, blood loss, postoperative
pain, intensive care unit and hospital stays are reduced
and that respiratory and shoulder function are less affected.

Lumbar Spine
Several methods have been described to reduce the exposure needed for posterior decompression surgery. The
most common is use of a surgical microscope. Microdiskectomy is the most commonly performed minimally invasive approach to partial disk removal (see chapter
44).
A newer endoscopic diskectomy technique uses a tubular working cannula. The fiberoptic image bundle is
housed in a sidewall of the cannula. Standard instruments, such as pituitary or Kerrison rongeurs, are inserted under direct vision. This technique has also been
used to treat cervical disk herniations (via laminoforaminotomy) and for microdecompression of lateral
recess lumbar spinal stenosis. The benefits of this technique over the more traditional microdiskectomy have
not yet been established. For most spine surgeons, microdiskectomy remains the safer technique.
The role of posterior endoscopic and minimally invasive techniques in the treatment of metastatic disease,
fractures, and infections is evolving. Although the anterior approach to metastatic disease is favored overall,
the use of an endoscope to assist posterolateral decompression may obviate the need for a second anterior surgery in patients undergoing posterior stabilization.
The medial branch of the dorsal primary ramus can
be injured when a midline approach is carried beyond
the facets and over the transverse processes. To minimize this type of injury, several newer techniques have
been developed that use a muscle-splitting approach in
the interval between the multifidus medially and the
longissimus laterally. This approach is similar to that described by Wiltse for far lateral diskectomy. Exploiting
this plane, endoscopic, fluoroscopic, or navigation system assistance allows for transpedicular instrumentation
through smaller incisions. Several variations, including

612

Orthopaedic Knowledge Update 8

interbody fusion techniques, have been described and
are marketed with specialized instruments by the various implant manufacturers. Long-term results of fusions
with these implants are not yet available.
Interestingly, much of the current interest in percutaneous pedicle screw technology comes from the increased use of anterior lumbar interbody cages, themselves part of the vanguard of minimally invasive spine
surgery. Increasingly, the late instability of fusion constructs resulting from the use of these threaded fusion
cages in motion segments with preserved disk height has
been described. Additionally, posterior stabilization has
been recommended. Translaminar facet screws represent another method of achieving posterior fixation
through small incisions with minimal tissue stripping.
Most truly endoscopic lumbar spine procedures are
anterior. Unlike the posterior approach, there are several critical anatomic structures that must be avoided
when approaching the spine anteriorly, including the inferior epigastric artery and vein, which lie deep to the
surface of the rectus muscle. Placement of endoscopic
ports through these structures is associated with postoperative hematoma formation. The ureters run obliquely
from the retroperitoneal space into the abdomen. The
left ureter runs obliquely in the mesentery deep to the
sigmoid and is not usually seen. The right ureter runs
obliquely over the right iliac into the pelvis and may occasionally be injured. The ureter will sometimes course
in midline directly over the L5-S1 disk space and may
be directly damaged by surgical exposures. Overretraction of the psoas muscle may injure the lumbar plexus.
Injury to the lumbar sympathetic chain, just anterior to
the L5-S1 disk, may result in retrograde ejaculation.
Contraindications to anterior spinal endoscopic surgery include peritoneal or pelvic infections, prior open
anterior spine surgery, or prior lower abdominal surgery
(such as colon resection or hysterectomy). Endometriosis also causes adhesions that can limit anterior spinal
endoscopy.
Anterior spinal endoscopy can be performed with or
without insufflation. Insufflation involves the use of carbon dioxide to inflate the abdominal cavity to more
completely visualize surrounding anatomic structures.
Insufflated surgery is a direct extension of conventional
laparoscopic surgery. It gives direct access to L5-S1,
L4-5, and occasionally to L3-4. The advantages of insufflation include organ retraction, more rapid exposure,
increased working space, and decreased bleeding. However, special instruments are required, including expensive trocars with diaphragms. Additionally, air may leak
during the procedure. Moreover, carbon dioxide insufflation of the peritoneal cavity increases mean arterial
pressure and decreases venous return. Carbon dioxide
absorption may lead to hypercapnia and decreased diaphragmatic excursion. Carbon dioxide embolism from
open veins has also been reported.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 50 Endoscopic and Minimally Invasive Spine Surgery

Figure 2 A, Photograph of the introduction of a working portal during thoracoscopy. B, Photograph of the initial thoracoscopic approach to the disk space.
With a gasless approach, the anatomic approach remains the same, but the working space is created by lifting the anterior abdominal wall with a fan retractor and
hydraulic arm (Laparolift, Origin Medsystems, Menlo
Park, CA). This technique decreases costs because conventional instruments are used. However, gasless surgery takes longer and is technically more difficult to
perform. Moreover, lateral vision is limited with the use
of this technique. A combined approach using insufflation for the initial spine exposure to place retractors and
Steinmann pins and the subsequent conversion to a
gasless/Laparolift procedure falls in the midrange in
terms of cost and ease of use.
Operating room positioning of the patient is critical
for these procedures. Drains including a Foley catheter
and nasogastric tube are placed. The patient is supine
with one or both arms tucked at the side or overhead
(for fluoroscopic control). Steep Trendelenburg positioning is often required; therefore, a special shoulder
harness or foot stirrups may also be needed. Fluoroscopy and endoscopy monitors must be in the direct line
of sight of spine and endoscopic surgeons. The preparation materials and drape must allow for bone graft harvest and conversion to an open procedure, if necessary.
The actual approach to the spine may be transperitoneal or retroperitoneal. In transperitoneal endoscopy,
the procedure begins with placement of the first (umbilical) portal. Once in place, the peritoneum is insufflated
to 15 mm with carbon dioxide. Secondary ports are then
placed under visual guidance, including a 17-mm suprapubic working portal. A superficial incision in the midline peritoneum from the sacral promontory to vascular
bifurcation is made, exposing the spine. Blunt dissection
in the retroperitoneal space is used to identify the middle sacral artery and the several medial branches of the
iliac veins, which are clipped. More blunt dissection will
expose the disk spaces, the levels of which are con-

American Academy of Orthopaedic Surgeons

firmed on fluoroscopy (Figure 2).

The retroperitoneal approach uses a potential space
behind the peritoneal sac. Advantages of this approach
include improved spinal access from T12 to S1. In that
the peritoneum contains the bowel loops, retraction is
facilitated. Additionally, conventional instruments may
be used and conversion from a pure percutaneous endoscopic to an endoscopically assisted anterior approach
may be performed if increased difficulty is encountered.
In retroperitoneal approaches, a left lateral decubitus positioning of the patient allows abdominal contents
to fall freely forward. A skin incision is made, the internal and external obliques are identified, and fibers are
spread bluntly. This exposes the transversalis fascia,
which is also incised. A finger may be used to bluntly
dissect in the retroperitoneal space along the twelfth rib
to the transverse process of T12 or L1. The potential
space may be increased by inserting an insufflating balloon or a gasless technique with a Laparolift or insufflated technique may be used. The blunt dissection is
carried posteriorly to the psoas fibers until the disk
space is identified.
Titanium cage or bone dowel implantation, which is
among the most commonly performed lumbar endoscopic techniques, is typically used for anterior column
fusion in patients with disk-related pain (Figure 3). Advantages include avoidance of extensor muscle stripping, more rigid anterior fusion, and lower pseudarthrosis rates from grafting under compression. Reports of
failure of these implants as stand-alone devices have
dimmed the initial enthusiasm.
The implementation of endoscopic approaches to
the anterior spine has fostered a renewed interest in decreasing the morbidity of traditional open anterior procedures. Various mini-open anterior paramedian musclesparing approaches to the lumbar spine have been
recently described. Proponents argue that this cosmeti-

613

Endoscopic and Minimally Invasive Spine Surgery

Figure 3 Postoperative AP radiograph of a patient who underwent endoscopic

transperitoneal titanium BAK cage (SpineTech, Minneapolis, MN) placement followed
by open posterior translaminar facet screw instrumentation.

Orthopaedic Knowledge Update 8

additional costs may be recouped with earlier patient
discharge, the potential benefits of a minimal approach
are not going to be realized by every surgeon. Because
endoscopic approaches limit visualization with a limited
field width, decreased three-dimensional perspective,
and a loss of tactile sense, surgeons should be prepared
for immediate conversion to an open procedure if necessary.
Endoscopic procedures will likely be performed
more commonly in the future. Current indications for
endoscopic spine surgery include minimally invasive access to the intervertebral disk (both anteriorly and posteriorly), dbridement of tumor and infection, releases
for deformity, and anterior interbody fusions. Placement
of spinal instrumentation anteriorly remains under evaluation. All of these procedures are at risk for achieving
overzealous acceptance before independent, randomized studies are done to demonstrate comparable outcomes and risk profiles with accepted techniques.

Annotated Bibliography
cally acceptable approach allows better tactile feedback
and overall safety than endoscopic techniques. In several centers, the initial enthusiasm for endoscopic techniques has gradually given way to a reversion toward
open surgery, but with far more attention given to cosmetics and the minimization of soft-tissue injury.

Summary
Most minimally invasive spine surgery involves a change
in approach, not a change in the surgical procedure itself. Therefore, the indications for surgical intervention
should not be relaxed merely because these procedures
may be performed endoscopically. As in any spine surgery, careful patient selection is paramount in predicting
successful outcomes. Few long-term data are available
for any of these endoscopic spine surgery techniques.
Although many of these procedures are promising, significant advantages over previous techniques have yet
to be demonstrated.
The advantages of endoscopic spinal surgery (improved surgical visualization through magnification and
lighting, decreased perioperative morbidity, and shortened hospital stays) must be counterbalanced with the
steep learning curve for this procedure. Initial experience with endoscopic spinal surgery may be associated
with higher complication rates and longer operating
times. A less efficacious technique should not be used
merely because it is endoscopic.
Endoscopic spine surgery techniques are also both
personnel and equipment intensive. Unlike open procedures, a second surgeon may be needed for an endoscopic procedure (to perform a second approach). Specialized and usually disposable instruments are also
required for endoscopic spine surgery. Although these

614

Percutaneous Disk Procedures

Biyani A, Andersson GB, Chaudhary H, An HS: Intradiscal electrothermal therapy: A treatment option in patients with internal disc disruption. Spine 2003;28:S8S14.
This review of the histologic, biomechanical, and clinical
results of IDET suggests that although the mode of effect is
not understood, IDET may be beneficial in some patients. In
those who did not have improvement of symptoms, little procedural risk was encountered.

Cohen SP, Larkin T, Abdi S, Chang A, Stojanovic M:

Risk factors for failure and complications of intradiscal
electrothermal therapy: A pilot study. Spine 2003;28:
1142-1147.
In this retrospective review of 79 patients undergoing
IDET, 48% of patients reported more than 50% pain relief at
6-month follow-up. In this series, there was a 10% complication rate, but most of the complications were transient and
self-limited. Obesity was a risk factor for failure of the procedure.

Freedman BA, Cohen SP, Kuklo TR, Lehman RA, Larkin P, Giuliani JR: Intradiscal electrothermal therapy
(IDET) for chronic low back pain in active-duty soldiers: 2-year follow-up. Spine J 2003;3:502-509.
In this consecutive case series assessing the use of IDET in
the management of chronic discogenic low back pain in 36 soldiers, 50% or greater pain reduction was reported by 47% of
patients at 6-month follow-up and 16% at 2-year follow-up.
The authors also reported that 20 of 31 soldiers (65%) had a
persistent decrease in their analog pain scores. Additionally, 7
of 31 soldiers (23%) went on to undergo spinal surgery within
24 months of undergoing IDET. The authors noted that their
reasonable early results diminished with time and that up to

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

20% of patients reported worsening of baseline symptoms at
final follow-up. They concluded that IDET should not be used
as a substitute for spinal fusion in the treatment of chronic discogenic low back pain in active-duty soldiers.

Chapter 50 Endoscopic and Minimally Invasive Spine Surgery

gery: An intraoperative EMG study. Eur Spine J 2002;
11:20-26.
This electromyographic study demonstrated that less mechanical irritation of the nerve root occurred with endoscopic
approaches than with traditional open surgery.

Lieberman IH, Dudeney S, Reinhardt MK, Bell G: Initial outcome and efficacy of kyphoplasty in the treatment of painful osteoporotic vertebral compression
fractures. Spine 2001;26:1631-1638.

Spruit M, Jacobs WC: Pain and function after intradiscal

electrothermal treatment (IDET) for symptomatic lumbar disc degeneration. Eur Spine J 2002;11:589-593.

This is a prospective study of kyphoplasty procedures that

were performed in 30 patients. The results demonstrate excellent outcomes and low complications.

This small series of prospectively evaluated patients failed

to show significant improvement of patient symptoms after
undergoing IDET.

Lindsay R, Silverman SL, Cooper C, et al: Risk of new

vertebral fracture in the year following a fracture.
JAMA 2001;285:320-323.

Yeung AT, Tsou PM: Posterolateral endoscopic excision

for lumbar disc herniation: Surgical technique, outcome,
and complications in 307 consecutive cases. Spine 2002;
27:722-731.

This article reviews the clinical consequences of vertebral

compression fractures and largely dispels the notion that these
injuries follow a benign, self-limited course.

Muramatsu K, Hachiya Y, Morita C: Postoperative magnetic resonance imaging of lumbar disc herniation:
Comparison of microendoscopic discectomy and Loves
method. Spine 2001;26:1599-1605.
The authors of this study compared postoperative MRI
studies and found that the effect of microendoscopic diskectomy on the cauda equina was comparable to that of open diskectomy. Furthermore, the postoperative images of the route of
entry failed to show that microendoscopic diskectomy is appreciably less invasive with respect to the paravertebral muscles.

Phillips FM, Ho E, Campbell-Hupp M, McNally T, Todd

Wetzel F, Gupta P: Early radiographic and clinical results of balloon kyphoplasty for the treatment of osteoporotic vertebral compression fractures. Spine 2003;
28:2260-2265.
This is an outcomes report of 29 patients who underwent
kyphoplasty in 37 separate sessions; a total of 61 vertebral
compression fractures were treated. In 30 of the 52 fractures
that were considered reducible, a mean 14.2 of correction was
noted. Significant pain relief was noted with minimal complications.

Saal JA, Saal JS: Intradiscal electrothermal treatment

for chronic discogenic low back pain: Prospective outcome study with a minimum 2-year follow-up. Spine
2002;27:966-973.
This 2-year follow-up study of the original designer series
demonstrated that the statistically significant improvement in
pain scores and SF-36 quality of life scores that was previously
reported increased between the 1- and 2-year observation
points.

Schick U, Dhnert J, Richter A, Knig A, Vitzthum HE:

Microendoscopic lumbar discectomy versus open sur-

American Academy of Orthopaedic Surgeons

In this retrospective review of 1-year outcomes in 307 consecutive patients with lumbar disk herniation who underwent
posterolateral endoscopic diskectomy, the response rate to the
questionnaire was 91%, and 90.7% of those patients reported
being satisfied with their surgical outcomes. The combined major and minor complication rate was 3.5%.

Endoscopic and Mini-Open Procedures

Anand N, Regan JJ: Video-assisted thoracoscopic surgery for thoracic disc disease: Classification and outcome study of 100 consecutive cases with a 2-year minimum follow-up period. Spine 2002;27:871-879.
In this retrospective review of prospectively collected
4-year follow-up data in 100 consecutive patients who underwent video-assisted thoracoscopic surgery, no permanent complications or spinal cord injuries were reported. The average
percentage of improvement in Oswestry scores was most significant in grade 4 patients (myelopathy, 60%), followed by
grade 3A patients (axial and thoracic radicular pain, 37%),
grade 3B patients (axial with leg pain, 28%), and grade 1 patients (pure axial, 24%). The authors concluded that the procedure was associated with significant clinical improvement in
most treated patients.

Escobar E, Transfeldt E, Garvey T, Ogilvie J, Graber J,

Schultz L: Video-assisted versus open anterior lumbar
spine fusion surgery: A comparison of four techniques
and complications in 135 patients. Spine 2003;28:729732.
This is a retrospective report comparing the outcomes of
anterior lumbar interbody fusion in 135 patients using one of
the following four approaches: transperitoneal endoscopy with
insufflation, retroperitoneal endoscopic surgery, mini-open
surgery, or traditional open approaches. The incidence of complications for video-assisted techniques was found to be consistent with that reported in the medical literature, but it was
higher than that for open techniques. The authors note that
they no longer use video-assisted techniques as a result of
these findings.

615

Endoscopic and Minimally Invasive Spine Surgery

Khoo LT, Beisse R, Potulski M: Thoracoscopic-assisted
treatment of thoracic and lumbar fractures: A series of
371 consecutive cases. Neurosurgery 2002;51(5 suppl):
104-117.
In this report of 371 patients with thoracic and thoracolumbar spine fractures who were treated with thoracoscopic
assistance, the severe complication rate was low (1.3%), with
only one instance of each of the following reported: aortic injury, splenic contusion, neurologic deterioration, cerebrospinal
fluid leak, and severe wound infection.

Pellis F, Puig O, Rivas A, Bag J, Villanueva C: Low fusion rate after L5-S1 laparoscopic anterior lumbar interbody fusion using twin stand-alone carbon fiber cages.
Spine 2002;27:1665-1669.
This study reported prospective data on 12 patients undergoing twin, stand-alone anterior cage placement. Although significant improvements in visual analog scores, Prolo scores,
and Waddell Disability Index scores were noted, the overall

616

Orthopaedic Knowledge Update 8

CT scan fusion rate at 2-year follow-up was 16.6% and was
deemed unacceptably low.

Classic Bibliography
Do HM: Magnetic resonance imaging in the evaluation
of patients for percutaneous vertebroplasty. Topics in
MRI 2000;14:235-244.
Grados F, Depriester C, Cayrolle G, et al: Long-term observations of vertebral osteoporotic fractures treated by
percutaneous vertebroplasty. Rheumatology (Oxford)
2000;39:1410-1414.
McLain RF: Endoscopically assisted decompression for
metastatic thoracic neoplasms. Spine 1998;23:1130-1135.
Silverman SL: The clinical consequences of vertebral
compression fractures. Bone 1993;13:S27-S31.

American Academy of Orthopaedic Surgeons

Chapter

51

Spinal Cord Injury

Kevin C. OConnor, MD
Eric K. Mayer, MD
Mitchel B. Harris, MD

Introduction
Over the past two decades, advances in spinal cord medicine have come about at an unprecedented pace. To
stay abreast of these advances, it is helpful to review
characteristics of the population affected by spinal cord
injury (SCI), their prognosis, testing procedures, and
major medical and rehabilitation sequelae that are standards of care.

Epidemiology
The annual incidence of SCI in the United States, not
including fatalities at the site of injury, is approximately
40 cases per million population, or approximately 11,000
new cases per year. The number of Americans with SCI
has been estimated to be between 183,000 and 230,000.
SCI primarily affects young adults with an average age
of injury of 32.1 years; 55% of all SCIs occur in people
between the ages of 16 to 30 years, during the most productive working/earning years. The average age of injury
has been increasing since the 1970s, mirroring the increase in the median age of the general population. The
fastest growing cohort with SCI is patients older than
age 60 years, who now represent 10% of new SCI patients. There remains a 4:1 male to female ratio that has
largely remained unchanged since the 1960s.
A significant trend has been observed in the racial
distribution of patients with SCI. Since 1990, AfricanAmericans and Hispanics have become disproportionately affected with new SCIs. Percentages of new SCIs
have risen for African-Americans from 5.7% in 1974 to
27.6% in 1990, and for Hispanics from 5.7% to 7.7%.
During the same time period, the percentage of Caucasians with new SCIs has decreased from 77.5% to 59.1%
(Figure 1).
Motor vehicle crashes still account for the largest
percentage of SCIs at a current rate of 38.5%, followed
by acts of violence (mostly gunshot wounds), falls, and
sports-related injuries (Figure 2). The proportion of injuries from falls and acts of violence has increased
steadily.

American Academy of Orthopaedic Surgeons

Over the past 20 years, tetraplegia has become less

common and paraplegia more common. Approximately
one half of all traumatic SCIs in the United States are
cervical lesions, and one third are thoracic. The most
common neurologic level of injury is C5, followed by
C4, and then C6. The most common level of paraplegia
is T12. There has also been a trend toward an increased
number of incomplete lesions (Figure 3).

Causes of Death
Overall, 85% of SCI patients who survive the first 24
hours after injury are still alive 10 years later. The most
common cause of death is diseases of the respiratory
system, with most of these resulting from pneumonia.
The second leading cause of death following SCI is
nonischemic heart disease. Deaths resulting from external causes such as subsequent unintentional injuries and
suicides and homicides (but not including multiple injuries sustained during the original accident) are the third
leading cause of death. The majority of these deaths are
the result of suicide. The fourth leading cause of death is
infectious and parasitic diseases (usually septicemia associated with decubitus ulcers, urinary tract or respiratory infections). Mortality rates are significantly higher
during the first year after injury than during subsequent
years.

Classification
In 1969, Frankel and associates described a five-grade
system for classifying traumatic SCI, divided between
complete and incomplete injuries. The amount of preserved motor or sensory function determined the specific Frankel classification. The Frankel classification
was replaced in 1992 by the American Spinal Injury Association (ASIA) Impairment Scale, which was revised
in 1996, and again in 2000. These standards subsequently
became known as the International Standards for Neurologic and Functional Classification of Spinal Cord Injury. The ASIA standards have gained widespread acceptance as the preferred classification system for SCI.
The initial neurologic examination serves as a baseline

619

Spinal Cord Injury

Orthopaedic Knowledge Update 8

Figure 1 Racial distribution of SCI.

Figure 2 Etiology of SCI since 1990.

Figure 3 Current percentage of injuries by ASIA classification (since 1991).

for evaluation over the first hours to days after injury;

however, neurologic status can change over the first few
days and is influenced by resuscitative procedures. The
period from 72 hours to 1 week after injury is the earliest time period after injury when detailed neurologic
evaluations can reliably be performed to predict neurologic recovery.
The ASIA sensory examination consists of 28 key
dermatomes, each separately tested for pinprick/dull
(with a safety pin) and light touch (with cotton) sensation on both sides of the body. A three-point scale from
0 to 2 is used, with the face serving as the normal reference point. A score of 0 represents absent sensation, including inability to distinguish the sharp and dull edge
of the pin; a score of 1 represents impaired sensation, including hyperesthesia; and a score of 2 is given when the
pin or cotton is felt in the same manner as when tested

on the face. The sensory level is the most caudal dermatome to have intact (2/2) sensation for both pinprick
and light touch on both sides of the body.
The ASIA motor examination consists of testing 10
key muscles (5 in the upper limb and 5 in the lower
limb) on each side of the body (Table 1). The patient is
supine during testing and strength is graded on a sixpoint scale from 0 to 5. The motor level is defined as the
lowest key muscle that has a grade of at least 3, provided that the segments above that level are graded as
5. Grade 4 is not considered normal, as it previously
was, unless the examiner judges that certain factors inhibited full effort, including pain, positioning, disuse, or
hypertonicity. In patients in whom there is no key muscle for a segment that has intact sensory dermatomes
(C2-C4, T2-L1, and S2-S5), the sensory level defines the
motor level.

620

American Academy of Orthopaedic Surgeons

Chapter 51 Spinal Cord Injury

Orthopaedic Knowledge Update 8

Table 1 | Key Muscle Groups Involved in ASIA Motor

Examination
Root Level

Muscle Group

C5
C6
C7
C8
T1

Elbow flexors
Wrist extensors
Elbow extensors
Long finger flexors
Small finger abductors

Root Level Muscle Group

L2
L3
L4
L5
S1

Hip flexors
Knee extensors
Ankle dorsiflexors
Long toe extensor
Ankle plantar flexors

The skeletal level of injury is defined as the spinal

level where, by radiographic examination, the greatest
vertebral damage is found. The neurologic level of injury is the most caudal level at which both motor and
sensory modalities are intact on both sides of the body.
Because of a poor correlation between vertebral injury
and function, the neurologic level of injury is preferred.
Because these levels can be different from side to side,
up to four levels (right C6 sensory, left C8 sensory, right
C5 motor, left C8 motor) may be needed to present a
clearer picture of the patients status.
Tetraplegia, previously called quadriplegia, is defined
as impairment or loss of motor or sensory function in
the cervical segments of the spinal cord caused by damage of the neural elements within the spinal cord. Tetraplegia results in impairment of function of the arms as
well as the trunk and legs. Paraplegia refers to impairment of motor or sensory function in the thoracic, lumbar, or sacral segments of the spinal cord secondary to
damage of the neural elements within the spinal cord.
Paraplegia results in impairment of the trunk and legs,
depending on the level of injury, but the arms are
spared. Plexus injury or injury to the peripheral nerves
outside the neural canal are not included. The terms
quadriparesis and paraparesis are discouraged because
they imprecisely describe incomplete lesions.
Complete injury is defined as the absence of sensory
or motor function in the lowest sacral segments, and incomplete injury is defined as preservation of motor or
sensory sensation below the neurologic level of injury
that includes the lowest sacral segments, also known as
sacral sparing. Sacral sparing is tested by light touch and
pinprick at the anal mucocutaneous junction (S4-5 dermatome) on both sides, as well as by testing for voluntary anal contraction and deep anal sensation. If any of
these sensations is present, even if impaired, the patient
has an incomplete injury. The zone of partial preservation refers to those dermatomes and myotomes caudal
to the neurologic level that remain partially innervated
and is used only in complete injuries. The ASIA Impairment Scale describes five levels of SCI severity
(Table 2).

American Academy of Orthopaedic Surgeons

Table 2 | ASIA Impairment Scale

A = Complete. No motor or sensory function is preserved in the sacral
segments S4-S5.
B = Incomplete. Sensory but no motor function preserved below the
neurologic level and includes the sacral segments S4-S5.
C = Incomplete. Motor function is preserved below the neurologic level,
and more than half of the key muscles below the neurologic level have a
muscle grade less than 3.
D = Incomplete. Motor function is preserved below the neurologic level,
and at least half of the key muscles below the neurologic level have a
muscle grade greater than or equal to 3.
E = Normal. Sensory and motor functions are normal.
A is a complete injury, while B through E represent incomplete injuries. For an individual to
receive a grade of C or D, the injury must be incomplete and have either voluntary anal
contraction or sparing of motor function more than three levels below the motor level

Spinal Cord Injury Syndromes

Central Cord Syndrome
The most common of the incomplete syndromes is central cord syndrome, which is characterized by motor
weakness of the upper extremities greater than the
lower extremities, in association with sacral sparing.
Central cord syndrome most frequently occurs in older
patients with cervical spondylosis and a hyperextension
injury, but can occur in any age group or with any mechanism. The mechanism of injury involves compression of
the cord during hyperextension, caused by an inward
bulging of the ligamentum flavum on an already narrowed canal.
Central cord syndrome has a typical pattern of recovery that begins with the lower extremities, followed
by bowel and bladder function, proximal upper extremities, and hand function. The prognosis is dependent on
the patients age, with those younger than age 50 years
having a much better prognosis for independence than
older patients. However, for patients with ASIA D tetraplegia, the prognosis for recovery of independent ambulation is excellent, even for those older than age 50
years.

Brown-Squard Syndrome
Brown-Squard syndrome involves hemisection of the
spinal cord and accounts for 2% to 4% of all traumatic
SCIs. Neurologically, there is an ipsilateral loss of position, light touch, vibration, and motor loss, and a contralateral loss of pain and temperature below the level
of the lesion. Neuroanatomically, this is explained by the
crossing of the spinothalamic tracts at the spinal cord
level and the crossing of the corticospinal and dorsal
columns at the medulla.
Overall, patients with Brown-Squard syndrome
have the best prognosis for functional outcome and potential for ambulation. Approximately 75% to 90% of

621

Spinal Cord Injury

patients ambulate independently upon discharge from
rehabilitation; patients are more likely to ambulate
when the upper limb is weaker than the lower limb on
the affected side.

Anterior Cord Syndrome

Anterior cord syndrome is associated with lesions that
affect the anterior two thirds of the spinal cord but preserve the dorsal columns. Although anterior cord syndrome can be associated with direct injury to the anterior cord, it is most commonly caused by lesions of the
anterior spinal artery, which supplies the anterior two
thirds of the spinal cord and can be injured through
thrombosis or embolism. There is inconsistent and variable loss of motor strength and pinprick sensation with
preservation of light touch, deep pressure, and proprioception. Prognosis is poor with less than a 20% chance
of motor recovery.

Conus Medullaris and Cauda Equina Syndromes

The conus medullaris is the terminal segment of the spinal cord. The segment above it is termed the epiconus
and it consists of spinal cord segments L4 to S1. Lesions
of the epiconus cause upper motor neuron dysfunction
of the lower lumbar and sacral segments. Recovery is
similar to that of other upper motor neuron SCIs. Lesions of the conus medullaris affect the S2-S5 segments
and there is a combination of upper and lower motor
neuron deficits because the cauda equina is often injured simultaneously. Frequently the lower motor neuron deficits predominate. With conus medullaris injuries,
recovery is limited.
Lesions below the L1 vertebral level injure the
cauda equina, producing motor weakness and atrophy
of the lower extremities with bowel and bladder involvement and areflexia. Cauda equina injuries have a
better prognosis for recovery than other SCIs, because
the roots are histologically peripheral nerves and regeneration can occur. Often, traumatic SCI will cause a
combination of conus and cauda equina injuries.

Acute Management of Spinal Cord Injury

Acute intervention after SCI may be surgical or nonsurgical. Nonsurgical intervention consists of pharmacologic and realignment techniques. The role of surgical
intervention for acute SCIs is still widely debated; however, there exists a significant body of information identifying the benefit of urgent surgical intervention in animal SCI models. Human clinical trials and retrospective
analysis have been unable to show a statistical improvement in neurologic outcome related to surgical intervention. According to one recent study, subacute and late
decompressive procedures have a major role in the
treatment of incomplete SCI.

622

Orthopaedic Knowledge Update 8

When the trauma patient is initially assessed at the
scene of injury, if there is any suspicion of SCI, additional attention is given to maintaining maximal oxygenation and normalization of blood pressure. These crucial
early interventions help to protect the fragile spinal
cord blood perfusion. Generally these interventions include nasal cannula administration of oxygen and vasopressors to maintain an optimal blood pressure. Crystalloids are initially administered; however, if adequate
pressures cannot be maintained in the SCI patient, pressors (dopamine) should be quickly added to resuscitation efforts. Blood pressure should be maintained at a
systolic pressure between 80 and 100 mm Hg to minimize the proliferation of secondary insults to the spinal
cord from hypoxemia.
The importance of maintaining adequate spinal cord
perfusion following injury is widely accepted though
largely unproven. There exists a hypothetical watershed
region of partial injury within the cord tissue that has
the potential for functional recovery if perfusion is
maintained and further trauma avoided. Healthy cord
tissue demonstrates autoregulation of blood flow capable of maintaining perfusion through a wide range of
systemic blood pressure. This autoregulatory function
can be lost following injury, leaving the watershed region vulnerable to fluctuations in systemic blood pressure. Maintenance of mean arterial pressure above
85 mm Hg has been recommended for at least 7 days
following injury. Initial efforts at volume expansion using crystalloid or colloid may be insufficient, and require subsequent administration of dopamine or neosynephrine.
Mortality in the acute phase of SCI is frequently secondary to pulmonary failure and shock. Shock is initially managed by the Advanced Trauma Life Support
protocol of crystalloid resuscitation, until it is clear that
supplementary pressors are required to maintain adequate blood pressure. Early and aggressive pulmonary
support is essential in the acute stages after injury. Pneumonia and atelectasis are difficult to avoid and represent the most common early complications. The frequency and severity of these occurrences is related to
the level of the SCI, as it relates to the remaining innervation of the respiratory muscles. In the patient with
acute SCI, respiratory support is often required.
The loss of sympathetic tone, found with lesions
above the T6 level, leads to unopposed vagal input. This
imbalanced neural activity can cause increased bronchial secretions and ultimately lead to mucus plug formation. Mucus plugs, an impaired cough reflex, and
compromised respiratory muscles all can cause early
respiratory arrest in the patient with acute SCI.
High-dose steroids are administered in the patient
with acute SCI. Intravenous administration of the glucocorticoid methylprednisolone in high doses (loading
dose 30 mg/kg bolus over 15 minutes followed by

American Academy of Orthopaedic Surgeons

Chapter 51 Spinal Cord Injury

Orthopaedic Knowledge Update 8

5.4 mg/kg hr) is recommended, with the length of time
of administration dependent on the time of presentation. If administration of the steroid is initiated within 3
hours of the SCI, administration for 23 hours is recommended, whereas if steroid administration is initiated
between 3 to 8 hours after SCI, the steroid should be
continued for 48 hours. The original clinical trials did
not recommend the use of methylprednisolone if the patient presents after 12 hours from the time of injury. Additionally, the study cohort did not include cord injuries
caused by ballistic/penetrating trauma or those with
nerve root injury. The studies supporting this protocol
have come under increasing scrutiny over the past few
years.
Shortcomings of the Second National Acute Spinal
Cord Injury Study (NASCIS II) include lack of data regarding the functional significance of apparently incremental improvements in neurologic scores, failure to
present raw study data, and absence of a clearly defined
initial study hypothesis. Follow-up post hoc subgroup
analysis suggests that placebo administration greater
than 8 hours following injury achieved an equivalent improvement in motor score in patients who had methylprednisolone administered within 8 hours of injury.
NASCIS II has also been criticized for lacking functional outcome measures. Currently the use of high-dose
steroids in the treatment of SCI is an off-label use of
this drug and has not been approved for this indication
by the Food and Drug Administration. High-dose steroid treatment also has been associated with an increased incidence of gastrointestinal bleeding as well as
an increased rate of infection.
The glucocorticoids are stabilizing agents that have
been shown to affect the neural membranes, preventing
uncontrolled intracellular calcium influx. Additional
stated positive effects include decreasing the lysosomal
enzyme effects and reducing secondary swelling and inflammation.
A second nonsurgical intervention is early indirect
reduction of the spinal canal via cranial tong traction.
The increasing availability of MRI has dampened some
of the enthusiasm for rapid closed reduction techniques,
as MRI has identified the presence of associated herniated intervertebral disks. Closed reduction of the traumatic deformity is directed toward restoration of the
spinal column to its premorbid alignment and reestablishment of the diameter of the bony spinal canal.
This reduction maneuver is applicable to cervical spine
injuries only. Additionally, before any significant amount
of traction (> 10 lb) can be applied through the tongs,
special effort must be made to rule out the possibility of
an associated occult craniocervical injury. For patients
with distractive injuries of the cervical spine, craniocervical dissociations, or fracture-dislocations associated
with ankylosing spondylitis, application of a holding device such as a halo is recommended in the urgent set-

American Academy of Orthopaedic Surgeons

ting. These injuries pose the risk of significant neurologic deterioration if undiagnosed or untreated.
Indirect reduction is infrequently performed in the
acute setting for the treatment of a thoracolumbar or
lumbar injury in the presence of a spinal cord, conus, or
cauda equina injury. Some physicians advocate early
open reduction and stabilization of thoracolumbar
fracture-dislocations with accompanying neural deficits.
Keeping in mind that most SCI models show irreversible cord pathophysiology within 6 to 8 hours, advocates
of early decompression, realignment, and stabilization
stress the importance of acute intervention. Such intervention, however, often is not possible because of
ground transportation of patients and hospital referral
networks.

Medical Complications of Spinal Cord Injury

Cardiac
The most important postinjury treatment issue is cardiovascular sequelae. These are often subdivided into direct
complications resulting from the injury itself (for example, hypotension, bradycardia, and autonomic dysreflexia) after spinal shock has resolved and indirect complications resulting from sequelae of immobilization and
the relatively sedentary lifestyle postinjury (for example, deep venous thrombosis [DVT], pulmonary embolism, and coronary artery disease).
Autonomic dysreflexia is characterized by a sudden
sympathetic discharge causing an exaggerated increase
in blood pressure, often with bradycardia, in response to
a noxious stimulus originating below the injury. This
syndrome usually occurs in 48% to 85% of patients with
SCI above the T6 level. Associated symptoms include
headache, mydriasis, piloerection, nasal congestion,
flushing, and sweating. The patient can also experience
no symptoms at all, with silent blood pressure elevation.
The most common cause of autonomic dysreflexia is
bladder distention, followed by fecal impaction, pressure
ulcers, ingrown toenails, fractures, and abdominal emergencies. Treatment includes putting the patient in a sitting position, removal of precipitating stimuli, and if the
cause cannot be found, use of antihypertensive agents
(such as nifedipine, nitrates, and angiotensin-converting
enzyme inhibitors).

Deep Venous Thrombosis

DVT has been reported in 15% to 80% of SCI patients
and usually develops during the first 2 weeks after injury. Residual motor function in the lower extremities
and the level of SCI contribute to the overall incidence.
The estimated incidence of pulmonary embolism is 2%
to 16%. All patients should receive prophylaxis for
DVT unless otherwise contraindicated. If prophylaxis is
delayed more than 72 hours, tests to exclude the presence of clots should be done. Anticoagulant prophylaxis

623

Orthopaedic Knowledge Update 8

Spinal Cord Injury

with either low-molecular-weight heparin or adjusted
dose unfractionated heparin should be initiated, unless
contraindicated, within 72 hours of injury. Vena cava filter placement is recommended in SCI patients for
whom anticoagulant prophylaxis has failed or who have
anticoagulation contraindications. Filters should also be
considered for SCI patients with high-level tetraplegia,
or for those with thrombosis of the inferior vena cava
despite the use of anticoagulants. Anticoagulant prophylaxis should continue for patients who are categorized
as motor incomplete on the ASIA impairment scale
while they are hospitalized and for 8 to 12 weeks in motor complete patients, depending on risk factors. Patients with chronic SCI remain at increased risk for
DVT, and reinstitution of prophylactic measures should
be considered for patients with chronic SCI who are immobilized and on bed rest for a prolonged period, or for
those who are to undergo surgical procedures.

Pulmonary
Pulmonary complications after SCI remain the leading
cause of death in patients with tetraplegia and paraplegia. Pneumonia accounts for 18.9% of deaths in the first
year after injury. The problems of pulmonary management can be classified into three main categories: secretion management, atelectasis (and its sequelae), and hypoventilation. At least two of the following treatments
should be available to patients: deep pulmonary suctioning, chest physiotherapy, assisted cough methods (for
example, quad cough), mechanical insufflationexsufflation, abdominal binders, frequent position
changes, incentive spirometry or resistive devices, and
positive pressure ventilation.

Bladder
The bladder is usually affected in one of two ways after
injury. A spastic bladder, or upper motor neuron bladder, fills with urine and a reflex automatically triggers
the bladder to empty. A flaccid bladder, or lower motor
neuron bladder, results when the reflexes of the bladder
muscles are slowed or absent. A flaccid bladder can become overdistended or stretched. Although the upper
tracts are at risk in both conditions, they are at greater
risk in a spastic bladder. Bladder sphincter dyssynergia
occurs when the sphincter muscles do not relax when
the bladder contracts, which may also place the upper
tracts at risk for injury. This may occur with upper motor neuron dysfunction. To treat the neurogenic bladder,
SCI patients are placed on a bladder management program, which will allow acceptable bladder emptying
with convenience and help to avoid bladder accidents
and infection, and long-term upper tract damage secondary to reflux. Common treatment options include
clean intermittent catheterization, an indwelling catheter, and for men, an external condom catheter.

624

Individuals with SCI are at a high risk for urinary

tract infection; complications caused by such infections
are the most common cause of SCI morbidity. Other
complications include bladder and renal calculi, renal
failure, and bladder carcinoma. Individuals with SCI
should have annual urologic evaluations including functional studies.

Bowel
Similar to the neurogenic bladder, the neurogenic bowel
may be either spastic or flaccid. Patients are placed on a
bowel program with predictable, regular, timely, and
thorough evacuation of the bowels without the occurrence of incontinence or complications. Pharmacologic
agents are not always needed for long-term use but can
be an effective adjunctive tool to facilitate the bowel
program. A common starting routine for patients with a
neurogenic bowel includes a stool softener three times
daily, a laxative at bedtime, and a suppository in the
morning after breakfast. A bowel routine after eating
makes use of the gastrocolic reflex. If a patient usually
has a bowel movement in the evening, then the laxative
can be given at noon and an enema or suppository in
the evening after dinner. It is difficult for patients with a
true lower motor neuron bowel to achieve continence
and a bulk laxative may be needed to limit loose stool
formation with diarrhea.

Skin
Pressure ulcers are a daily concern for patients with
SCI. It is estimated that up to 80% of individuals with
SCI will have a pressure sore during their lifetime, and
30% will have more than one pressure sore. Education
in pressure ulcer etiology (including pressure and
shear), use of appropriate equipment to help decrease
ulcers, and weight shifts, while seated and in bed is necessary. The treatment of pressure sores may include
dressings, bed rest, and surgery. All factors that contributed to pressure ulcer development must be eliminated
or minimized. Treatment can be very costly in terms of
lost wages and additional medical expenses.

Neuromusculoskeletal
After SCI, many changes occur in bone metabolism. An
imbalance between bone formation and bone resorption
rapidly develops, resulting in bone loss and osteoporosis.
Within the first 4 to 6 months, 25% or more of bone
mass is lost. By 16 months, bone mass homeostasis is
reached with bone mass at 50% to 70% of normal and
near the fracture threshold. Fractures are very common
in patients with SCI. Data from the Model SCI Systems
show that 14% of patients with SCI will have had a fracture within 5 years after injury. This percentage increases to 28% after 10 years, and 39% after 15 years.
The frequency of fractures increases with age and com-

American Academy of Orthopaedic Surgeons

Chapter 51 Spinal Cord Injury

Orthopaedic Knowledge Update 8

pleteness of injury, and is higher in women than men.
Most fractures are in the lower extremities and result
from falls; fractures can also occur with routine activities of daily living and range-of-motion exercises. Treatment of fractures in nonambulatory patients with SCI is
somewhat controversial. Circumferential plaster casts or
traction can cause skin breakdown, and surgical intervention for femoral shaft and distal femur fractures is
associated with a high complication rate because of
bone quality, risk of osteomyelitis, and recurrent bacteremia. Surgical intervention may be indicated when nonsurgical management fails to control rotational deformity, if vascular supply is jeopardized, or in proximal
femur fractures with severe spasticity. Often fractures
are treated with soft splints without bed rest, and a
wheelchair is used to provide functional positioning for
healing.
Heterotopic ossification is the formation of true
bone at ectopic sites and occurs in up to 50% of adults
with SCI. Ankylosis occurs in approximately 5% of patients. Most commonly, heterotopic ossification develops
between 1 to 4 months after injury and 90% develops
around the hip, followed by the knee and elbow. Description of the diagnosis and treatment of this condition can be found in chapter 53.
Spasticity is a common sequelae of upper motor
neuron SCI. It may enhance function and maintain muscle bulk, but when poorly controlled, spasticity may lead
to impaired function, pain, and decreased quality of life.
A full discussion of spasticity is found in chapter 52.

Prognosis After Spinal Cord Injury

Recovery of motor function has been documented for at
least 2 years after injury and changes in neurologic status may continue after 2 years. Recovery from SCI depends on the initial strength of the muscles, and most
importantly, on whether the injury is complete or incomplete. Vertebral displacement less than 30% and age
younger than 30 years at the time of injury are factors
associated with improved recovery. No correlation has
been found between degree of vertebral wedging, type
of fracture, or etiology of injuries. In evaluating changes
after SCI, MRI is the best modality to show intramedullary detail such as hematoma and edema. A hemorrhage
found with acute MRI correlates with the poorest prognosis, followed by contusion and edema. A normal MRI
scan correlates with the best prognosis. Although MRI
may augment the physical examination, it alone is not as
accurate a predictor of SCI recovery as the physical examination.
Ninety percent of patients with complete tetraplegia
will recover one root level of function. The initial
strength of the muscle is a significant predictor of
achieving antigravity strength and the rate of strength
recovery. In patients with no initial motor strength,

American Academy of Orthopaedic Surgeons

there is an improved prognosis of recovering antigravity

strength the faster the recovery begins. According to the
literature, muscles with grade 1 and 2 strength at
30 days have a 97% chance of recovery to 3/5 strength
at 1 year. In comparison, in the first muscle below the
neurologic level of injury with 0/5 strength at 1 month,
only 57% recover to at least 1/5 strength, and only 27%
recover to 3/5 or more strength at 1 year. Most upper
extremity recovery occurs during the first 6 months,
with the greatest rate of change during the first 3
months.
Upper extremity recovery in patients with incomplete tetraplegia is approximately twice as great as for
those with complete tetraplegia. For patients who are
initially sensory incomplete, the prognosis for motor recovery is much more favorable in those with pinprick
sparing than in those in whom light touch sensation
alone is spared.
For patients with complete paraplegia, 76% do not
show a change in neurologic level of injury from 1
month to 1 year after injury. Even without a change in
neurologic level, patients with incomplete paraplegia
have the best prognosis for lower extremity and functional recovery. Overall, 80% of patients with incomplete paraplegia regain strength to grade 3 or more in
hip flexors and knee extensors at 1 year. The absence of
any motor function at 1 month after injury is not an absolute indicator of poor motor recovery.
Most of the recovery after SCI is spontaneous, but it
may be enhanced by several factors. Active exercise can
promote motor recovery whereas immobilization may
impede motor recovery. Using body weight-supported
and interactive locomotor training, ambulation recovery
may be enhanced in patients with incomplete SCI. Serotonergic and neuroadrenergic drugs may enhance locomotor recovery, whereas drugs with gammaaminobutyric acid may impede ambulation.

Functional Outcomes
C1 through C4 Tetraplegia
Patients with C1 and C2 lesions may have functional
phrenic nerves. In these patients, implanted phrenic
nerve pacemakers can be used, and pacing of the diaphragms may be simultaneous or alternating. If secretions are not a problem, tracheostomies may be plugged
or discontinued.
Patients with C3 lesions have impaired breathing
and are often ventilator-dependent. They can shrug their
shoulders and have neck motion, which may permit the
operation of specially adapted power wheelchairs and
equipment (such as tape recorders, computers, telephones, page turners, automatic door openers, and other
environmental control units). Adaptive devices include
mouth (sip and puff) and voice activation controls, or
chin, head, eyebrow, or eye blink controls. Patients with

625

Spinal Cord Injury

C4 lesions may not require respiratory equipment beyond the initial acute care stage, but may have the same
functional equipment needs as ventilator-dependent patients.
In addition to powered wheelchairs, patients with
C1-C4 tetraplegia require assistance with all personal
care, turning, and transfer functions. Headrests, troughs
or a lapboard (for the upper extremities), and lifts may
be necessary. Bed surfaces with two or more segments
that are alternately inflated and deflated may be indicated for patients who do not have assistance for turning. Power recliners allow independent seated weight
shifts. Patients with partial C5 function may benefit
from adaptive equipment to enable feeding, writing, and
typing.

C5 Tetraplegia
Patients with C5 tetraplegia have functional deltoid
and/or bicep musculature. They can internally rotate and
abduct the shoulder, which causes forearm pronation by
gravity. They can externally rotate the shoulder and
cause supination and wrist extension and can flex the elbow. C5 tetraplegia patients require assistance to perform bathing and lower body dressing functions, for
bowel and bladder care, and for transfers. With the use
of adaptive equipment, C5 tetraplegia patients can feed
themselves, perform oral facial hygienic and upper body
dressing activities, operate some equipment (such as
computers, tape recorders, telephones), and participate
in leisure activities. They can propel manual wheelchairs
short distances on level surfaces. Powered wheelchairs
are needed for community distances and outdoor terrain.

C6 Tetraplegia
Patients with C6 tetraplegia have musculature that permits most shoulder motion, elbow bending, and active
wrist extension. Tenodesis orthoses support tenodesis
training early in recovery. Wrist-driven flexor hinge
splints permit pinching strength that is needed for catheterization and work skills. Short opponens orthoses
with utensil slots, writing splints, Velcro handles, and
cuffs permit feeding, writing, and oral facial hygiene. C6
tetraplegia patients can perform upper body dressing
without assistance; may seldom perform lower body
dressing without assistance; may seldom catheterize
themselves and perform their bowel program with assistive devices; can perform some transfers independently
with a transfer board; can turn independently with the
use of side rails; and can relieve pressure by leaning forward, alternating sides, or possibly by push-ups. Water
mattresses can lower pressure sufficiently to eliminate
the need for turning during the night. They can propel a
manual wheelchair short distances on level terrain, operate power wheelchairs, and may drive a specialized van.

626

Orthopaedic Knowledge Update 8

C7 to C8 Tetraplegia
Patients with C7 tetraplegia have functional triceps, can
bend and straighten their elbows, and also may have enhanced finger extension and wrist flexion. As a result,
these patients have enhanced grasp strength, which permits enhanced transfer, mobility, and activity skills. They
can turn and perform most transfers independently; can
propel a manual wheelchair on rough terrain and slopes,
and may therefore not need a powered wheelchair; can
drive a specialized van; and can perform most daily activities such as cooking and light housework, and therefore may occasionally live independently. They may,
however, require assistance for bowel care and bathing.
C8 tetraplegia patients have flexor digitorum profundus
function, which permits all arm movement, with some
hand weakness. They can propel a manual wheelchair
for community distances, including in and out of a car
and over curbs, and may even become wheelchair independent.

Ambulation After Spinal Cord Injury

Up to 90% of SCI patients who are initially sensory incomplete and motor complete, with preserved pinprick
sensation, will have sufficient motor recovery to walk at
the time of discharge from rehabilitation. Motor incomplete patients have a better prognosis for ambulation
than sensory incomplete patients. Patients who are able
to walk in the community have at least fair hip flexor
strength bilaterally and fair strength in at least one knee
extensor, so that the maximum bracing needed is one
long leg and one short leg orthosis. The Waters Ambulatory Motor Index as well as the ASIA Lower Extremity
Motor Score have been found to correlate and the determinants between motor power and walking ability
still were applicable. The Lower Extremity Motor Score
is determined by adding of the muscle grades of the 10
key muscles. All patients with incomplete injuries who
have a lower extremity motor score of greater than or
equal to 20 at 1 month will progress to ambulation by 1
year postinjury (Table 3).

Tendon Transfer Surgery

The basic principle of upper extremity muscle transfers
is to use functioning proximal musculature to control
distal parts, with minimal risk of loss of function. Arthrodesis may be performed in conjunction with tendon
transfers to stabilize a joint. Wrist fusion is generally
contraindicated in patients with tetraplegia because loss
of wrist extension interferes with manual wheelchair
propulsion and transfers. Upper extremity surgery is
usually not performed before the patient has completed
the rehabilitation program.
At the C5 level, patients have functional use of the
deltoids and elbow flexors. At this level, the most helpful procedures are transfer of the brachioradialis to the

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 51 Spinal Cord Injury

Table 3 | Lower Extremity Motor Score and Ambulation

review of current and future directions in rehabilitation

management. Spine 2001;26(suppl 24):S137-S145.

Incomplete Tetraplegia
Lower Extremity Motor Score at
Percentage Ambulatory at
1 Month
1 Year
1-9
21
10-19
63
> 20
100
Complete Paraplegia
1-9
45
Incomplete Paraplegia
0
33
1-9
70
10-19
100

This article reviews the medical literature to provide a

framework for predicting neurorecovery and functional outcomes after SCI based on injury severity. The authors conclude that within 72 hours to 1 month after SCI, it is possible
to predict with reasonable accuracy the magnitude of expected
recovery based on physical examination (and use of the ASIA
scale). Future directions for improving functional outcomes
through the use of novel interventions such as pharmaceutical
treatment, functional neuromuscular stimulation, standard
protocols to limit secondary injury, and functional training
methods (such as body weight support) to maximize activitydependent neuroplasticity are also addressed.

extensor carpi radialis brevis to restore wrist extension

(C6 level) and transfer of the deltoid to the triceps to
provide elbow extension (C7 level).
Upper extremity reconstructive surgery, or functional neuromuscular stimulation of the upper extremity, or both surgery and stimulation can improve function at the C6 level. Stimulation can be provided by
external, percutaneous, or implanted electrodes, by
shoulder motion using an external system, by key and
palmar grip and release, or by a bionic glove, an electrical stimulator garment that provides controlled grasp
and hand opening.
To restore lateral or key grip in patients without a
natural tenodesis, flexor pollicis longus (FPL) tenodesis
is surgically created by securing the proximal end of the
tendon to the distal radius. To preserve thumb interphalangeal joint flexion, a screw is inserted through the distal tip of the thumb and placed across the interphalangeal joint. When possible, transfer of an active muscle
to the FPL will also provide lateral pinch to the finger
flexors and will provide better function. To provide finger flexion in the C6 patient, tendon transfer to the
flexor digitorum profundus provides grasp and a firm
surface for lateral pinch. The flexor carpi radialis, pronator teres, and extensor carpi radialis longus have all
been used for lateral pinch and finger flexors.
Patients with C7 tetraplegia lack finger and thumb
flexion and intrinsic hand musculature. Transfer of the
brachioradialis to the FPL can achieve restoration of
thumb flexion. Restoration of finger flexion can be
achieved through transfer of the extensor carpi radialis
longus, flexor carpi ulnaris, or pronator teres to the
flexor digitorum profundus.

Annotated Bibliography
Burns AS, Ditunno JF: Establishing prognosis and maximizing functional outcomes after spinal cord injury: A

American Academy of Orthopaedic Surgeons

Fehlings MG, Sekhon LH, Tator C: The role and timing

of decompression in acute spinal cord injury: What do
we know? What should we do? Spine 2001;26(suppl 24):
S101-S110.
This article reviews the role of surgical reduction and decompression, particularly early surgery. A review of the current evidence available in the literature suggests that there is
no standard of care regarding the role and timing of surgical
decompression. The authors conclude that there are insufficient data to support overall treatment standards or guidelines
for this topic. The existence of biologic evidence from experimental studies in animals is noted, which shows that early decompression may improve neurologic recovery after SCI; however, it is concluded that the relevant time frame in humans
remains unclear.

Hurlbert RJ: The role of steroids in acute spinal cord injury: An evidence-based analysis. Spine 2001;26(suppl
24):S39-S46.
This is a literature review of methylprednisolone protocol
that attempts to evaluate the role of steroids in nonpenetrating (blunt) SCIs. From an evidence-based approach, the authors conclude that methylprednisolone cannot be recommended for routine use in acute nonpenetrating SCIs and that
prolonged administration (48 hours) of high-dose steroids is
not without risk and may be harmful to the patient. Until
more evidence is forthcoming, methylprednisolone should be
considered to have investigational (unproven) status only.

Kirshblum S, Campagnolo D, DeLisa J: Spinal Cord

Medicine. Philadelphia, PA, Williams and Wilkins, 2002.
This comprehensive resource of SCI medicine encompasses traumatic and nontraumatic disorders affecting the spinal cord. All aspects of the spinal cord including anatomy, epidemiology, classification, treatment and complications for
acute SCI, comprehensive rehabilitation topics, and aging with
SCI are reviewed. Various nontraumatic disorders of the spinal cord including motor neuron disorders, multiple sclerosis,
and postpolio syndrome are also covered.

National Spinal Cord Injury Statistical Center: Spinal

Cord Injury: Facts and Figures at a Glance, University

627

Spinal Cord Injury

of Alabama at Birmingham, Birmingham, AL. Available
at: www.spinalcord.uab.edu. Accessed May 2001.
The National Spinal Cord Injury Database has been in existence since 1973 and includes data from an estimated 13% of
new SCI patients in the United States. Since its inception, 25
federally funded Model SCI Care Systems have contributed
data to the National SCI Database. As of July 2004, the database contained information on 22,992 patients who sustained
traumatic SCI.

Papadopoulos SM, Selden NR, Quint DJ, Patel N,

Gillespie B, Grube S: Immediate spinal cord decompression for cervical spinal cord injury: Feasibility and outcome. J Trauma 2002;52:323-332.
Ninety-one consecutive patients with acute, traumatic cervical SCI (1990-1997) were prospectively studied to determine
the effect of immediate surgical spinal cord decompression on
neurologic outcome. The authors concluded that immediate
spinal column stabilization and spinal cord decompression,
based on MRI, may significantly improve neurologic outcomes.

Sekhon LH, Fehlings MG: Epidemiology, demographics,

and pathophysiology of acute spinal cord injury. Spine
2001;26(suppl 24):S2-S12.
This review examines the epidemiology, demographics, and
pathophysiology of acute SCI. The authors also discuss future
goals of increasing understanding of both primary and secondary mechanisms of injury, the roles of calcium, free radicals, sodium, excitatory amino acids, vascular mediators, and apoptosis.

Winslow C, Bode RK, Felton D, Chen D, Meyer PR Jr:

Impact of respiratory complications on length of stay
and hospital costs in acute cervical spine injury. Chest
2002;121:1548-1554.
This article attempts to determine if respiratory complications experienced during the initial acute-care hospitalization
in patients with acute traumatic cervical spinal injury are more
important determinants of the length of stay and total hospital
costs than level of injury. A retrospective analysis of an inception cohort (413 patients) for the 5-year period from 1993 to
1997 was performed. The results showed that both mean
length of stay and hospital costs increased monotonically with
the number of respiratory complications experienced. The authors concluded that the number of respiratory complications
experienced during the initial acute-care hospitalization for
cervical spine injury is a more important determinant of
length of stay and hospital costs than level of injury.

Orthopaedic Knowledge Update 8

Barbeau H, Ladouceur M, Norman K, Pepin A, Leroux
A: Walking after spinal cord injury: Evaluation, treatment, and functional recovery. Arch Phys Med Rehabil
1999;80:225-235.
Bracken MB, Shepard MJ, Collins WF, et al: A randomized, controlled trial of methylprednisolone or naloxone
in the treatment of acute spinal cord injury: Results of
the Second National Acute Spinal Cord Injury Study.
N Engl J Med 1990;322:1405-1411.
Bracken MB, Shepard MJ, Collins WR Jr, et al: Methylprednisolone or naloxone treatment after acute spinal
cord injury: One-year follow-up data: Results of the
second National Acute Spinal Cord Injury Study.
J Neurosurg 1992;76:23-31.
Bracken MB, Shepard MJ, Holford TR, et al: Administration of methylprednisolone for 24 or 48 hours or tirilazad mesylate for 48 hours in the treatment of acute
spinal cord injury: Results of the Third National Acute
Spinal Cord Injury Randomized Controlled Trial. National Acute Spinal Cord Injury Study. JAMA 1997;277:
1597-1604.
Brunette DD, Rackswold GL: Neurologic recovery following rapid spinal realignment for complete cervical
spinal cord injury. J Trauma 1987;27:445-447.
Cotler HB, Miller LS, DeLucia FA, Cotler JM, Dayne
SH: Closed reduction of cervical spine dislocations. Clin
Orthop 1987;214:185-199.
Delamarter RB, Sherman JE, Carr JB: Cauda equina
syndrome: Neurologic recovery following immediate,
early, or late decompression. Spine 1991;16:1022-1029.
Dolan EJ, Tator CH, Endrenyi L: The value of decompression for acute experimental spinal cord compression
injury. J Neurosurg 1980;53:749-755.
Frisbie JH: Fractures after myelopathy. J Spinal Cord
Med 1997;20:66-69.
Galandiuk S, Raque G, Appel S, Polk HC Jr: The twoedged sword of large-dose steroids for spinal cord
trauma. Ann Surg 1993;218:419-425.
George ER, Scholten DJ, Buechler CM, Jordan-Tibbs J,
Mattice C, Albrecht RM: Failure of methylprednisolone
to improve the outcome of spinal cord injuries. Am Surg
1995;61:659-663.

Classic Bibliography

Green D: Prevention of thromboembolism after spinal

cord injury. Semin Thromb Hemost 1991;17:347-350.

American Spinal Injury Association: International Standards for Neurological Classification of Spinal Cord Injury, revised 2002. Chicago, IL, American Spinal Injury
Association, 2002.

Guha A, Tator CH, Endrenyi L, Piper I: Decompression

of the spinal cord improves recovery after acute experimental spinal cord compression injury. Paraplegia 1987;
25:324-339.

628

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Merli GJ, Crabbe S, Doyle L: Mechanical plus pharmacological prophylaxis and deep vein thrombosis in acute
spinal cord injury. Paraplegia 1992;30:558-562.
Moberg EA: The present state of surgical rehabilitation
for the upper limb in tetraplegia. Paraplegia 1987;25:
351-356.
Nesathurai S: Steroids and spinal cord injury: Revisiting
the NASCIS II and NASCIS III Trials. J Trauma 1998;
45:1088-1093.
Parsons KC, Lammerts DP: Rehabilitation in spinal
cord disorders: 1. Epidemiology, prevention, and system
care of spinal cord disorders. Arch Phys Med Rehabil
1991;72(suppl):S293-S294.
Staas WE, Formal CS, Freedman MK, et al: Spinal cord
injury and spinal cord injury medicine, in DeLisa JA
(ed): Rehabilitation Medicine Principles and Practice, ed
3. Philadelphia, PA, Lippincott Raven, 1998, pp 12591292.

American Academy of Orthopaedic Surgeons

Chapter 51 Spinal Cord Injury

Stover SL, Fine PR: The epidemiology and economics of
spinal cord injury. Paraplegia 1987;25:225-228.
Tarlov IM, Klinger H: Spinal cord compression studies:
II. Time limits for recovery after acute compression in
dogs. AMA Arch Neurol Psychiatry 1954;71:271-290.
Tator CH, Duncan EG, Edmond VE, Lapczak LI, Andrews DE: Complications and costs of management of
acute spinal cord injury. Paraplegia 1993;31:700-714.
Vaccaro AR, An HS, Betz RR: The management of
acute spinal trauma: Prehospital and in-hospital emergency care. Instr Course Lect 1997;46:113-125.
Vaccaro AR, Daugherty RJ, Sheehan TP, et al: Neurologic outcome of early versus late surgery for cervical
spinal cord injury. Spine 1997;22:2609-2613.
Waters RL, Adkins RH, Yakura JS: Motor and sensory
recovery following complete tetraplegia. Arch Phys Med
Rehabil 1993;74:242-247.

629

Chapter

52

Miscellaneous Neurologic Diseases

Guy W. Fried, MD

Multiple Sclerosis
Multiple sclerosis (MS) is an inflammatory demyelinating disease of the central nervous system that causes
multiple focal lesions. The focal lesions and their neurologic effects are progressive and difficult to predict. MS
is twice as common in women than in men, is diagnosed
at an average age of 30 years, and is 10 times more prevalent in northern geographic areas such as the United
States and Canada than in Asia and Africa. MS is characterized by recurrent episodes of inflammation of the
myelin and it can affect any area in the central nervous
system. The inflammation and immune response destroy
the myelin, cause plaque formation, and disrupt nerve
conduction. The development and progression of MS
usually follows a set pattern according to the history of
exacerbations and remissions. Patients may suffer exacerbations following physical and emotional stressors; for
example, the stress of a surgical procedure. Good prognostic indicators for MS include having no motor findings at the time of presentation, resolution of early
flare-ups, an age younger than 30 years at onset, and
having limited cerebellar and pyramidal findings after 5
years.
Because of the complexity of its symptoms, the differential diagnosis of MS is broad. MS may appear as a
single subtle symptom or as a more significant constellation of symptoms. The differential diagnosis includes
central nervous system tumors, cerebrovascular accident, anterior horn cell disease, myasthenia gravis, and
collagen vascular diseases, as well as the secondary effects of Lyme disease and human immunodeficiency virus (HIV). Currently, MRI is the most accurate tool to
diagnose MS, with positive findings in 72% to 95% of
patients. White matter plaques are a characteristic finding. Gadolinium contrast can help distinguish old from
new lesions. Lumbar puncture may also be useful because spinal fluid characteristics may show increased
gamma globulin in 60% of patients with definite MS.
The oligoclonal band may be seen in the gamma globulin region on gel electrophoresis.

American Academy of Orthopaedic Surgeons

Pain and dysesthesias can often be a prominent feature of MS. More than half of the patients with MS have
a history of an acute or chronic pain disorder. These
symptoms are best treated with anticonvulsant or tricyclic antidepressant medications. When a patient has
pain, evaluation for correctable causes such as carpal
tunnel syndrome, cervical myelopathy, and mechanical
back pain is important.
Spasticity is a velocity-dependent increase in resistance to a passive stretch. Potential complications include reduced joint range of motion and contracture,
poor hygiene, predisposition to decubitus, pain, and an
impaired ability to use volitional motor power, resulting
in functional impairment. Spasticity can be beneficial in
some instances. Spasticity of the knee extensors can assist with stability during transfers and ambulation. The
Modified Ashworth Scale is a frequently used rating system for spasticity (Table 1).
If spasticity interferes with function, treatment may
be considered. Before treatment is initiated, however,
an increase in spasticity should be investigated for an
underlying cause. Any noxious stimulation such as bladder distension, bowel impaction, or an ingrown toenail
may increase a patients spasticity. The initial treatment
for spasticity is usually oral medications such as baclofen, benzodiazepines, tizanidine, and dantrolene sodium; however, these medications can have a sedating effect and may contribute to fatigue. Other treatments
include botulinum toxin serotypes A and B, which are
given as an intramuscular injection that binds at receptor sites inhibiting the release of acetylcholine. According to a 2001 study, botulinum toxin injections can be
quite useful in diminishing focal spasticity, and do not
have any cognitive or fatigue-related adverse effects.
However, the injections potentially can lead to temporary weakness and the cost per injection can begin at
several hundred dollars. Intrathecal baclofen also can be
useful in controlling severe lower extremity spasticity. A
surgically implanted pump allows precise titration to deliver varying doses of baclofen to correspond with the
need for spasticity control throughout the day. The in-

631

Miscellaneous Neurologic Diseases

TABLE 1 | Modified Ashworth Scale

Scale
0
1

1+

2
3
4

Description
No increase in muscle tone
Slight increase in muscle tone, manifested as a catch-andrelease or by minimal resistance at the end of range of
motion
Slight increase in muscle tone, manifested by a catch,
followed by minimal resistance throughout the remainder
(less than half) of the range of motion
More marked increase in muscle tone throughout most of
the range of motion, but the affected limb is easily moved
Considerable increase in muscle tone; passive range of
motion difficult
Affected limb is rigid

(Reproduced with permission from Black-Schaefer R: Stroke (Young), in Frontera W, Silver

J (eds): Essentials of Physical Medicine. Philadelphia, PA, Hanley and Belfus, 2002, pp
784-792.)

trathecal baclofen may allow the patient to take fewer

oral medications and reduce the overall side effect profile.
Orthotic treatment to support ankle and foot positioning as well as supportive management for impaired
functional abilities tend to be the primary rehabilitative
focus. Orthotic treatment may or may not be necessary
because of the variable presentation of MS. The patients degree of sensation, tone, and muscle strength, especially at the knee and ankle, must be considered.
Orthoses are typically made of metal, plastic, or a combination of these materials. The molded plastic anklefoot orthosis (AFO), which immobilizes the ankle in
gait, is the most commonly used. Plastic AFOs are
lighter and more easily concealed, which frequently
leads to better patient acceptance. Fluctuating edema is
better accommodated in a metal orthosis because it
does not closely approximate the skin and therefore allows the limb more flexibility. A limb with poor sensation is also better suited to a metal orthosis because it
can be manufactured with a soft leather shoe or inner
lining to avoid skin breakdown. A metal AFO also can
allow more adjustments for ankle joint control.
The location of the spasticity and how well it is controlled may determine the necessity of surgical procedures. The goal of surgical orthopaedic intervention is to
improve function and correct deformity. The split anterior tibial tendon transfer procedure is used to reduce
excessive inversion of the foot. The tendon of the tibialis
anterior is surgically split, allowing a portion to be attached to the cuneiform and cuboid bones to generate
an eversion force. MS will also frequently lead to an
equinovarus foot, which interferes with ambulation and
may require surgery. Knee flexion contractures can also
occur with severe spasticity. Orthopaedic hamstring

632

Orthopaedic Knowledge Update 8

lengthening procedures followed by postoperative casting can allow better range of motion. Hip flexion
contracture can accompany knee flexion contractures.
Iliopsoas tenotomy may be useful if standard range-ofmotion techniques fail.

Parkinsons Disease
Parkinsons disease is a chronic, degenerative, central
nervous system movement disorder characterized by
resting tremor and bradykinesia. Dementia may also develop in about 10% to 15% of patients. In Parkinsons
disease, the site of pathology is the basal ganglia and extrapyramidal motor system that is responsible for controlling upright posture, muscle tone, coordination, and
the initiation of automatic movements in the face and
body. There is also a deficiency of dopamine, which allows acetylcholine to become predominant and precipitate tremor. The classic tremor occurs at rest and is
abolished by movement; therefore, the disease is not initially disabling. The major motor disability is bradykinesia and akinesia as evidenced by difficulty initiating
movements and decreased associated motions. Clinical
observation will show a patient having moderate difficulty rising from a chair, often needing to use their arms
to push off. The patient will assume a forward-flexed
standing posture and will walk with a slow, moderately
wide-based gait with short steps. Patients may exhibit a
festinating gait, appearing to be falling forward as they
start off slowly and then move faster. The patients show
poor postural reflexes in which they do not reorient
their balance to environmental changes. Cogwheeling,
which represents a basal ganglia disorder of tone, is elicited by passive flexion and extension of a muscle, allowing the examiner to feel repetitive stops during rangeof-motion testing. Patients with Parkinsons disease may
also have a hypokinetic dysarthria, which presents as
limited intonation and affect. Deficits in posture, balance, and tone may lead to frozen joints and falls that
create most of the injuries (such as fractures) requiring
orthopaedic intervention. Postural instability, bradykinesia, and rigidity are the major predictors of falls. Patients
with Parkinsons disease can improve their ability to
move with exercises emphasizing range of motion, balance, gait, and fine motor dexterity; gains can be lost if
regular exercise is stopped. Maintenance exercises are
also useful for management of rigidity and bradykinesia.
The mainstay treatment of symptoms of Parkinsons
disease is medication. Centrally acting anticholinergic
medications can decrease tremors and saliva production. Amantadine potentiates the action of dopamine
and may help with akinesia and rigidity. Levodopa has
been found to be effective at improving bradykinesia
tremor and rigidity when added to carbidopa. The combination of levodopa and carbidopa allows more medication to enter the circulatory system and cross the

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

blood-brain barrier. It has been noted that a low protein
diet allows for more consistent results with levodopa.
Levodopa has also been noted to improve the hypokinetic dysarthria, allowing for faster and more natural lip
movement and thereby improving speech intelligibility.

Peripheral Polyneuropathy
Peripheral polyneuropathy is a common condition in orthopaedic patients. The lower motor neurons of the peripheral nervous system may be damaged in a variety of
diseases. Patients often will have abnormal sensation
that may be described as dull, prickling, numbing, or the
feeling of pins and needles. In addition, autonomic nervous system involvement may produce gastroparesis,
postural hypotension, and problems with regulation of
heart rate. It is important to start with a detailed history
and physical examination evaluating weakness, atrophy,
areflexia, and sensory loss. Physical assessment determines whether involvement is in a focal region of the
body associated with an entrapment site such as carpal
tunnel syndrome or whether it is a diffuse peripheral
condition. Electrodiagnostics can provide an added benefit in helping to clarify the nerve disease process. Needle electromyography and nerve conduction velocity
studies can help determine whether the neuropathic
process is predominantly axonal, demyelinating, or a
combination of both. These studies can also determine
whether the findings involve primarily motor or sensory
nerves, and whether they are symmetric versus multifocal. A careful family history of primary family members
can evaluate for inherited neuropathies, which can appear without a readily apparent clinical etiology.
Diabetes is the most common cause of peripheral
neuropathy. It affects multiple organ systems, including
the peripheral nervous system. The incidence of peripheral neuropathy in diabetic patients depends largely on
how it is defined. If it is defined by patient symptoms,
about half of the individuals are affected. If electrodiagnostic characteristics are considered, the incidence increases to 90%. Peripheral neuropathy is also associated
with end stage renal disease. Studies indicate that at the
time dialysis is initiated, as many as 65% of the patients
have a peripheral neuropathy; axonal damage is more
common than demyelination. The specific etiology is unclear but it appears to be related to toxin buildup because hemodialysis, peritoneal dialysis, and renal transplantation have led to an improvement in symptoms
and electrodiagnostic study results. Chronic alcohol ingestion leads to a similar type of peripheral neuropathy
affecting the axons and leading to significant demyelination. Abstinence from alcohol may improve symptoms.
Pernicious anemia is frequently caused by a deficiency
of gastric intrinsic factor leading to a malabsorption of
vitamin B12 from the ileum. In patients with pernicious
anemia, an axonal peripheral neuropathy is common

American Academy of Orthopaedic Surgeons

Chapter 52 Miscellaneous Neurologic Diseases

with dysesthesias. The peripheral neuropathy can be reversible if vitamin B12 is given early in the disease process.
Paraproteinemias including multiple myeloma, amyloidosis, lymphoma, Waldenstrms macroglobulinemia,
chronic lymphocytic leukemia, and nonmalignant gammopathy may also lead to a neuropathy. Frequently this
is a diffuse distal symmetrical problem. Multiple myeloma most commonly occurs in patients between 40
and 70 years of age. A distal sensorimotor involvement
will present in about 5% of the patients with multiple
myeloma. Radicular pain may occur with or without
vertebral fractures. Many demyelinating neuropathies
with a diffuse symmetric pattern are inherited; for example, Charcot-Marie-Tooth disease that presents with
a progressive symmetric distal muscular atrophy.
HIV is commonly associated with peripheral polyneuropathy. Fifty percent of patients with acquired immunodeficiency syndrome show symptoms of a peripheral polyneuropathy. The HIV infection may produce a
neuropathy, which is clinically indistinguishable from
Guillain-Barr syndrome. The presentation of the demyelinative neuropathy occurs early in the HIV infection
and may in fact be the presenting symptom. GuillainBarr syndrome is the most common disease producing
an acute generalized paralysis from an inflammatory demyelinating polyneuropathy. Patients present with distal
paresthesias and then a generalized weakness that
progresses. Cranial nerve involvement is not uncommon, with the facial nerve being affected in up to one
half of the patients. Early mortality in Guillain-Barr
syndrome is related to respiratory failure, and up to
30% of patients may require mechanical ventilation. Autonomic nervous system involvement will result in difficulties regulating blood pressure, heart rate, and body
temperature as well as bowel and bladder function.
Overall, about 16% of patients with Guillain-Barr syndrome will be left with a degree of permanent disability.
Electromyogram studies of the amplitude of the compound of motor action potential have a strong predictive value for ambulation. Children have a much greater
potential for recovery.
With all the neuropathies a thorough patient history
and physical examination followed by a search for reversible causes are critical. Musculoskeletal treatment emphasizes maintaining basic health status by monitoring autonomic functions including blood pressure, temperature,
and breathing. Musculoskeletal supports include a conditioning endurance program. Musculoskeletal weakness
can be addressed through range-of-motion activities and
orthotic devices such as a lightweight plastic AFO to
maintain function. Surgical treatment consisting of Achilles tendon lengthening, split anterior tibial tendon transfer, and plantar fascial release may prevent later problems.
Before performing surgical procedures, it should be determined whether the underlying process is progressive, such

633

Miscellaneous Neurologic Diseases

as Charcot-Marie-Tooth disease, whether there are reversible factors as in vitamin B12 deficiency, or ominous
factors such as HIV infection.

Postpolio Syndrome
Postpolio syndrome occurs in patients with a history of
acute polio. It has been estimated that 25% to 60% of
the patients who had acute polio may experience latent
effects of the disease. The characteristics of postpolio
syndrome include musculoskeletal pain, fatigue, new
muscle weakness or atrophy, respiratory impairment,
cold intolerance, and a decline in the ability to perform
activities of daily living. The specific cause of postpolio
syndrome is unknown; the etiology has been attributed
to pathophysiologic and functional causes. Pathophysiologic causes include chronic poliovirus infection, death
of the remaining motor neurons with aging, premature
aging, damage to the remaining motor neurons caused
by increased demands or secondary insults, and an
immune-mediated syndrome. Functional etiologies for
postpolio syndrome include greater energy expenditure
as a result of weight gain and muscle weakness caused
by disuse or overuse.
A mild conditioning program for patients with postpolio syndrome may be beneficial while avoiding any
overuse or excessive fatigue that can be detrimental.
Most orthopaedic needs are based on joint, muscle, or
back pain. Many patients require revision of orthotic
devices such as braces, canes, and crutches or new
orthotic devices to treat new symptoms. Common issues
include genu recurvatum, knee pain, back pain, degenerative arthritis, or arthralgia. Surgery for scoliosis or fractures may also be necessary to treat new conditions.

Annotated Bibliography
Multiple Sclerosis
Frohman EM: Multiple sclerosis. Med Clin North Am
2003;87:867-897.
MS is the most common disabling neurologic disease of
young people. It affects up to 450,000 people in the United
States. Substantial advances have been made in diagnosis and
treatment over the past decade. This excellent review article
helps in the formation of initial diagnostic and treatment
plans.

Jin YP, de Pedro-Cuesta J, Huang YH, Soderstrom M:

Predicting multiple sclerosis at optic neuritis onset. Mult
Scler 2003;9:135-141.
This prospective study examines patients who develop definite MS at the onset of optic neuritis. Correlations made with
MRI and oligoclonal bands in cerebrospinal fluid aid in the diagnosis. The study indicated a strong correlation between optic
neuritis leading to definite MS in patients with brain and spinal cord lesions.

634

Orthopaedic Knowledge Update 8

Joy JE, Johnston RB (eds): Multiple Sclerosis: Current
Status and Strategies for the Future. Washington, DC, National Academy Press, 2001, pp 29-69.
This review article is an excellent overview that highlights
the pathophysiology, diagnosis, prognosis, and treatment of
multiple sclerosis.

Keenan MA, Esquenazi A, Mayer NH: The use of laboratory gait analysis for surgical decision making in persons with upper motor neuron syndromes. Phys Med
Rehabil State Art Rev 2002;16:249-261.
A description of laboratory gait analysis and its use in
identifying muscle groups causing spasticity is presented. The
use of this information for presurgical planning is discussed.

Rammohan KW: Axonal injury in multiple sclerosis.

Curr Neurol Neurosci Rep 2003;3:231-237.
The pathophysiology and pathogenesis of MS are discussed. Axonal injury is recognized as an early occurrence in
the inflammatory lesions of MS. Neurologic functional impairment correlates best with axonal damage.

Shapiro RT: Pharmacologic options for the management

of multiple sclerosis symptoms. Neurorehabil Neural
Repair 2002;16:223-231.
This article reviews pharmacologic management of MS.
Immunomodulatory drugs that control the underlying disease
process but do not cure MS are reviewed. The article emphasizes that optimal disease treatment requires a multidisciplinary approach combining medication, rehabilitation, and
patient education.

Yablon SA: Botulinum neurotoxin intramuscular

chemodenervation: Role in the management of spastic
hypertonia and related motor disorders. Phys Med
Rehabil Clin N Am 2001;12:833-874.
Botulinum neurotoxin intramuscular chemodenervation
use is an important therapeutic tool in focal spasticity management. Botulinum toxin injections have been shown to be effective and safe. Techniques, dosage, targeted muscles, and functional goals are discussed in this review article.

Parkinsons Disease
Samii A, Nutt JG, Ransom BR: Parkinsons disease.
Lancet 2004;363:1783-1793.
Parkinsons disease is the most common serious movement disorder in the world, affecting about 1% of adults older
than 60 years. This review article covers diagnosis, treatment,
and pathogenesis.

Peripheral Polyneuropathy
Cannon A, Fernandez Castaner M, Conget I, Carreras
G, Castell C, Tresserras R: Type 1 diabetes mellitus in
Catalonia: Chronic complications and metabolic control
ten years after onset. Med Sci Monit 2004;10:CR185CR190.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

This review article confirms that diabetic microvascular
complications including polyneuropathy and retinopathy were
already present 10 years after the onset of diabetes.

Chapter 52 Miscellaneous Neurologic Diseases

LaBan MM, Martin T, Pechur J, Sarnacki S: Physical and
occupational therapy in the in the treatment of patient
with multiple sclerosis. Phys Med Rehabil Clin N Am
1998;9:603-614.

Postpolio Syndrome
Jubelt B: Post-polio syndrome. Curr Treat Options
Neurol 2004;6:87-93.
Postpolio syndrome describes the late manifestations that
occur in patients 30 to 40 years after acute poliomyelitis, such
as new weakness, muscle pain, joint pain, peripheral nerve
compression, fatigue, and cold intolerance. This review article
discusses the diagnosis and treatment plan.

Classic Bibliography
Cheng Q, Jiang GX, Press R, et al: Clinical epidemiology of Guillain-Barre syndrome in adults in Sweden in
1996-97: A prospective study. Eur J Neurol 2000;7:685692.

American Academy of Orthopaedic Surgeons

McDeavitt JT, Graziani V, Kowalske KJ, Hays RM: Neuromuscular disease: Rehabilitation and electrodiagnosis:
2. Nerve disease. Arch Phys Med Rehabil 1995;76:S10S20.
Meythaler JM, DeVivo MJ, Braswell WC: Rehabilitation
outcomes of patients who have developed GuillainBarr syndrome. Am J Phys Med Rehabil 1997;76:411419.
Miller RG, Peterson GW, Daube JR, Albers JW: Prognostic value of electrodiagnosis in Guillain-Barr syndrome. Muscle Nerve 1988;11:769-774.

635

Chapter

53

Traumatic Brain Injury and Stroke

Barbara J. Browne, MD
Jeanne G. Doherty, MD

Epidemiology
Traumatic brain injury is defined as any insult to the brain
caused by an external force, causing temporary or permanent impairments in physical function, cognitive ability,
and/or disturbance of behavioral and emotional function.
These impairments may cause total or partial functional
disability and psychosocial maladjustment. Two million
brain injuries occur in the United States each year. Approximately 50,000 patients die from acute brain injury.
Leading causes of traumatic brain injury include motor vehicle crashes (50%), falls (21%), gunshot wounds
(12%), and recreational injuries (10%). Alcohol is a factor in approximately 60% of injuries across all age
groups.
Approximately 750,000 Americans suffer strokes
each year; nearly one third of these patients die. Those
who survive experience significant levels of disability,
with hemiparesis as the most common impairment. Only
10% of stroke survivors experience a full recovery. Another 10% fail to improve. The remaining 80% have
varying degrees of neurologic impairment that will improve with rehabilitative intervention.

Mechanism of Injury
There are primary and secondary mechanisms of injury
in traumatic brain injury. Primary injury occurs immediately at the time of impact and is associated with
acceleration-deceleration and rotational forces. Secondary injury is the neurochemical and physiologic sequelae
of the primary brain insult, which occurs over hours to
days after the initial injury.
Primary injury includes skull fracture, intracranial
hemorrhage, cortical contusion, diffuse axonal injury (a
significant injury to the white matter of the brain), and
penetrating injury. The two basic types of skull fractures
are those of the cranial vault and the basilar skull. Fractures of the occipital condyles also may occur. Intracranial
hemorrhages include epidural and subdural hematomas
and intracerebral and subarachnoid hemorrhages.
Although skull fractures increase the risk of seizure
and intracranial hematoma, there is no correlation be-

American Academy of Orthopaedic Surgeons

tween the severity of the skull fracture and the severity

of the neurologic injuries. A cranial vault fracture may
be linear or stellate, and depressed or nondepressed. A
depressed skull fracture initially may be missed secondary to few or no neurologic symptoms. The mortality
rate in individuals with a depressed skull fracture is
11%. Approximately 11% of patients with a depressed
skull fracture will be left with permanent neurologic
deficits. Depressed skull fractures can be further subdivided into simple or closed (15%) and compound or
open (85%). A simple fracture is not associated with a
laceration. There is no evidence to suggest that surgical
elevation of a simple, depressed skull fracture reduces
the incidence of seizures or improves neurologic deficits
unless there is a skull fragment compressing the cerebral hemisphere. A compound, depressed skull fracture
is associated with a scalp laceration. These types of skull
fractures should be explored and surgically elevated to
reduce the risk of central nervous system infection. A
basilar skull fracture increases the risk of meningitis
with a dural tear, pneumocephalus, cerebrospinal fluid
fistula, and cranial nerve injury.
Secondary injuries include the effects of intracranial
pressure, cerebral edema, and hydrocephalus, which may
cause cerebral ischemia and hypoxia of brain tissue.
They may also include brain herniation syndromes,
which can cause significant damage to the brainstem.
The two major mechanisms causing stroke are ischemia (80%) and hemorrhage (20%). These two groups
can be further subdivided by location and cause. Ischemic
stroke has three main mechanisms: thrombosis, embolism,
and systemic hypoperfusion. Each of these mechanisms
can result from various underlying pathologic conditions.
Hemorrhages are most commonly associated with uncontrolled hypertension, aneurysms, and arteriovenous malformations.

Types of Deficits
Stroke and brain injury can cause mobility and self-care
deficits and behavioral, emotional, and cognitive dys-

637

Orthopaedic Knowledge Update 8

Traumatic Brain Injury and Stroke

function. Certain generalizations regarding types of deficits can be made based on the anatomic location of injury in the brain. Injuries involving the cerebral
hemispheres can result in contralateral paralysis, sensory changes, and visual field losses. Paralysis of the arm
tends to be more pronounced with middle cerebral artery lesions, whereas lower limb weakness is most pronounced with anterior cerebral artery lesions. Frontal
lobe injury often causes impairments of insight, judgment, logical reasoning, problem solving, memory, and
higher level executive functioning. It may also cause apathy, withdrawal from social interaction, emotional
swings, impulsiveness, and confabulation. Parietal lobe
injuries often cause agnosia (inability to recognize sensory stimuli), apraxia (inability to perform purposeful
movement despite adequate motor ability), aphasia,
visual-spatial disorders, and hemispatial neglect. Injury
to the dominant hemisphere can cause language and
speech disorders. Nondominant hemispheric injury may
cause deficits such as anosognosia (loss of awareness of
ones own deficits) and constructional apraxia (inability
to reproduce drawings or assemble and build structures
despite adequate motor ability). Emotional dysfunction
such as depression and indifference may occur. Temporal lobe injury may cause disturbances in behavior,
memory, and cognition. Dominant hemisphere injury
may cause fluent aphasia, alexia, and agraphia (inability
to express thoughts in writing). Anxiety, depression,
rage, and fear also may occur. Occipital lobe injury may
cause visual impairment or occipital blindness (acting
sighted despite blindness). Brainstem stroke generally
spares cognitive and language function but often shows
diverse symptoms. Impairments can involve both ipsilateral and contralateral functions, cranial nerve dysfunction, ataxia, vertigo, and dysphagia.
In the early period after a stroke, the affected limbs
are often flaccid with an absence of deep tendon reflexes.
Over several weeks, a progression from flaccidity to spasticity to normal tone may occur. Motor return frequently
begins as a synergistic contraction of multiple limb muscles rather than isolated action. Flexor activity predominates in the upper limb. A common pattern is scapular retraction and depression, shoulder adduction with internal
rotation, forearm pronation, and flexion at the elbow,
wrist, and fingers. Extensor overactivity is predominant in
the lower limb with adduction and internal rotation at the
hip, extension at the knee, and plantar flexion at the ankle
with supination at the foot. The entire recovery process is
variable and may halt at any stage. Early motor recovery
is usually seen in proximal muscles before distal or isolated muscle movement. Maximal neurologic and functional recovery has been shown to occur within 12 weeks
for most patients undergoing standard inpatient rehabilitation.

638

Complications
Spasticity
Spasticity is discussed in detail in chapter 52.

Contractures
A contracture is defined as a fixed loss of passive joint
range of motion (ROM) secondary to pathology of connective tissue, tendons, ligaments, muscles, joint capsule,
and/or cartilage. Contractures occur in up to 84% of patients and can be classified as arthrogenic, soft-tissue, or
myogenic. Arthrogenic contractures are caused by pathology of the intrinsic joint components, and cause restriction of ROM in all planes. Soft-tissue contractures
result in shortening of tendons, ligaments, and skin,
causing restriction of movement in one plane. Myogenic
contractures can be further classified as intrinsic or extrinsic. Intrinsic myogenic contractures are caused by a
primary disorder of muscle fibers, such as muscular dystrophy. In extrinsic myogenic contractures the muscle itself is histologically normal. Such contractures are secondary to muscles being placed in a shortened position
for extended periods of time such as occurs in brain injury. Factors that contribute to the occurrence of contractures include spasticity, immobility, prolonged bed
rest, weakness, improper positioning, pain, and heterotopic ossification (HO). Muscles such as the iliopsoas,
gastrocnemius, hamstrings, biceps, and tensor fascia lata
that cross two joints are at greatest risk of contracture.
Common locations in the lower extremity for contractures include ankle plantar flexors, hip flexors, and knee
flexors. Common locations for contractures in the upper
extremity are elbow flexors/supinators and shoulder
adductors/internal rotators.
The most important aspect of treatment of contractures is prevention. Early mobilization, daily ROM exercises, stretching, and the use of splints and orthotic devices are essential for preventing contractures. Proper
positioning in bed and the wheelchair must also be addressed. To avoid knee and hip flexion contractures in
bed, pillows should not be placed under the knees, and
lying in the prone position also can be helpful. To avoid
extreme shoulder adduction and internal rotation, pillows should be placed to keep the shoulder in a partially
abducted and externally rotated position. Elbows should
not be positioned in flexion and supination. Resting
night splints and bivalved casting can be used as a preventive measure. Ankle plantar flexion contractures are
common but preventable. In bed, the use of ankle-foot
orthoses ideally placed at 90 can help to prevent these
contractures. In addition, there are many wheelchair
modifications that can be used to alter a patients position and assist in preventing contractures. To encourage
normal lordosis of the lumbar spine and kyphosis of the
thoracic spine, the pelvis should be placed in a slightly
anterior tilted position. Armrests and lapboards are

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

used to prevent shoulder contractures. Reclining the
back support 10 and placing lateral trunk supports assists in trunk stabilization and proper alignment. Extensions or hip blocks placed laterally keep the pelvis in a
symmetrical position and align the lower extremities.
Leg straps prevent adduction of the lower extremities.
Footrest height can be adjusted to properly position the
ankle, knee, and hip. Despite all efforts to prevent contractures they may still develop. If so, treatment measures should be initiated immediately. Treatment includes physical, medical, and surgical intervention. A
significant aspect of physical intervention is an aggressive stretching program with sustained stretching techniques. To correctly stretch a joint, it is positioned at the
limit of ROM and sustained in that position. Caution
should be used with patients who have significant osteoporosis so that fractures are avoided. Before stretching, the use of therapeutic heat including ultrasound can
be very beneficial. Patients with impaired sensation
should be monitored closely. Serial casting applied to a
prestretched limb is a common treatment method for
joint contracture. The joint is held at its initial limit of
ROM. Casts are periodically removed and new casts
placed after increasing joint ROM by 5 to 10, the new
limit of ROM. The skin must be monitored for breakdown. Dynamic splints, which have movable parts to
counter contracting forces, are an alternative to serial
casting. Medical treatment includes maximizing treatment of spasticity and pain with modalities and medication (see chapter 52). Surgical intervention is reserved
for patients who are refractory to conservative treatment. Options include joint manipulation, tendon release, tendon lengthening, and joint capsule release.
Complications that can occur following tendon lengthening include reduction of muscle strength and altered
ROM. There is also a tendency for tendon shortening to
recur. Postoperative therapy is a key component to ensure the most favorable outcome.

Skin Conditions
Skin breakdown and pressure ulcers are preventable
complications following traumatic brain injury and
stroke. Risk factors include impaired cognition, decreased mobility in bed, diaphoresis, incontinence, infection, diabetes, malnutrition, anemia, muscle atrophy, and
impaired sensation. Spasticity and joint contractures are
significant contributing factors. The common mechanisms for skin involvement include pressure, shear, maceration, and friction. Preventive treatment includes
proper positioning, timely turning in bed, weight shifts
in the wheelchair, minimizing excessive perspiration and
urinary and fecal incontinence, proper transfer techniques, adequate nutrition, treatment of medical conditions such as diabetes and anemia, and appropriate
treatment of spasticity and joint contractures.

American Academy of Orthopaedic Surgeons

Chapter 53 Traumatic Brain Injury and Stroke

Shoulder Pain
Up to 85% of people with hemiplegia experience shoulder pain. Shoulder spasticity and subluxation are among
the most frequent causes of pain. Additional factors include adhesive capsulitis, impingement syndrome with
rotator cuff injury, complex regional pain syndrome
type 1 (CRPS 1), and brachial plexopathy. Effective
treatments to reduce spasticity-related muscle imbalance include electromyogram biofeedback with relaxation exercises, botulinum toxin, and phenol blocks.
Shoulder subluxation has been implicated as a contributing factor in pain, limited ROM, and function. The use
of arm slings can reduce pain and subluxation in some
patients and may also improve gait stability by limiting
detrimental displacement of the bodys center of gravity
seen in the hemiparetic gait. However, sling use can also
result in increased flexor tone that may lead to contracture formation, and may promote disuse of the extremity. Wheelchair lap trays or arm boards provide support
but may also overcorrect the subluxation and lead to
impingement syndromes. Corticosteroid injections can
be an effective treatment for adhesive capsulitis. Other
useful adjuncts include transcutaneous electrical nerve
stimulation to reduce subluxation and pain. Current trials of percutaneous intramuscular stimulation show
promise for enhanced efficacy. Pain that is thought to
originate centrally or related to CRPS 1 may respond to
gabapentin, tricyclic antidepressants, oral steroids, or
stellate ganglion blocks.

Systemic Complications
Dysphagia resulting in aspiration pneumonia occurs in
20% of stroke survivors and up to 45% of traumatic
brain injury survivors. A video fluoroscopic swallow
study provides additional sensitivity and identifies most
cases of dysphagia with aspiration. Factors that increase
the risk of aspiration pneumonia include nasogastric
feeding, tracheostomy, lethargy, emesis, and reflux. Specialized diets often involve thickened liquids with soft or
pureed solids. Patients believed to be at high risk for aspiration should have enteral feedings. A gastrostomy or
jejunostomy tube is inserted if prolonged enteral feeding is anticipated. Malnutrition is a possible complication of dysphagia.
Urinary dysfunction is often seen following stroke
and brain injury. Complications include urinary tract infection, neurogenic bladder, and less frequently, nephrolithiasis and urethral stricture.
Esophagitis, gastritis, ulcers, gastrointestinal hemorrhage, pancreatitis, diarrhea, and constipation may also
develop. An abnormal liver function test result is a common finding after brain injury. It may be drug induced
or related to trauma and often is influenced by some
premorbid factor.

639

Traumatic Brain Injury and Stroke

TABLE 1 |Brooker Classification Scale for Heterotopic

Ossification of the Hip
Class
I
II
III
IV

Description
Islands of bone with soft tissue
Bone spurs from the pelvis or proximal femur, leaving at
least 1 cm between bone surfaces
Bone spurs from the pelvis or proximal femur, reducing
the space between opposing surfaces to less than 1 cm
Bone ankylosis of the hip

(Reproduced with permission from Blount PJ, Bockenek WL: Heterotopic ossification, in
Frontera W, Silver J (eds): Essentials of Physical Medicine. Philadelphia, PA, Hanley and
Belfus, 2002, pp 569-574.)

Atelectasis, pneumonia, deep venous thrombosis

(DVT), pulmonary embolism, and pneumothorax are
possible complications after stroke and traumatic brain
injury. The more severely affected patients will require
ventilator support and many will require a tracheostomy. Cardiovascular complications include hypertension, arrhythmias, cardiomyopathy, myocardial infarction, and endocarditis.
DVT has been reported in up to 70% of unprotected
stroke patients and in approximately 20% of patients after traumatic brain injury. Risk factors include bed rest,
hemiplegia, flaccidity, older age, obesity, history of malignancy, and previous thrombosis. An additional risk
factor in the posttraumatic brain-injured population is
lower extremity fracture. The risk of DVT appears lowest in patients capable of ambulating a distance of more
than 100 feet. Differential diagnosis includes cellulites,
fracture, tumor, CRPS, and HO. A full discussion of
treatment of DVT is found in chapter 51, Spinal Cord
Injury.

Seizure Disorder
A seizure occurs in 5% to 25% of stroke survivors. The
highest rates follow hemorrhagic, cortical, and embolic
strokes. Seizure disorder following traumatic brain injury occurs in 1.5% of patients with mild injuries, 2.9%
with moderate injuries, and 17% of those with severe injuries. These posttraumatic seizures can be classified as
immediate (occurring within hours of the injury), early
(first detected within the first week of injury), and late
(occurring more than 1 week after injury). It is estimated that 12% to 15% of new onset seizures following
a traumatic brain injury occur 10 to 30 years after the
acute injury. Risk factors for posttraumatic seizures include depressed skull fractures, prolonged posttraumatic
amnesia, missile injuries with penetration and retained
metal fragments, loss of consciousness for more than 24
hours, focal neurologic signs on initial examination, intracerebral hemorrhage, diffuse brain contusion, cortical
injury with subcortical extension, advanced age, and rec-

640

Orthopaedic Knowledge Update 8

reational drug use. Previously there has been controversy over the need for prophylactic treatment of posttraumatic seizures. It has been shown that prophylactic
treatment is not indicated for more than 1 week after
traumatic brain injury unless there is documented seizure activity following injury, a history of seizure disorder, or multiple risk factors. A seizure occurring 2 weeks
or more after stroke probably represents scar formation
and is usually treated with long-term prophylaxis.

Heterotopic Ossification
HO is a musculoskeletal complication that may occur
following traumatic brain injury and less frequently after stroke. HO is described as new bone formation in
nonskeletal tissue, located periarticularly. HO is not specific to traumatic brain injury and stroke; it also can develop following a spinal cord injury, burn, fracture, total
joint arthroplasty, or trauma to muscle or joints. HO is
defined as neurogenic when it occurs secondary to traumatic brain, stroke, or spinal cord injury. The reported
incidence of HO after traumatic brain injury varies depending on the study and ranges from 11% to 75%.
There is an increased incidence to 76% in patients with
severe traumatic brain injury. Approximately 20% to
30% of patients have clinically significant loss of ROM
from HO and 10% to 15% have complete ankylosis. The
Brooker classification system has been used to describe
HO of the hip (Table 1).
Risk factors for developing HO subsequent to traumatic brain injury include the presence of fractures, especially those of the long bones; prolonged coma (more
than 2 weeks); spasticity; and decreased ROM. HO usually develops within 1 to 6 months of injury but it may
develop as early as 2 weeks or as late as 12 months after
injury. The clinical presentation of HO is variable. Most
commonly it presents initially as decreased joint ROM
or increased spasticity. A patient may also present with
a low-grade fever, erythema, pain, swelling, or tenderness. Less frequently HO may present with a nerve
compression, vascular compression, or lymphedema. It
is most frequently found at the hips, then the shoulders
and elbows, then the knees. In the hip, HO can form inferomedially, anterolaterally, or posteriorly. In the knee
it is usually found anteromedially, in the shoulder it usually forms inferomedially, and in the elbow it forms
along the medial collateral ligament. Alternative diagnoses include DVT, cellulitis, CRPS, septic joint, hematoma, or tumor. The gold standard for diagnosing
HO is the three-phase radionuclide bone scan. Phase I
and II of the bone scan can detect the condition as early
as 2 to 3 weeks after the onset and phase III may be
positive in 4 to 8 weeks. The bone scan usually normalizes 7 to 12 months from the time of initial diagnosis. Although its sensitivity is high, the disadvantage of using

American Academy of Orthopaedic Surgeons

Chapter 53 Traumatic Brain Injury and Stroke

Orthopaedic Knowledge Update 8

the bone scan for diagnosing HO is its lack of specificity. Serum alkaline phosphatase is another test used in
the diagnosis of HO, with elevated levels associated with
clinically significant HO. Normal levels range from 38 to
126 IU. Serum alkaline phosphatase levels begin rising
to the upper limits of normal within the first 2 weeks
and may begin to exceed normal values within the first
3 weeks from the initial onset of HO; peak levels are
reached at approximately 10 weeks and elevation may
persist for up to 6 months. Serum alkaline phosphatase
levels parallel the clinical course of HO. Elevated serum
alkaline phosphatase levels also are not specific. Levels
may be elevated in clinical findings such as hepatic dysfunction or fractures. If there is a suspicion of HO based
on elevated serum alkaline phosphatase levels, a threephase bone scan should be performed for confirmation.
Treatment of HO includes physical therapy, medication, and surgical intervention. Physical therapy includes
early aggressive ROM exercises, stretching, and joint
mobilization. There is no evidence to suggest that this
intervention increases the formation of HO because of
an inflammatory response. At times, manipulation under
anesthesia is necessary for active ROM of a joint that
has not responded to conservative physical therapy.
Medication options for treatment of HO include etidronate disodium and nonsteroidal anti-inflammatory
drugs (NSAIDs). Etidronate disodium is a biphosphonate that limits ossification by blocking the formation of
hydroxyapatite crystals. The prophylactic oral dose is 20
mg/kg/day for 2 weeks, then 10 mg/kg/day for 10 weeks.
The treatment oral dose is 20 mg/kg/day for 6 months or
20 mg/kg/day for 3 months, then 10 mg/kg/day for 3
months. Adverse effects of etidronate disodium include
nausea, vomiting, diarrhea, and abdominal discomfort; it
can also contribute to osteoporosis and impaired fracture healing. NSAIDs reduce bone formation by inhibiting prostaglandin synthetase. They have been shown to
be effective in preventing HO following total hip arthroplasty and HO resection. Indomethacin is the
NSAID most often used for treatment. NSAIDs have
not been proven specifically to decrease the incidence
of neurogenic HO.
Indications for surgical intervention for HO include
impaired positioning, which causes difficulty with sitting,
lying, hygiene, ambulation, and activities of daily living;
and ankylosed joints, which cause skin breakdown, profound pain, and progressive nerve compression. Some
studies have indicated that patients with good neuromuscular control in general have the best postoperative
functional outcome. Those with a poor neurologic recovery and persistent spasticity have a higher incidence of
recurrence and less functional improvement in the limb.
The surgery may be a wedge or complete resection. It is
essential to defer the surgery until the bone formation is
mature, at least 18 to 24 months from the initial diagnosis, to limit excessive bleeding. Cautious dissection, ade-

American Academy of Orthopaedic Surgeons

quate hemostasis, neurovascular bundle isolation, and

adequate bone mass exposure are other measures to reduce morbidity. Postoperative complications include
hemorrhage, sepsis, and recurrence of HO. Postoperative prophylaxis to prevent recurrence includes medications such as etidronate disodium or NSAIDs. Also, low
dose radiation of 600 to 750 rads may be an option.
There have been some studies exploring the use of a
postoperative continuous passive ROM machine. Therapy for ROM begins when pain and swelling have subsided, usually 10 to 14 days postoperatively.

Fractures
There are certain complications such as fractures and
nerve injuries that occur specifically after traumatic
brain injury. Fractures occur in approximately 34% of
patients; 11% are estimated to be occult fractures. Common types include those of the pelvis, hip, knee, shoulder, and cervical spine fractures and dislocations.
Screening diagnostic studies including radiographs of
the cervical and thoracic spine, pelvis, hips, and long
bones should be obtained in comatose patients, especially those with high velocity injuries. Bone scan should
also be considered as a screening study. Treatment of
these fractures should include early fixation because of
the strong possibility that these patients may eventually
become agitated and confused. Delayed treatment of
these fractures may become more difficult because of
the increased risk for the development of hypertonicity
and spasticity. Early stabilization is associated with
fewer pulmonary complications, decreased use of pain
medications, fewer joint contractures, decreased mortality, and shortened hospital stays. It also allows for earlier mobilization and thus facilitates earlier rehabilitation.

Nerve Injury
Peripheral nerve injuries occur in approximately 34% of
patients with severe traumatic brain injuries; 11% are
occult injuries. Causes include direct trauma, improper
positioning, postoperative complications, and HO. Cranial nerve injuries may also occur and are associated
with the direct trauma.

Assessment
There are multiple assessment tools used to measure
both severity of brain injury and function after injury.
The Glasgow Coma Scale is the gold standard for measuring severity of injury in the acute stage. One score totaling from 3 to 15 is obtained based upon eye opening,
verbal response, and best motor response (Table 2). Injury is classified as severe if the score is 3 to 8, moderate
from 9 to 12, and mild from 13 to 15. Loss of consciousness for more than 6 hours is indicative of a severe injury. Posttraumatic amnesia is memory loss following in-

641

Orthopaedic Knowledge Update 8

Traumatic Brain Injury and Stroke

TABLE 2 |Glasgow Coma Scale

Scale
6
5
4

Eye Opening

Verbal Response

Spontaneous

Oriented
Confused

3
2
1

To speech
To pain
No response

Inappropriate
Incomprehensible
No response

Motor Response
Obeys commands
Localizes to pain
Withdraws from
pain
Flexor posturing
Extensor posturing
No response

(Reproduced with permission from Burke DT: Traumatic brain injury, in Frontera W, Silver J
(eds): Essentials of Physical Medicine. Philadelphia, PA, Hanley and Belfus, 2002, pp 806812.)

jury. A brain injury is mild if it lasts less than 1 hour

after initial injury and severe if it persists for more than
24 hours.

Outcome Measures
Although there are a variety of assessment tools used in
patients with brain injury, many do not have outcome
predictive value. Among the tools used to predict outcome are the Glasgow Outcome Scale, Disability Rating
Scale, Functional Independence Measures, Functional
Assessment Measures, and Galveston Orientation Assessment Test.
Several general characteristics can predict a good
outcome in the brain-injured patient: limited trauma,
posttraumatic amnesia for less than 4 weeks, loss of consciousness for less than 2 weeks, Glasgow Coma Scale
score greater than 5, age younger than 60 years, strong
support system, more highly educated, and premorbid
higher intelligence. A poorer outcome is predicted with
recurrent injury, a mass lesion, anoxia, elevated intracranial pressure, hypotension, history of alcohol or drug
use, premorbid disability, violent etiology, poor work
history, and premorbid psychiatric history.

Rehabilitative Intervention
Effective rehabilitation of the multiple impairments
arising from stroke and brain injury requires a team approach. Traditional team members include a physician,
nurse, psychologist, and physical, occupational, speech,
and recreational therapists. Experienced case management is crucial for overall coordination, communication,
and successful discharge. Any restrictions such as weight
bearing or ROM must be identified before beginning
rehabilitation. Conventional approaches include ROM
exercises, mobilization activities, and teaching compensatory skills. Newer promising treatments of stroke and
brain injury are focused on central nervous system recovery using periods of intensive active motor training.
Theoretical explanations for success of these newer

642

treatments include avoiding learned nonuse of the affected side and brain reorganization. Other promising
treatments include treadmill gait training with bodyweight support, which can promote balance and gait efficiency, and neuromuscular electrical stimulation.

Annotated Bibliography
Epidemiology
Petrilli S, Durufle A, Nicolas B, Pinel JF, Kerdoncuff V,
Gallien P: Prognostic factors in the recovery of the ability to walk after stroke. J Stroke Cerebrovasc Dis 2002;
11:330-335.
This article presents a prospective study of 93 stroke patients; factors predictive of future ambulation are identified.

Complications
Blount PJ, Bockenek WL: Heterotopic ossification, in
Frontera W, Silver J (eds): Essentials of Physical Medicine. Philadelphia, PA, Hanley and Belfus, 2002, pp 569574.
A comprehensive description of HO including definition,
diagnosis, and treatment is presented in this chapter.

Brashear A, Gordon MF, Elovic E, et al: Intramuscular

injection of botulinum toxin for the treatment of wrist
and finger spasticity after a stroke. N Engl J Med 2002;
347:395-400.
This article discusses a randomized, blinded, and controlled study that demonstrates the effectiveness of botulinum
toxin to reduce spasticity and improve ROM.

Farmer SE, James M: Contractures in orthopedic and

neurological conditions: A review of causes and treatment. Disabil Rehabil 2001;23:549-558.
This article reviews the scientific literature from 1966 to
2000 on the development and treatment of contractures in
neurologic and orthopaedic conditions. Information is presented on predisposing factors, muscle physiology, and the effectiveness of various treatment modalities for contractures.

Gardner MJ, Ong BC, Liporace F, Koval KJ: Orthopedic

issues after cerebrovascular accident. Am J Orthop
2002;31:559-568.
A review of the most common orthopaedic problems and
deformities experienced after stroke is presented. Muscle
spasticity, contractures, and HO are discussed along with
methods of prevention and surgical intervention.

Melamed E, Robinson D, Halperin N, Wallach N, Keren

O, Groswasser Z: Brain injury-related heterotopic bone
formation: Treatment strategy and results. Am J Phys
Med Rehabil 2002;81:670-674.
In this study, the results of 12 excisions of HO in 9 patients
were assessed. Functional improvement was noted in all patients 1 year after intervention. There was no evidence of re-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

currence. Recommendations were made regarding indications
for excision of HO.

Assessment
Burke DT: Traumatic brain injury, in Frontera W, Silver
J (eds): Essentials of Physical Medicine, Philadelphia,
PA, Hanley and Belfus, 2002, pp 806-812.
This chapter presents an overview of a number of functional assessment tools used for patients with posttraumatic
brain injury.

Rehabilitative Intervention

Chapter 53 Traumatic Brain Injury and Stroke

recovery in both acute and chronic hemiparesis using more intense periods of active use of paretic limbs is explored. Potential efficacy with other impairments such as dystonia, aphasia,
and phantom pain are addressed.

Yu D, Chae J: Neuromuscular stimulation for treating

shoulder dysfunction in hemiplegia. Crit Rev Phys Med
Rehabil 2002;14:1-23.
A review of several randomized controlled trials of neuromuscular stimulation to treat shoulder subluxation and pain in
hemiplegic patients is presented. Potential causes of shoulder
pain are reviewed.

Chae J: Neuromuscular electrical stimulation for motor

relearning in hemiparesis. Phys Med Rehabil Clin N Am
2003;14:S93-S109.

Classic Bibliography

This article examines the types of neuromuscular electrical

stimulation technologies being used and their clinical usefulness in aiding upper limb functional recovery after hemiparesis.

Bonyke C, Boake C: Principles of brain injury rehabilitation, in Braddom R, Buschbacher R, Dumitru D, et al

(eds): Physical Medicine and Rehabilitation. Philadelphia, PA, WB Saunders, 1996, pp 1027-1052.

Hesse S, Werner C, Frankenberg S, Bardeleben A:

Treadmill training with partial body weight support after stroke. Phys Med Rehabil Clin N Am 2003;14:S111S123.
This article explores treadmill training with body-weight
support as a means to enhance gait recovery after stroke.
Technical aspects of using a treadmill gait-trainer are covered.

Snels IA, Dekker JH, van der Lee JH, Lankhorst GJ,
Beckerman H, Bouter LM: Treating patients with hemiplegic shoulder pain. Am J Phys Med Rehabil 2002;81:
150-160.
A literature review of 14 studies is presented, exploring
the causes and treatment of hemiplegic patients with shoulder
pain.

Taub E, Uswatte G, Morris DM: Improved motor recovery after stroke and massive cortical reorganization following Constraint-Induced Movement therapy. Phys
Med Rehabil Clin N Am 2003;14(suppl 1):S77-S91.
This article reviews newer theories on brain plasticity and
the effects of learned nonuse. The potential for greater motor

American Academy of Orthopaedic Surgeons

Garland DE: A clinical perspective on common forms

of acquired heterotopic ossification. Clin Orthop 1991;
263:13-29.
Garland D, Keenan MA: Orthopedic strategies in the
management of the adult head injured patient. Phys
Ther 1983;63:2004-2009.
Garland DE, Rhoades ME: Orthopedic management of
brain-injured adults: Part II. Clin Orthop 1978;131:111122.
Ippolito E, Formisano R, Farsetti P, Caterini R, Penta F:
Excision for the treatment of periarticular ossification
of the knee in patients who have a traumatic brain injury. J Bone Joint Surg Am 1999;81:783-789.
Langhorne P, Stott DJ, Robertson L, et al: Medical complications after stroke: A multicenter study. Stroke 2000;
31:1223-1229.
Woo BH, Nesathural S (eds): The Rehabilitation of People With Traumatic Brain Injury. Boston, MA, Blackwell
Science Inc, 2000, pp 13-17, 35-43, 95-99.

643

Chapter

54

Limb Amputation and Prosthetic

Rehabilitation
Michael C. Munin, MD
Gary F. Galang, MD

Incidence and Etiology

The most recent (1993) National Health and Human
Services data show that there are about 1.5 million amputees in the United States, with 50,000 new amputations occurring each year. About 5% of amputations are
tumor related, 15% result from trauma, and the remaining approximately 80% are vascular in etiology, with
diabetes-related ischemic vascular disease as the major
cause. In the United States there was a 27% increase in
the overall rate of dysvascular amputations from 1988 to
1996, with a large part of the increase occurring in geriatric and minority populations. Traumatic and cancerrelated amputations decreased by one half within the
same period. The average age for patients with amputation resulting from vascular etiologies is between age 51
to 69 years, whereas the average patient age for traumarelated amputations is 21 to 30 years. The overall male
to female ratio is approximately 2:1.
Most adult limb amputations occur in the lower extremity (80%); partial foot amputations comprise 50%
and transtibial and transfemoral amputations about
25% each. Upper extremity amputations account for
the remaining 20% of all amputations, with most resulting from work-related injuries involving the dominant
arm in healthy younger patients. The most common upper limb amputations are at the transradial level (57%),
followed by transhumeral amputations (23%).
Congenital abnormalities are the most common
cause of limb deficiencies for patients younger than age
10 years. Causes of congenital amputation include intrapartum exposure to drugs and radiation (10%) and genetic deficits (20%); the etiology of the remaining 70%
is unknown. Congenital upper limb absence is usually
seen at the level of the transverse upper third of the left
radius.
Limb salvage or reconstruction has recently become
an increasingly viable alternative to amputation for patients with a traumatic or oncologic etiology. A recent
multicenter study comparing surgical reconstruction
with limb amputation after major trauma (type IIIB,
IIIC tibial fractures) reported that the outcomes for pa-

American Academy of Orthopaedic Surgeons

tients who underwent reconstruction were not significantly different from those who underwent amputation.
However, reconstruction was associated with a higher
risk of complications, additional surgeries, and rehospitalization.

Surgical Technique and Functional Outcomes

Bone
In general, ambulation and prosthetic control are optimized with the preservation of the knee joint and maintenance of a longer residual bone length, which provides
better pressure distribution that minimizes skin breakdown with prosthetic wear. A longer residual bone
length also provides a longer torque arm, improving
control and decreasing the metabolic demands of prosthetic ambulation. In geriatric dysvascular amputees,
90% of transtibial amputees will successfully use a prosthesis compared with 25% or fewer of transfemoral amputees. Postoperative surgical wound healing is relatively higher at the transfemoral level (70% to 98%)
compared with the transtibial level (30% to 90%) in
dysvascular patients. This finding suggests that the transfemoral level may be more suitable for geriatric patients
with fewer activity demands. Younger patients can have
excellent outcomes at the transfemoral level with appropriate prosthetic prescriptions and training.
At the transtibial level, the ideal surgical margin is at
the musculocutaneous junction of the gastrocnemius
muscle. Longer bone lengths provide less soft-tissue
padding and poorer blood supply for the myocutaneous
flap. Furthermore, because of the height requirements
of the foot and ankle assembly, longer bone lengths may
preclude the addition of important components such as
a vertical shock pylon and rotator unit. The recommended transtibial residual limb length is 12.5 cm to
17.5 cm or approximately 2.5 cm of bone length per 30
cm of body height. A residual limb length that measures
33% to 50% of the intact tibial length from the medial
tibial plateau to the medial malleolus is also recommended. In transfemoral amputations, a 5-cm stump
length measured from the greater trochanter is fitted as

645

Orthopaedic Knowledge Update 8

Limb Amputation and Prosthetic Rehabilitation

a hip disarticulation. A transfemoral residual bone
length between 50% to 75% of the intact femur measured from the greater trochanter to the lateral femoral
condyle is considered optimal.
Beveling the cut cortical bone at the site of amputation is recommended regardless of surgical level. Ideally,
a 45 angle may prevent pressure on the anterior skin
flap. In transtibial amputations, sectioning the fibula 1.2
cm shorter than the tibia is also recommended. Complete removal of the fibula and adjacent muscle bulk
was previously performed for short residual limbs; however, this practice is not currently recommended because retention of the fibular head is important for anchoring the total contact socket within the residual limb.

ever, soft-tissue coverage is less than attained in posterior flap closures.

Nerve
Careful attention should be given to the treatment of
the transected nerves in an attempt to prevent symptomatic neuroma formation to minimize future pain.
Careful traction followed by transection allows the
nerve to recoil within the soft tissues of the residual
limb. Silicone capping, grafting with an epidural nerve
sheath, or a venous graft may prevent disorganized
nerve regeneration and sprouting. However, most techniques currently in use do not eliminate neuroma formation and more studies are required to determine the
best long-term outcomes.

Soft Tissue
Adequate padding and coverage to the distal bone is
achieved by muscle stabilization techniques and skin
flaps with the goal of attaining a muscular, cylindrical,
residual limb to provide ideal prosthetic fitting. Muscle
stabilization through myoplasty and tension myodesis
not only provides more soft-tissue coverage but also
acts as a replacement insertion that facilitates muscle
contraction and movement. In tension myodesis,
transected muscle groups are sutured to bone under
physiologic tension. In myoplasty, the muscle is sutured
into soft-tissue fascia or opposing muscle groups. In
younger more active patients who need firmer stabilization, a myodesis is preferred. In transfemoral amputations, it has been reported that the absence of adductor
magnus myodesis results in a loss of 70% of hip adduction power. Furthermore, myodesis prevents the anterolateral drift of the distal femur caused by muscular imbalance from the hip abductors that are unaffected by
the transfemoral amputation. Both procedures are relatively contraindicated in ischemic limbs because they
may further compromise the already marginal blood
supply.
The type of skin and soft-tissue flaps are determined
by the vascular supply of the distal limb. In transtibial
amputations, blood supply in the posterior and medial
aspects of the leg is more abundant than in the anterolateral region. In ischemic limbs, a long posterior myocutaneous flap and a short or even absent anterior flap
is recommended. A posterior flap that measures 1 cm
more than the diameter of the leg at the level of the
bone division is recommended. To preserve all intact
vascular connections between muscle and skin, dissection along tissue planes is avoided and myocutaneous
flaps are used. Occasionally, to provide more coverage, a
vascular graft can be taken from the other leg.
The fish mouth or guillotine transtibial amputation
is indicated for patients with ischemic limbs or those
who require delayed primary wound closure. The procedure uses equal length anterior and posterior flaps; how-

646

Preoperative Consultation
Optimal care begins before surgery. When amputation is
considered, patient consultation with a physiatrist, physical therapist, prosthetist, and a functioning amputee is
necessary to facilitate a smoother transition into prosthetic training. The physiatrist coordinates all the efforts
of the rehabilitative team, provides input on the recommended surgical level of amputation and limb length, informs the patient about the anticipated postoperative
rehabilitation protocol and outcomes, and explains potential problems. The physical therapist can initiate preprosthetic training before surgery by educating the patient on conditioning and range-of-motion exercises for
the affected extremity. The prosthetist can provide valuable input on the latest advances in rehabilitation technology, concentrating on limitations so as not to create
false expectations. A functioning amputee provides the
patient with psychological reassurance that a health professional may not be able to convey.

Immediate Postoperative Treatment

Wound Protection and Coverage
Wound protection and edema control can be achieved
simultaneously with a postoperative compressive dressing such as an elastic stockinette (stump shrinkers) or
elastic bandages. Elastic bandages require proper application to avoid proximal constriction from a tourniquet
effect. An elastic stockinette tends to be easier to apply
and provides reasonable control of edema. Some patients may experience pain as the stockinette is pulled
across the incision line and function better with elastic
bandages. A nonremovable plaster cast can be placed
immediately after the amputation for protection; however, this technique prevents wound inspection and can
result in pistoning and skin breakdown when residual
limb edema subsides. A removable rigid dressing made
of thermoplastic provides wound coverage and edema
control and allows wound inspection and dressing
changes. It can be fabricated to resemble a socket to im-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

prove residual limb shape or it can be formed into a bivalve or clamshell design. This approach can be used for
patients who need protection from falling and accidental contact while waiting for the residual limb to mature.

Range of Motion
A range-of-motion and contracture prevention protocol
is integral in the early postoperative course. Lying
prone, bedside range-of-motion exercises, and early mobilization can prevent knee flexion and hip abduction/
external rotation contractures that are seen in lower extremity amputees. It has been shown that early
contracture was independently associated with the inability to complete an inpatient prosthetic rehabilitation
program soon after amputation surgery.

Postamputation Pain
The treatment of pain has important functional and
prognostic implications. Pain is present in up to 70% of
amputees immediately after surgery and in 50% after 5
years. Along with ambulation distance, pain is the most
important amputation-specific determinant of healthrelated quality of life. Pain is experienced as hyperalgesia (a stronger or earlier withdrawal response to noxious
stimuli) or allodynia (the sensation of pain from nonnoxious stimuli). Phantom limb pain is defined as the
nociceptive sensation of the amputated limb.
Generally, postamputation pain can be attributed to
local and biomechanical factors affecting the residual
limb or to the changes to the peripheral and central nervous systems caused by the amputation. Local factors
include ischemia, infection, wound dehiscence, and skeletal abnormalities (such as ectopic bone formation, excess fibular length, and inadequate tibial beveling). Neuropathic pain is often manifested as residual limb pain
secondary to a neuroma, pain in the missing limb (phantom limb pain), or sympathetically driven pain.
The treatment of neuropathic pain after amputation
does not produce perfect results and often requires a
multidisciplinary approach. Pain management protocols
include pharmacologic agents, mechanical, surgical, and
physical modalities (exercise), psychological counseling,
and behavioral approaches including imagery techniques.
The antiepileptics, tricyclic antidepressants, and opioids are the most commonly used oral medications for
postamputation neuropathic pain. Gabapentin, an antiepileptic agent with both gamma-aminobutyric acid and
glutamate antagonist properties, is shown to decrease
neuropathic pain and is relatively well tolerated. Desipramine, a tricyclic antidepressant, could also be considered a first-line drug. The role of opioids in the treatment of neuropathic pain is controversial but remains
clearly indicated in acute postamputation pain. Other
medications that are used in neuropathic pain disorders

American Academy of Orthopaedic Surgeons

Chapter 54 Limb Amputation and Prosthetic Rehabilitation

include -2 agonists (clonidine), N-methyl D-aspartate
receptor antagonists (amantidine, dextromethorphan),
anti-inflammatory agents, magnesium sulfate, or cannabinoids.
Preoperative analgesia with intrathecal anesthetics
or opioids has been beneficial in patients with longstanding preamputation pain. This regimen is initiated
preoperatively and continued for up to 3 days after amputation before converting to an oral regimen. Neuroma
excision with the combination of funiculectomy, epineural sleeve suture ligation, or silicone capping is being
performed but long-term outcome studies are not yet
available. Local neurolysis with phenol/glycerin has provided relief in patients for whom neuroma excision has
been unsuccessful. Ultrasound for identification and
guidance during neurolysis also has been used with
moderate success. Cortical, thalamic, and dorsal column
stimulators implanted to stimulate neuroinhibitory pathways can provide neuromodulation of neuropathic pain
but not nociceptive pain. Intrathecal pumps that infuse
clonidine, anesthetics, or opioids have been implanted in
patients with intractable pain.
Mechanical stimulation by gentle residual limb tapping and the application of transcutaneous electrical or
vibratory stimulation have been the usual physical modalities for treating postamputation pain. The use of
transcutaneous nerve stimulation has had moderate success and can be applied directly to the site of pain, on
the contralateral limb, or even auricularly. Prosthetic
wear and early mobilization promotes afferent signals
that increase cortical reorganization and may relieve
neuropathic pain in many patients.

Prosthetic Training
Prior research has supported both immediate prosthetic
training and delayed prosthetic fitting. Vascular transtibial amputees treated with a rigid intrasurgical plaster
cast followed by early postoperative prosthetic limb fitting have been compared with those referred to an amputee clinic for fitting several weeks after hospital discharge. A significant decrease in the number of total
hospitalization days was noted in the group who had
early prosthetic limb fitting, implying increased cost effectiveness with this approach. In another study, outcomes for patients with immediate prosthetic transtibial
fitting were examined and no significant variation was
found in terms of local necrosis or infection. In a more
recent study, a 68% success rate in early prosthetic fitting was reported despite strict inclusion criteria that
only admitted patients who it was believed would benefit from this approach. Based on this data, a postoperative delay of about 3 weeks before beginning prosthetic
rehabilitation was recommended, although some patients will require a preprosthetic rehabilitation program.

647

Limb Amputation and Prosthetic Rehabilitation

Early functional training includes donning and maintenance of the prosthetic device, residual limb desensitization, strengthening, gradual and monitored increased
prosthetic wear, and gait retraining. Strengthening of
the quadriceps, glutei, and hamstring muscles must be
done as these muscle groups compensate for the loss of
the ankle and/or knee. The duration of prosthetic wear
should be gradual, beginning with 30-minute intervals
two times a day and increasing to several hours at a
time with frequent skin checks. Gait training is started
with parallel bars to facilitate weight acceptance on the
residual limb. Weight bearing is gradually increased on
the prosthetic limb. Gait retraining is initiated using a
step to gait pattern leading to a step through gait pattern. It is important to achieve a smooth gait pattern
with equal step-lengths to avoid gait deviations that increase metabolic energy expenditure. The early establishment of an energy-efficient walking pattern has considerable long-term implications that affect function and
independence.

Prosthetic Components and Prescription

The technology involved in the development of prosthetic components has evolved significantly in the past
decade. The incorporation of titanium, carbon fiber, and
other metal alloys has increased the rigidity and tensile
strength of the components without adding weight. The
use of silicone liners and sockets composed of flexible
plastics along with suction suspension has generally increased comfort and decreased skin breakdown. Polycentric and multiaxial joints and terminal devices with
pneumatic, hydraulic, and computer-driven controls
have contributed to a smoother and more energyefficient gait pattern. The goal of prosthetic wear has
gone beyond community ambulation to educational, vocational, and recreational activities as well.

Partial Foot Amputation

The amputation of a single toe, except for the big toe,
results in minimal loss of function. Toe spacers are inserted to fill the void and to prevent further deformities
(varus or valgus) in the remaining toes. The amputation
of the first ray significantly affects push-off; therefore, a
long steel shank and a rocker bottom is usually provided as compensation. For transmetatarsal amputations, a custom molded insole with a toe filler is recommended.
Proximally, the Lisfranc amputation is a transmetatarsal disarticulation whereas a Chopart amputation is a
disarticulation at the midtarsal joint through the talonavicular and calcaneocuboid joints. Both of these procedures may result in a significant equinovarus deformity with anterior weight bearing through the scar line,
predisposing to skin breakdown over time. Early postoperative rigid dressing and Achilles tendon lengthening

648

Orthopaedic Knowledge Update 8

are done to prevent the equinus deformity. For this level
of amputation, a special shoe or a slipper-like prosthesis
or a combination of both is provided. A ground-reacting
ankle-foot orthosis with anterior and posterior shells is
prescribed in a muscularly imbalanced and pressuresensitive foot to provide maximal control.
The Boyd amputation excises all the tarsal bones except the calcaneus. Because of resultant residual limb
problems, it is not a popular procedure and is used primarily for the pediatric congenital amputee. The Symes
amputation is an ankle disarticulation with a heel flap
attached securely to the distal tibia to provide weight
bearing on the residual limb. Cosmetic concerns may
limit acceptance of this procedure because the distal residual limb has a bulbous shape. The Symes prosthesis
extends to the proximal tibia but has a removable medial wall to allow the bulbous distal limb to enter the
socket. The foot component is similar to that used in a
transtibial amputation but with a lower profile to accommodate the residual limb.

Transtibial Components
Foot and Ankle Assembly
The solid ankle cushioned heel exemplifies the earlier
foot assemblies. It consists mainly of a semirigid wooden
keel surrounded by a resilient material concentrated at
the heel. At heel strike, energy is absorbed with the
compression of the heel assembly and a plantar flexion
moment is simulated. A stable base of support is provided by the keel, which hyperextends at the metatarsophalangeal line at heel-off. During the swing phase,
the unloaded toe region reverts to its neutral position.
The use of this design is limited to home ambulators or
those patients with limited insurance coverage.
Dynamic response feet store energy at heel strike
and transmit the forces to the keel during heel-off, providing recoil. Examples include the Seattle (Model and
Instruments Works, Inc, Seattle, WA), the Carbon Copy
II (Ohio Willow Wood Co, Mt. Sterling, OH), the Quantum (Hosmer Dorrance Corp, Campbell, CA), and the
Flex (Ossur North America, Aliso Viejo, CA) foot designs. Because they provide a better spring for running
and jumping, these prosthetic foot designs are recommended for more active individuals.
Articulated foot assemblies provide motion at the
anatomic location of the ankle, better accommodate uneven surfaces, and absorb torsional forces reducing
torque to the limb by the socket. Designs are either single axis, which provide dorsiflexion and plantar flexion,
or multiaxis, which provide motion in the dorsiflexionplantar flexion and inversion-eversion planes. The College Park (College Park Industries, Frasier, MI), Luxon
(Otto Bock North America, Minneapolis, MN), and En-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

dolite (Endolite North America, Centerville, OH) designs are common examples of multiaxis foot assemblies.
Heel height adjustability is a feature developed to
address cosmetic issues in patients who use footwear of
varying heel heights. This option is often incorporated
into the foot assembly. The Century 22 (Otto Bock
North America) foot offers manual adjustment of varying heel heights from 35 to 50 mm.
Occasionally, a torque absorber is placed between
the shank and the foot assembly to absorb torsional
forces as the prosthesis twists around the residual limb
during ambulation. This modification is recommended
for patients with skin conditions or for those who participate in sport activities or vocations that require significant foot rotation.
Pylon
The pylon is the interphase between the foot assembly
and the socket and is classified as having an endoskeletal or exoskeletal design. Endoskeletal pylons consist of
a rigid and unyielding central support surrounded by a
polyurethane, soft cosmetic cover to allow for a more
natural appearance. Substitution of isolated components
(socket, shank, and foot) is also possible without discarding the entire prosthesis. This situation is ideal in
early prosthetic rehabilitation when a temporary or intermediate prosthesis is provided or in the pediatric
population to allow for limb growth. Recent improvements in material construction using titanium and carbon fiber have significantly increased the weight limit
for these pylons to as much as 350 lb.
Exoskeletal or crustacean shanks are solid blocks of
hard plastic or wood molded to resemble the intact limb
and hollowed in the middle to reduce weight. The exterior surfaces are laminated to provide waterproofing.
Although more rigid and stronger than the endoskeletal
pylon designs, they are heavier and revisions often entail discarding the entire prosthesis. Their use is limited
to relatively obese patients or people such as farmers or
outdoorsmen who are constantly exposed to environments that could corrode the more intricate components
of the endoskeletal pylon.
Recent designs have also incorporated the pylon
with the foot assembly. The Re-Flex Vertical Shock pylon (Ossur North America) integrates a foot with a vertical shock-absorbing pylon that returns energy both in
the vertical and sagittal planes.
Socket
The socket acts as an interphase between the residual
limb and the prosthesis and applies weight-supporting
forces to the residual limb. In socket fabrication,
pressure-tolerant areas are built up for more contact
and pressure-sensitive areas are relieved to minimize
contact. Pressure-tolerant areas include the patellar ten-

American Academy of Orthopaedic Surgeons

Chapter 54 Limb Amputation and Prosthetic Rehabilitation

don, the pretibial muscles, the gastrocnemius/
gastrocnemius-soleus complex muscles, the popliteal
fossa, the lateral flat aspect of the fibula, and the medial
tibial flare. The pressure-sensitive areas include the tibial crest, tubercles, condyles, the fibular head, the distal
tibia and fibula, and the hamstring tendons.
The patellar tendon-bearing socket is the most frequently used. Although it is designed to put substantial
weight on the patellar tendon, its intimate contact with
the entire residual limb provides even distribution of
pressure with minimal distal end bearing. The trim line
extends anteriorly to the lower patella level, mediolaterally to the femoral condyles, and posteriorly below the
level of the hamstring tendon insertions.
For shorter residual limbs and those with mediolateral instability, a patellar tendon supracondylar socket is
more appropriately prescribed. A supracondylar socket
extends above the medial and lateral femoral condyles
using the bony ridges to suspend the prosthesis. It provides better mediolateral knee support and is appropriate for shorter residual limbs. A supracondylar/
suprapatellar suspension uses the same design but
extends the anterior trim line to envelop the patella for
added support. This particular feature also helps counteract genu recurvatum forces.
Suspension Systems
Mechanical suspension devices are external devices that
are attached to anchor on bony prominences to provide
stabilization. They are indicated early after amputation
when there is significant volume fluctuation along the
residual limb secondary to edema. They also are useful
in older amputees with poor hand dexterity and poor
residual limb bulk and tone. Examples include: (1) A suprapatellar suspension strap that is fastened to the medial and lateral socket walls passing over the superior
border of the patella. This device offers some resistance
to knee hyperextension forces and is relatively easy to
put on and take off; however, it is prone to mild pistoning and is not indicated for short, painful residual limbs
with mediolateral instability. (2) A thigh corset secured
to the socket by sidebars and a knee joint assembly is
useful to take weight off of the residual limb and to provide better rotational stability. (3) Anteriorly, a fork
strap can be attached to a waist or pelvic belt to add stability in the anteroposterior plane and to help suspend
the prosthesis, especially in knee flexion. It is indicated
for shorter residual limbs and for obese patients who
cannot use a suprapatellar strap. (4) Over-the-knee
sleeves are fabricated from a flexible material such as
neoprene, which extends from the proximal end of the
prosthesis to the distal thigh.
Atmospheric or suction types of suspension are generally preferred over mechanical types to minimize pistoning and shearing forces. However, they require adequate hand and visual function to apply. Furthermore,

649

Limb Amputation and Prosthetic Rehabilitation

suction suspensions may provide undue stress to a healing surgical wound, thus promoting delayed wound
healing or wound dehiscence; they are typically withheld early in the postoperative period until volume fluctuations have stabilized. One example is a prefabricated,
closed-end silicone elastomer liner that is flexible
enough to adapt to the irregular surfaces in the residual
limb and to promote an airtight environment to achieve
suspension. This device is rolled proximally up and over
the knee and is secured to the socket by means of a
shuttle lock mechanism, which consists of a pin at the
distal end of the liner and a locking mechanism located
at the bottom of the socket. Early in the postamputation
period when the residual limb volume fluctuates significantly, the pin and shuttle lock mechanism can be deferred in lieu of a supracondylar strap. The Iceross (Ossur North America) and Alpha (Ohio Willow Wood)
liners are examples of silicone liners. Another example
is the Vacuum Assisted Socket System (Otto Bock
North America), which consists of a silicone liner, a suspension sleeve, and an air evacuation pump that creates
an elevated vacuum of 15 mm Hg between the liner and
the socket wall. This design controls volume fluctuation
and promotes a more secure and intimate fit between
the residual limb and socket.

Transfemoral Components
Socket Design
The two main transfemoral socket designs are termed
quadrilateral and ischial containment. The quadrilateral
socket has its posterior border under the ischial tuberosity and buttocks for weight bearing, with the anterior
border providing a posteriorly directed force to control
the femur. Therefore, it is narrow anteroposteriorly and
wide mediolaterally. An ischial containment socket
should contain the ischial tuberosity and apply counter
pressure from the lateral wall of the socket. It is narrow
mediolaterally and wider anteroposteriorly. Quadrilateral socket designs are useful for obese patients,
whereas more active patients may benefit from the ischial containment design that maintains the femur in
adduction, allowing the gluteal musculature to generate
maximal tension at an ideal resting length. Material construction can be either rigid (wood, plastic laminates),
flexible, or a combination of both (Scandinavian or
Icelandic-Swedish-New York sockets). The outer hard
socket can be windowed posteriorly and anteriorly to allow improved comfort with sitting. In this situation, a
flexible inner liner is used in contact with the skin.
Suspension Device
The transfemoral suspension devices can either be mechanical or atmospheric. A pelvic belt and hip joint offers maximal stabilization and is best suited to those
with short residual limbs, significant mediolateral stabil-

650

Orthopaedic Knowledge Update 8

ity, or lack of rotational control. Less restrictive suspension would include a Silesian belt or total elastic suspension belt, which is composed of neoprene with
reinforced elastic bands running at oblique angles posteriorly and anteriorly around the waist. A silicone liner
with a distal pin and shuttle lock similar to that found in
transtibial devices can be used. This component adds
length and in combination with the knee unit may cause
an uneven knee axis (prosthetic thigh longer than intact
thigh), leading to cosmetic and biomechanical deficits. A
suction suspension is ideal for patients with mature residual limbs. Surface tension, negative pressure, and
muscle contractions suspend the prosthesis from the
limb without the need for a waist belt. Suction is generated by placing the residual limb into the socket with a
nylon pull sock or using a lubricating lotion, then expelling air with a one-way valve.
Knee Assembly
Knee assemblies consist of a joint or bolt that allows
flexion and extension, an extension stop that limits hyperextension, and a friction device that provides
smoother motion along the assembly. General classifications are based on axis and control, either on swing or
stance phase.
A single axis knee assembly acts as a hinged joint
and is preferred for its simplicity, reliability, minimal
maintenance, and cost. It is also ideal for children because growth entails frequent changes in prosthetic
components. A polycentric or multiaxis knee is more
physiologic and uses multiple bar linkages along the
bolt to allow a shifting mechanical axis during flexion
and extension. This feature provides more stability while
allowing for improved performance at higher walking
speeds.
Control is provided either at the stance or swing
phase. Stance phase control is provided for patients with
knee instability who cannot control knee flexion. A
manual locking knee is used to maximize stability; however, this eliminates knee motion throughout the gait
cycle and is recommended for limited household ambulators. Stance phase weight-activated control is used for
those with slightly better overall knee stability who
need support during upright activities.
The presence and type of swing phase control is determined by the patients ability to vary cadence. A constant friction device is used for elderly patients who
walk at a constant speed. This unit has a split bushing
and clamp around the knee bolt, which provides constant friction throughout the swing phase. The amount
of friction is controlled by a friction adjusting screw that
tightens or loosens the bushing. For active patients who
want to vary walking speeds and run, hydraulic or pneumatic swing phase control devices are used. Hydraulic
and pneumatic mechanisms adjust resistance to changes

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Table 1 | The Centers for Medicare and Medicaid Services

Functional Levels of Ambulation
Level
K0
K1

K2

K3

K4

Amputee and Prosthetic Characteristics

Patient cannot transfer independently
Prosthesis will not enhance quality of life or mobility
Patient has the ability or potential to use a prosthesis for
transfers or ambulation on level surfaces with a fixed cadence
Patient has limited or unlimited household ambulation
A solid ankle cushioned heel foot or a single axis foot; a manual locking, single axis or polycentric knee unit (with or
without stance phase weight-activated control); or any
socket or suspension design is allowed
Patient has the ability to traverse low-level environmental
barriers such as curbs, stairs, or uneven surfaces with a
constant cadence
All K1 prosthetic components are allowed with the addition of
flexible keels or multiaxial feet, rotators, and torque absorbers
Patient has the ability or potential for ambulation with variable cadences
Patient can traverse most environmental barriers
May have vocational, therapeutic, or exercise activity that
demands prosthetic use beyond simple locomotion
All K1 and K2 prosthetic components are allowed with the
addition of hydraulic/pneumatic knees and dynamic or
shock absorbing pylons
Patients have the ability or potential for prosthetic use that
exceeds basic ambulation
Patients exhibit high impact and energy levels typical of the
growing child, active adult, or athlete
All existing prosthetic components are allowed in this functional level

in the rate of motion and this improves knee control

and provides a smoother gait pattern.
The C-Leg
In 1999, a microprocessor knee unit called the C-Leg
(Otto Bock North America) was introduced in the
United States. This knee unit uses force sensors in the
pylon that detect knee angles and loading forces at the
foot and ankle that are then sampled by onboard microprocessors 50 times per second. The C-Leg was reported
to allow patients to have a significant reduction in oxygen consumption during slower speed ambulation and
to achieve a wider range of cadences compared with a
conventionally controlled hydraulic single axis knee.
However, randomized trials comparing this technology
to multiaxis hydraulic knee designs are needed to determine optimal patient selection.

Knee Disarticulation
Knee disarticulations involve the removal of the tibia
and fibula at the knee with suturing of the patellar ten-

American Academy of Orthopaedic Surgeons

Chapter 54 Limb Amputation and Prosthetic Rehabilitation

don to the cruciate ligaments. This procedure is done on
patients with traumatic injuries and very short residual
tibiae above the tubercle, or on elderly dysvascular patients without prosthetic potential who need to bear
weight on the residual limb for transfers. The knee disarticulation prosthesis consists of a modified quadrilateral
transfemoral socket with some ischial weight bearing
and a four-bar linkage polycentric knee that may result
in uneven knee axes.

Functional Levels
The Centers for Medicare and Medicaid Services have
identified functional levels of ambulation with corresponding components deemed appropriate for each
level of activity (Table 1). This classification system is
only applicable to patients with single lower extremity
amputation. It is also possible for a classification to be
upgraded if initial expectations are exceeded.

Common Gait Abnormalities in Lower Limb

Prosthetic Rehabilitation
Transtibial Deviations
Excessive knee flexion during heel strike could be
caused by a foot placed in dorsiflexion, an excessively
stiff heel cushion, anterior translation of the socket in
relation to the pylon, or a residual limb flexion contracture. Insufficient knee flexion can be caused by plantar
flexion of the foot, an excessively soft heel cushion, posterior displacement of the socket in relation to the pylon, or weak quadriceps on the residual limb.
Lateral thrust occurs when the pylon is placed medially, creating a varus moment. The patient usually reports concomitant pain and pressure on the medial femoral condyle and lateral distal tibia. Medial thrust is an
analogous problem with the pylon placed laterally, causing a valgus moment. Patients often report pain on the
lateral femoral condyle and medial distal tibia. These
deviations are corrected by repositioning the pylon to
the socket in the coronal plane.

Transfemoral Deviations
Lateral trunk bending is a trunk lean to the prosthetic
side during the stance phase. It occurs because of an abducted socket, a hip abduction contracture, insufficient
lateral support of the prosthetic socket, a short prosthesis, or weak ipsilateral hip abductors.
Circumduction is a curvilinear motion of the prosthesis during the swing phase and is caused by a functionally longer prosthetic limb length, which creates difficulty in clearance. Common causes include an ill-fitting
socket that pistons or does not fully accommodate the
residual limb, a manual locking knee, a device that
causes excessive knee friction, or a foot in plantar flexion.

651

Orthopaedic Knowledge Update 8

Limb Amputation and Prosthetic Rehabilitation

Table 2 | Metabolic Cost of Ambulation Per Level and Nature of Amputation

Amputation Level
Symes
Traumatic transtibial
Vascular transtibial
Traumatic transfemoral
Vascular transfemoral

Metabolic Cost
Increased 15%
Increased 25%
Increased 40%
Increased 68%
Increased 100%

(Data from Czerniecki JM: Rehabilitation in limb deficiency: Gait and motion analysis. Arch
Phys Med Rehabil 1996;77:S3-S8.)

more, 95% of the survivors wore prostheses and averaged

80 hours of use per week. An earlier study involving
younger amputees showed a 79% job reintegration rate.
Age at the time of amputation, comfort of the prosthesis,
education level, and lowering of the physical demands in
the workplace were positive predictors for return to work.
Pain did not appear to be associated with return to work
but was directly associated with work-related satisfaction.

Upper Limb Amputation and Prosthetic

Components
Levels of Amputation

Vaulting is described as excessive plantar flexion of

the intact foot during the prosthetic swing phase and occurs because of a functionally longer prosthetic limb
length for reasons similar to those that create a circumduction moment.
Medial or lateral heel whip is often caused by rotational malalignment of the knee apparatus in comparison to the tibial components. A knee unit placed in excessive external rotation leads to a medial whip; a knee
unit placed in excessive internal rotation results in a lateral whip.

Long-Term Prosthetic Rehabilitation Issues

Metabolic Requirements
The metabolic requirements of ambulation increase
proportionally with decreased length of the amputated
limb, the number of amputated joints, and the number
of amputated limbs. Furthermore, vascular amputees
have higher metabolic requirements than their traumatic counterparts. Amputees, however, offset the increased metabolic demands by choosing a slower selfselected walking pace. Patients should not be denied
prosthetic fitting and training on the basis of cardiopulmonary limitations unless these limitations are profound. The metabolic cost of ambulation based on the
level and nature of amputation is shown in Table 2.

Functional Outcomes and Work Reintegration

The long-term survival rate for patients after dysvascular lower extremity amputation is 50% to 70% at 2
years, and 30% to 40% at 5 years with a postoperative
mortality rate of 10% to 30%. Heart disease is the most
common cause of death (51%), followed by carcinomatosis (14%) and cerebrovascular accidents (6%). The
risk of contralateral limb amputation reaches 15% to
20% in 2 years after the initial amputation and the
reamputation rate of the ipsilateral limb in a transtibial
dysvascular patient ranges from 4% to 30%.
Traumatic amputees show more favorable outcomes.
In a 2001 study, a 9% acute admission mortality rate with
only a 3.5% 10-year mortality rate was found. Further-

652

The transradial amputation is the most common and

preferred upper extremity amputation. The residual
limb length can be classified as either long (55% to
90%), medium (35% to 54%), or short (0 to 34%)
based on the length of the intact forearm measured
from the medial epicondyle to the distal ulna or radius.
Because of component weight and length considerations, the long transradial residual limb provides the
longest lever arm most suitable for a body-powered
prosthesis. A medium length transradial residual length
provides enough clearance for externally powered prosthetic components. The short transradial limb is difficult
to suspend and may promote deficits in elbow strength
and range of motion. Elbow disarticulations may pose
cosmetic concerns because of limited options in externally-powered elbow units available for this amputation
level.
Transhumeral amputations are also classified by residual limb length (short, medium, and long), but the
prosthetic options and rehabilitative interventions are
similar at each level. Shoulder disarticulations and forequarter amputations are reserved for tumor excision
surgeries because of the difficulty in providing a prosthesis with adequate suspension.

Prosthetic Control
Upper extremity prosthetic control is usually either
body powered, externally powered, or a combination of
both. Body-powered control is provided by the intact
movements of the residual limb that connects to cables
to flex and extend the elbow or open and close the terminal device. The movements include scapular abduction; chest expansion; shoulder depression, flexion and
abduction; and elbow flexion. Myoelectric controls use
the electrical activity generated by muscle contractions
to control the flow of energy from a battery to a motor
controlling the terminal device or elbow unit. Comparative studies between myoelectric and body-powered
hands showed no significant difference in terms of performance that could limit application of this technology
to specialized situations.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Prosthetic Components
The terminal device attempts to approximate the complex
functions of the hand and is available in different designs
based on the users preference. The devices can either be
hooks that provide lateral pinch or hands that provide a
three-jaw chuck pinch. They can be passive or purely cosmetic, body powered, or externally powered. Bodypowered designs are either voluntary opening or voluntary closing; the selection depends on the patients
anticipated use (for example, an amputee who plans to
work in a factory would benefit from a voluntary closing
terminal device, which would not get caught in the assembly line). Terminal devices cannot provide the sensory
feedback and dexterity of an intact hand. Myoelectricpowered hands allow for a proportional grasp. Slip control systems have microprocessors that maintain constant
pressure on the object to prevent slippage.
Wrist designs can be manually or externally controlled, and most provide passive supination and pronation with a friction lock to control rotation when lifting
heavier objects. Quick disconnect wrists allow for easy
interchange of terminal devices. Elbow mechanisms are
either internal or external and can also be passive, body
powered, or externally powered. Body-powered designs
are controlled with mechanical cables through movements of the residual limb or can be manually locked by
the contralateral hand, chin, or the ipsilateral shoulder.
Electrical elbows are operated by electrical switches or
myoelectric impulses.
Prosthetic socket design also has progressed with the
development of lightweight and durable materials. Flexible thermoplastics provide better fit and comfort and
are used in the internal layer, which comes in close contact with the residual limb. Carbon fiber has replaced
wood or laminated plastics in the external layer. Mechanical suspension devices usually consist of harness
systems that anchor across the shoulders. Fabric liners as
well as silicone suction suspension systems are also
available for the upper extremity prostheses.

Annotated Bibliography
Incidence and Etiology
Bosse M, MacKenzie EJ, Kellam JF, et al: An analysis of
outcomes of reconstruction or amputation after leg
threatening injuries. N Engl J Med 2002;347:1924-1931.
Limb salvage has replaced amputation as the primary surgical treatment in severe limb trauma. This prospective cohort
study examines the long-term outcomes of significant lower
extremity injury (grade III tibial or ankle fractures, severe dysvascular and soft-tissue injury) treated either with amputation
or limb salvage. The functional outcomes are similar between
the two treatment options; however, longer hospital stays, increased number of complications, and more surgeries were
found in the limb-salvage group.

American Academy of Orthopaedic Surgeons

Chapter 54 Limb Amputation and Prosthetic Rehabilitation

Dillingham TR, Pezzin LE, MacKenzie EJ: Limb amputation and limb deficiency: Epidemiology and recent
trends in the United States. South Med J 2002;95:
875-883.
This study provides a comprehensive perspective on the
epidemiology of limb amputations in the United States using
linear regression techniques on data from 1988 to 1996. Dysvascular amputations have increased to 27%, accounting for
82% of all limb loss. Rates of trauma and cancer-related amputations have declined by one half and the incidence of congenital deficiencies has remained stable.

Immediate Postoperative Treatment

Bone M, Critchley P, Buggy DJ: Gabapentin in postamputation phantom limb pain: A randomized, doubleblind, placebo-controlled, cross-over study. Reg Anesth
Pain Med 2002;27:481-486.
This study used a randomized, double blind, placebocontrolled crossover methodology. Subjects with severe phantom limb pain were enrolled in a 6-week course of gabapentin
therapy titrated by increments of 300 mg to a maximum of
2,400 mg per day versus the control group. Significant improvement in terms of pain intensity was found in patients
treated with gabapentin as compared with those given the placebo. However, no differences were noticed in terms of rescue
medication (codeine/paracetamol) required, sleep disturbance,
or function as measured by the Barthel index.

Ernberg LA, Adler RS, Lane J: Ultrasound in detection

and treatment of painful stump neuroma. Skeletal
Radiol 2003;32:306-309.
The use of ultrasound has shown considerable success in
the assessment of amputation stump neuromas. This study investigates the usefulness of ultrasound guided localization for
steroid injection of amputation stump neuromas.

Munin MC, Espejo-De Guzman MC, Boninger ML,

Fitzgerald SG, Penrod LE, Singh J: Predictive factors for
successful early prosthetic ambulation among lowerlimb amputees. J Rehabil Res Dev 2001;38:379-384.
In an effort to predict successful outcomes with early prosthetic rehabilitation, demographic and medical factors were
analyzed in a group of lower extremity amputees admitted to
an inpatient rehabilitation facility. Sixty-eight percent of the
patients met criteria for successful ambulation at time of discharge. The absence of lower extremity contractures and
longer length of inpatient rehabilitation stay were significantly
related to successful prosthetic ambulation.

van der Schans CP, Geertzen JH, Schoppen T, Dijkstra

PU: Phantom pain and health-related quality of life in
lower limb amputees. J Pain Symptom Manage 2002;24:
429-436.
This study analyzed the determinants of health-related
quality of life in a population of 437 lower extremity amputees
using the RAND-36 Item Health Survey (Dutch Language
Version) questionnaire. In general, the most important

653

Limb Amputation and Prosthetic Rehabilitation

amputation-specific determinants of health-related quality of
life were walking distance and stump pain. Amputees with
phantom limb pain had poorer quality of life than those with
only phantom limb sensation.

Prosthetic Components and Prescription

Carnesale PG: Amputation of the lower extremity, in
Canale ST (ed): Campbells Operative Orthopedics. Philadelphia, PA, Mosby, 2003, pp 575-595.
This chapter examines the different levels of lower extremity amputations including their indications, techniques,
and postoperative treatment. Noticeable differences occur in
approaches taken in the nonischemic and ischemic lower extremity.

Schmaltz T, Blumentritt S, Tsukishiro K, Kocher L, Dietl

H: Energy efficiency of trans-femoral amputees walking
on computer-controlled prosthetic knee joint: C-LEG.
Otto
Bock
Website.
Available
at:
http://
www.ottobockus.com/products/lower_limb_prosthetics/
c-leg_energy.pdf. Accessed June, 2004.
This online article reviews a study of the energy expenditure needed for walking by six transfemoral amputees using
several prosthetic devices including the C-Leg with electronic
controls.

Long-Term Prosthetic Rehabilitation Issues

Schoppen T, Boonstra A, Groothoff JW, deVries J,
Goeken LN, Eisma WH: Employment status, job characteristics and work-related health experience of people
with a lower limb amputation in the Netherlands. Arch
Phys Med Rehabil 2001;82:239-245.
This study examined the occupational status of lower limb
amputees in the Netherlands and compared the health experience of working and nonworking amputees. Patients who
stopped working because of the amputation had a worse
health experience compared with those who continued to
work. Patients who later returned to work, reported problems
(such as finding a suitable job or obtaining workplace modifications) stemming from the long delay between the amputation and the return to work.

Classic Bibliography

Orthopaedic Knowledge Update 8

Czerniecki JM: Rehabilitation in limb deficiency: 1. Gait
and motion analysis. Arch Phys Med Rehabil 1996;
77(suppl 3):S3-S8.
Edelstein JE, Berger N: Performance comparison
among children fitted with myoelectric and bodypowered hands. Arch Phys Med Rehabil 1993;74:376380.
Esquenazi A: Upper limb amputee rehabilitation and
prosthetic restoration, in Braddom RL (ed): Physical
Medicine and Rehabilitation. Philadelphia, PA, WB
Saunders, 2000, pp 263-278.
Gottschalk F: Transfemoral amputation: Biomechanics
and surgery. Clin Orthop 1999;361:15-22.
Kane TJ III, Pollak EW: The rigid versus soft postoperative dressing controversy: A controlled study in vascular
below-knee amputees. Am Surg 1980;46:244-247.
Leonard EI, McAnelly RD, Lomba M, Faulkner VW:
Lower limb prosthesis, in Braddom RL (ed): Physical
Medicine and Rehabilitation. Philadelphia, PA, WB
Saunders, 2000, pp 279-311.
Leonard JA, Meier RH: Upper and lower extremity
prosthesis, in DeLisa JA, Gans BM (eds): Rehabilitation
Medicine: Principles and Practice. Philadelphia, PA,
Lippincott-Raven, 1998, pp 669-697.
Pezzin LE, Dillingham TR, MacKenzie EJ: Rehabilitation and the long-term outcomes of persons with
trauma-related amputations. Arch Phys Med Rehabil
2000;81:292-300.
Pinzur MS, Littooy F, Osterman H, Wafer D: Early postsurgical prosthetic limb fitting in dysvascular belowknee amputees with a pre-fabricated temporary limb.
Orthopedics 1988;11:1051-1053.
Yuksel F, Kislaoglu E, Durak N, Ucar C, Karacaoglu E:
Prevention of painful neuromas by epineural ligatures,
flaps and grafts. Br J Plast Surg 1997;50:282-185.

Carabelli RA, Kellerman WC: Phantom limb pain: Relief by application of TENS to contralateral extremity.
Arch Phys Med Rehabil 1985;66:466-467.

654

American Academy of Orthopaedic Surgeons

Chapter

55

Musculoskeletal Rehabilitation
Tom G. Mayer, MD
Joel Press, MD

Levels of Care
Nonsurgical care for patients with injuries to the spine
and extremities can be classified into three distinct levels of treatment. Timing of the care is dependent on the
diagnosis and anticipated healing time from the inciting
event.

Primary Care
Primary care is usually provided during an acute stage
of an injury with pain control as the primary focus;
avoidance of deconditioning is also a consideration. The
duration of the period of primary care depends on the
type of injury and can range from 10 to 14 days in patients with mild sprains, strains, and lacerations and
from 8 to 12 weeks in patients with complex fractures
and dislocations. Treatment modalities include thermal
(heat/cold) applications, pain medication and muscle relaxants, immobilization, bed rest, traction, and injection
methods.

Secondary Care
Secondary care is usually appropriate in the postacute
phase of an injury with the goal of providing reactivation to prevent long-term physical deconditioning and
psychosocial changes that could extend beyond the normal healing period. This phase can begin when an injury
has undergone sufficient partial healing and/or stabilization (through surgery, bracing, casting, or tissue healing)
to permit progressive motion and strengthening exercises. Active joint mobilization and strengthening of the
involved para-articular muscles are the primary modalities; treatment may be assisted with bracing, manipulation, thermal modalities, medication, and injections. In
most patients, secondary care is the last component of
musculoskeletal rehabilitation and is administered after
an injury is treated nonsurgically or with surgery during
the acute stage of injury.

Tertiary Care
Tertiary care is the final phase of musculoskeletal rehabilitation and is needed for only a small percentage of

American Academy of Orthopaedic Surgeons

patients who have chronic disabling musculoskeletal

pain that does not respond to early surgical and/or nonsurgical intervention. Because this group of patients is
often characterized by a complex mix of physical deconditioning, psychosocial and socioeconomic barriers to
recovery, and some resistance to care, a multidisciplinary approach is usually needed. Some patients require a functional restoration approach, which includes
intensive physical training along with a cognitive behavioral disability management program aimed at increased
productivity. Measurement of function, narcotic detoxification, psychotropic medication, and infrequent injections to relieve pain may be components of this approach. At the other extreme, palliative pain
management usually involves a more passive role for
the patient with an emphasis on pain-relieving injections
and long-term use of opiates accompanied by a psychosocial focus on deemphasizing the pain experience. Neuroablative procedures and devices such as drug pumps
and spinal cord stimulation also may be used. Chronicity of symptoms is usually established by 6 months after
injury; however, more functionally-based programs may
be appropriate within 3 to 4 months of symptom onset if
significant psychosocial or treatment resistance is evident.

Primary Rehabilitation
The two main objectives of primary rehabilitation are to
control pain and to prepare the musculoskeletal system
for proper healing from injury. Pain control can be accomplished through the use of medications, physical
modalities, injections, and occasionally bracing or relative immobilization (fracture management). No single
pain medication is effective for all injuries. With any
medication, knowledge of the mechanism of action, side
effect profile, and interactions with other medications is
essential for proper use. Acetaminophen (less than 4
mg/day) is an excellent first-line analgesic medication
for pain because of its low cost. Serious adverse effects
are rare except for liver toxicity, which may occur with
prolonged use at a high dosage; particularly in associa-

655

Musculoskeletal Rehabilitation
tion with substantial alcohol intake. Comparisons of effectiveness with nonsteroidal anti-inflammatory drugs
(NSAIDs) are inconsistent. NSAIDs, all of which are
analgesic, antipyretic, and anti-inflammatory, show no
significant differences among the various available compounds. Some patients reported a marked preference
and variation in efficacy of different NSAIDs, thus warranting a trial of a second or third class of medication if
one class provides no pain relief. An adequate trial of
NSAIDs may be 2 to 3 weeks. Several rare, serious adverse effects including clinical hepatitis, aplastic anemia,
and agranulocytosis can occur. Gastrointestinal adverse
effects are the most common and occur in approximately 25% of patients taking NSAIDs, whereas silent
endoscopically demonstrated lesions occur in as many
as 60% of patients. The overall risk for serious gastrointestinal bleeding in patients treated with NSAIDs
is 1 per 1,000 patients, with the risk significantly greater
in patients older than 65 years.
Other medications used for acute pain symptoms
with muscle spasms are muscle relaxants. These medications are centrally acting drugs, which produce nonspecific sedation that accounts for their muscle relaxation
effect. Although peripherally acting muscle relaxants exist (such as dantrolene sodium), these medications are
not used for musculoskeletal disorders because of potential severe adverse effects. Muscle relaxants have
been found to be more effective than a placebo in the
relief of symptoms of acute musculoskeletal disorders.
Oral corticosteroids also may be useful as a strong antiinflammatory agent for patients with radicular symptoms in the cervical and/or lumbar region. Short-term
use (7 to 10 days) or corticosteroids taken at a high dosage (30 to 40 mg prednisone or equivalent) have not
been associated with major adverse effects. Opioid analgesics, on occasion, can be used for acute pain symptoms. Opioid analgesics act primarily by binding opiate
receptors in the central nervous system and can be associated with tolerance, toxicity, addiction, and illicit use
with long-term administration. Even short-term use of
these medications should be undertaken with caution
because of an association of adverse effects including
demotivation, early reactive hyperalgesia, and early dependency problems in a select group of patients. Although more potent than NSAIDs and acetaminophen,
in two of three clinical trials, narcotic analgesics were
found to be no more effective than these medications in
relieving pain. The dosage schedule should be defined
and use limited to patients whose pain is unresponsive
to alternative medications.
Physical agents including ultrasound, electrical stimulation, and heat and cold have been used to promote
tissue healing, increase circulation, decrease inflammation, and reduce pain. Although physical agents are frequently used for symptomatic relief, these passive modalities do not appear to have any effect on clinical

656

Orthopaedic Knowledge Update 8

outcomes. It is essential to understand indications and
contraindications of specific modalities when prescribing these agents. In general, short-term use (1 to
3 weeks) of physical modalities may be appropriate for
an acute musculoskeletal problem or a flare-up of a
chronic condition. No single modality has been shown
to be superior to others for relief of musculoskeletal
pain. Prolonged use of these passive modalities should
be discouraged. Newer treatments such as vertebral axial decompression and continuous passive motion have
been purported to alleviate acute and chronic pain. Until clinical studies confirm these findings, use of these
treatments should be viewed as another physical modality and treated accordingly.
Injections of a variety of medications in various anatomic locations can be an adjunct to treating acute painful musculoskeletal conditions. Epidural injections have
been shown to be effective in reducing radicular pain;
however, the results in some controlled, prospective
studies are variable. The benefit of facet joint and sacroiliac injections is controversial. They may provide some
short-term pain relief that serves as an adjunct to other
treatments (for example, manipulation, mobilization,
and exercise) by facilitating joint movement in otherwise hypomobile joints or segments. Trigger point injections may provide some temporary relief for tight, painful muscle spasms to allow earlier activation of the
musculoskeletal system. Multiple, repeat injections without concomitant activation of the patient is probably of
little benefit.

Secondary Rehabilitation
Secondary rehabilitation focuses on restoring function
to the musculoskeletal system once initial pain symptoms have subsided and tissue healing has been initiated. No single component of musculoskeletal rehabilitation is effective for every disorder; most disorders
require multiple components to provide a comprehensive program. Understanding the roles and skills of chiropractors, physical therapists, and other health care
providers is critical to avoid overuse and abuse of any
one treatment. Cornerstones of restoring function are
activation of the patient to prevent the sequelae of immobility and exercise to restore muscle flexibility, muscle balance, and coordination. Initially, exercises are emphasized in nonpainful ranges and planes of motion.
Manual treatments and mobilization of restricted joints
and soft tissues, either by chiropractors, osteopaths, therapists, or physicians, are often initiated before beginning
focused strengthening programs. Injections for pain control (for example, local corticosteroids and epidural injections) may play some role in this phase of rehabilitation if the injections are used as an adjunct to increasing
the patients active participation in therapy or exercise.
Exercise programs for musculoskeletal rehabilitation

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

should focus not only on mobility and absolute muscle
strength but also on muscle balance (agonist and antagonist strength ratios), kinetic chain issues (proximal and
distal segment interplay with an injured segment), and
muscle endurance or ability to function without fatiguing. Exercise programs should focus on strengthening
weaker structures, with minimal aggravation of pain
symptoms. Nonspecific exercise programs for nonspecific diagnoses yield nonspecific and often unsuccessful
results.
Repetitive flexion-biased exercises for a patient with
a typical posterolateral disk herniation may increase
symptoms. Exercises that emphasize centralization of
pain symptoms to the lumbar spine in patients with low
back and leg symptoms may prove useful. The benefit of
determining the directional preference and proper plane
for flexion for each patient with back pain before initiating exercise treatment has been well documented. A
recent study of patients with acute low back pain has
shown that a classification-based exercise program improved disability and returns to work status in the exercise group compared with a control group followed for 1
year. Aerobic conditioning has been shown to be beneficial in terms of improving aerobic capacity, muscle
strength, and flexibility in patients with nonactive to
moderately active rheumatoid arthritis and osteoarthritis of the hip and/or knee. Short periods of immobilization immediately after surgery may be required for adequate tissue healing. However, during the postoperative
period, and especially following anterior cruciate ligament reconstruction, early rehabilitation to prevent loss
of range of motion and muscle strength has been shown
to improve functional outcome.
Physical conditioning programs, called work conditioning or work hardening programs, attempt to improve work status and function; however, there is no evidence of their efficacy for acute back pain. Emphasis
should be placed on use of programs that focus on return to work. Patients should participate in restoration
programs that simulate the activities of daily living that
they need to perform, and emphasize improvement in
function. Programs that emphasize pain relief alone and
that use short-term passive modalities are not effective.
If no improvement is seen in a 4- to 6-week period with
physical conditioning or work hardening programs,
other options such as tertiary rehabilitation should be
considered.

Tertiary Rehabilitation
Multidisciplinary Assessment
Because the diagnosis in patients with chronic pain or
disability may be multifactorial, the involvement of
many health professionals is often required. Before instituting a tertiary rehabilitation program, a multidisciplinary assessment of the patients treatment options

American Academy of Orthopaedic Surgeons

Chapter 55 Musculoskeletal Rehabilitation

should take place. This assessment should include all appropriate imaging tests, supportive electrodiagnostic
tests, and diagnostic injection procedures as needed.
Electromyography, nerve conduction velocity tests,
MRI, CT, myelography, diskography, selective nerve
root block, peripheral nerve block, and sympathetic
block tests may be included. Careful surgical decisionmaking and discussions with the patient and family
about surgical options should then take place. If the patient agrees to surgery, it should proceed with appropriate postoperative secondary care. Tertiary rehabilitation
is the last resort when all reasonable surgical interventions have been exhausted or the patient is unwilling to
elect additional surgery.
As a prelude to the tertiary rehabilitation process,
special assessment of physical and functional capacity
and psychosocial status is needed. From a physical perspective, patients have usually already undergone many
weeks or months of physical therapy guided by a variety
of health professionals and limited by resistance, fear,
and inhibition of function (often termed fearavoidance). During tertiary rehabilitation, the quantification of physical function by objective measures may
help to guide the treatment approach by setting parameters for exercise, by providing feedback on progress to
the patient, and by leading to better assessment of permanent impairment. Measurements of the mobility and
strength around the injured joint(s) or spinal region(s),
aerobic capacity, and the functional capacity of performing various activities of daily living (such as lifting, gripping, climbing) are important. Such tests are often
termed physical or functional capacity evaluations and
may involve goniometers, inclinometers, or threedimensional digitizers to measure mobility; isometric,
isoinertial, or isokinetic strength testing devices; bicycle
or treadmill ergometers; and a variety of lifting protocols and devices. In unmotivated patients or those with
chronic pain, measurements of functional capacities may
not reflect their true physiologic functional abilities.
Psychosocial assessment is also an essential prerequisite for planning tertiary rehabilitation. Traditionally,
clinicians have searched for causes of pain, seeking a
physical basis for pain complaints that, once identified,
could be eliminated or blocked. When no organic basis
was identified, a psychological cause was assumed,
hence the term psychogenic pain. However, the concept
of a simple dichotomy, that pain is physical or psychological, is inadequate. It is more accurate and clinically
effective to identify a variety of psychosocioeconomic
barriers to recovery or risk factors creating entitlement,
resistance, or fear-avoidance. A variety of simple validated questionnaires have been developed that can be
used to assess pain, disability, health status, and depression (Table 1). A psychiatric diagnosis can be made using the Structured Clinical Interview for the Diagnostic
and Statistical Manual of Mental Health Disorders,

657

Musculoskeletal Rehabilitation

TABLE 1 | Questionnaires Available for Patient Physical and

Psychological Assessment
Pain Drawing Questionnaire
Oswestry Low Back Pain Disability Questionnaire
Roland-Morris Disability Questionnaire
Medical Outcomes Study 36-Item Short Form
Million Disability Questionnaire

fourth edition. Significant psychopathology has been

noted in chronic pain patients. A psychological interview conducted by a clinical psychologist who considers
identified risk factors for ongoing disability may be useful. Personality changes may be manifested by anger,
hostility, noncompliance, and resistance to the efforts of
the therapeutic team. Minor head injuries, organic brain
dysfunction, and a history of alcohol or drug use can
produce cognitive errors and dysfunctions, which could
create clinical management problems and make the patient refractory to education.
Finally, it is imperative that the orthopaedic surgeon
is aware of the influence of the disability system, which
may create a variety of incentives and disincentives to
expected behaviors in treatment. Knowledge of the patients involvement with the workers compensation system (state or federal) and the key factors associated
with this system is needed for proper assessment. It
should be determined if the patient is receiving current
temporary total disability benefits, impairment or disability benefits, vocational rehabilitation benefits, or is
approaching a variety of financial and medical end
points (termed maximum medical improvement or permanent and stationary status). The case manager, adjustor, or attorney for the injured worker may be a valuable resource for information about case status. Other
financial benefits from private short- or long-term disability insurers, federal disability systems (Supplemental
Security or Social Security Disability benefits), or from
a variety of third parties (such as auto or product liability insurance) may affect patient motivation and behavior.

Functional Restoration Approaches

Functional restoration approaches to tertiary rehabilitation are oriented toward recovery from disability as well
as pain control. The physical and psychological components of disability are known to be a source of much of
the disability and it is believed that regaining greater physical capacity, particularly in the injured weak link, combined with multimodal disability management will ultimately reduce pain perception. A temporary increase in
pain during treatment is considered acceptable. Such programs have increased in popularity over the past 2 decades, with multiple outcome studies available in the lit-

658

Orthopaedic Knowledge Update 8

erature. These programs use the measurement of physical
and functional capacity to guide the rehabilitation approach quantitatively, limiting the amount of exercise so
that patients neither exert excessive efforts nor do so little that their time is wasted. Quantification removes some
of the subjectivity inherent in allowing patients to remain
in treatment who are giving only a negligible effort while
claiming they are doing their best.A supportive psychological and case management program must provide education about pain control and stress management techniques, provide treatment to resolve the disability,
reintegrate the patient into the work force, become involved in school or training as needed, and help in case
settlement efforts. Modalities that may enhance mobility
or help to control pain, such as intra-articular injections,
transcutaneous electrical nerve stimulation, and antiinflammatory or psychotropic medications may prove to
be valuable adjuncts to such programs. Objectively documented outcomes that show return to productivity are
the measure of success. Outcome measures for functional
restoration programs include work status (for example, return to work and work retention), future healthcare system use (for example, additional surgery to the injured
area, persistent healthcare-seeking behavior, number of
visits to new providers), recurrent injury after work return
(for example, new claims and lost time) and case closure
(workers compensation and third-party claims). Repeated use of pain or disability questionnaires may be useful, but should become the primary outcome measure only
if the patient has no outstanding disability issues.

Palliative Pain Management

Many pain management specialists and pain clinics focus on palliation of pain rather than restoration of function. Efforts usually are focused on helping the patient
reduce stress and tension while accepting a relatively
nonfunctional lifestyle. Multidisciplinary assessment
may result in recommendations for procedures that may
not be beneficial to functional recovery, but which may
ameliorate pain. These may include a variety of neuroablative procedures, radiofrequency neurotomies in the
spine, spinal cord stimulation, or the use of intrathecal
drug pumps. Treatment programs generally focus on
psychological interventions, with only minimal physical
rehabilitation. Although the physical component of
functionally-oriented rehabilitation usually has greater
intensity and more elements of supervision than secondary care options, physical components in palliative pain
management rely more on elements of primary care
(manipulation, acupuncture, massage, and thermal modalities) or very light exercise (aqua therapy, light
stretching). Contributing to the limited physical aspects
of these programs is a tendency to rely on narcotics as
an active component of pain control, with acceptance of
long-term narcotic use as a permanent component of

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

the patients medical care. Although psychotropic drugs
and NSAIDs may also be used, the type of narcotic and
dosage tends to determine much about the patients ultimate physical and psychological well-being. The regular use of continuous release narcotics (for example,
methadone, meperidine hydrochloride, hydromorphone
hydrochloride), or narcotic pumps is a rapidly growing
trend. Psychologic therapies focus on stress management and learning techniques to cope with inactivity
and lack of societal productivity, even in younger individuals. Federal disability benefits (Supplemental Security or Social Security Disability benefits) provide a
safety value for those with preretirement disabilities,
with the number of patients receiving such payments increasing from four million to nine million in the United
States over the past 20 years. These benefits have increased from $40 billion to $100 billion over the same
time period, now accounting for 5% of the US federal
budget. Outcome measurement may be problematic, because the patients self-assessment of pain and health
status may be markedly influenced by the treatment
team and requirements for continued narcotic use.

Summary
Primary rehabilitation focuses on control of painful
symptoms and prevention of sequelae of extensive immobility. Secondary rehabilitation stresses early reactivation of the patient with emphasis on stabilizing the injured area while improving flexibility, strength,
endurance, and coordination skills. Tertiary care is reserved for patients with chronic disabling musculoskeletal pain that require a more comprehensive and often
multidisciplinary approach to improve function even if
some pain symptoms persist. Palliative care is provided
when functional progress is no longer deemed feasible.

Annotated Bibliography
General
McGill SM: Low Back Disorders: Evidence-Based Prevention and Rehabilitation. Human Kinetics, Champaign, IL, 2002.
This book includes information on epidemiologic studies
on low back disorders, relevant functional anatomy and normal and injury mechanics of the lumbar spine, scientifically
based approaches to back pain prevention at work, and low
back rehabilitation.

Primary Rehabilitation
Fritz JM, Delitto A, Erhard RE: Comparison of
classification-based physical therapy with therapy based
on clinical practice guidelines for patients with acute
low back pain. Spine 2003;28:1363-1371.
Seventy-eight patients randomly received therapy based
on a classification system or clinical practice guidelines. For
patients with acute, work-related low back pain, the use of a

American Academy of Orthopaedic Surgeons

Chapter 55 Musculoskeletal Rehabilitation

classification-based approach resulted in improvement in disability and return-to-work status after 4 weeks as compared
with therapy based on clinical guidelines.

Secondary Rehabilitation
Mayer T, Polatin P, Smith B, et al: Spine rehabilitation:
Secondary and tertiary nonoperative care. Spine J 2003;
3(suppl 3):28S-36S.
This Contemporary Concepts Review presents a position
statement of the North American Spine Society Board and
summarizes aspects of secondary and tertiary rehabilitation
specific to spinal disorders in greater detail.

van Tulder MW, Malmivaara A, Esmail R, Koes BW: Exercise therapy for low back pain, in The Cochrane Library (Update Software on CD-ROM), Issue 3, 2003.
Thirty-nine randomized controlled trials were identified.
Exercise therapy was shown to be more effective than the
usual care given by general practitioners and equally effective
as conventional physiotherapy for chronic low back pain. Exercises may be helpful for patients with chronic low back pain
to facilitate earlier return to normal daily activities and work.

Tertiary Rehabilitation
Anagnostis C, Mayer T, Gatchel R, Proctor TJ: The Million Visual Analog Scale: Its utility for predicting tertiary rehabilitation outcomes. Spine 2003;28:1051-1060.
When a validated disability outcome questionnaire is
given to a group of patients with chronic disabling spinal disorders before and after tertiary rehabilitation, excellent predictive value for socioeconomic outcomes (work status, health
system usage, recurrent injury 1 year after treatment) is identified, particularly in patients who report high levels of disability
immediately after treatment.

Dersh J, Gatchel RJ, Polatin P: Chronic spinal disorders

and psychopathology: Research findings and theoretical
considerations. Spine J 2001;1:88-94.
This prospective study of the prevalence of psychopathology in a large sample of chronically disabled patients with
work-related spinal disorders uses a validated clinicianadministered instrument for psychiatric diagnosis. The Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders, fourth edition was used.

Jouset N, Fanello S, Bontoux L, et al: Effects of functional restoration versus 3 hours per week of physical
therapy: A randomized controlled study. Spine 2004;29:
487-494.
A functional restoration multidisciplinary approach featuring physical training and disability management proved better
than therapy alone in a randomized controlled trial assessed
by quantifiable outcomes.

Proctor TJ, Mayer TG, Gatchel RJ, McGeary DD: Unremitting health-care-utilization outcomes of tertiary re-

659

Musculoskeletal Rehabilitation
habilitation of chronic musculoskeletal disorders. J Bone
Joint Surg Am 2004;86:62-69.
Comparison of patients who persistently seek health care
after tertiary rehabilitation with those who do not reveals that
persistent healthcare seekers demonstrate poor outcomes in
work-related injuries that lead to higher societal costs and decreased worker productivity.

Schonstein E, Kenny D, Keating J, Koes B, Herbert RD:

Physical conditioning programs for workers with back
and neck pain: A Cochrane systematic review. Spine
2003;28:E391-E395.
A meta-analysis by the Cochrane collaboration of randomized trials of chronic back pain in work-related injuries
demonstrates that physical conditioning programs that incorporate a cognitive behavioral approach reduce work loss.

Classic Bibliography
Bendix AE, Bendix T, Haestrup C, Busch E: A prospective, randomized 5-year follow-up study of functional
restoration in chronic low back pain patients. Eur Spine
J 1998;7:111-119.
Hazard RG, Fenwick JW, Kalisch SM, et al: Functional
restoration with behavioral support: A one-year prospective study of patients with chronic low-back pain.
Spine 1989;14:157-161.
Jordan KD, Mayer TG, Gatchel RJ: Should extended
disability be an exclusion criterion for tertiary rehabilitation? Socioeconomic outcomes of early versus late
functional restoration in compensation spinal disorders.
Spine 1998;23:2110-2117.

660

Orthopaedic Knowledge Update 8

Mayer T, Gatchel R, Polatin P: Occupational Musculoskeletal Disorders: Function, Outcomes and Evidence. Lippincott Williams & Wilkins, Philadelphia, PA, 1999.
Mayer T, Gatchel R, Polatin P, Evans T: Outcomes comparison of treatment for chronic disabling work-related
upper extremity disorders and spinal disorders. J Occup
Environ Med 1999;41:761-770.
Mayer T, McMahon MJ, Gatchel RJ, Sparks B, Wright
A, Pegues P: Socioeconomic outcomes of combined
spine surgery and functional restoration in workers
compensation spinal disorders with matched controls.
Spine 1998;23:598-606.
Mayer TG, Gatchel RJ, Mayer H, Kishino N, Keeley J,
Mooney V: A prospective two-year study of functional
restoration in industrial low back injury: An objective
assessment procedure. JAMA 1987; 258:1763-1767.
Mazanec D: Medication use in sports rehabilitation, in
Kibler WB, Herring SA, Press JM (eds): Functional Rehabilitation of Sports and Musculoskeletal Injuries.
Gaithersburg, MD, Aspen, 1998, pp 71-79.
OSullivan PB, Phyty GD, Twomey LT, Allison GT: Evaluation of specific stabilizing exercise in the treatment of
chronic low back pain with radiologic diagnosis of
spondylolysis or spondylolisthesis. Spine 1997;22:
2959-2967.

American Academy of Orthopaedic Surgeons

Chapter

56

Genetic Diseases and Skeletal

Dysplasias
William G. Mackenzie, MD
R. Tracy Ballock, MD

Introduction
The recent advances in the fields of human and mouse
genetics and molecular biology have led to rapid
progress in understanding the etiology and pathogenesis
of many human skeletal dysplasias and other genetic
diseases affecting the skeleton. Some of these new findings relate to the diagnosis and treatment of children
with inherited disorders affecting the skeleton.

Trisomy 21 (Down Syndrome)

Duplication of a portion of the long arm of chromosome
21 occurs once every 800 to 1,000 live births and results in
the disorder known as Down syndrome. The most common chromosomal abnormality in humans, Down syndrome can be diagnosed prenatally by amniocentesis.
Postnatal diagnosis can be made by recognition of the
characteristic facies (upward slanting eyes, epicanthal
folds, and a flattened profile) as well as the single transverse flexion crease in the palm (simian crease).Although
mental retardation is usually associated with Down syndrome, the degree of mental deficiency is variable. Other
abnormalities may include congenital heart disease,
duodenal atresia, hypothyroidism, and hearing loss.
Children with Down syndrome may develop musculoskeletal problems as a result of the increased ligamentous laxity that occurs with this condition. Approximately
10% of patients will exhibit asymptomatic atlantoaxial instability. Therefore, it is recommended that children with
Down syndrome have screening flexion-extension radiographs of the cervical spine obtained before athletic participation. Spinal cord compression is rare, and surgical intervention is reserved for children who exhibit symptoms
of myelopathy (Figure 1).
In addition to C1-C2 instability, children with Down
syndrome also may develop hip instability. Unlike typical hip dysplasia in which the shallow acetabulum allows
progressive migration of the femoral head out of the
socket, in children with Down syndrome the hip may
dislocate out of an acetabulum that may be only mildly
dysplastic. This is possible because of the degree of ligamentous laxity present. Brace treatment for 6 to 8

American Academy of Orthopaedic Surgeons

months to achieve stability has been recommended for

children younger than 6 years. Surgical intervention for
this type of hip dislocation has a high failure rate and
has not been shown to improve outcomes.
Other musculoskeletal disorders occurring in children with Down syndrome are also the result of excess
ligamentous laxity and joint instability. Patellofemoral
subluxation and dislocation may develop in the absence
of symptoms. Soft-tissue reconstruction alone is frequently unsuccessful in preventing recurrent instability.
Pes planovalgus and hallux valgus are also common and
usually respond to shoe wear modifications.

Turners Syndrome
In 1 of every 3,000 live births, a single X chromosome
(XO) is present instead of the normal XX or XY combination, resulting in Turners syndrome. Patients are phenotypically females with short stature, a webbed neck,

Figure 1 MRI showing spinal cord compression in a 12-year-old girl with Down syndrome. An os odontoideum (arrow) has become lodged between the ring of C1 and the
dens. The patient had hyperreflexia but was otherwise neurologically intact.

663

Genetic Diseases and Skeletal Dysplasias

Table 1 | Diagnostic Criteria for Neurofibromatosis 1

Six or more caf-au-lait spots whose greatest diameter is 5 mm in prepubertal and 15 mm in postpubertal patients
Two or more neurofibromas of any type or one plexiform neurofibroma
Axillary freckling
Optic glioma
Two or more Lisch nodules (iris hamartomas)
A distinctive osseous lesion
A first-degree relative with neurofibromatosis 1

Orthopaedic Knowledge Update 8

in children age 6 years or younger. Nondystrophic scoliosis in younger children can modulate into the dystrophic type over several years. Radiographic features of
dystrophic scoliosis include scalloping of the vertebral
end plates, foraminal enlargement, and penciling of the
ribs. These dystrophic curves are notoriously resistant to
brace treatment and will progress if not treated by early
anterior and posterior fusion. Preoperative MRI is essential to identify areas of dural ectasia and intraspinal
neurofibromas.

Hereditary Multiple Exostosis

and a low hairline. Girls with Turners syndrome do not
pass through puberty or develop secondary sexual characteristics because of the lack of sex steroid hormones.
Cubitus varus is a common finding in Turners syndrome
but rarely requires treatment. Scoliosis may also develop and management is similar to that for idiopathic
curves.

Neurofibromatosis
Neurofibromatosis (NF) is divided into two distinct clinical entities, NF1 and NF2. NF1 is the most common single gene disorder, occurring once in every 3,000 births,
and it results from a mutation in the gene encoding a
protein now known as neurofibromin. Neurofibromin
helps regulate cell growth through modulation of the
Ras signaling pathway. The diagnosis of NF1 relies on
identification of up to six clinical criteria (Table 1).
Common clinical findings include caf-au-lait macules,
axillary freckles, Lisch nodules of the iris, and neurofibromas. Malignant transformation of a neurofibroma to
a neurofibrosarcoma results if a somatic mutation occurs in the remaining normal copy of the gene. Therefore, neurofibromas that enlarge suddenly or become
painful should be managed as potential sarcomas.
The typical bone lesion in neurofibromatosis is an
anterolateral bowing deformity of the tibia that may
progress to pseudarthrosis (Figure 2). Prophylactic bracing with a total contact orthosis is recommended to diminish the likelihood of pseudarthrosis formation. Once
a pseudarthrosis is established, bone grafting with intramedullary fixation is the initial treatment. For persistent pseudarthroses, either a vascularized bone graft or
bone transport by distraction osteogenesis may be required for healing. Although amputation and prosthetic
fitting for recalcitrant pseudarthrosis may result in improved lower extremity function over these salvage
techniques, this is not commonly performed.
Scoliosis occurs commonly in children with neurofibromatosis, and it is classified as either dystrophic or
nondystrophic. Nondystrophic curves resemble idiopathic scoliosis and are managed in a similar fashion.
Dystrophic curves are short and sharp, occurring over
four to six spinal levels, and represent 80% of scoliosis

664

Hereditary multiple exostosis (HME) is an autosomal

dominant disorder with a prevalence of 1 in 50,000 patients. Bony projections with cartilage caps develop near
the ends of multiple long bones (Figure 3). These exostoses continue to grow until skeletal maturity is
reached, and they may cause partial growth inhibition in
the adjacent physis, resulting in limb deformity, limblength discrepancy, and occasionally subluxation of an
adjacent joint. Although initially a benign lesion, the
risk of transformation to a malignant chondrosarcoma
or osteosarcoma has been estimated at 0.5% to 3%,
which is higher than the general population risk.
HME is caused by mutations in EXT1 or EXT2,
members of a newly described family of putative tumor
suppressor genes that encode glycosyltransferases.
EXT1 and EXT2 proteins have been localized by immunohistochemical techniques to the Golgi apparatus of
the cell, where they form a protein complex that is responsible for the biosynthesis of heparan sulfate glycosaminoglycans. The consequence of the diseasecausing mutations is loss of heparan sulfate on the cell
surface, where the molecule frequently functions as a
coreceptor for peptide growth factors to increase binding affinity.
Although EXT genes are ubiquitously expressed, the
disease appears to only result in aberrant proliferation
of growth plate chondrocytes. Histologically, abnormal
cytoskeletal inclusions consisting of actin and -actinin
are found in chondrocytes comprising the cartilage cap.
The relationship between this observation and the aberrant chondrocyte proliferation is unclear, however. Recently it has been demonstrated that the Drosophila
EXT1 homolog tout-velu is necessary for the diffusion
of the hedgehog protein in the developing wing tissue of
the fly by affecting the transduction of the hedgehog
signal across adjacent cells. Therefore, EXT mutations
may also affect the function of the Indian hedgehog
protein in the growth plate, whose role is to modulate
the rate at which chondrocytes stop proliferating and
differentiate into hypertrophic cells.
Current genetic observations indicate that HME
may not be a true chondrodysplasia, but rather a neoplastic condition. According to the Knudsen two-hit

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 56 Genetic Diseases and Skeletal Dysplasias

Figure 2 Radiograph showing anterolateral bowing in a child with neurofibromatosis, which progressed to congenital pseudarthrosis of the tibia.
model of tumor suppressor gene inactivation in cancer,
both copies of the gene must be inactivated to abolish
the normal tumor suppressor activity. In hereditary cancer, the first hit usually consists of a germline mutation,
whereas the second hit is the inactivation of the remaining wild-type copy through a somatic mutation. This
model has recently been applied to patients with HME.
An inherited haploinsufficiency (malfunction of one of
the two working copies of a gene within the cell) combined with a subsequent loss of function in the remaining copy of the gene through a somatic mutation is required for osteochondroma formation. The growth
dysregulation that ensues then predisposes the cell to
further genetic alterations, resulting in chondrosarcoma
in a small percentage of patients.

Skeletal Dysplasias
Achondroplasia and Related Disorders
Several chondrodysplasia phenotypes result from mutations in the fibroblast growth factor receptor 3 gene, in-

American Academy of Orthopaedic Surgeons

Figure 3 Typical radiographic appearance of multiple osteochondromas around the

knees in a patient with hereditary multiple exostosis.

665

Genetic Diseases and Skeletal Dysplasias

Orthopaedic Knowledge Update 8

Figure 4 Typical clinical (A) and radiographic (B) findings in a child with achondroplasia.
cluding achondroplasia, hypochondroplasia, thanatophoric dysplasia, and severe achondroplasia with
developmental delay and acanthosis nigricans dysplasia.
These disorders are closely related and represent a continuum of severity.
Achondroplasia
Achondroplasia, the most common form of dwarfism, is
an autosomal dominant disorder that is caused by a single nucleotide substitution. More than 90% of instances
of achondroplasia result from a sporadic mutation. The
specific mutation in achondroplasia converts either guanine to arginine or guanine to cysteine at position 380 in
the transmembrane domain of the protein, resulting in a
glycine to arginine substitution. This single amino acid
substitution not only causes stabilization of the fibroblast growth factor receptor protein and its accumulation on the cell surface, but also results in uncontrolled,
prolonged ligand-dependent activation of the receptor.
The result of this sustained fibroblast growth factor receptor activity is growth retardation in the proliferative
zone of the growth plate, leading to decreased bone
length. The most profound effect on the skeleton is in
the areas of greatest endochondral growth (humerus
and femur), resulting in the characteristic rhizomelia.
The rhizomelic pattern of shortening in children with

666

achondroplasia involves the proximal limb bones, including the humerus and femur, whereas children with
other disorders have mesomelia (shortening of the forearm and leg) and acromelia (shortening of the hands
and feet). The diagnosis is often made prenatally by ultrasound; if not identified prenatally, this disorder is
identified at birth by the presence of rhizomelic shortening, a trunk of normal length, macrocephaly, frontal
bossing, a depressed nasal bridge, and trident hands
(Figure 4). Elbow flexion contractures are the result of
bowing of the distal humerus and posterior radial head
subluxation/dislocation. Classic radiographic findings include shortening of the long bones, progressive narrowing of the interpedicular distance through the lumbar
spine, squared iliac wings, horizontal acetabula, and narrow sacrosciatic notches.
Foramen magnum and upper cervical stenosis is a
life-threatening disorder that affects these children at
birth and in early life. Cervicomedullary cord compression can result in hypotonia, delayed development,
weakness, and apnea. The central apnea can be complicated by obstructive components, including abnormal
nasopharyngeal development and enlarged tonsils and
adenoids. Routine perinatal screening with MRI is controversial. Symptomatic children are typically assessed
with MRI and a sleep study. Surgical management in-

American Academy of Orthopaedic Surgeons

Chapter 56 Genetic Diseases and Skeletal Dysplasias

Orthopaedic Knowledge Update 8

Figure 5 A and B Radiographs showing spontaneous correction of thoracolumbar kyphosis.

cludes decompression and occasionally shunting for hydrocephalus. Unlike other forms of skeletal dysplasias,
cervical instability is rare.
Thoracolumbar kyphosis occurs commonly in infants. It is thought to be secondary to truncal hypotonia
and weakness and an enlarged head. Avoidance of
propped sitting is advocated. The kyphosis typically resolves as children achieve independent ambulation
(later than in children of average stature with achondroplasia who typically have a mean walking age range of
18 to 24 months) (Figure 5).
Bracing is reserved for those children with severe
progressive kyphosis and wedging of the apical vertebra. Anterior and posterior spinal fusion is indicated in
older children with persistent severe kyphosis, but it is
rarely required. Lumbar spinal stenosis is common
(30% of children are symptomatic in the second decade), and it is caused by interpedicular narrowing,
short pedicles, disk bulging, and hyperlordosis. In older
patients, the stenosis is aggravated by degenerative
changes. Lumbosacral decompression is recommended
for patients with symptomatic spinal stenosis. Spinal fusion is required in addition to decompression in patients

American Academy of Orthopaedic Surgeons

with thoracolumbar kyphosis and those with significant

degenerative changes. Scoliosis is unusual in this patient
population.
Genu varum is typical in over 90% of patients with
achondroplasia. The deformity may occur at multiple
levels, including the distal femur, the proximal and distal
tibia, and through the lateral ligaments of the knee. Realignment osteotomies are recommended for patients
with progressive, symptomatic genu varum. Correction
of the associated internal tibial torsion is done concurrently.
Short stature in achondroplasia is moderately severe
and can result in functional limitations. Teenagers with
achondroplasia have concerns about their body image.
Adult studies indicate that function and self-image in
patients with achondroplasia are similar to those of
average-statured adults. Growth hormone has been advocated in growing children, but its use is controversial
because the improvement in final height is variable. Extended limb lengthening to increase stature has been
used commonly in Europe, but is more controversial in
North America. Long-term effects on the adjacent articular surfaces have not been well studied.

667

Genetic Diseases and Skeletal Dysplasias

Orthopaedic Knowledge Update 8

clude rhizomelic shortening, macrocephaly, platyspondyly, and severe restrictive lung disease resulting
from a small thoracic cavity.
Severe Achondroplasia With Developmental Delay and
Acanthosis Nigricans Dysplasia
Children with severe achondroplasia with developmental delay and acanthosis nigricans dysplasia exhibit extreme short stature, developmental delay, acanthosis nigricans, and genu varum.

Dysplasias Secondary to Type II Collagen

Abnormalities
This spectrum of disorders includes spondyloepiphyseal
dysplasia congenita, spondyloepimetaphyseal dysplasia,
and Kniest dysplasia.

Figure 6 Photograph showing typical clinical appearance of a child with spondyloepiphyseal dysplasia congenita.

Hypochondroplasia
Hypochondroplasia is an autosomal dominant disorder
with clinical features and radiographic findings similar
to those associated with achondroplasia, but to a milder
degree. The diagnosis is rarely apparent before age
2 years and typically results from investigation of short
stature. Musculoskeletal problems that present for management in hypochondroplasia include lumbar spinal
stenosis, genu varum, and short stature. Management of
hypochondroplasia is the same as for the disorders associated with achondroplasia.
Thanatophoric Dysplasia
Thanatophoric dysplasia is a severe, usually lethal disorder resulting from mutations in the fibroblast growth
factor receptor 3. Clinical features of this dysplasia in-

668

Spondyloepiphyseal Dysplasia
Spondyloepiphyseal dysplasias are characterized by
short stature secondary to a short trunk and short limbs.
Many different types of this disorder exist, with the
most common type being the congenita form. This is
caused by mutations in COL2A1 (collagen type II 1
chain). This gene encodes type II collagen, which is
found primarily in cartilage and in the vitreous humor,
locations consistent with the phenotype of spondyloepiphyseal dysplasia congenita and other disorders such as
Kniest dysplasia and type I Sticklers syndrome. Radiographic changes include abnormal spinal development
with odontoid hypoplasia and platyspondyly, abnormal
formation of the long bone epiphyses with variable
metaphyseal involvement, and generalized delay in epiphyseal ossification. The diagnosis is made at birth. Patients have marked short stature with a very short trunk
and often a barrel-shaped chest (Figure 6). Lumbar lordosis is typical, and progressive kyphoscoliosis occurs.
Retinal detachment, severe myopia, and sensorineural
hearing loss are common in childhood and adult life.
Atlantoaxial instability resulting from odontoid hypoplasia and ligamentous laxity must be evaluated early in
life and monitored on a regular basis with flexionextension lateral C-spine radiographs and/or MRI. Surgical management is indicated in patients with significant
instability or cervical myelopathy. Progressive kyphoscoliosis in the growing child can be managed with a brace;
however, with progression, surgical management is usually
required. Lower extremity malalignment is common in
these children. Coxa vara, genu valgum, valgus alignment
of the distal tibia, and planovalgus foot deformities are
typical. These deformities can result in significant gait abnormalities consisting of increased lumbar lordosis, a waddling gait, and a crouch gait with the knees knocking together. The coxa vara is often progressive and is difficult
to assess radiographically because of the delayed capital
femoral ossification (Figure 7).

American Academy of Orthopaedic Surgeons

Chapter 56 Genetic Diseases and Skeletal Dysplasias

Orthopaedic Knowledge Update 8

When considering realignment of the lower extremities, deformity at the hip, knee, and ankle must be considered and managed together. Abnormal epiphyseal
development typically results in early osteoarthritis of
the weight-bearing joints, and most young adults require
total joint arthroplasty. There is no evidence that realignment osteotomies will delay the need for joint arthroplasty.
An uncommon tarda form of spondyloepiphyseal
dysplasia is milder and presents later in the first decade
of life. This form has X-linked inheritance and is caused
by a mutation in the SEDL gene, which is thought to be
involved in vesicle transport from the endoplasmic reticulum to the Golgi apparatus. Patients with this type of
spondyloepiphyseal dysplasia present with deformities
that are milder but similar to those of patients with
spondyloepiphyseal dysplasia congenita. In this patient
population, the upper cervical spine must be assessed,
scoliosis may develop, and osteoarthritis frequently occurs in later life.
Kniest Dysplasia
Kniest dysplasia is an autosomal dominant disorder
caused by mutations in COL2A1. The trunk is disproportionately shorter than the limbs in patients with Kniest dysplasia, and the face is flattened with midface hypoplasia. Myopia, retinal detachment, deafness, joint
contractures, limb malalignment, and kyphoscoliosis can
occur throughout life in this patient population. The radiographic features include dumbbell-shaped long bones
with broad metaphyses and irregular dysplastic epiphyses. Joint stiffness and pain typically result from deterioration of the articular cartilage.

Metaphyseal Chondrodysplasia
Although many forms of metaphyseal dysplasia have
been described, metaphyseal involvement with short
stature and bowing of the legs is a common feature to
all. The most common Schmid type is autosomal dominant and is caused by a type X collagen mutation
(COL110A1). This type is the mildest in this group, with
patients typically exhibiting moderate short stature, a
waddling gait, and genu varum. The diagnosis is usually
made in early childhood. The radiographic features resemble those seen in patients with rickets and include
metaphyseal irregularity and flaring with widening of
the physes.
The Jansen type is a rare autosomal dominant disorder caused by a mutation in the parathyroid hormone
receptor gene that regulates the differentiation of
growth plate chondrocytes. Severe short stature with deformity is typical, and some affected children have hypercalcemia, hypercalciuria, and hyperphosphaturia.
The McKusick type is an autosomal recessive disorder (also called cartilage-hair hypoplasia) caused by a

American Academy of Orthopaedic Surgeons

Figure 7 Radiograph showing coxa vara in a child with spondyloepiphyseal dysplasia

congenita.

mutation of RMRP, a nuclear gene (RNA component of

mitochondrial RNA processing endoribonuclease). Patients with this disorder have fine, sparse body hair and
significant complications, including impaired cellular immunity, anemia, Hirschsprungs disease, and malignancy.
Patients with metaphyseal chondrodysplasia may need
surgical management of short stature and limb malalignment (Figure 8).

Pseudoachondroplasia
Pseudoachondroplasia is an autosomal dominant disorder caused by a mutation in the gene encoding for cartilage oligomeric matrix protein (COMP). COMP is an
extracellular calcium-binding glycoprotein belonging to

669

Orthopaedic Knowledge Update 8

Genetic Diseases and Skeletal Dysplasias

Figure 9 Radiographic appearance of the hips of a patient with multiple epiphyseal

dysplasia.

Figure 8 Radiographic appearance of the knees of a patient with Jansen metaphyseal chondrodysplasia.
the thrombospondin family, and it is involved in chondrocyte migration and proliferation. The COMP molecule is composed of five flexible arms with large globular domains at the end of each arm, resembling a
bouquet of flowers. Mutations affecting the type III repeat region or C-terminal domain of the protein result
in decreased calcium binding caused by a structural
change in the protein. Approximately 30% of patients
have an in-frame deletion mutation, resulting in four aspartic acid residues instead of five at amino acids 469
through 473 of the protein. It is interesting to note that
mutations in the COMP gene have also been discovered
in patients with multiple epiphyseal dysplasia. This suggests that pseudoachondroplasia and multiple epiphyseal dysplasia, although originally described as distinct
disorders, are now recognized as part of a disease spectrum.
Children with pseudoachondroplasia have a short
trunk and short limb dysplasia that is not usually diagnosed until early childhood. Atlantoaxial instability is
common. Generalized ligamentous laxity is present, and
this is particularly noticeable in the hands (the fingers
are short and hypermobile). Radiographic features include shortening of the long bones with irregular, expanded metaphyses and small fragmented epiphyses.
Evidence of platyspondyly can be observed with unique
anterior projections. Genu valgum, genu varum, or
windswept deformities of the lower extremities can also

670

be observed. Limb alignment procedures are indicated

for patients with severe progressive deformities that interfere with function, but there is a high rate of recurrence because of ligamentous laxity and abnormal bone
growth. Early arthritis resulting in hip arthroplasty occurs in about one third of patients with this disorder by
the fourth decade of life.

Multiple Epiphyseal Dysplasia

Multiple epiphyseal dysplasia (MED) describes a spectrum of autosomal dominant disorders characterized by
epiphyseal dysplastic changes resulting in mild to moderate short stature, genu valgum, and early onset osteoarthritis. Short, hyperextensible fingers are present in
patients with severe involvement. The diagnosis is usually made in midchildhood, with reports of pain and
joint stiffness in patients with short stature. Radiographs
demonstrate delayed epiphyseal ossification with subsequent irregular development of the epiphyses. The spine
is usually not affected. The clinical phenotypes are variable, ranging from mild epiphyseal abnormalities in the
hips that can lead to a misdiagnosis of bilateral LeggCalv-Perthes disease to severe widespread epiphyseal
dysplasia with joint contractures, osteoarthritis, and
short stature.
In the hips of children with MED, capital femoral ossification is delayed with subsequent fragmented ossification, and it eventually coalesces to form an intact but
small ossific nucleus. Although the radiographic appearance in the fragmented stage appears similar to that
seen in patients with Legg-Calv-Perthes disease, the
typical stages of sclerosis, collapse, fragmentation, and
reossification are not apparent (Figure 9). Osteonecrosis
can occur in the hips of patients with MED, thereby

American Academy of Orthopaedic Surgeons

Chapter 56 Genetic Diseases and Skeletal Dysplasias

Orthopaedic Knowledge Update 8

complicating the radiographic appearance. Management
in this circumstance is similar to the containment methods used for patients with Legg-Calv-Perthes disease.
Painful, stiff joints are managed by avoidance of activities that result in significant joint stress, administration
of nonsteroidal anti-inflammatory drugs, and aquatic
and physical therapy. The typical genu valgum can be
managed by staple hemiepiphysiodesis or osteotomy.
Progressive osteoarthritis usually results in total joint arthroplasty.
Three separate genetic loci have been linked to
MED. As noted previously, mutations causing all clinical
forms of MED have been identified in the gene encoding COMP. The phenotype of the mild pseudoachondroplasia patients overlaps with those of MED. Other mutations have been identified in COL9A2, which encodes
the 2 chain of type IX collagen. Type IX collagen is a
nonfibrillar heterotrimeric molecule with three chains
encoded by three different genes. It is a structural component of hyaline cartilage, intervertebral disks, and the
vitreous body of the eye. Type IX collagen decorates the
surface of type II collagen molecules to which it is covalently cross-linked. Its function is postulated to involve mediating the interaction of type II collagen with
other extracellular matrix components in cartilage.
The phenotype of the COL9A2 mutants seems to be
milder without hip involvement compared with the
more severe disease caused by mutations in COMP. All
mutations in COL9A2 described to date result in splicing errors that eliminate exon 3, and hence delete 12
amino acids from the N-terminal portion of the molecule. This may affect the structure and function of the
molecule in mediating interactions between type II collagen molecules and other extracellular matrix components.
Recently, an unclassified form of MED has been
linked to COL9A3 in a four-generation family with autosomal dominant disease. This is the first diseasecausing mutation to be identified in COL9A3, which encodes another of the three chains of type IX collagen.
The phenotype of the COL9A3 mutants overlaps significantly with COL9A2 mutants, but it differs by the presence of hip involvement.

Figure 10 Radiograph showing genu valgum in a patient with Ellis-van Creveld syndrome.

gum resulting from characteristic deficiency of the anterolateral tibial epiphyses (Figure 10). Realignment typically requires femoral and tibial osteotomies.
Ellis-van Creveld syndrome has recently been linked
to a new gene named EVC, which encodes a protein
that has no homology to known proteins. It is expressed
at higher levels in the distal limb than the proximal limb
in human embryonic tissue, and it is also expressed in
the developing vertebral bodies, ribs, heart, kidneys, and
lungs. Although the structure of the EVC gene product
includes both putative nuclear localization signals and a
transmembrane domain, its function is currently unknown.

Ellis-van Creveld Syndrome

Ellis-van Creveld syndrome is an autosomal recessive
disorder characterized by postaxial hand polydactyly,
short stature with mesoacromelic shortening, sparse,
thin hair, and dysplastic nails and teeth. Severe congenital heart disease is seen in approximately 60% of children with this disorder. The diagnosis is typically made
at birth. The radiographic features include the postaxial
polydactyly, fusion of the capitate and hamate, and medial spikes projecting from the iliac bones. Children with
Ellis-van Creveld syndrome develop severe genu val-

American Academy of Orthopaedic Surgeons

Dyschondrosteosis (Leri-Weill Syndrome)

Dyschondrosteosis (Leri-Weill syndrome) is a common
dominant disorder resulting in mild short stature, mesomelic shortening, and Madelungs deformity. The radiographic features are typical of Madelungs deformity,
with bowing of the radius and dorsal dislocation of the
ulna. Mutations in the short stature homeobox (SHOX)
gene have recently been implicated as the cause of dyschondrosteosis. Located at the very tip of the short arms
of both sex chromosomes, the SHOX gene encodes a

671

Genetic Diseases and Skeletal Dysplasias

homeobox-containing DNA transcription factor. SHOX
defects also appear to be associated with growth failure
in Turners syndrome, although it is not clear why female patients with Turners syndrome do not develop
dyschondrosteosis.

Disorders Caused by Abnormalities in Genes

Important in Normal Skeletal Development
This spectrum of disorders includes cleidocranial dysostosis and nail-patella syndrome.
Cleidocranial Dysostosis
Cleidocranial dysostosis is an autosomal dominant disorder with subtle clinical features consisting of a broad
forehead, delayed closure of the anterior fontanel, increased ability to appose the shoulders, shortening of
the middle phalanges of the third through fifth fingers,
delayed eruption of the permanent dentition, and mild
short stature. The diagnosis can be made at birth but is
often delayed. Radiographic features include multiple
wormian bones in the skull, with delayed closure of the
sutures and anterior fontanel. The clavicles are small
and absent, and there is widening of the symphysis pubis. The disorder is caused by a mutation of core binding
factor 1 (CBFA1). CBFA1 is a gene that is important
in the induction of osteoblast differentiation. The clinical features result in abnormal development of the anterior midline skeleton. Coxa vara can occur, and if symptomatic or progressive, it can be corrected with a
proximal femoral osteotomy.
Nail-Patella Syndrome
Nail-patella syndrome or osteoonychodysplasia is an autosomal dominant disorder characterized by aplasia or
hypoplasia of the patellae and dysplastic nails, with the
thumbnails being most commonly involved. Mutations
of the Lim homeobox transcription factor 1 (LMX1B)
gene cause this syndrome. LMX1B is involved in determining normal dorsoventral patterning in the developing limb bud, and it is expressed in the eyes and kidneys.
Up to 30% of affected individuals develop renal failure
by the fourth decade of life. Glaucoma is another complication associated with this disorder. The radiographic
features include small or absent patellae and iliac horns.
Posterior dislocation of the radial head limits elbow extension. Knee deformity and patellar malalignment may
require soft-tissue releases and realignment osteotomies.

Disorders Caused by Abnormalities in Genes That Play

a Role in the Processing of Proteins
This spectrum of disorders includes diastrophic dysplasia and a large group of lysosomal storage diseases
called mucopolysaccharidoses.

672

Orthopaedic Knowledge Update 8

Diastrophic Dysplasia
Diastrophic dysplasia is an autosomal recessive disorder
characterized by marked short stature and progressive
deformity. The diagnosis can be made prenatally by ultrasound and by findings at birth, including a cleft palate, tracheomalacia, hitchhiker thumbs, and bilateral
equinocavovarus or skewfoot deformities. Cervical kyphosis is typically present in the perinatal period and in
later childhood, joint contractures develop, and kyphoscoliosis is common. Cauliflower ear deformities resulting from cystic swelling in the cartilage of the ear
commonly develop after birth.
Mutations in the diastrophic dysplasia sulfate transporter gene have been identified as causing diastrophic
dysplasia. The diastrophic dysplasia sulfate transporter
gene is ubiquitously expressed and encodes a protein
that facilitates the transport of sulfate across the cell
membrane. The disease primarily affects cartilage because of the importance of negatively charged sulfate
groups in the function of proteoglycan molecules, which
is to maintain the hydration and compressive strength of
cartilage. One in 70 Finnish citizens are carriers of a mutant diastrophic dysplasia sulfate transporter gene.
Cervical kyphosis often resolves spontaneously. Progressive deformity and/or spinal cord compression requires surgical management. Kyphoscoliosis can also occur in the thoracolumbar spine; if progressive, it
typically requires anterior and posterior spinal fusion.
Early hip development is normal, but most children
with this disorder develop hip flexion contractures with
epiphyseal irregularity. End stage osteoarthritis is managed with total joint arthroplasty. Patients with diastrophic dysplasia often have flexion contractures of
the knees, valgus alignment, and patellar dislocation.
Surgical realignment of the knees can be difficult in
these patients. Congenital foot deformities include the
equinocavovarus feet or skew feet. The traditional surgical management results in rigid deformities with a high
rate of recurrence, of which recurrent equinus is the
most common. Supramalleolar osteotomies are often
needed for realignment.
Mucopolysaccharidoses
The mucopolysaccharidoses consist of a large group of
lysosomal storage diseases. Abnormal function of lysosomal enzymes results in the intracellular accumulation
of partially degraded compounds. The mucopolysaccharidoses are subdivided by the enzymatic deficiency and
the accumulation of these partially degraded compounds (Table 2). The subtypes are inherited in an autosomal recessive manner, with the exception of Hunters
syndrome, which is a sex-linked recessive disorder.
The clinical features vary depending on the severity
of the subtype. Corneal clouding, deafness, hepatosplenomegaly, and cardiovascular abnormality all are
present in Hurlers syndrome, and children with this dis-

American Academy of Orthopaedic Surgeons

Chapter 56 Genetic Diseases and Skeletal Dysplasias

Orthopaedic Knowledge Update 8

Table 2 | Mucopolysaccharidoses Subtypes

Subtype

Cause

Prognosis

Type I H (Hurlers syndrome)

Type I HS (Hurler-Scheie syndrome)
Type I S (Scheies syndrome)
Type II (Hunters syndrome)
Type III (Sanfilippos syndrome)
Type IV (Morquios syndrome)

Alpha-L-iduronidase deficiency

Type I H: Death in the first decade of life

Type I HS: Death in third decade of life
Type I S: Good survival
Death in second decade of life
Death in second decade of life
More severe involvement in patients with type IV A
than in those with type IV B; survival into
adulthood is possible
Poor survival with severe form
Poor survival

Type VI (Maroteaux-Lamy syndrome)

Type VII (Slys syndrome)

Sulpho-iduronate-sulphatase deficiency
Multiple enzyme deficiency
Type A (galactosamine-6-sulfate-sulphatase
deficiency)
Type B (beta-galactosidase deficiency)
Arylsulphatase B deficiency
Beta-glycuronidase deficiency

order usually die in the first decade of life. Morquio syndrome, in contrast, has a much better prognosis, and
there is only mild corneal clouding with deafness, no
hepatosplenomegaly, and only mild cardiovascular abnormalities. Biochemical analysis of the urine can lead
to the diagnosis of the specific mucopolysaccharidoses.
Specific enzyme activity known to be abnormal can be
detected in skin fibroblast culture and prenatally using
chorion villous sampling. All of the subtypes lead to
short stature; patients with Morquio syndrome are the
most severely affected. Although there are common radiographic findings among this group of disorders, it is
not possible to differentiate the various types based on
radiographic features alone. The skull is enlarged with a
thick calvarium. The ribs are broader anteriorly than
posteriorly. The vertebral bodies are ovoid when immature, but in time they develop platyspondyly. In patients
with Morquio syndrome, an anterior beak develops at
the thoracolumbar junction (Figure 11). Kyphoscoliosis
is common. Epiphyseal ossification is delayed, and
marked deformity of the joints can develop. The second
through fifth metacarpals are narrowed at their proximal ends and the phalanges are bullet-shaped.
Atlantoaxial instability resulting from odontoid hypoplasia and ligamentous laxity is very common, particularly in patients with Morquio syndrome. Soft-tissue
deposition in this area also results in further narrowing
of the spinal canal. Children with this disorder must be
very carefully evaluated for clinical signs of cervical myelopathy and for any evidence of atlantoaxial instability.
Treatment is by surgical stabilization of this area and
decompression if required. Thoracolumbar kyphosis
with anterior wedging is commonly seen in patients with
Morquios syndrome. Bracing may be required; if it is
progressive, anterior and posterior fusion is indicated.
Severe hip deformity is common in this patient population, and proximal femoral and periacetabular osteotomies are typically used to realign the hip. Genu valgum
can be secondary to distal femoral or proximal tibial
valgus. If severe and interfering with function, realign-

American Academy of Orthopaedic Surgeons

Figure 11 Radiograph showing the presence of characteristic vertebral anterior

beaking in a patient with Morquio syndrome.

ment can be achieved by osteotomies or staple hemiepiphysiodesis in the growing child.

Bone marrow transplantation has been used successfully to treat patients with Hurlers syndrome and other
mucopolysaccharidoses. Although there has been improvement in the coarse facies and hepatosplenomegaly,
the neurologic abnormalities and skeletal deformities
persist. Extensive research in the areas of gene therapy
and enzyme replacement is ongoing.

673

Orthopaedic Knowledge Update 8

Genetic Diseases and Skeletal Dysplasias

Annotated Bibliography
General Reference
Flynn M, Pauli R: Double heterozygosity in bone
growth disorders: Four new observations and review.
Am J Med Genet 2003;121A:193-208.
This article reviews the complex progeny with double heterozygosity and reports that matings between individuals with
short stature are common.

Unger S: A genetic approach to the diagnosis of skeletal

dysplasia. Clin Orthop 2002;401:32-38.
The author recommends that molecular investigation of a
skeletal dysplasia should be undertaken only after a limited
differential diagnosis is established.

children with congenital pseudarthrosis of the tibia: The

effect of treatment. J Pediatr Orthop 1998;18:381-386.
The authors of this study report that patients with healed
congenital pseudarthrosis of the tibia demonstrated muscle
weakness and inefficient gait compared with patients who
were treated by amputation.

Parisini P, Di Silvestri M, Greggi T, Paderni S, Cervellati

S, Savini R: Surgical correction of dystrophic spinal
curves in neurofibromatosis: A review of 56 patients.
Spine 1999;24:2247-2253.
In this study of patients with neurofibromatosis who underwent surgical correction of dystrophic spinal curves, the authors report that the failure incidence of posterior instrumentation and fusion alone was 53% of patients, and for combined
anterior fusion and posterior fusion it was 23%.

Trisomy 21 (Down Syndrome)

Doyle JS, Lauerman WC, Wood KB, Krause DR: Complications and long-term outcomes of upper cervical
spine arthrodesis in patients with Down syndrome.
Spine 1996;21:1223-1231.

Hereditary Multiple Exostosis

Duncan G, McCormick C, Tufaro F: The link between
heparan sulfate and hereditary bone disease: Finding a
function for the EXT family of putative tumor suppressor proteins. J Clin Invest 2001;108:511-516.

In this outcome study of patients with Down syndrome

who underwent upper cervical spine arthrodesis, 11 of 15 patients (73%) had 23 major complications, including nonunion,
loss of reduction, neurologic deterioration, late subaxial instability, infection, and wound dehiscence.

The authors of this article review the biology of EXT proteins in vivo and explore the possible roles of these proteins in
normal bone development and the formation of exostoses.

Winell J, Burke SW: Sports participation of children

with Down syndrome. Orthop Clin North Am 2003;34:
439-443.

Koziel L, Kunath M, Kelly OG, Vortkamp A: Ext1dependent heparan sulfate regulates the range of Ihh
signaling during endochonral ossification. Dev Cell 2004;
6:801-813.

The authors report that patients with Down syndrome can

actively participate in athletic activities if their orthopaedic
problems are appropriately managed.

Exostosin 1 is necessary for the synthesis of heparan sulfate chains of proteoglycans. These regulate the signaling of
several growth factors. The authors also found that the loss of
tout velu in Drosophila inhibits Hedgehog movement.

Turners Syndrome
Batch J: Turner syndrome in childhood and adolescence.
Best Pract Res Clin Endocrinol Metab 2002;16:465-482.
The authors provide a review of the clinical features of patients with Turners syndrome and discuss management issues.

Ogata T: SHOX haploinsufficency and its modifying factors. J Pediatr Endocrinol Metab 2002;15:1289-1294.

Skeletal Dysplasias
Aldegheri R, DallOca C: Limb lengthening in short
statured patients. J Pediatr Orthop B 2001;10:238-247.
In this article, the authors provide a good review of limb
lengthening for stature.

This article describes the clinical findings and pathophysiology of SHOX mutations in patients with Turners syndrome.

Cooper S, Flaitz C, Johnston D, Lee B, Hecht J: A natural history of cleidocranial dysplasia. Am J Med Genet
2001;104:1-6.

Neurofibromatosis

This is a review of orthopaedic and other abnormalities

observed in patients with cleidocranial dysplasia. The authors
also provide management recommendations.

Durrani AA, Crawford AH, Choudry SN, Saifuddin A,

Morley TR: Modulation of spinal deformities in patients
with neurofibromatosis type I. Spine 2000;25:69-75.
In this review of 91 patients with neurofibromatosis and
spinal deformity, the authors report that nondystrophic curves
may modulate into dystrophic curves, particularly before age
7 years.

Karol LA, Haideri NF, Halliday SE, Smitherman TB,

Johnston CE II: Gait analysis and muscle strength in

674

Dalvie S, Skinner J, Vellodi A, Noorden M: Mobile thoracolumbar gibbus in Morquio Type A: The cause of
paraparesis and its management. J Pediatr Orthop B
2001;10:328-330.
In this article, the authors report that supine spine MRI
may underestimate spinal stenosis and cord compression because of mobility of the thoracolumbar kyphosis in patients
with Morquio syndrome.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Gomez P, Garcia O, Ginebreda M, Gairi T, Vilarrubius
G: Lumbar canal stenosis in achondroplasia: Prevention
and correction of lumbosacral lordosis. An Esp Pediatr
2001;54:126-131.
The authors of this article report that femoral lengthening
improves lumbar hyperlordosis and reduces the symptoms of
lumbar spinal stenosis.

Helenius I, Remes V, Tallroth K, Peltonen J, Possa M,

Paavilainen T: Total hip arthroplasty in diastrophic dysplasia. J Bone Joint Surg Am 2003;85-A:441-447.
The authors of this article report that the rate of implant
survival is good and that Harris hip scores increased significantly in patients with diastrophic dysplasia who underwent
total hip arthroplasty.

Kitoh H, Kitakoji T, Kurita K, Katoh M, Takamine Y:

Deformities of the elbow in achondroplasia. J Bone
Joint Surg Br 2002;84:680-683.
The authors of this article report that posterior bowing of
the distal humerus (primarily) and posterior dislocation of the
radial head result in loss of elbow extension.

Morgan K, Rehman M, Schwartz R: Morquios syndrome and its anaesthetic considerations. Paediatr
Anaesth 2002;12:641-644.
The authors of this article provide a review of anesthetic
care for children with Morquio syndrome.

Munns C, Glass I, LaBrom R, et al: Histopathological

analysis of Leri-Weill dyschondrosteosis: Disordered
growth plate. Hand Surg 2001;6:13-23.
The authors of this article report that the zone of dyschondrosteosis is characterized by marked disruption of the normal
physeal arrangement.

Remes V, Marttinen E, Poussa M, Helenius I, Peltonen

J: Cervical spine in patients with diastrophic dysplasia:
Radiographic findings in 122 patients. Pediatr Radiol
2002;32:621-628.
The authors of this article report that the most common
alignment of the cervical spine in adulthood is cervical lordosis and that degenerative changes are common. As the prevalence of spina bifida occulta is high, the authors recommend
that care be taken during the surgical approach for patients
with diastrophic dysplasia.

Remes V, Poussa M, Peltonen J: Scoliosis in patients

with diastrophic dysplasia: A new classification. Spine
2001;26:1689-1697.

American Academy of Orthopaedic Surgeons

Chapter 56 Genetic Diseases and Skeletal Dysplasias

The authors of this article report that scoliosis is common
among patients with diastrophic dysplasia; types include early
progressive in 10% of patients, idiopathic-like in 50%, and
mild nonprogressive in 40%.

Stanley G, McLoughlin S, Beals RK: Observations on

the cause of bowlegs in achondroplasia. J Pediatr
Orthop 2002;22:112-116.
The authors of this study report that tibia vara increased
during growth in patients with achondroplasia and that
younger children with this disorder tended to have proximal
tibial varus and older children tended to have distal tibial
varus. Increased fibular length was thought to play a role.

Classic Bibliography
Bailey JA II: Orthopaedic aspects of achondroplasia.
J Bone Joint Surg Am 1970;52:1285-1301.
Bethem D, Winter RB, Lutter I, et al: Spinal disorders of
dwarfism: Review of the literature and report of eighty
cases. J Bone Joint Surg Am 1981;63:1412-1425.
Fairbank T: Dysplasia epiphysealis multiplex. Br J Surg
1947;34:325.
Hastbacka J, Superti-furga A, Wilcox WR, Rimoin DL,
Cohn DH, Lander ES: Sulfate transport in chondrodysplasia. Ann N Y Acad Sci 1996;785:131-136.
Kopits SE: Orthopaedic complications of dwarfism. Clin
Orthop 1976;114:153-179.
Mackenzie WG, Bassett GS, Mandell GA, Scott CI Jr:
Avascular necrosis of the hip in multiple epiphyseal dysplasia. J Pediatr Orthop 1989;9:666-671.
Peltonen JL, Hoikka V, Poussa M, Paavilainen T, Kaitila
I: Cementless hip arthroplasty in diastrophic dysplasia.
J Arthroplasty 1992;7(suppl):369-376.
Poussa M, Merikano J, Ryoppy S, Marttinen E, Kaitila I:
The spine in diastrophic dysplasia. Spine 1991;16:881887.
Remes V, Marttinen E, Poussa M, et al: Cervical kyphosis in diastrophic dysplasia. Spine 1999;24:1990-1995.
Ribbing S: Studien uber Hereditaire multiple Epiphysenstorungen. Acta Radiol 1937;1(suppl):34.
Ryoeppy S, Poussa M, Merikanto J, Marttinen E, Kaitila
I: Foot deformities in diastrophic dysplasia: An analysis
of 102 patients. J Bone Joint Surg Br 1992;74:441-444.

675

Chapter

57

Neuromuscular Disorders in Children

John F. Sarwark, MD
Arash Aminian, MD
David E. Westberry, MD
Jon R. Davids, MD
Lori A. Karol, MD

Myelomeningocele
Myelomeningocele (spina bifida, myelodysplasia) is the
most common major birth defect, with an incidence in
the United States ranging from 0.6 per 1,000 to 0.9 per
1,000 births. This neural tube defect results from embryologic failure of closure of neural crests during the neurulation phase of the spine in the third to fourth week
after fertilization. The failure results in a cerebrospinal
fluid-filled swelling of dura and arachnoid with spinal
nerve roots contained in the sac. In 85% to 95% of patients, the disorder is caused by dietary deficiency of
folate. Daily supplementation of 0.4 mg (400 g) of folic
acid before conception and during the pregnancy reduces the risk of neural tube defects significantly.
Prenatal diagnosis via enzyme elevation of maternal
serum -fetal protein has an accuracy rate of 60% to
95% for screening neural tube defects. An improved ultrasonographic technique for prenatal diagnosis also has
allowed informed prenatal assessment for possible elective termination. Cesarean section is the preferred
method of delivery when the diagnosis is known because it avoids trauma to the large myelomeningocele
and its neural elements. Fetal sac closure surgery has
been postulated to improve neurologic outcome. In one
study of 59 patients, intrauterine myelomeningocele repair reduced the incidence of hindbrain herniation (4%
versus 50%) and the incidence of shunt-dependent hydrocephalus (58% versus 92%). Functional level was
unchanged. Currently, multicenter prospective studies to
assess risk and benefits of intrauterine myelomeningocele repair continue.
Latex allergy, sensitivity, or anaphylaxis affects 20%
to 70% of patients with myelomeningocele. The etiology
is proposed as a result of multiple exposures to latex.
Therefore, routine precautions, including latex-free environments, should be taken to limit latex exposure.

Orthopaedic Considerations
The goal of orthopaedic treatment is to maximize function; ambulation is one main goal. Ambulation ability is
related to the level of the last intact motor root, which

American Academy of Orthopaedic Surgeons

often correlates with the last intact laminar arch (Table

1). Ambulation is generally predicted by age 2 years. A
child with myelomeningocele who is not a community
ambulator by the age of 6 years is not likely to reach
this goal.
Serial, well-documented orthopaedic and neurologic
examinations are key components in detecting and
avoiding progressive deterioration caused by neurologic
loss. Major areas of concern include loss of motor function and rapid progression of scoliosis. Common causes
of increasing spasticity and loss of function include hydrocephalus, shunt malfunction, tethered cord, syringomyelia, and posterior fossa compression from ArnoldChiari malformation.

Hip Deformities
Hip flexion contracture in children with myelomeningocele is caused by muscle imbalance (weak extensors/
strong flexors), spasticity as seen in tethered cord patients, or habitual sitting posture. During the first
2 years of life, hip flexion contracture decreases except
in high thoracic-level patients. Surgery is usually not indicated in patients younger than 2 years of age. In patients with high thoracic lesions, flexion contracture of
30 to 40 may be tolerated. Greater than 30 to 40 of
contracture will result in impairment of standing ability,
short stride length, and increased lumbar lordosis. In patients with low lumbar motor deficit, hip flexion contracture of more than 20 causes decreased walking ability as a result of anterior pelvic tilt, decreased velocity,
and increased demand of the upper extremities.
Treatment of hip subluxation in patients with myelomeningocele depends on their functional level and
the physical demand on the hip. Gait symmetry often
corresponds more to the absence of hip contractures
and less to the presence of hip dislocation. Therefore,
hip reduction is unnecessary in the low-demand hip.
Management of most hips focuses on contracture release and/or realignment osteotomies to prevent bracing
difficulties and spinal deformities except where significant asymmetry is present. In children with community

677

Orthopaedic Knowledge Update 8

Neuromuscular Disorders in Children

Table 1 | Motor Level and Functional Status: Myelomeningocele

Group

Lesion Level

Muscle Involvement

Function

Ambulation

Thoracic/high lumbar

No quadriceps function

Sitter
Possible household ambulator with
RGO

Low lumbar

Quadriceps and medial

hamstring function, no
gluteus medius, maximus

Household/community ambulator
with KAFO or AFO

Sacral

Quadriceps and gluteus

medius function
No gastrosoleus strength

Community ambulator with AFO,

UCBL, or none
Community ambulator with AFO,
UCBL, or none

Some degree until age 13 years with

HKAFO, RGO
95% to 99% wheelchair dependent as
adults
Require AFO and crutches, 79% community
ambulators as adults, wheelchair for long
distances; significant difference between
L3 and L4 level, medial hamstring
needed for community ambulation
94% retain walking ability as adults

High sacral

Low sacral

Good gastrocnemius-soleus
strength, normal gluteus
medius, maximus

Walk without support but require AFO, have

gluteus lurch and excessive pelvic
obliquity and rotation during gait
Walk without AFO, gait close to normal

RGO = reciprocating gait orthosis; UCBL = University of California/Berkley Lab (orthosis); KAFO = knee-ankle-foot orthosis; AFO = ankle-foot orthosis; HKAFO = hip-knee-ankle-foot orthosis

ambulation skills, the instability is addressed aggressively with a goal of achieving concentric reduction and
acetabular coverage for the high functioning child.

Knee Contractures and Torsional Deformities

Knee flexion contractures and extension contractures
are the most common knee deformity in children. Studies have shown that increased knee flexion during gait
has a high energy cost. Correction of knee flexion contracture of 20 or more is indicated. In some patients,
hamstring lengthening is insufficient and a posterior
capsulotomy of the knee may be required. Supracondylar extension osteotomy of the distal femur is recommended in older ambulatory children with fixed flexion
deformity of 20 or more after failed radical knee flexor
release.
Knee extension contractures are much less common
than flexion contractures. Treatment with serial casting
to achieve 90 of flexion is successful in most patients. If
nonsurgical treatment fails, V-Y lengthening of the
quadriceps is favored.
Moderate and severe torsional deformities impede
walking and place abnormal stress on the knee. A study
of 25 patients with lumbosacral myelomeningocele
showed that increased external tibial torsion increased
clinical valgus knee stress. A thigh-foot angle greater
than 20 significantly increased this stress and supramalleolar tibial/fibular osteotomy is indicated.
Internal tibial torsion is not an acquired deformity in
a child younger than 5 years of age with spina bifida. It is

678

present at birth and may lead to a significant in-toeing

gait. In patients age 5 years or younger, treatment is observation; orthotic devices may be used. If the condition
persists in a child older than 5 years of age, an external
supramalleolar tibial/fibular derotation osteotomy is indicated.

Foot Deformities
The goal of treatment is to achieve a supple, braceable,
and plantigrade foot. Approximately 30% of children
with myelomeningocele have a rigid clubfoot at birth.
Initial treatment includes serial casting, but correction is
rarely achieved by nonsurgical treatment. Comprehensive posteromedial lateral release is indicated at or before walking age.
Acquired equinus deformity occurs more frequently
in children with high lumbar and thoracic-level lesions.
Prevention is attempted by bracing and physical therapy. Heel cord resection is indicated to achieve a plantigrade and braceable foot. In patients with mild deformities who are high functioning, Achilles tendon resection
can be performed. With severe deformity, a radical posterior release including capsulotomy is required.
Coronal plane malalignments include valgus deformity of the ankle or the hindfoot and are commonly
seen in patients with L4-5 level deformity; this condition
can lead to difficulty with brace fitting and pressure
sores over the medial malleolus. Radiographs help determine the location of the deformity. Ankles in mild
valgus are treated with percutaneous screw hemi-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 57 Neuromuscular Disorders in Children

epiphysiodesis. In severe deformities and in older children, supramalleolar closing wedge osteotomy is indicated. For hindfoot valgus, a medial sliding osteotomy of
the calcaneus with displacement of 50% of the width of
the fragment is recommended.

rate of pseudarthrosis. Surgical complications include

wound infection (6% to 15%), fixation failure (15%),
and pseudarthrosis (7% to 28%). Patients have an improved sitting balance after surgery, but ambulation may
be more difficult.

Scoliosis

Cerebral Palsy

Scoliosis is related to the degree of motor paralysis. Paralytic spinal deformities are expected in 5% of patients
with sacral level function, 25% at L5, 60% at L4, 70% at
L3, 80% at L2, and more than 90% at L1 and higher.
Congenital lumbar kyphosis is a severe spinal deformity
occurring in 10% to 15% of patients. This deformity
does not respond to bracing. Indications are variable
among surgeons. Lumbar kyphosis may have deleterious
effects on pulmonary function (because of abdominal
compression and thoracic hypokyphosis) and sitting balance, and can progress to skin ulcerations because of the
prominent gibbus. Newer approaches to surgical management, including vertebral subtraction or decancellation procedures with instrumentation in young patients,
and kyphectomy in the older patient with instrumentation, are complex and carry significant risks for morbidity and mortality.
Physical examination of patients with spinal deformity includes assessment of pelvic obliquity, joint contractures, and leg length inequality as potential reversible causes of spinal deformity. Clinical findings of
tethered cord syndrome must be considered in the assessment. Treatment with an orthotic device is advised
for moderate scoliotic curves. Studies have shown the
greatest curve progression before the age of 15 years,
and average curve progression of 5 per year for curves
greater than 40. Intraspinal pathology such as tethering
of the spinal cord and syringomyelia contribute to curve
progression. Although most curves progress, bracing has
a beneficial temporary effect of delaying definitive spinal fusion until adult sitting height is achieved and also
supports the trunk in a functional position for those patients with imbalance and hypotonia. Patients who use
spinal orthotic devices may find independent gait reduced because of increased energy expenditure and balance disturbances. In wheelchair-dependent patients, increasing spinal deformity may compromise sitting
balance and lead to pressure sores. Prevention of sitting
imbalance is correlated with prevention of an unbalanced spine to curvature of less than 40 and pelvic
obliquity to less than 25.
Indications for spinal fusion and instrumentation in
scoliosis are progression of curve greater than 50, poor
sitting balance, and pulmonary compromise not controlled by bracing. In a study of 29 patients with severe
thoracolumbar and lumbar scoliosis, combined anterior
and posterior instrumentation gave the best correction
of the deformity and pelvic obliquity, and reduced the

Epidemiology

American Academy of Orthopaedic Surgeons

The cerebral palsies (CPs) are the most common causes

of physical disability in children. They can be defined as
a heterogeneous group of upper motor neuron impairment syndromes caused by chronic brain abnormalities.
Causes are multifactorial and include antenatal and
postnatal insults to the developing brain. The incidence
ranges from 1 to 3 per 1,000 live births. The rate of CP
in children weighing less than 1,500 g at birth is 70 times
higher compared with those weighing 2,500 g or more.
The incidence and prevalence of CP has remained constant over the past 40 years. Improved obstetrical care
has resulted in fewer neonatal insults, but improved
neonatal care has resulted in the decreased mortality of
more preterm infants. Although the cerebral lesion is
nonprogressive, musculoskeletal deformities and impairment can be progressive in the growing child.
Major types of CP are described as spastic, dystonic,
ataxic, and hypotonic. The spastic variety is the most
common, with mixed varieties now being increasingly
diagnosed. The clinical spectrum of CP ranges from very
minor impairment as seen in a patient with mild hemiplegia with minimal gait deviations to significant cognitive deficits and motor functional limitations as seen in
children with severe total body involvement. Patients
with severe CP often have comorbidities such as seizures, recurrent pneumonias, and difficulty eating, often
requiring feeding tubes.
Care of a child with CP involves a multidisciplinary approach with emphasis on early intervention therapies,
tone management, use of orthotic devices, and musculoskeletal surgery. Input from orthopaedists, physical and
occupational therapists, neurologists, neurosurgeons, pediatricians, and orthotists are required to maximize the function and activities of daily living of a child with CP. Treatment is focused during the early years of growth on tone
management, prevention of contractures, aids for walking,
and multilevel bony and soft-tissue surgery to improve
gait.
Literature on the treatment of adult CP patients is
limited. A recent survey showed that many adults with
CP continue to have spasticity and pain-related issues,
which contribute to the deteriorating ability to ambulate. Independent walking and other forms of supported
ambulation are often lost upon reaching adulthood.
Care of adult CP patients is limited by lack of adult
health and rehabilitation services.

679

Orthopaedic Knowledge Update 8

Neuromuscular Disorders in Children

Outcomes Assessment
With the recent focus on evidence-based medicine and
the development of best practice guidelines, greater attention has been focused on the assessment of outcomes
in multiple domains (technical, functional, patient and
family satisfaction, and cost). Recent studies of CP patients have had variable success using measurement instruments such as the Child Health Questionnaire and
Pediatric Outcomes Data Collection Instrument and in
attempting to assess outcomes of treatment. A single
study found considerable variability in the reliability
and validity of these different instruments for patients
with CP.

Bone Mineral Density

Multiple factors may adversely affect bone mineral density in children with CP. Impaired weight bearing and
short-term immobilization after orthopaedic surgery
contribute to low bone mineral density. Poor nutrition,
low calcium intake, and anticonvulsants play a negative
role as well. A recent controlled clinical trial showed
that pamidronate is a safe and effective agent to increase bone mineral density in children with CP.

Treatment Modalities
Botulinum Toxin
Botulinum toxin A is a protein polypeptide chain that
irreversibly binds to the cholinergic terminals at the
neuromuscular junction and effectively inhibits release
of acetylcholine from the synaptic vesicles. The toxin
can be injected locally into spastic muscles, causing a
rapid onset of weakness that may last for 3 to 6 months.
The most common locations benefiting from botulinum
toxin A injection are spastic ankle plantar flexors, posterior tibialis, hamstrings, hip adductors, and wrist flexors.
The benefits of repeated injections are still unclear.
Risks of repeat injections include lessening of the positive effect of the toxin as well as possible antibody formation. A 2000 study of repeat botulinum A toxin injections into calf muscles of patients with spastic equinus
showed similar effects after the first and second injections, with a decrease in the duration of response after
the third and fourth injections. The long-term benefits of
botulinum toxin have yet to be determined.
Baclofen
Baclofen is a -aminobutyric acid agonist that acts peripherally and centrally at the spinal cord level to impede the release of excitatory neurotransmitters that
cause spasticity. Large doses of oral baclofen are required to detect changes in spasticity, often with the unwanted side effect of sedation. Intrathecal baclofen allows for larger doses to reach the target tissues in the
spinal cord. A synchronized implantable infusion pump
is now available. Complications, which may be life

680

threatening, occur in approximately 20% of patients and

are primarily related to breakage of the catheter, catheter dislodgment, infection, and skin breakdown at the
insertion site. Development of scoliosis in patients with
baclofen pumps remains an important concern. Most
pumps are implanted in patients with severe spasticity
that interferes with function and positioning. A longterm review of this device showed benefits in terms of
decreased pain and spasm relief, better sleep, independence, and improved ease of care.

Physical Therapy
Physical therapy is the mainstay of nonsurgical treatment in children with motor dysfunction. Critical periods for intensive therapy include the early ambulatory
years and the immediate postsurgical rehabilitation
phase. Debate continues regarding the method, indications, and value of therapies offered to patients. Currently, precise objective and collaborative goal setting
has been emphasized. However, a recent randomized
controlled study failed to show any long-term benefit in
intensive, goal-directed therapies versus the traditional
forms of therapy.
Recent emphasis has been placed on the weakness
seen in spastic muscles. A retrospective analysis failed to
show a true relationship between the degree of spasticity and strength of the muscle. There was a trend toward
increased spasticity and less strength in distal muscle
groups compared with proximal muscle groups.

Orthotic Devices
Ankle-foot orthoses (AFOs) are commonly prescribed
to improve gait and stability, prevent deformity, and to
protect surgically-treated limbs. In patients with spastic
diplegia, solid AFOs have been shown to improve ankle
kinetics, whereas floor reaction AFOs are effective in
correcting crouch type gait secondary to ankle plantar
weakness. AFOs may be used initially to prevent equinus contractures, and are often used to prevent recurrence of deformity after botulinum toxin injection and
casting during treatment of dynamic or mild myostatic
deformities of the ankle plantar flexors. The AFOs must
extend to just below the knee and have a rigid ankle,
leaf spring, or hinged design to prevent equinus deformity. Supramalleolar designs are ineffective at preventing equinus. Prevention of equinus has been shown to
improve walking speed and stride length for most children. The efficacy of AFOs to help in overcoming functional limitations and preventing contractures is yet to
be established.

Alternative Therapies
Electrical Stimulation
Threshold electrical stimulation is based on lowintensity, long-duration electrical transcutaneous stimu-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

lation of select muscle groups. It is theorized that electrical stimulation promotes local blood flow, increasing
the size of atrophic muscles and perhaps their function.
Early studies show improved motor performance and
lower incidence of falls after 3 months of treatment.
However, two recent randomized trials failed to show
any benefit in motor or ambulatory function in children
with CP.
Hyperbaric Oxygen Therapy
Evidence for the effectiveness of hyperbaric oxygen
therapy treatment in children with CP is largely anecdotal. Known risks include oxygen toxicity (tremors or
seizures), barotrauma (tympanic membrane rupture),
myopia, and pneumothorax. A randomized multicenter
trial showed that hyperbaric oxygen therapy had no
greater effect on patients compared with breathing
slightly pressurized air.
Horseback Riding
The postulated effects of hippotherapy involve the
movement of the horses hind legs and pelvis to produce
a three-dimensional input to the patient. Ambulatory
pelvic motion is therefore mimicked while riding horseback. Potential benefits include improvement in overall
gross motor function, walking, running, and jumping. Using the Gross Motor Function Measure, a single study of
children with CP showed minimal improvement after 18
weeks of hippotherapy.

Neurosurgery
Selective dorsal rhizotomy (SDR) selectively sections
dorsal rootlets from L1 to S2. Heightened responses to
afferent impulses from muscle spindles are interrupted
with an end result of decreased spasticity. Concerns remain regarding the effects of SDR on muscle strength.
Original indications for SDR were limited to ambulatory patients with diplegia. Short-term results have
shown reduced spasticity and improved function when
SDR is followed by intensive physical therapy. There is
a lack of consensus in the literature regarding patient
indications, surgical techniques, surgical approach, and
determination of which rootlets to section. Complications, although rare, include bowel or bladder incontinence, dysesthesia, dural leaks, and risk of musculoskeletal deformities including hip subluxation and
spondylolysis, spondylolisthesis, and scoliosis. A metaanalysis of three randomized clinical trials showed functional improvement and reduced spasticity in carefully
selected CP patients after SDR and physical therapy
when compared with physical therapy alone. Other
studies have shown the need for additional orthopaedic
procedures in a significant percentage of patients who
have undergone SDR.

American Academy of Orthopaedic Surgeons

Chapter 57 Neuromuscular Disorders in Children

Gait Analysis
Gait analysis continues to evolve as a diagnostic and research tool. Using computers, cameras, reflective markers, in-floor force plates, dynamic electromyography, and
pedobarographs, movement in multiple planes can be
captured and analyzed. Preoperative assessment of ambulatory patients provides information across multiple
joints and lends direction to appropriate procedures
such as muscle tendon lengthening and rotational osteotomies.
A recent longitudinal study using gait analysis evaluated the natural progression of gait in children with CP.
Gait function was shown to deteriorate over a 4-year
span with respect to temporal/stride measures, passive
range of motion, and kinematic parameters. This deterioration was not seen in a similar cohort who had undergone orthopaedic intervention.
The application of quantitative gait analysis in clinical decision making is in evolution. Current controversies are focused on the reliability and variability of data
collection and interpretation between centers. Technologic advances in motion analysis and greater experience with the clinical applications of the data will increase the use of gait analysis.

Management of Specific Musculoskeletal

Complications
Spine
Spinal deformity is more commonly seen in severely affected CP patients with quadriplegia. The typical long
C-shaped thoracolumbar curve is resistant to bracing
and often progresses relentlessly. The spinal deformity
leads to difficulty with seating systems and unequal balance secondary to pelvic obliquity. Associated comorbidities are commonly present and the decision to pursue surgical correction must be done cautiously. Often,
these patients have significant comorbidities such as seizure disorder, poor nutritional status, and respiratory
compromise. Aggressive preoperative and postoperative
care is required. Current techniques use segmental fixation with Galveston fixation to the pelvis. Severe curves
may require an additional anterior procedure to achieve
adequate correction. A recent review of skeletally immature patients (with open triradiate cartilage) with
neuromuscular scoliosis following posterior spinal fusion to the pelvis revealed minimal evidence of subsequent progression of the deformity caused by anterior
overgrowth, also known as the crankshaft phenomenon.
Hip
Hip subluxation secondary to spastic hip adductors and
flexors may become evident in children between the
ages of 2 to 4 years and is typically seen in children with
severe spasticity. AP radiographs of the pelvis should be
obtained at regular intervals to monitor for subluxation.

681

Neuromuscular Disorders in Children

Typical findings include acetabular dysplasia (usually superior or posterosuperior), increased femoral anteversion, and increased femoral neck-shaft angle. Soft-tissue
release of the adductors shows a failure rate of 40%.
Osteotomies of the proximal femur (varus-producing) in
addition to acetabular osteotomies improve the coverage and stability of the developing hip and are typically
done after age 4 years. Combined procedures may carry
an increased risk of proximal femoral osteonecrosis.
Untreated hip dislocation in the older child or young
adult remains a difficult diagnostic and therapeutic challenge. Recent studies show that not all dislocated hips
are painful and not all painful hips are dislocated. Determining the degree of pain from a dislocated hip remains difficult. Long-term benefits of salvage operations
such as proximal femoral resection or hemiarthroplasty
are yet to be determined. For patients with unilateral
hip dislocation and the absence of spinal deformity, hip
fusion may be an alternative procedure, with a recent
review showing diminished pain and postural improvement following hip fusion in a series of older patients
with CP.
Knee
Spastic hamstring muscles increase knee flexion during
the stance phase of gait. Untreated knee flexion contractures and worsening crouch gait can lead to increased loads across the patellofemoral joint, resulting
in pain and late arthritis. Decreased dynamic range and
poor transition from stance to swing secondary to rectus
femoris spasticity are commonly seen. Treatment consists of medial hamstring lengthening with concomitant
posterior transfer of the rectus tendon. Knee flexion
contractures following hamstring lengthening can be addressed using serial stretch casting during the postoperative period. Lateral hamstring release is reserved for
the older patient with resistant knee flexion contractures or for patients with previous medial hamstring release. Risks of lateral hamstring lengthening include excessive weakness leading to knee hyperextension during
the stance phase. Extension osteotomy can be performed at the level of the distal femur for resistant contractures.
Ankle
Equinus is the most common deformity in CP and
causes adverse effects during standing and gait. For children younger than 3 years of age, developmental therapy and orthotic devices are used. From ages 3 to 6
years, a combination treatment with serial casting, botulinum toxin, and orthotic devices is recommended. Definitive muscle-tendon surgery is preferably delayed until at least age 6 years for persistent equinus deformity
to minimize risk of recurrence. Surgical options vary depending on the degree of deformity and contracture. Selective gastrocnemius-soleus complex recession is per-

682

Orthopaedic Knowledge Update 8

formed more commonly. Nonselective Achilles tendon
lengthening is reserved for more severe myostatic deformities. There is greater risk of overlengthening with
Achilles tendon lengthening, which can result in disruption of the ankle plantar flexion-knee extension couple,
leading to a crouch gait pattern.
Foot
Planovalgus and hallux valgus deformities are common
in patients with CP. In nonambulatory patients, a shoeable plantigrade foot is the goal of treatment. In ambulatory patients, treatment is directed toward correction
of skeletal lever arm deficiency to maximize ankle plantar flexion function and to facilitate bracing. Surgical
treatment involves posterior calf muscle lengthening
and skeletal realignment by calcaneal lengthening
and/or subtalar arthrodesis. Hallux valgus deformities
often require metatarsophalangeal fusion in patients
with CP to prevent recurrence.
Equinovarus foot deformity is seen in the setting of
a spastic posterior tibialis tendon. For flexible deformities, surgical considerations include posterior calf
lengthening, intramuscular or Z-lengthening of the posterior tibial tendon, or split posterior or anterior tibial
tendon transfers. A recent long-term follow-up survey
showed a failure rate in 75% of patients with low-level
ambulatory diplegia and quadriplegia undergoing these
procedures before age 8 years.
Transverse Plane
In some ambulatory patients with CP, increased femoral
anteversion fails to resolve with growth, resulting in lever deficiency at the level of the hip. Assessing the degree of pathologic femoral anteversion requires some
combination of clinical examination, two-dimensional
CT, and gait analysis. Compensatory actions during the
gait cycle include internal rotation of the hip and external rotation of the pelvis to restore the lever arm. Correction of these deviations can be achieved following
correction of the increased anteversion by derotational
osteotomy of the femur. This procedure can be performed proximally by an intertrochanteric or subtrochanteric osteotomy, or distally above the metaphyseal
flare. A prospective study comparing both methods
showed excellent clinical results with either technique. If
this condition is not corrected by the teenage years, patients may develop compensatory increased external tibial torsion. In such cases, the transverse plane malalignment may need to be corrected at both the femoral and
tibial levels. Tibial rotation osteotomy is best performed
at the supramalleolar level to minimize the risk of compartment syndrome and neurologic injury. Simultaneous
procedures will necessitate an aggressive postoperative
physical therapy regimen to enable the patient to return
to their preoperative functional level.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 57 Neuromuscular Disorders in Children

Figure 1 Patient with Duchenne muscular dystrophy show the Gowers maneuver.
A, The prone position. B, The bear position. C, Moving the hands up the thighs to
help upright the trunk and augment knee
extension. D, The upright position. (Reproduced from Sussman M: Duchenne muscular dystrophy. J Am Acad Orthop Surg
2002;10:138-151.)

Muscular Dystrophy
The mutation that causes Duchenne muscular dystrophy
is a point deletion of a segment of the dystrophin gene,
located at Xp21 on the X chromosome. This produces a
frame shift resulting in all messenger RNA distal to the
deletion coding for a nonsense protein. Thus, formation
of the dystrophin protein is absent. Dystrophin is critical
to the stability of the cell membrane. Becker muscular
dystrophy resembles Duchenne muscular dystrophy in
that it is also inherited in an X-linked recessive manner.
A less significant mutation in the dystrophin gene does
not result in a frame shift, and therefore allows production of smaller amounts of an inferior dystrophin. The
onset of weakness in patients with Becker muscular dystrophy is delayed usually until the second decade, and
life expectancy is longer. This difference in the mutation
results in a significant difference in disease severity such
that the amount and quality of dystrophin produces a
spectrum of disease.
Duchenne muscular dystrophy can be diagnosed using DNA analysis in at least two thirds of patients.

American Academy of Orthopaedic Surgeons

Duchenne muscular dystrophy should be suspected in

boys whose laboratory values show extremely high levels of serum creatine phosphokinase (measuring 10 to
200 times normal), which leaks across the defective cell
membrane into the serum. The diagnosis is definitively
made by absent dystrophin staining of muscle biopsy
specimens in patients who do not have discoverable mutations.
Because of the relatively common presentation of
the clumsy child, diagnosis of Duchenne muscular dystrophy is often missed for up to 2 years following presentation. Early diagnosis is extremely important to
families because of the possibility of genetic counseling
for future offspring. Clinical features seen in young boys
with Duchenne muscular dystrophy are a waddling gait,
toe walking, falling, and difficulty climbing stairs. Walking is usually delayed until age 18 to 24 months. Unlike
in patients with spinal muscular atrophy, deep tendon
reflexes are present. Gowers sign is positive, as the
child uses the hands to walk up the legs when rising
from the floor because of proximal muscle weakness

683

Neuromuscular Disorders in Children

Figure 2 Physical findings in patients with Duchenne muscular dystrophy. A 5-year-old boy with Duchenne muscular dystrophy and marked
pseudohypertrophy of the calves. (Reproduced from Sussman M: Duchenne muscular dystrophy. J Am Acad
Orthop Surg 2002;10:138-151.)

(Figure 1). Documentation of the Gowers sign should

be mandatory in all boys evaluated for weakness or gait
disturbances. Calf pseudohypertrophy develops as muscle is replaced by fibroadipose tissue (Figure 2). Gait is
wide-based, and lumbar lordosis is increased.
Research is underway in gene therapy using adenovirus vectors in animal models to import the normal
dystrophin gene. The production of an immune response
to introduction of the gene and reaction to the viral vectors are complications of this therapy. Stem cell and myoblast transplant are also under investigation.
Corticosteroids have been shown to acutely improve
muscular strength, and to slow progressive weakening in
intermediate follow-up. Ambulation may be prolonged
and the deterioration of pulmonary function slowed
with corticosteroid therapy. The preferred steroid and
mode of administration is debated. Although positive
effects on skeletal muscle have been shown, side effects
such as obesity and osteopenia are of concern.
Advances in pulmonary care have resulted in longer
life spans. A recent study revealed that patients who use
nighttime ventilation have up to 50% survival at age
25 years.

Orthopaedic Manifestations
Hip abductor and hamstring releases, Achilles tendon
lengthening or release with posterior tibialis tenotomy
or transfer may prolong ambulation in those patients
whose ability to walk is deteriorating. Timing of the surgery is very important because those children who have
already become nonambulatory will not regain the ability to walk after surgery. In a recent study, 42% of boys
who underwent multilevel releases with posterior tibia-

684

Orthopaedic Knowledge Update 8

lis tendon transfer to prolong ambulation would choose
to have the surgery again. Willingness to use kneeankle-foot orthoses and compliance with postoperative
physical therapy must be evaluated preoperatively. Isolated foot surgery consisting of posterior tibialis tendon
transfer (combined with Achilles tendon lengthening
and toe flexor tenotomy) corrected equinovarus, allowed patients to continue to wear shoes, and had high
patient and family satisfaction.
Scoliosis is a major problem in the nonambulatory
patient with Duchenne muscular dystrophy. Shortly after transitioning to wheelchair use, up to 95% of patients develop scoliosis, which is progressive in almost
all patients. Bracing is contraindicated because it does
not prevent progression, and may delay surgery until a
time when the pulmonary status of the child precludes
safe surgery. Forced vital capacity has been found to decrease 4% yearly, and an additional 4% for each 10 of
thoracic deformity. Treatment is early posterior spinal
fusion, recommended after a curve reaches 20 to 30 of
magnitude in boys without significant cardiomyopathy,
and whose preoperative pulmonary function tests reveal
forced vital capacity of at least 35% of normal levels.
Most authors recommend fusion with segmental instrumentation from the upper thoracic spine to the sacrum
or pelvis, although there are reports of maintenance of
correction following fusion to L5 in patients with minimal (< 10) preoperative pelvic obliquity, especially with
the use of pedicle screw fixation distally. Outcomes
studies suggest improved quality of life in boys who
have undergone spinal fusion compared with their peers
who have not had surgery.
Fractures are more prevalent in patients with Duchenne muscular dystrophy, occurring in up to 21% of
affected boys. Frequent falls, combined with osteopenia,
lead to the increased frequency. Loss of mobility occurs
in up to 20% of ambulatory boys with Duchenne muscular dystrophy as a result of fractures.

Spinal Muscular Atrophy

Spinal muscular atrophy represents a group of diseases
characterized by progressive symmetric muscular weakness caused by the loss of anterior horn cells. Spinal
muscular atrophy is inherited as an autosomal recessive
trait. A mutation in the survival motor neuron gene on
chromosome 5q can be detected in 94% of patients. The
survival motor neuron gene is critical to RNA metabolism and is a mediator of apoptosis (programmed cell
death). Spinal muscular atrophy can be diagnosed using
DNA analysis or by muscle biopsy. Muscle enzyme levels are usually normal. Prenatal diagnosis is possible.
Subtypes of spinal muscular atrophy are distinguished by the age of onset and clinical severity. Type I,
known as Werdnig-Hoffmann or acute infantile spinal
muscular atrophy, is a lethal condition presenting in pa-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

tients 6 months of age or younger. The babies exhibit little or no movement of their extremities, with characteristic tongue fasciculations and absent deep tendon
reflexes. Most die of respiratory infections before the
age of 2 years. Longer survival is possible with tracheostomy or noninvasive ventilation. Orthopaedic treatment is limited to splinting of fractures that may occur
at birth.
Patients with type II or chronic infantile spinal muscular atrophy, present between 6 and 12 months of age.
These babies have muscle weakness that is more advanced in the legs than in the arms. They achieve head
control, and most can sit, but they cannot walk. Survival
into adulthood is common.
Type III spinal muscular atrophy is known as
Kugelberg-Welander syndrome and has an age of onset
between 2 and 15 years, when proximal muscle weakness becomes noticeable. Patients ambulate as children,
but become wheelchair users in adulthood.

Orthopaedic Manifestations
Currently there is no effective medical treatment for
spinal muscular atrophy. Orthopaedic treatment is
needed for scoliosis, which is widespread in patients
with type II and III. Soft orthotic devices can delay but
not prevent surgery in children younger than 10 years of
age. Posterior spinal fusion is recommended for most
progressive curves, with fusion to the pelvis in nonambulatory patients.
Hip dislocation and subluxation occur in 62% of
hips in patients with type II spinal muscular atrophy,
and less frequently in type III patients. Although open
reduction of the hip can be performed, even with concomitant pelvic osteotomy, redislocation frequently occurs because of the inherent muscle weakness. Longterm studies show that nearly all hip dislocations in
patients with spinal muscular atrophy remain painless;
surgery should be avoided in nonambulatory patients.

Friedreichs Ataxia
Friedreichs ataxia is characterized by involvement of
the cerebellum and the spinal cord pathways. Clinically,
patients present with the classic triad of ataxia, loss of
deep tendon reflexes, and an extensor Babinski response. Associated medical conditions include hypertrophic cardiomyopathy, and in some patients, diabetes
mellitus. The average age of onset is 12 years, and death
occurs on average 25 years after diagnosis because of
cardiac deterioration.
Friedreichs ataxia is inherited in an autosomal recessive fashion. The disease is caused by the presence of
an expanded guanine-adenosine-adenosine trinucleotide
repeat in both copies of the frataxin gene on chromosome 9. The normal number of repeats is less than 33,
whereas patients with Friedreichs ataxia have 66 to

American Academy of Orthopaedic Surgeons

Chapter 57 Neuromuscular Disorders in Children

1,500 abnormal guanine-adenosine-adenosine repeats.
The number of added repeats may correlate with the severity of disease or age of onset. Frataxin is a mitochondrial protein, involved in iron metabolism and oxidative
stress. Knowledge of the role of frataxin has led to the
initiation of treatment with antioxidants such as Coenzyme Q, which has been shown in preliminary studies to
decrease the rate of cardiac deterioration but not the
ataxia.

Orthopaedic Manifestations
Patients with Friedreichs ataxia often present for the
evaluation of frequently falling and have cavus or
cavovarus feet. Surgery to prolong ambulation is performed in selected patients when the position of the
foot exacerbates the instability during gait. Scoliosis occurs in more than 80% of patients. Curve types vary between idiopathic-looking curves to long, sweeping thoracolumbar curves. Increased thoracic kyphosis is
common. Progression is linked to early age at diagnosis
and to curve magnitude. Bracing is ineffective at preventing progression, and may interfere with mobility.
Curves that stabilize at less than 40 at skeletal maturity
rarely progress. Surgery is indicated for curves of 60 or
greater. Posterior spinal fusion from the upper thoracic
to the lower lumbar spine is effective.

Charcot-Marie-Tooth Disease
Charcot-Marie-Tooth disease is a clinically and genetically heterogeneous group of hereditary motor sensorineuropathies. Traditionally, Charcot-Marie-Tooth disease has been divided into demyelinating forms that
slow nerve conduction velocity (Charcot-Marie-Tooth
types 1, 3, and 4), and axonal forms that decrease the
compound muscle action potential (Charcot-MarieTooth type 2). Overlap between these two forms exists.
To date, 10 genes causing various forms of CharcotMarie-Tooth disease have been identified. These genes
are expressed either by Schwann cells and/or accompanying neurons. The most common mutation involves
overexpression of the peripheral myelin protein 22 gene
on chromosome 17, producing autosomal dominant
Charcot-Marie-Tooth disease type 1A. Other genes
commonly disturbed include the myelin protein zero
gene, and connexin 32, which is present in the x-linked
Charcot-Marie-Tooth disease.
In most patients the disease manifests during the
second decade of life. Cavovarus foot deformity is frequently the presenting symptom. Physical examination
is notable for lower limb areflexia, calf atrophy, an increased longitudinal arch, and clawing of the toes.

Orthopaedic Manifestations
The first line of treatment of mild cavus and cavovarus
is the use of orthotic devices. AFOs can improve gait in

685

Orthopaedic Knowledge Update 8

Neuromuscular Disorders in Children

patients with flexible deformities. Patients with CharcotMarie-Tooth disease frequently have significant foot
drop during swing phase, and this is improved by use of
AFOs.
Surgical treatment of the cavovarus foot deformity is
based on the flexibility of the deformity. In patients with
mild flexible hindfoot varus, plantar fascia release and
first metatarsal dorsal closing wedge osteotomy, with or
without transfer of the peroneus longus to peroneus
brevis, can improve the deformity. The posterior tibialis
tendon can be transferred to the dorsum of the midfoot
to improve swing phase foot drop. When the hindfoot is
rigid, calcaneal osteotomy is necessary. After correction
of the hindfoot deformity is achieved, metatarsal osteotomies to align the forefoot may be performed. Alternatively, midfoot dorsal closing wedge osteotomy improves
cavovarus by correcting midfoot and forefoot equinus.
In severe cases, triple arthrodesis is necessary to obtain
and maintain correction. Clawing of the toes may be improved with transfer of the toe extensors to the metatarsal necks. Although initial results following cavovarus
surgery are usually satisfactory, progression of the neurologic disease often leads to recurrence of deformity.
Hip dysplasia is also associated with Charcot-MarieTooth disease. Ambulatory patients may report symptoms during adolescence. Symptomatic acetabular dysplasia can be treated by pelvic osteotomies. Patients
with hereditary motor sensorineuropathies are more
prone to sciatic nerve palsy after pelvic surgery.
Scoliosis occurs more frequently in patients with
Charcot-Marie-Tooth disease. Left thoracic and kyphotic curves distinguish the radiographic appearance of
many of these curves from idiopathic scoliosis. Although
bracing rarely is effective, surgical correction can be performed without an increase in complications.

Annotated Bibliography
Myelomeningocele
Gabrieli AP, Vankoski S, Dias L, et al: Gait analysis in
low lumbar myelomeningocele patients with unilateral
hip dislocation or subluxation. J Pediatr Orthop 2003;23:
330-334.
Gait symmetry using motion analysis of 20 patients with
low lumbar myelomeningocele with unilateral hip dislocation
or subluxation and no scoliosis corresponded to the absence of
hip contractures and had no relation to the presence of hip
dislocation. The authors concluded that reduction of the hip is
unnecessary.

Farmer DL, von Koch CS, Peacock WJ, et al: In utero repair of myelomeningocele: Experimental pathophysiology, initial clinical experience, and outcomes. Arch Surg
2003;138:872-878.
Fetoscopic repair of neural tube defects before 22 weeks
gestation is physiologically and technically feasible. Surgical

686

repair of 13 patients showed one third were spared the need

for a shunt at 1 year of age, but improvement in neurologic
function was unclear.

Nolden MT, Sarwark JF, Vora A, Grayhack JJ: A

kyphectomy technique with reduced perioperative morbidity for myelomeningocele kyphosis. Spine 2002;27:
1807-1813.
Evaluation of 11 patients with myelomeningocele and
lumbar kyphosis showed subtraction vertebrectomy with posterior instrumentation to be a safe and efficacious technique
for correction and stabilization of myelomeningocele kyphosis
in young patients.

Scoliosis
Berven S, Bradford DS: Neuromuscular scoliosis:
Causes of deformity and principles for evaluation and
management. Semin Neurol 2002;22:167-178.
In a review of the literature, outcomes in patients with
neuromuscular deformity who were treated with a combined
anterior and posterior fusion showed improvement of the major curve of up to 60%, more effective correction of pelvic
obliquity, and a decreased pseudarthrosis rate.

Trivedi J, Thomson JD, Slakey JB, Banta JV, Jones PW:

Clinical and radiographic predictors of scoliosis in myelomeningocele. J Bone Joint Surg Am 2002;84-A:13891394.
In a retrospective review of patients with myelomeningocele over a 5-year period with an average follow-up of 9.4
years, curves of less than 20 often resolved. Clinical motor
level, ambulatory status, and last intact laminar arch were predictive factors for development of scoliosis.

Cerebral Palsy
Anderson C, Mattsson E: Adults with cerebral palsy: A
survey describing problems, needs, and resources, with
special emphasis on locomotion. Dev Med Child Neurol
2001;43:76-82.
This article presents a report of a survey of adult CP patients discussing natural history and progression. Of those
polled, 77% had difficulty with spasticity, 18% had full disability pension, 35% reported decreased walking ability, and 18%
had pain every day.

Bell KJ, Ounpuu S, DeLuca PA, Romness MJ: Natural

progression of gait in children with cerebral palsy.
J Pediatr Orthop 2002;22:677-682.
Using clinical measures and gait analysis, gait function in
patients with CP decreased longitudinally compared with the
group who had orthopaedic intervention.

Bower E, Michell D, Burnett M, Campbell M, McLellan

D: Randomized control trial of physiotherapy in 56 children with cerebral palsy followed for 18 months. Dev
Med Child Neurol 2001;43:4-15.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Patients in aim and goal directed therapy showed no advantage in acquisition of gross motor function or performance
compared with patients undergoing more traditional forms of
therapy.

Chapter 57 Neuromuscular Disorders in Children

No relationship between strength and spasticity in patients
with CP was identified.

Flynn JM, Miller F: Management of hip disorders in patients with cerebral palsy. J Am Acad Orthop Surg 2002;
10:198-209.

Saraph V, Zwick EB, Zwick G, Steinwender C, Steinwender G, Linhart W: Multilevel surgery in spastic diplegia: Evaluation by physical examination and gait
analysis in 25 children. J Pediatr Orthop 2002;22:150157.

This review article discusses surgical options for hip disorders in CP. Childhood and adult hip pathology with treatment
options are discussed.

Clinical, kinematic, and kinetic improvements are shown

following multilevel bony and soft-tissue procedures in ambulatory CP patients.

Graham K, Selber P: Musculoskeletal aspects of cerebral palsy. J Bone Joint Surg Br 2003;85:157-166.

Smucker JD, Miller F: Crankshaft effect after posterior

spinal fusion and unit rod instrumentation in children
with cerebral palsy. J Pediatr Orthop 2001;21:108-112.

A comprehensive review of the role of the orthopaedic

surgeon in the management of CP is detailed. The etiology,
natural history, pathophysiology, and surgical strategies are
discussed.

Henderson RC, Lark RK, Kecskemethy HH, Miller F,

Harcke HT, Bachrach SJ: Bisphosphonates to treat osteopenia in children with quadriplegic cerebral palsy: A
randomized, placebo-controlled clinical trial. J Pediatr
2002;141:644-651.
This small clinical trial demonstrated increased bone mineral density following administration of bisphosphonates in
nonambulatory children with CP.

McLaughlin J, Bjornson K, Temkin N, et al: Selective

dorsal rhizotomy: Meta-analysis of three randomized
controlled trials. Dev Med Child Neurol 2002;44:17-25.
A consistent reduction in spasticity in seen following selective dorsal rhizotomy (SDR). With respect to functional outcome, the difference between SDR alone and SDR with physical therapy was minimal.

Pirpiris M, Trivett A, Baker J, Rodda G, Nattrass R, Graham H: Femoral derotation osteotomy in spastic diplegia. J Bone Joint Surg Br 2003;85:265-272.
Proximal and distal femoral derotation osteotomy was
found to be equally effective for correction of increased femoral anteversion and intoeing gait. Distal osteotomy may provide faster rehabilitation and a decrease in surgical complications.

Rosenbaum PL, Walter SD, Hanna SE, et al: Prognosis

for gross motor function in cerebral palsy: Creation of
motor development curves. JAMA 2002;288:1357-1363.
Evidence-based prognostication concerning gross motor
progress in children with CP is discussed. Patterns of gross
motor development are illustrated relating to the severity of
motor impairment.

Ross S, Engsberg J: Relation between spasticity and

strength in individuals with spastic diplegic cerebral
palsy. Dev Med Child Neurol 2002;44:148-157.

American Academy of Orthopaedic Surgeons

In skeletally immature children with cerebral palsy, posterior spinal fusion alone with unit rod instrumentation was successful in maintaining correction of neuromuscular scoliosis.
Crankshaft deformity was not identified in this series of 50 patients treated with isolated posterior procedures.

Sterba J, Rogers B, France A, Vokes D: Horseback

riding in children with cerebral palsy: Effect on gross
motor function. Dev Med Child Neurol 2002;44:301-308.
A small clinical trial evaluated the effect of recreational
horseback riding therapy on Gross Motor Function Measure
scores. An increase of 7.6% in these scores was seen after
18 weeks of horseback riding therapy and returned to control
levels 6 weeks following therapy. A minimal increase in walking, running, and jumping ability was seen after horseback
riding therapy.

Vitale MG, Levy DE, Moskowitz AJ, et al: Capturing

quality of life in pediatric orthopaedics: Two recent
measures compared. J Pediatr Orthop 2001;21:629-635.
Cross-sectional application of Child Health Questionnaire
and Pediatric Outcomes Data Collection Instrument among
varying diagnoses. The instruments were able to distinguish
physical from psychosocial health problems.

Muscular Dystrophy
Bentley G, Haddad F, Bull TM, et al: Treatment of scoliosis in muscular dystrophy using modified Luque and
Harrington-Luque instrumentation. J Bone Joint Surg
Br 2001;83:22-28.
Sixty-four patients with Duchenne muscular dystrophy
and 33 patients with spinal muscular atrophy had posterior
spinal fusion with sublaminar wiring. Fusion to the pelvis is
recommended.

Chan KG, Galasko CS, Delaney C: Hip subluxation and

dislocation in Duchenne muscular dystrophy. J Pediatr
Orthop B 2001;10:219-225.
Nineteen of 54 boys had subluxated or dislocated hips,
usually related to pelvic tilt.

687

Orthopaedic Knowledge Update 8

Neuromuscular Disorders in Children

Connolly AM, Schierbecker J, Renna R, et al: High dose
weekly oral prednisone improves strength in boys with
Duchenne muscular dystrophy. Neuromuscul Disord
2002;12:917-925.

Freidreichs Ataxia

Improvement in muscle strength was seen up to 2 years

following twice weekly prednisone therapy with fewer side effects than daily administration.

Molecular genetic research has led to the discovery of mutations in the frataxin gene. The link between genotype and
phenotype has been studied. Treatment is under investigation
based on frataxins role in cellular responses to oxidative
stress.

Eagle M, Baudouin SV, Chandler C, et al: Survival in

Duchenne muscular dystrophy: Improvements in life expectancy since 1967 and the impact of home nocturnal
ventilation. Neuromuscul Disord 2002;12:926-929.
Using nighttime noninvasive ventilation, survival averaged
50% at 25 years of age in boys without severe cardiomyopathy.

McDonald DG, Kinali M, Gallagher AC, et al: Fracture

prevalence in Duchenne muscular dystrophy. Dev Med
Child Neurol 2002;44:695-698.
Twenty-one percent of boys with Duchenne muscular dystrophy sustained fractures; 48% of the fractures occurred in
ambulatory patients as a result of falls, and 20% of boys
ceased walking because of the fracture.

Scher DM, Mubarak SJ: Surgical prevention of foot deformity in patients with Duchenne muscular dystrophy.
J Pediatr Orthop 2002;22:384-391.
Plantigrade shoeable feet were maintained after tibialis
posterior transfer, Achilles tendon lengthening, and toe flexor
releases.

Sengupta DK, Mehdian SH, McConnell JR, et al: Pelvic

or lumbar fixation for the surgical management of scoliosis in Duchenne muscular dystrophy. Spine 2002;27:
2072-2079.
Increasing pelvic tilt was not seen at 3.5-year follow-up after posterior spinal fusion with pedicle screw fixation to L5 for
mild curves.

Sussman M: Duchenne muscular dystrophy. J Am Acad

Orthop Surg 2002;10:138-151.
A detailed review of current treatment and research on
Duchenne muscular dystrophy is presented.

Spinal Muscular Atrophy

Bach JR, Baird JS, Plosky D, et al: Spinal muscular atrophy type 1: Management and outcomes. Pediatr
Pulmonol 2002;34:16-22.
Although tracheostomy prolonged survival in patients
with type 1 spinal muscular atrophy, noninvasive ventilation
preserved speech.

Sporer SM, Smith BG: Hip dislocation in patients with

spinal muscular atrophy. J Pediatr Orthop 2003;23:10-14.
Thirty-nine of 41 patients with spinal muscular atrophy
and hip dislocations were pain free at 18-year follow-up.

688

Lynch DR, Farmer JM, Balcer LJ, et al: Friedreich ataxia: Effects of genetic understanding on clinical evaluation and therapy. Arch Neurol 2002;59:743-747.

Charcot-Marie-Tooth Disease
Berger P, Young P, Suter U, et al: Molecular cell biology
of Charcot-Marie-Tooth disease. Neurogenetics 2002;4:115.
The specific mutations in 10 genes linked with subtypes of
Charcot-Marie-Tooth disease are discussed.

Sammarco GJ, Taylor R: Cavovarus foot treated with

combined calcaneus and metatarsal osteotomies. Foot
Ankle Int 2001;22:19-30.
Fifteen feet affected by Charcot-Marie-Tooth disease underwent combined calcaneal and metatarsal osteotomies with
improvements in foot scores and radiographs.

Classic Bibliography
Beaty JH, Canale ST: Orthopaedic aspects of myelomeningocele. J Bone Joint Surg Am 1990;72:626-630.
Botto LD, Moore CA, Khoury MJ, Erickson JD: Medical progress: Neural tube defects. N Engl J Med 1999;
341:1509-1519.
Brinker MR, Rosenfield SR, Feiwell E, Granger SP,
Mitchell DC, Rice JC: Myelomeningocele at the sacral
level: Long term outcomes in adults. J Bone Joint Surg
Am 1994;76:1293-1300.
Daher YH, Lonstein JE, Winter RB, et al: Spinal deformities in patients with Charcot-Marie-Tooth disease: A
review of 12 patients. Clin Orthop 1986;202:219-222.
Damiano DL, Abel MF: Functional outcomes of
strength training in spastic cerebral palsy. Arch Phys
Med Rehabil 1998;79:119-125.
Delp SL, Zajac FE: Force and moment generating capacity of the lower extremity muscles before and after
tendon lengthening. Clin Orthop 1992;284:247-259.
DeLuca PA, Davis RB, Ounpuu S, Rose S, Sirkin R: Alterations in surgical decision making in patients with cerebral palsy based on three-dimensional gait analysis.
J Pediatr Orthop 1997;17:608-614.
Drennan JC: Current concepts in myelomeningocele.
Instr Course Lect 1999;48:543-550.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Galasko CS, Delaney C, Morris P: Spinal stabilization in
Duchenne muscular dystrophy. J Bone Joint Surg Br
1992;74:210-214.
Greene WB: Treatment of hip and knee problems in
myelomeningocele. Instr Course Lect 1999;48:563-574.
Hensinger RN, MacEwen GD: Spinal deformity associated with heritable neurological conditions: Spinal muscular atrophy, Friedreichs ataxia, familial dysautonomia, and Charcot-Marie-Tooth disease. J Bone Joint
Surg Am 1976;58:13-24.
Hoffman EP, Kunkel LM: Dystrophin abnormalities in
Duchenne/Becker muscular dystrophy. Neuron 1989;2:
1019-1029.
Koman LA, Mooney JF III, Smith BP, Walker F, Leon
JM: Botulinum toxin type A neuromuscular blockade in
the treatment of lower extremity spasticity in cerebral
palsy: A randomized, doubled-blind, placebo-controlled
trial: BOTOX Study Group. J Pediatr Orthop 2000;20:
108-115.
Kunkel LM: Analysis of deletions in DNA in patients
with Becker and Duchenne muscular dystrophy. Nature
1986;322:73-77.
Labelle H, Tohme S, Duhaime M, et al: Natural history
of scoliosis in Friedreichs ataxia. J Bone Joint Surg Am
1986;68:564-572.
Lieber RL: Skeletal muscle adaptability: I. Review of
basic properties. Dev Med Child Neurol 1986;28:390-396.
Lim R, Dias L, Vankoski S, Moore C, Marinello M, Sarwark J: Valgus knee stress in lumbosacral meylomeningocele: A gait analysis evaluation. J Pediatr Orthop
1998;18:428-433.
Mazar J, Menelaus MB, Dickens DR, Doig WG: Efficacy
of surgical management for scoliosis in myelomeningocele: Correction of deformity and alteration of functional status. J Pediatr Orthop 1986;6:568-575.

Chapter 57 Neuromuscular Disorders in Children

Part I. The effect of rectus femoris transfer location on
knee motion. J Pediatr Orthop 1993;13:325-330.
Ounpuu S, Muik E, Davis RB III, Gage JR, Deluca PL:
Rectus femoris surgery in children with cerebral palsy.
Part II: A comparison between the effect of transfer and
release of the distal rectus femoris on knee motion.
J Pediatr Orthop 1993;13:331-335.
Perry J: Determinants of muscle function in the spastic
lower extremity. Clin Orthop 1993;288:10-26.
Read L, Galasko CS: Delay in diagnosing Duchenne
muscular dystrophy in orthopaedic clinics. J Bone Joint
Surg Br 1986;68:481-482.
Reimers J: Functional changes in the antagonists after
lengthening of the agonists in cerebral palsy. Clin
Orthop 1990;253:30-37.
Rose SA, Deluca PA, Davis RB III, Ounpuu S, Gage
JR: Kinematic and kinetic evaluation of the ankle after
lengthening of the gastrocnemius fascia in children with
cerebral palsy. J Pediatr Orthop 1993;13:727-732.
Skaggs DL, Rethlefsen SA, Kay RM, Dennis SW, Reynolds RA, Tolo VT: Variability in gait analysis interpretation. J Pediatr Orthop 2000;20:759-764.
Smith AD, Koreska J, Moseley CF: Progression of scoliosis in Duchenne muscular dystrophy. J Bone Joint Surg
Am 1989;71:1066-1074.
Strauss DJ, Shavelle RM, Anderson TW: Life expectancy of children with cerebral palsy. Pediatr Neurol
1998;18:143-149.
Sutherland DH, Davids JR: Common gait abnormalities
of the knee in cerebral palsy. Clin Orthop 1993;288:139147.

Muller EB, Nordwall A: Brace treatment of scoliosis in

children with myelomeningocele. Spine 1994;19:151-155.

Talipan N, Bruner JP, Hernandez-Schulman M, et al: Effects of intrauterine myelomeningocele repair on central
nervous system structure and function. Pediatr
Neurosurg 1999;31:183-188.

Mubarak SJ, Morin WD, Leach J: Spinal fusion in Duchenne muscular dystrophy: Fixation and fusion to the sacropelvis? J Pediatr Orthop 1993;13:752-757.

Walker JL, Nelson KR, Heavilon JA, et al: Hip abnormalities in children with Charcot-Marie-Tooth disease.
J Pediatr Orthop 1994;14:54-59.

Mubarak SJ, Wenger DR, Valencia F: One-stage correction of the spastic dislocated hip: Use of pericapsular acetabuloplasty to improve coverage. J Bone Joint Surg
Am 1992;74:1347-1357.

Wetmore RS, Drennan JC: Long-term results of triple

arthrodesis in Charcot-Marie-Tooth disease. J Bone
Joint Surg Am 1989;71:417-422.

Ounpuu S, Muik E, Davis RB III, Gage JR, Deluca PL:

Rectus femoris surgery in children with cerebral palsy:

American Academy of Orthopaedic Surgeons

Williams JJ, Graham GP, Dunne KB, Menelaus MB:

Late knee problems in myelomeningocele. J Pediatr
Orthop 1993;13:701-703.

689

Chapter

58

Pediatric Hematology
Michael T. Busch, MD
Tim Schrader, MD
Gary M. Lourie, MD

Hemophilia
Hemophilia is a group of genetic bleeding disorders that
affects about 20,000 Americans. Hemophilia is inherited
in about two thirds of patients; the condition results
from spontaneous genetic mutations in one third of patients. The most common deficiencies involve factor
VIII (classic hemophilia or hemophilia A) and factor IX
(Christmas disease or hemophilia B). Because the genes
for these deficiencies are carried on the X chromosome,
they are usually recessive disorders affecting males. Females, who are genetic carriers of the disease, can have
mild deficiencies and may become symptomatic after
trauma or surgery. von Willebrands disease is a deficiency or abnormality of von Willebrands factor, a large
protein that is responsible for the adherence of platelets
to damaged endothelium and acts as a carrier protein
for the factor VIII molecule. von Willebrands disease
can be inherited in either an autosomal dominant or recessive manner, and it affects almost an equal number
of males and females. Other less common clotting factor
deficiencies and disorders of platelets may need to be
considered when evaluating patients for spontaneous
musculoskeletal bleeding that is spontaneous or occurs
after trauma or surgery.
The hallmark symptom of severe hemophilia (factor
levels 1% of normal) is spontaneous bleeding. In these
patients, bleeding occurs without any recognizable
trauma. Recurrent hemarthrosis, or repeated bleeding
into the joints, usually begins after a child starts to walk.
In children, the ankles and elbows are more commonly
affected than the knees and shoulders. Orthopaedic surgeons may encounter these frequently undiagnosed disorders in patients who experience excessive intraoperative, postoperative, or posttraumatic bleeding.
Preoperative screening histories should query for frequent gum bleeding, epistaxis, excessive bruising, and
menorrhagia. A family history should also be obtained.
Because mild deficiencies may not result in significant
symptoms, a screening history should be obtained for all
patients before surgery, particularly those undergoing
more demanding procedures such as spinal fusion. The

American Academy of Orthopaedic Surgeons

possibility of an undiagnosed coagulopathy should be

considered in patients with compartment syndrome after relative minor injuries and when excessive bleeding
is encountered during or after surgery.
Although several substances in blood produce an inflammatory reaction from the synovium, iron is thought
to be the primary stimulant. In response to the inflammation, the synovium develops villi that are highly vascular and friable (Figure 1). These villi are then prone to
spontaneous hemorrhage with normal activity or minor
trauma. Once inflamed, the synovium also releases
chondrolytic substances and other factors that alter the
normal homeostasis between osteoblasts and osteoclasts, resulting in chondrolysis, periarticular osteopenia,
and cyst formation.
Hemophilic arthropathy can develop in patients as
young as 3 years of age. Synovitis must be treated early
and aggressively to prevent hemophilic arthropathy.
This involves medical treatment, including varying
degrees of a combination of prolonged factor replacement, physical therapy, and activity modifications. If
these measures are not enough to control bleeding, inflamed synovium may become so thick and abundant
that control can only be gained from a synovectomy
(Figure 2, A).

Prophylaxis
By infusing exogenous clotting factor concentrates on a
regular basis (typically two to three times per week), the
trough concentrations can be kept above 1% (patients
typically bleed when levels are below 1%). Although
primary prophylaxis is a good strategy to prevent joint
bleeding, there are numerous issues related to cost and
difficulty of ongoing administration.

Demand Therapy
The common approach is to treat hemarthroses as they
occur. The first dose administered after a joint bleed
should elevate the deficient clotting factor level to 80%
of normal. This decays to 5% in 48 hours (the half-life
for factor VIII is 12 hours), and an additional 40% dose

691

Pediatric Hematology

Figure 1 Histologic section of the hemosiderotic synovitis that results from recurrent
hemarthrosis in a patient with severe hemophilia. Note the highly vascular villi with
little supporting collagen, making this area highly prone to bleeding with minor trauma
or even normal joint motion.
is given on the second and third days after a bleed so
that levels remain above 5% for 5 days. Rest, ice, compression, and elevation are instituted immediately to reduce the severity of the bleed. As swelling resolves and
motion returns, the child gradually resumes activities.
If a joint bleeds more than three times in 6 months,
it becomes prone to developing a chronic synovitis. Secondary prophylaxis is instituted for 3 months. If bleeding stops and the synovitis resolves, the options are to
continue secondary prophylaxis indefinitely or to resume demand therapy. If the bleeding continues on prophylaxis or if the synovitis fails to resolve after
3 months, the synovitis is considered recalcitrant to
medical management and a synovectomy becomes necessary. The primary indications for synovectomy are persistent synovitis or bleeding despite 3 months of prophylaxis and resumption of joint bleeding within 1 year of
successful prophylaxis.
Synovectomy is therefore a means of rescuing the
joint from chronic synovitis, and it is a key treatment
modality for the successful use of demand therapy. The
primary objective when performing a synovectomy in
these joints is to remove most of the friable villous layer
of the synovium. Perioperative hemostasis is provided
by factor administration, allowing the synovium to heal.
The resultant synovial lining is smoother and less prone
to repetitive injury.

Radionuclide Synovectomy
By injecting a radioactive pharmaceutical agent into an
affected joint, the hypertrophic synovium is ablated, restoring a smoother and less friable surface. In the
United States, the most commonly used substance is P32
chromic phosphate. This substance releases primarily
radiation, which only penetrates a few millimeters into

692

Orthopaedic Knowledge Update 8

tissue. Although there is still some theoretic risk of subsequent malignant transformation after radionuclide
synovectomy, only one associated instance has been reported in a child with hemophilia, and there are no reports of growth disturbances.
The procedure is typically done as an outpatient
procedure in the radiology suite, and local anesthesia is
usually adequate. An arthrogram is always done to confirm intra-articular placement of the needle tip and to
look for venous extravasation (which usually occurs in
joints acutely inflamed by a bleed just before the procedure and is a contraindication to proceeding). The dosage varies depending on the joint and the size of the patient. Steroids are simultaneously injected to reduce the
inflammatory response that can accompany the treatment. A small-gauge needle is used to minimize leakage
up the needle track, and pressure is applied afterward.
Follow-up liver and spleen scans are used to look for
systemic leakage, but the results of these tests are rarely
positive. Patient activities are reduced for 2 weeks, and
physical therapy is not usually needed. Radionuclide
synovectomy generally reduces the frequency of bleeds
by 50% of patients with hemophilia. Radionuclide synovectomy is ideally suited for patients with hemophilia
who have developed an inhibitor, which is an immunoglobulin G-mediated antibody response to the clotting
factor.

Arthroscopic Synovectomy
Arthroscopic synovectomy offers some advantages over
radionuclide synovectomy in that the diseased synovium
is physically removed and lesions of articular cartilage
can be dbrided simultaneously (Figure 2, B). This procedure can be done on an outpatient basis using a continuous infusion pump to fully correct the deficient clotting factor before surgery and through the fourth
postoperative day. Infusion is then continued every
other day for at least 6 to 12 weeks. The most common
complication, loss of motion, is associated with preexisting arthritis (significant lost motion already present or
radiographic changes of joint space narrowing) or early
postoperative bleeding. Arthroscopic synovectomy requires compliance with a rigorous preoperative and
postoperative regimen of factor replacement and physical therapy; therefore, a committed family and experienced multidisciplinary team are keys to success. Joints
in children as young as 3 years can be arthroscopically
treated, and early intervention significantly improves
outcomes. Overall, an 80% reduction in hemarthroses
can be expected, and virtually all patients should experience improvement.
Although the direct costs of radionuclide synovectomy are less than those of arthroscopic synovectomy,
radionuclide synovectomy carries a higher rate of repeat
procedures, and arthroscopic synovectomy may be more

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 58 Pediatric Hematology

Figure 2 A, Arthroscopic view of the anterior compartment in the elbow of an 8-year-old boy with hemophilic (hemosiderotic) synovitis. Note the vascular villi produced by this
proliferative synovitis. B, Arthroscopic view of the same region of the anterior elbow compartment seen near the completion of an arthroscopic synovectomy. Note that most of
the villi have been removed, but the capsule remains intact. (Reproduced with permission from Dunn AL, Busch MT, Wyly JB, Sullivan KM, Abshire TC: Arthroscopic synovectomy
for hemophilic joint disease in a pediatric population. J Pediatr Orthop 2004;24:414-426.)
effective, especially in more severely affected joints. Arthroscopic synovectomy also avoids the long-term risks
of radiation exposure. Regardless of the type of synovectomy, the key to success is early intervention. Most
patients with severe hemophilia who are managed with
on-demand therapy protocols will develop a target
joint by adolescence, and an early synovectomy rescue is essential to minimize the risk of premature arthritis.

Sickle Cell Disease

Normal hemoglobin is composed of and chains. The
gene for globulin is located on chromosome 11, and a
single substitution of valine for glutamic acid at the
sixth amino acid position causes an abnormally functioning hemoglobin (the gene for hemoglobin S) to be
produced. Heterogeneous individuals have sickle cell
trait, and homozygous individuals have sickle cell disease. Other abnormalities can occur in the and globulin chains and cause sickle cell disorders, including
sickle cell thalassemia, sickle cell hemoglobin D, and
sickle cell hemoglobin E.
Sickle cell trait is common in western Africa, where
it provides some protection against malaria. In the
United States, about 10% of the African-American population carries the hemoglobin S gene, and 1 in 600 has
sickle cell disease. Sickle cell disease primarily affects
blacks, but it can occur in other ethnic groups.

American Academy of Orthopaedic Surgeons

Heterozygous individuals have a normal gene for hemoglobin A and a gene for hemoglobin S. Carriers are
typically asymptomatic because hemoglobin A still accounts for about 55% to 60% of an individuals total hemoglobin. Only rarely do individuals with sickle cell
trait display symptoms. Homozygous individuals with
two genes for hemoglobin S, however, have no hemoglobin A, and every organ and tissue in the body can be adversely affected.
As oxygen tension decreases, the hemoglobin S molecules polymerize within the red cells and lead to alteration in red cell shape, membrane changes, cellular dehydration, decreased deformability, and shorter life span
of the red cells. Tissues with low oxygen tension and end
vessel flow are particularly vulnerable. Musculoskeletal
symptoms account for 80% of all hospital admissions.
In infancy, individuals with sickle cell disease are
protected by fetal hemoglobin (hemoglobin F). In infants age 4 to 6 months with sickle cell disease, the
amount of hemoglobin F begins to decrease and symptoms can occur. Dactylitis (hand-foot syndrome) typically presents between the ages of 6 months and 4 years.
This condition is characterized by swelling and tenderness in the hands or feet, and occurs in about 70% of all
individuals with sickle cell disease. Dactylitis is thought
to represent a vaso-occlusive episode. Treatment consists of analgesia, oxygen, and hydration, with symptoms
typically lasting 3 to 7 days. Osteomyelitis must remain
in the differential diagnosis, and aspiration can be used
to aid in determining the proper treatment. Dactylitis is

693

Orthopaedic Knowledge Update 8

Pediatric Hematology
rare after age 6 years, when hematopoietic marrow disappears in the digits. Forty percent of individuals can
have recurrent dactylitis.
The incidence of sickle cell crises in patients with
sickle cell disease is around 0.8 episodes per year. Hydroxyurea has been shown to decrease the number of
painful episodes in adults, and a multicenter trial of hydroxyurea in children is currently underway.
Although individuals with sickle cell disease are
more prone to infection, the incidence of osteomyelitis
remains quite low (< 1% per year), and bone infarcts
are 50 times more common. Nonetheless, differentiating
between osteomyelitis and bone infarcts can be difficult
because their signs and symptoms are quite similar.
Pain, swelling, redness, warmth, stiffness, and elevated
laboratory parameters are common. Erythrocyte sedimentation rates are unreliable in patients with sickle
cell disease because of the altered red cells. Blood cultures should be included in routine work-up, and bone
aspiration should be done in patients with suspected
sickle cell disease. Several imaging modalities have been
studied to help differentiate between infection and infarct, including bone scans and MRI. Ultrasonography
may be less expensive and equally effective. A recent report has identified serum procalcitonin concentration as
a negative predictor of serious musculoskeletal infection. Common pathogens include Staphylococcus aureus, Salmonella, and Streptococcus pneumoniae. Treatment consists of antibiotics, hydration, oxygenation, and
surgical drainage of abscesses. Reactive arthritis and
septic arthritis can also occur.
Bone marrow hyperplasia can cause bony changes in
patients with sickle cell disease. Osteopenia in the vertebral bodies can lead to bulging of the intervertebral
disks, compression fractures, and progressive kyphosis.
Long bone growth can be affected, and pathologic fractures can occur. Growth retardation is common in patients with sickle cell disease, with height and weight significantly lower than average at 2 years of age. Skeletal
maturity can also be delayed. Osteonecrosis of the femoral and humeral heads commonly occurs, with prevalences of 30% and 6%, respectively. Treatment options
include restricted weightbearing, core decompression,
polymethylmethacrylate injection, vascularized fibular
grafts, hemiresurfacing, and total joint arthroplasty. Joint
arthroplasty is complicated by higher rates of infection
and loosening.

Thalassemia
The thalassemia syndromes comprise a heterogeneous
group of hemolytic anemias that result from mutations
that affect globulin synthesis. Normal hemoglobin is
composed of and chains; therefore, and thalassemias can occur. Individuals with thalassemia major are
homozygous. There is very low to no hemoglobin A pro-

694

duction, which results in a severe hemolytic, microcytic,

hypochromic anemia. Bone marrow hyperplasia ensues,
causing bone fragility, growth retardation, and an increased rate of fractures. Slipped capital femoral epiphysis and early osteoarthritis occur at increased rates.
Radiographic findings include widened marrow spaces,
thinned cortices, severe osteopenia, and hair-on-end
appearance of the skull. Extramedullary hematopoiesis
in the vertebrae can cause paraparesis by compression
of the spinal cord. Increased body iron loads can cause
tissue fibrosis, resulting in cardiomyopathy, diabetes
mellitus, and adrenal, parathyroid, and thyroid hypofunction. Splenectomy in childhood lessens the transfusion requirements, but it increases the already higher
than normal rate of infections. Treatments of individuals
with thalassemia include early transfusion therapy
(maintaining hemoglobin > 9 g/dL), iron chelation, bone
marrow transplantation, splenectomy, and supportive
medical management.

Leukemia
Acute leukemia is the most common malignancy in
childhood, accounting for nearly one third of instances
of cancer in children. Acute lymphocytic leukemia accounts for approximately 80% of these instances. With
current therapies, approximately 80% of children with
acute lymphocytic leukemia now survive.
Although leukemia is primarily a disease of the bone
marrow, any organ can be infiltrated by the malignant
cell, and this accounts for the highly variable clinical
presentation. The musculoskeletal system is often involved, including pain in the extremities, back pain, osteomyelitis, septic arthritis, or fracture. Twenty percent
of patients present with a primary report of limp or extremity pain, and up to 50% of patients have these
symptoms as secondary complaints at the time of initial
diagnosis. Although long bone pain is the most common
symptom, the spine may be involved. This becomes an
important consideration in the differential diagnosis of
the child presenting with back pain. At the onset of the
disease, 10% of children have normal peripheral blood
counts, making the diagnosis even more challenging.
MRI may be helpful because it is very sensitive to marrow changes caused by infiltration and secondary infarct.
Painful joint swelling and fever are common presenting symptoms of the musculoskeletal manifestations of
leukemia. These symptoms may mimic sepsis or acute
onset of a juvenile idiopathic arthritis (formerly called
juvenile rheumatoid arthritis). Typically, patients with
juvenile idiopathic arthritis present with morning stiffness and pain localized to the joints, whereas patients
with leukemia more often present with night pain and
nonarticular bone pain. The radiographs of patients with
juvenile idiopathic arthritis may show joint effusions

American Academy of Orthopaedic Surgeons

Chapter 58 Pediatric Hematology

Orthopaedic Knowledge Update 8

and mild osteopenia surrounding joints, whereas the radiographic changes suggestive of acute leukemia include
diffuse osteopenia, metaphyseal bands, periosteal new
bone formation, geographic lytic lesions, sclerosis, mixed
sclerosis/lysis, and permeative destruction.
Once considered rare, acute megakaryoblastic leukemia now accounts for about 12% of all instances of
acute myeloid leukemia in children. Most instances of
acute megakaryoblastic leukemia are difficult to diagnose because of their complex clinical presentation and
unusual bone marrow morphologic features. In children,
acute megakaryoblastic leukemia is often confused with
metastatic solid tumors or myelodysplastic syndrome.
These abnormalities (bilaterally symmetric periostitis
and osteolytic lesions) differ markedly from those commonly reported in association with pediatric acute leukemias.
Treatment protocols for leukemia often include significant doses of steroids because of their antilymphocytic effects. MRI shows changes consistent with areas
of osteonecrosis in approximately 25% of children receiving high-dose steroid regimens. Children younger
than 10 years are generally not affected. One third or
more of those affected develop symptomatic complications of weight-bearing joints. When these complications
do occur, treatment options are limited, and end stage
joint disease can develop.

Thrombocytopenia With Absent Radii

Thrombocytopenia with absent radii (TAR) syndrome is
an autosomal recessive, rare disorder characterized by
congenital hypomegakaryocytic thrombocytopenia and
bilateral absence of the radius, with shortening of the
ulna and a normal thumb. This syndrome develops between the fourth and eighth weeks of gestation when
the radii develop and platelet production begins. Significant complications that can occur in infancy secondary
to thrombocytopenia include intracranial hemorrhage,
anemia, and frequent occurrences of epistaxis. Characteristic hematologic findings include hypomegakaryocytic thrombocytopenia, periodic leukemoid reactions,
and eosinophilia. Mortality in early infancy approaches
40% of patients; however, if the child is able to survive
the first 1 to 2 years of life, the thrombocytopenia usually resolves and the musculoskeletal problems become
the chief manifestation of TAR.
The upper extremity anomalies include bilateral absence of the radii with radial deviation at the wrist, a
shortened and bowed ulna, and normal thumbs. Recently, an anomalous muscle spanning the elbow with its
origin on the humeral shaft at the deltoid tubercle and
insertion into the hand and carpal bones (termed the
brachiocarpalis) has been described and its excision has
been recommended to negate the strong radial deviation to the hand.

American Academy of Orthopaedic Surgeons

Lower extremity anomalies occur in up to 80% of

patients, and they include severe genu varum, instability,
and patellar abnormalities, which has prompted some to
rename this syndrome TARK, with the letter K signifying involvement of the knee. Other associated deformities include hip dysplasia, pes valgus, talipes equinovarus, synostosis in the metatarsals, and syndactyly
involving the toes.
Management of patients with TAR syndrome includes hematologic support, including platelet transfusions until the patients condition stabilizes, usually
around 2 years of age. Musculoskeletal management includes radiographs for patients with occult hip dysplasia. Varus angulation and internal tibial rotation usually
recur after corrective osteotomy, which has prompted
some to advocate a more conservative management approach using simple adaptive devices and powered mobility aids. Foot and ankle reconstruction including posteromedial release and/or heel cord lengthening have
been recommended. The goals of surgical treatment of
the upper extremity deformities are to obtain and maintain correction of the radial deviation. The primary indication for upper extremity surgery correction is persistent wrist contracture and reconstruction of functionlimiting thumb deficiency.

Annotated Bibliography
Hemophilia
Butler RB, McClure W, Wulff K: Practice patterns in
haemophilia A therapy: A survey of treatment centres
in the United States. Haemophilia 2003;9:549-554.
This study surveyed 52 hemophilia centers with a total of
4,129 patients receiving treatment. Among patients with severe hemophilia, 49% were receiving on-demand treatment,
whereas 44% were receiving some form of prophylaxis (13%
primary, 20% secondary, and 11% tertiary). Primary prophylaxis was the most common type in children younger than
5 years, who comprised 25% of this age group. In children age
6 to 18 years, 58% were receiving some type of prophylactic
regimen, whereas on-demand treatment was most frequent
among adult patients.

Dunn AL, Busch MT, Wyly JB, Abshire TC: Radionuclide synovectomy for hemophilic arthropathy: A comprehensive review of safety and efficacy and recommendation for a standardized treatment protocol. Thromb
Haemost 2002;87:383-393.
This review article compiles the published experience to
date using radionuclide synovectomy for hemophilic joint disease. A suggested treatment protocol is presented.

Dunn AL, Busch MT, Wyly JB, Sullivan KM, Abshire

TC: Arthroscopic synovectomy for hemophilic joint disease in a pediatric population. J Pediatr Orthop 2004;24:
414-426.

695

Pediatric Hematology
This study provides data on 47 pediatric patients who underwent arthroscopic synovectomy (40 ankles, 22 elbows, 9
knees, and 2 shoulders) for hemophilic joint disease. The median patient age at time of surgery was 10.3 years, and median
follow-up was 79 months. The authors report that joints with
sufficient follow-up data showed a median bleeding frequency
decline of 84% (P < 0.001).

Journeycake JM, Miller KL, Anderson AM, Buchanan

GR, Finnegan M: Arthroscopic synovectomy in children
and adolescents with hemophilia. J Pediatr Hematol
Oncol 2003;25:726-731.

Orthopaedic Knowledge Update 8

absent radius (TAR) syndrome and discussion of its
clinical significance, in American Society for Surgery of
the Hand Meeting Abstracts, Phoenix, AZ, 2002, p 13.
Available at: http://www.assh.org. Accessed October 6,
2004.
The authors of this retrospective review of 16 patients with
thrombocytopenia absent radius syndrome show the presence
of an anomalous muscle spanning the elbow (termed the brachiocarpalis) and emphasize its contribution to the radial deformity. They recommend excision in patients with progressive
radial deviation.

In this study, 28 arthroscopic synovectomies (11 knees, 12

ankles, and 5 elbows) were done on 26 joints in 20 patients
with hemophilia. The authors report that the frequency of hemarthrosis diminished significantly in the first year and was
maintained for up to 5 years in all but three joints.

Classic Bibliography

Sickle Cell Disease

Chambers JB, Forsythe DA, Bertrand SL, Iwinski HJ,

Steflik DE: Retrospective review of osteoarticular infections in a pediatric sickle cell age group. J Pediatr
Orthop 2000;20:682-685.

Scott LK, Grier LR, Arnold TC, Conrad SA: Serum procalcitonin concentration as a negative predictor of serious bacterial infection in acute sickle cell pain crisis.
Med Sci Monit 2003;9:CR426-CR431.
In this preliminary study of 24 patients with sickle cell disease, pain crisis, and acute inflammation, procalcitonin levels
were measured and levels less than 2 ng/mL were reported to
have a good negative predictive value of serious infection.

Skaggs DL, Kim SK, Greene NW, Harris D, Miller JH:

Differentiation between bone infarction and acute osteomyelitis in children with sickle-cell disease with use
of sequential radionuclide bone-marrow and bone scans.
J Bone Joint Surg Am 2001;83:1810-1813.
This is a study of 70 patients with sickle cell disease and
acute bone pain. The findings suggest that serial radionuclide
bone marrow and bone scan scintigraphy can provide some
differentiation between osteomyelitis and bone infarction.

Leukemia
States LJ: Imaging of metabolic bone disease and marrow disorders in children. Radiol Clin North Am 2001;
39:749-772.
In this study, the author reports that MRI provides detailed information about bone marrow and is gaining an increasingly important role in the management of disorders of
bone marrow infiltration.

Thrombocytopenia With Absent Radii

Arnold WD, Hilgartner MW: Hemophilic arthropathy:

Current concepts of pathogenesis and management.
J Bone Joint Surg Am 1977;59:287-305.

Christensen CP, Ferguson RL: Lower extremity deformities associated with thrombocytopenia in absent radius syndrome. Clin Orthop 2000;375:202-206.
Hedberg VA, Lipton JM: Thrombocytopenia with absent
radii: A review of 100 cases. Am J Pediatr Hematol
Oncol 1988;10:51-64.
Heinrich SD, Gallagher D, Warrior R, Phelan K, George
VT, MacEwen GD: The prognostic significance of the
skeletal manifestations of acute lymphoblastic leukemia
of childhood. J Pediatr Orthop 1994;14:105-111.
Mattano LA Jr, Sather HN, Trigg ME, Nachman JB: Osteonecrosis as a complication of treating acute lymphoblastic leukemia in children: A report from the Childrens Cancer Group. J Clin Oncol 2000;18:3262-3272.
McLaurin TM, Bukrey CD, Lovett RJ, Mochel DM:
Management of thrombocytopenia-absent radius (TAR)
syndrome. J Pediatr Orthop 1999;19:289-296.
Schoenecker PL, Cohn AK, Sedgwick WG, Manske PR,
Salafsky I, Millar EA: Dysplasia of the knee associated
with the syndrome of thrombocytopenia and absent radius. J Bone Joint Surg Am 1984;66:421-427.

Carter PR, Mills J, Ezaki M: (Abstract) Anatomical description of an anomalous muscle in thrombocytopenia

696

American Academy of Orthopaedic Surgeons

Chapter

59

Shoulder and Humerus: Pediatrics

Ann Van Heest, MD

Introduction
Familiarity with the spectrum of shoulder disorders seen
in children is necessary for the treatment of pediatric
upper extremity disorders. For the child whose symptoms prevent normal use of the arm, a complete evaluation should include the shoulder girdle because children
often are unable to provide details on the history of
their disorder and have difficulty localizing symptoms.
A general classification of pediatric shoulder disorders is based on congenital, developmental, or acquired
etiology. Congenital shoulder disorders are present at
birth and are the result of abnormal fetal formation and
development. Developmental disorders are caused by
growth disturbances such as growth plate dysfunction,
neuromuscular disorders, systemic disease, or manifestations of a syndromic disorder. Acquired disorders are
most commonly secondary to trauma, infection, or tumor. This classification system helps provide a general
framework for understanding pediatric shoulder disorders.

Sprengels Deformity
Sprengels deformity is the congenital failure of descent
of the scapula from the embryonic level opposite the
fifth cervical vertebra to its final normal position, with
its superior border at the seventh cervical vertebra and
with its inferior angle at the level of the sixth rib. In up
to 50% of patients, an associated omovertebral bar has
been described, consisting of a fibrous, cartilaginous, or
bony connection between the superior angle of the
scapula and the cervical vertebral spinous process, lamina, or transverse process. Additionally, omoclavicular
bars have been described. Other common associated
anomalies requiring investigation including scoliosis,
spina bifida, rib anomalies, Klippel-Feil syndrome, abnormal musculature, foot deformities, torticollis, facial
asymmetry, and pulmonary and kidney disorders. The
incidence of these disorders is nearly equal on the right
and left sides; bilaterality is reported in 10% to 30% of
patients. The condition is slightly more common in girls.

American Academy of Orthopaedic Surgeons

Diagnosis of Sprengels deformity primarily has

been made based on a physical examination; physical
characteristics include a webbed neck, a palpable scapula into the cervical region, and a loss of range of motion. Radiographs confirm the diagnosis. Recently, prenatal diagnosis through the use of ultrasound has been
described.
Recommended treatment is based on the Cavendish
classification. Grade 1 deformity is very mild with level
glenohumeral joints and no visible deformity when the
patient is dressed; no functional impairment is present
and treatment is not warranted. Grade 2 deformity is
mild with level glenohumeral joints with a visible lump
in the neck when the patient is dressed. If the cosmetic
deformity is not acceptable, surgical treatment would include extraperiosteal resection of the scapular prominence including resection of the tethering omovertebral
bar if present. Hypertrophic scars are common; therefore, parents need to be aware that the bump will be replaced with a scar. Grade 3 deformity is moderate with
2 to 5 cm of shoulder joint elevation compared with the
contralateral side, and grade 4 is a severe deformity with
the position of the scapula near the occiput. Surgical
treatment involves not only resection of the superior angle of the scapula and the tethering omovertebral bar if
present, but also derotation of the scapula and relocation to a more caudal position.
The age at which surgical intervention should be undertaken has been discussed by various authors, and the
general agreement is for intervention when the child is
at least 3 years of age, which is when the child is old
enough to tolerate the extensive nature of the procedure yet young enough to maximize correction through
subsequent growth. With caudal repositioning of the
scapula, there is a risk of compressive injury to the brachial plexus. This risk is greatest in a child older than
8 years and can be lessened by concomitant clavicular
osteotomy.
Surgical techniques described for scapular relocation
include: (1) the Woodward procedure, which relocates
the scapula by detachment and caudal relocation of the
midline origin of the parascapular muscle; (2) the Green

697

Shoulder and Humerus: Pediatrics

procedure, which relocates the scapula by extraperiosteal detachment of scapular muscles at their scapular
insertion with reattachment after caudal relocation of
the scapula using scapular traction cables; and (3) correction by vertical scapular osteotomy. Concomitant
clavicular osteotomy or morcellization is recommended
to reduce the incidence of neurovascular injury and to
provide anterior release allowing greater correction of
the deformity.
The Woodward procedure is the most common surgical technique used for correction of a moderate or severe Sprengels deformity. A recent report has confirmed previous findings of cosmetic and functional
improvement with this procedure. The age of the patient
and the presence of an omovertebral bone did not influence the results. Associated cervical spine anomalies
were indicative of a negative prognosis.

Pseudarthrosis of the Clavicle

Pseudarthrosis of the clavicle in children usually is
present at birth, with parents reporting a prominent
bump at the midportion of the clavicle. Physical examination reveals little or no pain on palpation of mobile
clavicle segments with a prominence in the midportion.
Shoulder range of motion is normal or hypermobile. The
prominent bump in the midclavicle often becomes more
apparent with loss of infantile adipose tissue. If the child
is examined at a later age, asymmetry of the shoulder
with drooping of the affected side caused by fragment
hypermobility may be present. Difficulties with shoulder
function are rare, although the recent literature has reported the occurrence of thoracic outlet symptoms in
adolescence.
Radiologic evaluation shows lack of bone continuity
in the middle third of the clavicle without evidence of
reactive bone. The presence of minimal pain and lack of
callus formation distinguishes this condition from a
birth fracture. Pseudarthrosis of the clavicle can be distinguished from cleidocraniodysostosis because pseudarthrosis is more commonly unilateral with absence of
cranial or pelvic involvement. Associated cervical ribs or
elevated first ribs may be present in pseudarthrosis of
the clavicle. The lesion is usually on the right side; leftsided occurrence is usually in association with dextrocardia. Bilateral occurrences have been reported.
Most children with congenital pseudarthrosis of the
clavicle are asymptomatic and do not require surgical
intervention. Surgery to establish union is reserved for
children with an unacceptable bump, pain, or shoulder
dysfunction. Surgical treatment options include resection of the fibrous pseudarthrosis and sclerotic bone
ends with suturing of the bone ends, with bone grafting,
and/or internal fixation.

698

Orthopaedic Knowledge Update 8

Glenoid Hypoplasia
Normal glenoid ossification occurs through consolidation of the secondary ossification centers of the superior
glenoid, inferior glenoid, and base of the coracoid. Failure of normal ossification of the inferior glenoid epiphysis is a rarely diagnosed entity. The patient has limited
abduction, mild axillary webbing, and aplasia or hypoplasia of the glenoid. Several cases of familial involvement with variable penetrance have been reported.
Lack of ossification of the inferior glenoid appears to
cause such minimal disability that this condition is often
undiagnosed, and is most commonly seen as an asymptomatic incidental finding or as part of other congenital
deficiencies.

Birth Brachial Plexus Palsy

The most common pattern of birth brachial plexus injury involves C5 and C6 neurologic injury with complete
or partial paralysis of the shoulder girdle muscles. Weakness of the deltoid, biceps, supraspinatus, and infraspinatus results in an adduction internal rotation deformity
caused by the unopposed pull of the pectoralis major,
subscapularis, teres major, and latissimus dorsi. Secondary skeletal changes include retroversion of the glenoid,
posterior subluxation and medial flattening of the humeral head, and elongation with prominence of the acromion. Early radiographic changes of immature ossification of the epiphysis compared with the contralateral
side and scapular winging are shown in Figure 1.
During infancy, treatment of the shoulder dysfunction associated with brachial plexus injury is aimed at
prevention of both internal rotation contracture and
secondary bone deformity during the neurologic recovery period. Brachial plexus exploration with neurolysis,
nerve repair, or nerve grafting may be indicated if biceps function has not returned by 6 months of age. If
shoulder muscle weakness persists with external rotation deficiency, anterior release or lengthening of the
tight anterior structures (pectoralis major, subscapularis,
anterior capsule) is recommended, coupled with augmentation of the deficient posterior structures (latissimus dorsi and teres major to rotator cuff tendon transfer).
Children with a history of upper trunk brachial
plexus injury with incomplete return of shoulder external rotation usually should be monitored for development of skeletal changes of the glenohumeral joint. Radiographic evaluation for glenohumeral dysplasia has
been described using plain radiographs, ultrasound,
MRI, CT (Figure 2), and arthrography. Although the
method of radiographic evaluation remains controversial, the necessity of monitoring children for glenohumeral dysplasia is not disputed. The natural history of
this disease has been well defined; therefore, progressive
flattening of the humeral head, posterior subluxation,

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 59 Shoulder and Humerus: Pediatrics

Figure 1 Early radiographic changes associated with glenohumeral dysplasia in a child with upper trunk birth brachial plexus injury are shown. These changes include immature
ossification of the epiphysis compared with the contralateral side with scapular winging. Secondary skeletal changes include retroversion of the glenoid, posterior subluxation,
medial flattening of the humeral head, and elongation with prominence of the acromion.

eventual dislocation of the joint, and increased retroversion of the glenoid is highly probable. Surgical intervention should be done using tendon transfer surgery before significant glenohumeral dysplasia develops.
Recent studies have shown that loss of passive external
rotation with the arm in adduction is an early finding on
physical examination that glenohumeral dysplasia is developing. If fixed skeletal deformity of the glenohumeral
joint exists, a humeral osteotomy with external rotation
of the distal fragment is recommended (Figure 3).
Recent studies also have shown that a posterior dislocation can occur in association with birth brachial
plexopathy even in infancy (younger than 1 year of
age). Whether this condition represents an early manifestation of a severe muscle imbalance or whether the
condition occurred acutely at the time of birth is currently unknown. A high index of suspicion for posterior
dislocation is essential in the evaluation of an infant
with brachial plexus injury at birth. If diagnosed, an
open reduction may be necessary.

Clavicle Fractures
A recent study that prospectively screened newborn infants for clavicle fractures reported the incidence of 5 frac-

American Academy of Orthopaedic Surgeons

Figure 2 A CT scan more effectively evaluates glenohumeral dysplasia. Radiographic

measurements should include glenoid version and the percent of humeral head coverage.

699

Shoulder and Humerus: Pediatrics

Orthopaedic Knowledge Update 8

Figure 3 Severe fixed glenohumeral dysplasia in the skeletally mature adolescent occurs with untreated residual upper trunk dysfunction. Because of the severity of the
dysplasia and loss of passive joint motion, a humeral rotation osteotomy was necessary for treatment. A and B, Preoperative radiographs. C, CT scan. D, Postoperative radiograph.

tures per 1,000 live births. The fractures occurred equally

on the right and left side and occurred equally in males
and females. All fractures occurred during vaginal deliveries. No clavicle fractures occurred with breech presentation. Risk factors for fracture included large birth
weight, shoulder dystocia, mechanically assisted delivery,
and prolonged gestational age. One in 11 newborns with
a clavicle fracture had an associated brachial plexus palsy.

700

During the first decade of life, 88% of clavicle fractures are midshaft and 55% are nondisplaced. A recent
report of surgical treatment of clavicle fractures indicated that over a 21-year period, only 15 patients with
clavicle fractures required surgical stabilization. Surgical
treatment was indicated primarily in the older child with
a displaced, unstable fracture pattern. Intramedullary
elastic nails or plates and screws are options for internal

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

fixation. However, surgery is very rarely needed for pediatric clavicle fractures.
After an isolated nondisplaced midshaft clavicle
fracture, the risks of nonunion, malunion, or neurovascular complications are exceptionally rare; therefore,
follow-up beyond the initial diagnostic visit is rarely
necessary.

Proximal Humerus and Shaft Fractures

Fractures of the proximal humerus are relatively rare
(0.5% of all childhood fractures), compared with the
more common distal radius, supracondylar elbow, and
forearm shaft fractures. Salter-Harris type I injuries are
most common in neonates and children younger than
age 5 years; metaphyseal fractures are most common in
children 5 to 11 years old; and Salter-Harris type II fractures are most common in those older than 11 years.
Diagnosis of proximal humeral fractures in the newborn may be difficult because these fractures must be
distinguished from septic arthritis, brachial plexus injuries, and clavicle fractures in the infant with pseudoparalysis of the arm. Plain films may not be helpful in the
diagnosis because the proximal humerus epiphysis does
not ossifiy until after 6 months of age. MRI, ultrasound,
or aspiration/arthrogram may be necessary. As the child
approaches skeletal maturity, pain with arm dysfunction,
a history of a fall on an outstretched hand, or a direct
blow to the shoulder region along with positive radiographs confirm the diagnosis.
Fractures of the proximal humerus are usually
treated using closed techniques. The need for a closed
reduction is determined by the extent of displacement
and the remodeling remaining, based on the childs age.
General guidelines include up to 70 of angulation in
the child younger than 5 years old; 40 to 70 of angulation in the child 5 to 12 years old; and up to 40 of angulation in the child older than 12 years.
When reduction cannot be achieved, or when reduction is lost as the arm is moved down to the side, percutaneous pinning is usually recommended. Open reductions are reserved for special circumstances such as in
patients with associated injuries, open injuries, concomitant glenohumeral dislocation, or significant muscle interposition. In patients with open injuries, the use of external fixation has been used with good results.

Septic Arthritis and Osteomyelitis

Septic arthritis of the shoulder joint remains an uncommon and difficult diagnosis that requires a high index of
suspicion. The disease usually involves very young infants and has a high prevalence of associated infections.
In a recent review of the literature, elevated body temperature was present in 90% of pediatric patients with
septic arthritis and was the most constant finding of the
physical examination. An increased erythrocyte sedi-

American Academy of Orthopaedic Surgeons

Chapter 59 Shoulder and Humerus: Pediatrics

mentation rate (that may rise above 100 mm/hr) and increased white blood cell count also may be seen.
The initial radiographs are frequently normal; ultrasonography supports the diagnosis in some patients by
showing an accumulation of fluid inside the joint space.
Aspiration of synovial fluid from the affected glenohumeral joint is necessary to evaluate the offending
pathogen. In a recent review, false-negative Gram stains
occurred in approximately 90% of patients, whereas
synovial fluid cultures showed the pathogen in 88% of
patients. Blood cultures were positive in 90% of pediatric patients. The most common isolated pathogen was
Staphylococcus aureus, which accounted for 41% of infections. Gram-negative bacilli, which accounted for
about 20% of infections, are more prevalent in the pediatric population, especially in neonates.
Pyogenic shoulder arthritis should be treated with
intravenous antibiotics after synovial fluid aspiration
and joint irrigation. Irrigation (using a wrist arthroscope) or arthrotomy may be chosen.
The close proximity of the proximal humeral metaphysis to the shoulder joint capsule, as well as the perforating vessels from the physis to the metaphysis of the
proximal humerus, necessitate investigation to determine if septic arthritis of the shoulder joint coexists with
osteomyelitis of the proximal humerus. About half of affected children have both conditions. Treatment of septic arthritis in these children should include intravenous
antibiotics and drilling of the metaphyseal bone of the
proximal humerus at the time of joint washout. The
presence of persistent pain, swelling, and fever in children who have had an arthrotomy or arthroscopy for
septic arthritis of the shoulder may indicate concurrent
osteomyelitis.
Certain systemic diseases such as sickle cell disease
tend to be more closely associated with the development of osteomyelitis. Tuberculous osteomyelitis of the
proximal humerus has been described, with the diagnosis confirmed by tissue biopsy and culture. The presence
of Mycobacterium tuberculosis is confirmed by its
growth at body temperature on the agar in the
Lwenstein-Jensen culture medium, or can be seen histologically with caseating granulomas. The importance
of tissue diagnosis cannot be overemphasized. A high
degree of suspicion in patients from immigrant populations is appropriate, as osteomyelitis is more common in
this patient group.

Annotated Bibliography
Sprengels Deformity
Chinn DH: Prenatal ultrasonographic diagnosis of
Sprengels deformity. J Ultrasound Med 2001;20:693-697.
This article describes the use of ultrasound for prenatal diagnosis of Sprengels deformity.

701

Orthopaedic Knowledge Update 8

Shoulder and Humerus: Pediatrics

Khairouni A, Bensahel H, Csukonyi Z, Desgrippes Y,
Pennecot GF: Congenital high scapula. J Pediatr Orthop
B 2002;11:85-88.
This recent review of the Woodward procedure confirms
the cosmetic and functional improvements achieved. Seventynine percent of patients had good or excellent results.

Pseudarthrosis of the Clavicle

Lorente Molto FJ, Bonete Lluch DJ, Garrido IM: Congenital pseudarthrosis of the clavicle: A proposal for
early surgical treatment. J Pediatr Orthop 2001;21:689693.
This article presents a review of six children with congenital pseudarthrosis (ages 18 months to 4 years). The patients
were surgically treated with bone graft and internal fixation;
healing of the pseudarthrosis was obtained in all patients in 6
to 8 weeks.

Glenoid Hypoplasia
de Bellis U, Guarino A, Castelli F: Glenoid hypoplasia:
Description of a clinical case and analysis of the literature. Chir Organi Mov 2001;86:305-309.
This article discusses the skeletal changes associated with
glenoid hypoplasia and describes a case of unilateral glenoid
hypoplasia.

Birth Brachial Plexus Palsy

Hoeksma AF, Ter Steeg AM, Dijkstra P, Nelissen GHH,
Beelen A, de Jong BA: Shoulder contracture and osseous deformity in obstetrical brachial plexus injuries.
J Bone Joint Surg Am 2003;85-A:316-322.
A high prevalence of glenohumeral dysplasia on plain radiographs is associated with internal rotation contracture of
the shoulder following birth brachial plexus injury. A loss of
passive external rotation of the shoulder was highly associated
with osseous deformity of the glenohumeral joint.

Clavicle Fractures
Calder JD, Solan M, Gidwani S, Allen S, Ricketts
DM: Management of paediatric clavicle fractures: Is
follow-up necessary? An audit of 346 cases. Ann R Coll
Surg Engl 2002;84:331-333.
This study recommends that there is no need for follow-up
of children with isolated, uncomplicated midshaft clavicle fractures; these fracture patients can be discharged after their first
assessment in fracture clinic.

Kubiak R, Slongo T: Operative treatment of clavicle

fractures in children: a review of 21 years. J Pediatr Orthop 2002;22:736-739.
This report on the surgical treatment of clavicle fractures
showed that, during a 21-year period, only 15 patients with
clavicle fractures required surgical stabilization. Surgery was
indicated primarily in the older child with a displaced, unstable fracture pattern. Intramedullary elastic nails were the preferred method of internal fixation.

Septic Arthritis and Osteomyelitis

Forward DP, Hunter JB: Arthroscopic washout of the
shoulder for septic arthritis in infants: A new technique.
J Bone Joint Surg Br 2002;84:1173-1175.
The use of a wrist arthroscope for arthroscopic washout of
septic arthritis of the shoulder in three children younger than
3 years of age is described. Full recovery was achieved with a
single intervention.

Monach PA, Daily JP, Rodriguez-Herrera G, Solomon

DH: Tuberculous osteomyelitis presenting as shoulder
pain. J Rheumatol 2003;30:851-856.
Tuberculous osteomyelitis of the proximal humerus in children is described, along with a review of the relevant literature.

Classic Bibliography
Moukoko D, Ezaki M, Wilkes D, Carter P: Posterior
shoulder dislocation in infants with neonatal brachial
plexus palsy. J Bone Joint Surg Am 2004;86A:787-793.
Of 134 patients with neonatal brachial plexus palsy, the diagnosis of posterior shoulder dislocations was made in 11 patients (8%). Diagnosis was made on clinical examination and
confirmed by ultrasonography at an average patient age of 6
months (range, 3 to 10 months).

Pearl ML, Edgerton BW, Kon DS, et al: Comparison of

arthroscopic findings with magnetic resonance imaging
and arthrography in children with glenohumeral deformities secondary to brachial plexus birth palsy. J Bone
Joint Surg Am 2003;85-A:890-898.
The natural history of brachial plexus injury leading to internal rotation contracture, which subsequently leads to glenohumeral dysplasia, is documented within the first 2 years of
life. The glenohumeral dysplasia was evaluated comparatively
by MRI, arthrography, and arthroscopy.

702

Bennek J: The use of upper limb external fixation in

paediatric trauma. Injury 2000;31(suppl 1):21-26.
Cavendish ME: Congenital elevation of the scapula.
J Bone Joint Surg Br 1972;54:395-408.
Sprengel: Die angeborene Verschieburg des Schulterblattes nach oben. Arch Klin Chir 1891;42:545.
Goddard NJ, Fixsen JA: Rotation osteotomy of the humerus for birth injuries of the brachial plexus. J Bone
Joint Surg Br 1984;66:257-259.
Grogan DP, Love SM, Guider KJ, Ogden JA: Operative
treatment of congenital pseudarthrosis of the clavicle.
J Pediatr Orthop 1991;11:176-180.
Hoffer M, Wickenden R, Roper B: Brachial plexus birth
palsies: Results of tendon transfers to the rotator cuff.
J Bone Joint Surg Am 1978;60:691-695.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 59 Shoulder and Humerus: Pediatrics

Lossos IS, Yossepowitch O, Kandel L, Yardeni D, Arber

N: Septic arthritis of the glenohumeral joint: A report of
11 cases and review of the literature. Medicine (Baltimore) 1998;77:177-187.

Webb LX, Mooney JF: Fractures and dislocations about

the shoulder, in Green NE, Swiontkowski MF (eds):
Skeletal Trauma in Children, ed 2. Philadelphia, PA, WB
Saunders, 1998, pp 319-341.

McBride MT, Hennrikus WL, Mologne TS: Newborn

clavicle fractures. Orthopedics 1998;21:317-320.

Waters PM, Smith GR, Jaramillo D: Glenohumeral deformity secondary to brachial plexus birth palsy. J Bone
Joint Surg Am 1998;80:668-677.

Schnall SB, King JD, Marrero G: Congenital pseudarthrosis of the clavicle: A review of the literature and surgical results of six cases. J Pediatr Orthop 1988;8:316321.

American Academy of Orthopaedic Surgeons

Woodward JW: Congenital elevation of the scapula:

Correction by release and transplantation of muscle origins. J Bone Joint Surg Am 1961;43:219-228.

703

Chapter

60

Elbow: Pediatrics
John M. Flynn, MD
Roger Cornwall, MD

Pediatric Elbow Fractures

Pediatric elbow fractures are common; however, they
are challenging to evaluate and treat even for the experienced orthopaedist. The elbow of a young child is
composed predominantly of unossified cartilage. In
older children, multiple centers of ossification make
evaluation difficult. Associated neurovascular injuries
frequently occur and may significantly impact the urgency and method of treatment. In general, the pediatric
elbow has a relatively limited remodeling capacity. The
elbow physes grow more slowly than those in the proximal humerus and distal radius. As a hinge joint, the elbow has a poor tolerance for coronal plane deformities
and has a particular propensity for loss of motion. These
factors help to explain the disproportionate number of
medical liability claims and the anxiety related to pediatric elbow fracture treatment.
Initial evaluation of a child with an injured elbow
should begin with a general examination and a search
for other sites of injury. Next, the involved upper extremity should be examined carefully for the point of
maximal tenderness in the area with maximum swelling.
Any upper extremity deformity or a break in the skin
integrity should be noted. Because of the high prevalence of neurovascular injuries, a careful neurovascular
examination, including a check of the anterior interosseous nerve, is essential. The carrying angle of the
uninjured side should be noted, especially in a child
with a supracondylar humerus fracture.
Radiographic evaluation should include the bones
and joints above and below the site of injury. The presence of a posterior fat pad is an important sign of a possible occult elbow fracture. Radiographs should be studied closely for signs of a secondary injury.

Supracondylar Fractures
Supracondylar fractures, which represent 60% of elbow
fractures in children, are classified based on the extent
and direction of the displacement of the distal fragment;
98% are extension injuries. The modified Gartland classification system is used to describe the extent of injury.

American Academy of Orthopaedic Surgeons

Type I fractures are minimally displaced, type II are displaced with an intact posterior cortex, and type III fractures have a completely displaced distal fragment. In
posteromedial fractures (occurring in approximately
75% of patients), the radial nerve is at risk, whereas in
posterolateral fractures, the brachial artery and median
nerve are at risk. About 2% of supracondylar fractures
are flexion injuries with disruption of the posterior periosteum. The flexion pattern is considered much more
difficult to treat with standard closed pinning techniques. A recent study of 29 flexion-type supracondylar
humerus fractures reported 86% good or excellent results with casting for minimally displaced fractures, and
with closed reduction and percutaneous pinning for all
type II and III flexion injuries.

Treatment
The goal of treatment of a pediatric supracondylar humerus fracture is to restore alignment to a position
where there is no varus malalignment and the anterior
humeral line intersects the capitellar ossification center.
Reduction and stabilization should be done without
causing iatrogenic nerve injury or using elbow hyperflexion to a degree that will compromise distal perfusion. The use of a long arm cast for 3 weeks is satisfactory treatment for type I fractures, and for type II
fractures in which the anterior humeral line intersects
the capitellar ossification center and Baumanns angle is
acceptable. Closed reduction and percutaneous pinning
is indicated for any fracture that does not meet these
conditions. The trend toward pinning most type II and
III supracondylar humerus fractures has dramatically
reduced the incidence of clinically important malunions.
Because the technique of reduction and hyperflexion of
the elbow is now avoided, Volkmanns ischemic contracture is rare.
Initial evaluation includes a careful neurologic and
vascular examination and AP and lateral radiographs of
the distal humerus, with separate views of the entire
forearm to check for associated injuries. The urgency of
surgery was analyzed in a recent study of 158 well-

705

Orthopaedic Knowledge Update 8

Elbow: Pediatrics

Figure 1 Two pinning techniques for displaced supracondylar humerus fractures are
cross pins (A) and divergent lateral entry pins (B). These techniques both have been
shown to be clinically effective and equally stable to extension varus and valgus testing. The lateral pins do not put the ulnar nerve at risk for injury.
perfused type III fractures. The authors found no correlation between delay to surgery and the complication
rate. However, children with a type III supracondylar
fracture that will be pinned should be admitted to the
hospital and carefully monitored for signs of neurovascular compromise or compartment syndrome.
Before fracture reduction, any entrapped soft tissue
should be dislodged by gently milking the brachialis
muscle and other soft tissue from the fracture site. The
fracture is reduced and held in hyperflexion while percutaneous pins are placed. Recent reports of iatrogenic
ulnar nerve injury from the use of a medial pin have focused attention on using only lateral entry pins. Although cadaver models have not shown such a technique to be as biomechanically stable as the use of
crossed pins, lateral entry pinning combined with casting
has been shown to be clinically effective. In a recent
biomechanical study in a synthetic bone model, divergent lateral pins had similar stability to crossed pins on
tests of extension and on tests of varus and valgus alignment, but not for axial stress (Figure 1). If there is concern for instability after placing two divergent lateral
entry pins, a third lateral pin can be added. The fracture
fixation should be tested for stability after pinning by
moving the elbow through a range of flexion and extension, and by carefully stressing the fracture in rotation
and varus and valgus, before splinting or casting in 70
to 80 of elbow flexion. The percutaneous pins are removed if radiographs show satisfactory callus at 3- to
4-week follow-up.

Neurovascular Compromise
Vascular compromise has been documented in about
10% to 20% of type III extension supracondylar humerus fractures. In these injuries, the brachial artery can
sustain an injury ranging from a minor intimal tear to
complete arterial disruption. Because there is a rich collateral circulation around the elbow, complete distal is-

706

chemia is rare. The absence of a palpable pulse is less

important than the integrity of distal perfusion. An initial angiogram is not warranted because the location of
the injury is known to be the fracture site. When treating a supracondylar humerus fracture without a palpable pulse, the first step is anatomic closed reduction with
percutaneous pinning. Usually, a pulse that can be detected by Doppler ultrasonography will be present
shortly after pinning. If this is the case, the fracture
should be splinted in 40 to 50 of flexion and the child
should be admitted to the hospital for 24 hours of observation. If the hand remains poorly perfused after anatomic reduction and pinning, a vascular consultation is
warranted.
Volkmanns ischemic contracture is now a rare complication. A recent study evaluated forearm compartment pressures in supracondylar humerus fractures in
29 children. Pressures in the deep volar compartment
were significantly more elevated than in other compartments and fracture reduction did not have an immediate
effect on pressures. Most importantly, the investigators
found that compartment pressures rose with elbow flexion greater than 90. Also, deep volar compartment
pressures greater than 30 mm often existed without clinical symptoms and without the subsequent development
of compartment syndrome. The orthopaedist should be
alert for forearm pain after a supracondylar humerus
fracture, or for any indication of worsening pain (or analgesic requirements) in a patient after closed reduction
and percutaneous pinning.
A neurologic deficit is noted in about 20% of patients with supracondylar humerus fractures. The anterior interosseous nerve is the most common site of
nerve injury in extension injuries. In the flexion supracondylar pattern, the ulnar nerve is the most likely to be
injured. Most nerve injuries associated with supracondylar humerus fractures will resolve within 6 months. If an
iatrogenic ulnar nerve injury is noted after use of a medial pin, some investigators have noted that no intervention is warranted; satisfactory resolution of the injury
often occurs. However, most investigators believe that
upon recognition, the offending medial pin should be
removed and replaced to a safe position.

Malunion
The instability after reduction of severely displaced supracondylar fractures results from failure to obtain good
apposition of the distal medial and lateral columns.
There is posterior rotation of the distal fragment, then a
tilt into varus. Although the deformity may become increasingly apparent as elbow extension is regained, the
deformity does not truly worsen over time. Although cubitus varus was considered primarily a cosmetic problem, studies have shown an increased incidence of ulnar
neuropathy, late lateral condyle fractures, and postero-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 60 Elbow: Pediatrics

Figure 2 Lateral condyle stages of displacement. A, Type I: A fracture in which

the fracture line does not completely
cross the cartilaginous surface of the distal humerus. B, Type II: A complete fracture of the cartilaginous surface without
displacement from the joint. C, Type III: A
complete fracture with displacement of
the lateral condylar fragment. (Adapted
with permission from Jakob R, Fowles J,
Rang M, et al: Observations concerning
fractures of the lateral humeral condyle in
children. J Bone Joint Surg Br 1975;57:
430-436.)
lateral rotatory instability with this condition. Most osteotomy techniques and fixation strategies have been
described in small studies. A recent report describes the
use of an external fixator to stabilize the fragment after
an oblique, incomplete medial opening wedge osteotomy. The complication rate for all techniques tends to
be quite high. Currently, there is no clear standard technique for the treatment of cubitus varus after a pediatric
supracondylar humerus fracture.

Lateral Condyle Fractures

Because lateral humeral condyle fractures in children
involve both the joint and growth cartilage, they are associated with a host of complications including nonunion, malunion, loss of motion, osteonecrosis, and cubitus valgus. The most clinically valuable classification
system involves stages of displacement (Figure 2). The
oblique radiograph is most helpful in assessing the extent of displacement. A recently published study of the
MRI evaluation of lateral condyle fractures with 0 to
3 mm of displacement confirmed previous arthrogram
studies and concluded that the stability of the fracture is
related to the integrity of the cartilage hinge. Lateral
condyle fractures with up to 2 mm of displacement can
be treated with casting and close observation. An AP,
lateral, and oblique radiograph of the distal humerus
with the cast removed should be obtained after about
5 days and again at about 2 weeks to ensure that there
is no progressive displacement. If displacement is seen,
reduction and pinning is the recommended treatment.
Closed reduction and percutaneous pinning is the
preferred treatment for fresh fractures that are only displaced from 2 to 5 mm, are relatively stable, and can be
reduced to the anatomic position. Open reduction and
internal fixation is usually required in the treatment of
lateral condyle fractures. The technique involves a standard lateral approach to the distal humerus with no posterior dissection (the blood supply to the lateral condyle
is posterior). An anatomic reduction is ensured both visually and radiographically and two percutaneous Kir-

American Academy of Orthopaedic Surgeons

schner wires are placed to stabilize the fracture. The

pins are removed at 4 weeks, followed by 2 additional
weeks of casting. Pins left in for longer than 4 weeks are
increasingly likely to lead to irritation or infection. A recent study evaluated the rate of healing in 55 patients
with lateral condyle fractures. The authors found that
radiographic evidence of complete bony healing was not
present until 6 weeks after injury, even in children between 1 and 5 years of age.
Lateral condyle fractures are one of the few types of
fractures in children in which delayed union or nonunion is not unusual. If open reduction to an anatomic
realignment is done more than 3 weeks after injury, osteonecrosis may result if extensive soft-tissue stripping
is performed to achieve the reduction. In a recent review of 11 children who had open reduction and internal fixation without soft-tissue stripping, none of the
children had osteonecrosis. The authors recommended
that nonunions with greater than 1 cm of displacement
be fixed without soft-tissue stripping and without restoration to the anatomic position. It was found that despite fixation in a nonanatomic position, overall elbow
alignment and range of motion were good. Late lateral
condyle nonunion is asymptomatic in some patients,
whereas others develop cubitus valgus and a delayed ulnar nerve palsy. Some authors have recommended a supracondylar medial closing wedge osteotomy and transposition of the ulnar nerve for symptomatic patients.

Medial Epicondyle Fractures

Fractures of the medial epicondyle account for 7% to
11% of all pediatric elbow fractures. These fractures result from avulsion of the epicondyle by the flexorpronator muscles and medial collateral ligament during
valgus stress or elbow dislocation. In patients with elbow dislocation where spontaneous or manipulative reduction entraps the medial epicondyle in the joint, the
medial epicondyle ossification center may be obscured
by the distal humerus or mistaken as the trochlear ossification center. Careful evaluation of radiographs is im-

707

Elbow: Pediatrics
portant whenever a medial elbow injury is suspected
in a child.
Treatment recomendations for medial epicondyle
fractures are somewhat controversial. Reduction and
pin or screw fixation of fractures displaced more than 2
to 5 mm is recommended in three circumstances: if a
medial epicondyle is entrapped in the joint; in a fracture
with associated ulnar nerve dysfunction; or for displaced
fractures in children who place high physical demands
on their elbow. The latter group is difficult to define;
however, gymnasts and throwing athletes are included.
A recent study retrospectively reviewed the results (at
more than 30-year follow-up) of 42 patients with medial
epicondyle fractures displaced more than 5 mm. Patients treated with cast immobilization had similar functional results to those treated surgically, although radiographic union was much more common in the group
treated surgically. Those treated with excision did
poorly. Based on these findings, the authors questioned
surgical treatment of fractures displaced as much as 15
mm, and did not recommend primary surgical excision.

Medial Condyle Fractures

True medial condyle fractures are rare, accounting for
only 1% to 2% of pediatric elbow fractures. Diagnosis
may be difficult in the young child, especially before the
appearance of the trochlear ossification center. However, the fracture typically involves a metaphyseal bony
fragment, clearly differentiating it from a medial epicondyle fracture. Treatment principles for medial
condyle fractures are similar to those for lateral condyle
fractures. Nondisplaced fractures may be treated with
cast immobilization, whereas displaced fractures should
undergo closed or open reduction with internal fixation.
Arthrography or MRI may be helpful in determining articular displacement and guiding the choice of treatment. A recent study of 21 patients with medial condyle
fractures showed that such fractures are associated with
a high rate of complications, including delayed or
missed diagnosis, osteonecrosis of the trochlea, nonunion, fixation failure, and significant elbow stiffness.

Olecranon Fractures
Olecranon fractures account for approximately 4% to
6% of pediatric elbow fractures and are associated with
other elbow fractures in up to 41% of patients. Olecranon fractures in children are similar to those in adults,
in that they are usually intra-articular and metaphyseal.
However, unlike in adults, many olecranon fractures in
children are minimally displaced and suitable for closed
immobilization. Fractures with more than 2 mm of articular displacement should be reduced and stabilized with
internal fixation. A recent report described a new fixation technique using a pin with a threaded tip and an
adjustable locking device at the opposite end. This tech-

708

Orthopaedic Knowledge Update 8

nique was biomechanically equivalent to tension band
wiring without requiring exposure of the fracture site. A
recent review of 39 patients with pediatric olecranon
fractures with a mean follow-up of 32 years found encouraging results. Poor results were seen in only three
patients, two of whom had associated radial head fractures. The presence of associated injury may be more
important than the degree of displacement of the olecranon fracture in determining outcome.

Proximal Radius Fractures

Although not a common injury, fractures of the proximal radius in children result in many complications. In
contrast to similar fractures in adults, however, fractures
through the neck of the radius are more common than
radial head fractures. A variety of treatment algorithms
exist for radial neck fractures based on the degree of
displacement. Radial neck fractures with less than 3 mm
of translation and less than 30 of angulation can be
treated with a brief period of immobilization for comfort, followed by early motion. More displaced or angulated fractures should be reduced. Closed reduction is
often successful; if it fails, reduction with the use of a
percutaneous Kirschner wire manipulation can be done.
If these methods of reduction fail, and angulation is
greater than 60, open reduction should be performed.
Fixation is usually achieved with the use of Kirschner
wires across the fracture but not the joint. A similar algorithm was used to treat 24 patients in a recent study.
All of the fractures were reduced to less than 15 of angulation, and all patients had excellent functional and
radiographic results at 14- to 25-year follow-up.
Radial head fractures, unlike radial neck fractures,
are intra-articular and deserve anatomic alignment.
Open reduction is often required to achieve this goal.

Monteggia Fractures
Monteggia lesions include a fracture of the proximal
ulna with a radial head dislocation. The ulnar fracture
also may be a plastic deformation. A Monteggia fracture
should be suspected whenever an isolated ulna fracture or radial head dislocation is seen. In a child, closed
reduction of the ulna usually reduces the radial head,
and the fracture often can be treated with a long arm
cast. However, open reduction and internal fixation of
the ulna is required when an anatomic reduction of the
ulna, radiocapitellar joint, and proximal radioulnar joint
cannot be obtained or maintained by closed methods.
Previous reports have discouraged attempted open
reduction of a chronic radial head dislocation with or
without annular ligament reconstruction because of the
high rate of redislocation and other complications. A recent study reviewed 22 patients treated for chronic posttraumatic radial head dislocation. The authors concluded that ulnar osteotomy to correct the malunion

American Academy of Orthopaedic Surgeons

Chapter 60 Elbow: Pediatrics

Orthopaedic Knowledge Update 8

Figure 3 A schematic drawing of the distal humeral proximal radius and ulnar relationship in various pediatric elbow injuries. A, Normal elbow. B, Elbow dislocation. C, Lateral
condyle fracture. D, Supracondylar fracture. E, Distal humeral epiphyseal separation. (Adapted with permission from Delee JC, Wilkins KE, Rogers LF, Rockwood CA: Fractureseparation of the distal humeral epiphysis. J Bone Joint Surg Am 1980;62:46-51.)

Displaced fractures through the distal humeral physis

can be difficult to diagnose. In the infant, the only sign
may be malalignment of the radial and ulnar shafts relative to the distal humerus. This injury is often associated
with battered child syndrome (Figure 3). Ultrasound,
MRI, arthrography, and stress radiographs can help clarify the diagnosis of a transphyseal fracture. Treatment is
similar to that for supracondylar fractures.

injuries. Noninvasive measures such as splinting and

physical therapy are successful in many patients. The
technique of open capsular release has been used extensively in adults; some authors have reported its use in
children. A report on elbow release in 13 adolescent patients with posttraumatic stiffness showed than an average increase of 54 was produced in the flexion extension arc. Only one patient lost motion. Nine patients
(69%) achieved at least a 25 to 120 arc of motion. A
recent review of 37 patients receiving similar treatment
showed only a 25 to 30 improvement in arc of motion;
a useful (30 to 130) arc was achieved only in 46% of
patients. After an extended period of observation, surgical release can be a valuable tool in the treatment of
stiffness after complex pediatric elbow injuries. The patient and family should be counseled regarding expected
improvements in arc of motion.

Floating Elbow

Annotated Bibliography

Ipsilateral fractures of the distal humerus and distal

forearm are called pediatric floating elbow. Commensurate with the severity of the trauma is an increased risk
of clinically significant swelling, neurovascular compromise, and forearm compartment syndrome. In a recent
study of 16 patients with a supracondylar humerus fracture and a displaced distal radius fracture, surgical
fixation of both fractures to avoid the high rate of potentially devastating complications (particularly compartment syndrome) was recommended.

Supracondylar Fractures

was essential in treating missed Monteggia fractures, if a

normal radial head contour remained. A similar report
of 15 patients who underwent open reduction of chronic
radial head dislocations and ulnar osteotomies found
good to excellent functional results in 12 patients, although most lost some degree of pronation postoperatively.

Transphyseal Distal Humerus Fractures

Elbow Contractures
Although children are generally less prone to posttraumatic joint stiffness than adults, not all children develop
normal or even functional range of motion after elbow

American Academy of Orthopaedic Surgeons

Battaglia TC, Armstrong DG, Schwend RM: Factors affecting forearm compartment pressures in children with
supracondylar fractures of the humerus. J Pediatr
Orthop 2002;22:431-439.
The authors measured the forearm compartment pressures
of 29 children with a type II or III supracondylar humerus
fracture. Pressures in the deep volar compartment were significantly elevated compared with the pressure in other compartments. There were also significantly higher pressures closer to
the elbow within each compartment. Fracture reduction did
not have an immediate effect on pressures. Most importantly,
flexion beyond 90 produced significant pressure elevation.
The authors also found that pressures greater than 30 mm Hg
may exist without clinical evidence of compartment syndrome.

709

Elbow: Pediatrics
They concluded that to avoid unnecessary elevation of pressures, elbows should not be immobilized in greater than 90 of
flexion.

De Boeck H: Flexion-type supracondylar elbow fractures in children. J Pediatr Orthop 2001;21:460-463.

Twenty-nine children with a flexion-type supracondylar
humerus fracture were reevaluated with an average follow-up
of 6.3 years. Seven children with a flexion-type supracondylar
humerus fracture were treated in a cast and 22 children with a
displaced flexion-type fracture were treated with closed reduction and percutaneous pinning. Good or excellent results were
noted in 86.2% of the patients. All patients were satisfied with
the end result and had normal use of the elbow. This article reports on a large number of relatively rare flexion-type supracondylar fractures and highlights the good results that can be
obtained without an open reduction, even in patients with displaced and unstable fractures.

Koch PP, Exner GU: Supracondylar medial open wedge

osteotomy with external fixation for cubitus varus deformity. J Pediatr Orthop B 2003;12:116-122.
The authors describe a technique in which an anteromedial approach and a medial opening wedge osteotomy with external fixation are used to treat cubitus varus in children after
a malunion and following a supracondylar humerus fracture.
The mean valgus correction was 21.75. At 2-year follow-up,
three of the four children had symmetric elbow alignment and
one had slight residual varus deformity. The authors concluded
that this is a good technique for treating cubitus varus deformity allowing correction of both valgus and flexion.

Lee ST, Mahar AT, Meisen D, Newton PO: Displaced

pediatric supracondylar humerus fractures: Biomechanical analysis of percutaneous pinning techniques.
J Pediatr Orthop 2002;22:440-443.
This biomechanical study of different supracondylar pinning techniques used a pediatric synthetic bone model. The
authors concluded that divergent lateral pins had similar stability compared with cross pins in extension, varus, and valgus
testing. In axial testing, cross pins were more stable. Divergent
lateral pins also had statistically greater stability than parallel
lateral pins under varus and valgus loading.

Leet AI, Frisancho J, Ebramzadeh, E: Delayed treatment of type 3 supracondylar humerus fractures in children. J Pediatr Orthop 2002;22:203-207.

Orthopaedic Knowledge Update 8

The authors report on 25 instances of cubitus varus deformity (average 15) subsequent to a pediatric distal humeral
fracture or congenital anomaly in 24 patients who developed
posterolateral rotatory instability of the elbow two to three
decades afterward. Surgery was performed on 22 limbs, and
only 3 limbs had persistent instability at 3-year follow-up. The
authors propose that the repetitive external rotation torque on
the ulna, resulting from the medial displacment of the mechanical axis, the olecranon, and the triceps line of pull, can
stretch the lateral collateral ligament complex and lead to posterolateral rotatory instability in patients with cubitus varus
deformity. Surgical correction can relieve symptoms of instability.

Skaggs DL, Cluck MW, Mostofi A, Flynn JM, Kay RM:

Lateral-entry pin fixation in the management of supracondylar fractures in children. J Bone Joint Surg Am
2004;86:702-707.
The authors reported on 124 consecutive Gartland types II
and III supracondylar humerus fractures treated with two or
three lateral entry pins. The only complication reported was a
pin tract infection. There was no loss of reduction, cubitus
varus, hyperextension, loss of motion, or iatrogenic nerve palsies, and no additional surgeries. The authors concluded that
the most important technical points for assuring fixation with
lateral-entry pins were: separating the pins at the fracture site;
engaging the medial and lateral columns proximal to the fracture and sufficient bone in both the proximal segment and the
distal fragment; and using three pins to provide additional stability when necessary.

Lateral Condyle Fractures

Cardona JI, Riddle E, Kumar SJ: Displaced fractures of
the lateral humeral condyle: Criteria for implant removal. J Pediatr Orthop 2002;22:194-197.
The authors reviewed 55 lateral condylar fractures treated
at their institution over a 10-year period, with special attention
to the rate of healing of the fracture. They concluded that in
50 patients, the fracture was clinically and radiographically
healed at 6 weeks. Four patients showed radiographic healing
at 5 weeks and one patient at 4 weeks.

Horn DH, Herman MJ, Crisci K, Pizzutillo PD,

MacEwan GD: Fracture of the lateral humeral fracture
condyle: Role of the cartilage hinge in fracture stability.
J Pediatr Orthop 2002;22:8-11.

A retrospective review of 158 patients with well-profused

type III supracondylar humerus fractures showed no correlation between an increase in time to surgical intervention and
the complication rate. The average time from injury to surgical
treatment was 21.3 hours.

The authors studied 16 patients with a lateral condyle fracture who had both radiographs and MRI. All unstable fractures had complete fractures on MRI scans. Ten of the 12 patients with radiographically stable injuries had incomplete
fractures on MRI scans. The authors concluded that the stability of the lateral humeral condyle fractures is related to the integrity of the cartilage hinge.

ODriscoll SW, Spinner RJ, McKee MD, et al: Tardy posterolateral rotatory instability of the elbow due to cubitus varus. J Bone Joint Surg Am 2001;83:1358-1369.

Wattenbarger JM, Gerardi J, Johnston CE: Late open

reduction internal fixation of lateral condyle fractures.
J Pediatr Orthop 2002;22:394-398.

710

American Academy of Orthopaedic Surgeons

Chapter 60 Elbow: Pediatrics

Orthopaedic Knowledge Update 8

The authors reviewed the records of 11 children who had a
malunion or nonunion of a lateral condyle fracture treated
more than 3 weeks after injury and found no patients with osteonecrosis. In fractures with greater than 1 cm of displacement, the fragment position was minimally improved by surgical treatment, but final alignment and range of motion were
good. The authors concluded that the risk of osteonecrosis
with open reduction of lateral condyle fractures at greater
than 3 weeks is reduced if no tissue is stripped off the fracture
fragment posteriorly. Even children without an anatomic reduction had functional arms and little or no pain.

Medial Epicondyle Fractures

Farsetti P, Potenza V, Caterini R, Ippolito E: Long-term
results of treatment of fractures of the medial humeral
epicondyle in children. J Bone Joint Surg Am 2001;83-A:
1299-1305.
The authors retrospectively reviewed 42 patients with medial epicondyle fractures that were displaced more than 5 mm.
Nineteen patients were treated with cast immobilization without reduction, 17 underwent open reduction and internal fixation, and 6 patients had fragment excision. The functional results at 30-year follow-up were identical for the first two
groups, but worse for the excision group. The authors argue
that nonsurgical treatment is as effective as open reduction
and internal fixation for fractures with displacement of more
than 5 mm.

Medial Condyle Fractures

Leet AI, Young C, Hoffer MM: Medial condyle fractures
of the humerus in children. J Pediatr Orthop 2002;22:2-7.
The authors reviewed a series of 21 children with medial
condyle fractures. MRI was used to make the diagnosis or assess displacement in four children. Of the nine patients with
displaced fractures that were treated surgically, four had complications, including two patients who lost reduction and required revision with open reduction and internal fixation. Of
the 12 patients who were treated in a cast, 2 developed nonunions and 1 developed osteonecrosis of the trochlea. Almost
half of the children had some impairment in range of motion.

Olecranon Fractures
Caterini R, Farsetti P, DArrigo C, Ippolito E: Fractures
of the olecranon in children: Long-term follow-up of 39
cases. J Pediatr Orthop B 2002;11:320-328.
Thirty-nine patients with olecranon fractures in childhood
were followed at a mean of 32 years. Thirty-four patients had
good results and three had poor results. All had minimally displaced olecranon fractures and other associated fractures. Arthritis was rare. The authors concluded that associated injuries
may be a stronger predictor of outcome than the degree of
displacement in patients with olecranon fractures.

Gicquel P, Maximin MC, Boutemy P, et al: Biomechanical analysis of olecranon fracture fixation in children.
J Pediatr Orthop 2002;22:17-21.

American Academy of Orthopaedic Surgeons

The authors describe a novel fixation technique for olecranon fractures in children. A pin with a threaded tip is inserted
antegrade across the fracture site, without opening the fracture. A locking device is placed proximally on the pin to provide compression. A biomechanical analysis found this technique as strong as tension band wiring. The authors
recommended this new technique over tension band wiring
because it obviates the need to expose the fracture site.

Proximal Radius Fractures

Malmvik J, Herbertsson P, Josefsson PO, Hasserius R,
Besjakov J, Karlsson MK: Fracture of the radial head
and neck of Mason types II and III during growth: A
14-25 year follow-up. J Pediatr Orthop B 2003;12:63-83.
Twenty-four patients with angulated radial head and neck
fractures during childhood were evaluated at a mean
follow-up of 19 years. Two patients ultimately required radial
head excision. Of the remainder, all but three patients had received closed treatment. All patients had less than 15 of angulation after primary treatment. Nineteen patients had no
pain at follow-up, although elbow flexion was typically limited
compared with the uninjured side.

Waters PM, Stewart SL: Radial neck fracture nonunion

in children. J Pediatr Orthop 2001;21:570-576.
The authors describe nine children with radial neck nonunions; eight had associated elbow injuries. Eight children underwent open reduction and fixation, and one had repeated
closed reduction attempts with percutaneous fixation. In each
case, fixation failed and led to loss of reduction. Four patients
were pain free, but all had restriction of motion. Secondary
open repair of the nonunion did not guarantee good functional outcome.

Monteggia Fractures
Horii E, Nakamura R, Koh S, Inagaki H, Yajima H, Nakao E: Surgical treatment for chronic radial head dislocation. J Bone Joint Surg Am 2002;84-A:1183-1188.
The authors reviewed 22 patients treated for chronic posttraumatic radial head dislocation. Thirteen patients underwent
open reduction without ulnar osteotomy, and seven had redislocation of the radial head. In a subsequent group of nine patients who had an ulnar osteotomy, the radial heads of two patients were subluxated at follow-up. Both patients had radial
head abnormalities preoperatively. The authors concluded that
ulnar osteotomy to correct malunion is essential in treating
missed Monteggia fractures if a normal radial head contour
remains.

Kim HT, Park BG, Suh JT, Yoo CI: Chronic radial head
dislocation in children: Part 2. Results of open treatment
and factors affecting final outcome. J Pediatr Orthop
2002;22:591-597.
The authors describe 15 cases (14 patients) of open reduction and reconstruction of chronic radial head dislocations, 12
of which were posttraumatic. Results were classified as excellent in 10 elbows, good in 2, fair in 2, and poor in 1. One pa-

711

Elbow: Pediatrics
tient developed a radioulnar synostosis, and 10 lost some degree of pronation postoperatively. All radial heads in the
posttraumatic patients remained reduced at follow-up. The authors argue that the results of their series should encourage attempts at reducing chronically dislocated radial heads, especially in posttraumatic patients.

Floating Elbow
Ring D, Waters PM, Hotchkiss RN, Kasser JR: Pediatric
floating elbow. J Pediatr Orthop 2001;21:456-459.
The authors reviewed a series of 16 patients with pediatric
floating elbow injuries. All but one patients supracondylar humerus fracture was treated with pin fixation. In 10 patients, the
distal forearm fractures were treated with closed reduction
and cast immobilization. Of these 10 patients, 2 developed
compartment syndrome with 1 patient having a subsequent
Volkmann ischemic contracture. An additional 4 of these 10
patients required cast removal because of excessive swelling
and impending compartment syndrome. The remaining six patients were treated with pin fixation of their distal forearm
fractures, and were not placed in circumferential casts. None
developed compartment syndrome. The authors recommended
surgical treatment of all fractures in this injury pattern to
avoid the high rate of potentially devastating complications associated with cast immobilization.

Elbow Contractures
Bae DS, Waters PM: Surgical treatment of posttraumatic
elbow contracture in adolescents. J Pediatr Orthop 2001;
21:580-584.
The authors describe 13 adolescents who had elbow capsular release for posttraumatic stiffness. An average increase of
54 was achieved in the flexion-extension arc. Nine patients
(69%) achieved at least a 25 to 120 arc of motion; one patient lost motion. The authors expressed enthusiasm for surgical release of elbow contractures in adolescents.

Stans AA, Maritz NG, ODriscoll SW, Morrey BF: Operative treatment of elbow contracture in patients 21years of age or younger. J Bone Joint Surg Am 2002;84A:382-387.
The authors describe surgical elbow releases in 37 patients
younger than 21 years of age. Twenty-eight elbow contractures
were posttraumatic. An average increase of only 25 to 30 of
motion was obtained in the flexion-extension arc, regardless of
the etiology of the contracture. Fewer than half of the patients
achieved a flexion extension arc of greater than 30 to 130.
The authors warn of the limited gains that can be expected
from elbow releases in children compared with those in adults.

Classic Bibliography
Archibeck MJ, Scott SM, Peters CL: Brachialis muscle
entrapment in displaced supracondylar humerus fractures: A technique of closed reduction and report of initial results. J Pediatr Orthop 1997;17:298-302.

712

Orthopaedic Knowledge Update 8

Bernstein SM, McKeever P, Bernstein L: Percutaneous
reduction of displaced radial neck fractures in children.
J Pediatr Orthop 1993;13:85-88.
Cramer KE, Green NE, Devito DP: Incidence of anterior interosseous nerve palsy in supracondylar humerus
fractures in children. J Pediatr Orthop 1993;13:502-505.
Culp RW, Ostermann AL, Davidson RS, Skirven T, Bora
FW Jr: Neural injuries associated with supracondylar
fractures of the humerus in children. J Bone Joint Surg
Am 1990;72:1211-1215.
Flynn JC: Nonunion of slightly displaced fractures
of the lateral humeral condyle in children: An update.
J Pediatr Orthop 1989;9:691-696.
Foster DE, Sullivan JA, Gross RH: Lateral humeral
condylar fractures in children. J Pediatr Orthop 1985;5:
16-22.
Fowles JV, Slimane N, Kassab MT: Elbow dislocation
with avulsion of the medial humeral epicondyle. J Bone
Joint Surg Br 1990;72:102-104.
France J, Strong M: Deformity and function in supracondylar fractures of the humerus in children variously
treated by closed reduction and splinting, traction, and
percutaneous pinning. J Pediatr Orthop 1992;12:494-498.
Gaddy BC, Strecker WB, Schoenecker PL: Surgical
treatment of displaced olecranon fractures in children.
J Pediatr Orthop 1997;17:321-324.
Garbuz DS, Leitch K, Wright JG: The treatment of supracondylar fractures in children with an absent radial
pulse. J Pediatr Orthop 1996;16:594-596.
Keenan WN, Clegg J: Variation of Baumanns angle with
age, sex and side: Implications for its use in radiological
monitoring of supracondylar fracture of the humerus in
children. J Pediatr Orthop 1996;16:97-98.
Kissoon N, Gapin R, Gayle M, Chacon D, Brown T:
Evaluation of the role of comparison radiographs in the
diagnosis of traumatic elbow injuries. J Pediatr Orthop
1995;15:449-453.
Lincoln TL, Mubarek SJ: Isolated traumatic radial head
dislocation. J Pediatr Orthop 1994;14:454-457.
Pirone AM, Graham HK, Krajbich JI: Management of
displaced extension-type supracondylar fractures of the
humerus in children. J Bone Joint Surg Am 1988;70:641650.
Roye DP Jr, Bini SA, Infosino A: Late surgical treatment of lateral condylar fractures in children. J Pediatr
Orthop 1991;11:195-199.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Sabharwal S, Tredwell SJ, Beauchamp RD, et al: Management of pulseless pink hand in pediatric supracondylar fractures of the humerus. J Pediatr Orthop 1997;17:
303-310.
Williamson DM, Coates CJ, Miller RK, Cole WG: Normal characteristics of the Baumann (humerocapitellar)

American Academy of Orthopaedic Surgeons

Chapter 60 Elbow: Pediatrics

angle: An aid in assessment of supracondylar fractures.
J Pediatr Orthop 1992;12:636-639.
Zionts LE, McKellop HA, Hathaway R: Torsional
strength of pin configuration used to fix supracondylar
fractures of the humerus in children. J Bone Joint Surg
Am 1994;76:253-256.

713

Chapter

61

Forearm, Wrist, and Hand:

Pediatrics
Peter M. Waters, MD
Alexander D. Mih, MD

Cerebral Palsy
Deformities of the forearm, wrist, and hand, limited
function, and decreased sensibility are common in patients with cerebral palsy. Limited motor function occurs
with poor release and grasp function resulting from
flexor spasticity and contractures, combined with the effects of weak extension of the fingers and wrist. Limited
pinch from thumb-in-palm deformity occurs because of
intrinsic adductor and flexor spasticity and contractures.
Discriminatory sensibility is deficient in more than 50%
of these children. Poor voluntary control of the upper
extremity limits functional placement of the hand in
space; many of these children have visual and cognitive
abnormalities that further impair hand function. At best,
most patients with cerebral palsy have assistive hand
function.
Upper extremity classification systems for patients
with cerebral palsy have been used to assess function.
The House classification of function has nine levels extending from 0 (no use of the extremity) to 8 (complete
spontaneous use). In this schema, the nine levels are further classified into four subgroups based on patient
function: no use (level 0), passive assist (levels 1 to 3),
active assist (levels 4 to 6), and spontaneous use (levels
7 and 8) (Table 1). Because spasticity varies with stress,
growth, and central nervous system changes, it may be
difficult to accurately define a patients level of function
based on any single observation. Surgical planning and
outcome assessment should be based on the level of
function determined preoperatively and postoperatively.
Surgery will not create a normal hand but will improve
assistive function and cosmesis. The goals of surgery
need to be realistic and obtainable.

Treatment
In broad terms, treatment options include observation
of the patients growth and development, use of therapy
(including splints), consideration of the need for injections (such as phenol or botulinum toxin), and surgical
reconstruction of the forearm, wrist, and hand. Botulinum toxin is currently the most common form of neuro-

American Academy of Orthopaedic Surgeons

muscular blockade injection. Injections into the pronator, flexor carpi ulnaris, and adductor pollicis muscles
are most often performed. Aggressive therapy should be
used to stretch agonistic muscle-tendon units and
strengthen antagonists. Antibody formation to botulinum toxin will limit its effectiveness in certain patients.
In patients with cerebral palsy, surgery improves
level of function and cosmesis in the hemiplegic patient,
and ease of nursing care in the patient with quadriparesis while lessening the risk of skin breakdown. The best
surgical candidates are patients with hemiplegia and
good voluntary control, sensibility, and motivation. The
principle of surgery is to correct muscle imbalance by
lengthening or releasing tight, spastic muscles and augmenting weak, stretched muscles by tendon transfers
and tenodesis procedures. Unstable joints need to be
stabilized by soft-tissue or arthrodesis procedures to
maximize outcome of tendon reconstruction. The patient and family must understand that surgery will not
alleviate all functional deficiency or repair all cosmetic
defects of the hand. Even the best outcome will still result in deficiencies of function, cosmesis, and sensibility.
However, in properly selected patients, surgery will
clearly improve function and patient satisfaction.
Forearm hyperpronation significantly limits hand
function in patients with hemiplegia. Release or rerouting of the pronator teres through the interosseous membrane is effective in improving function. Transfer of the
flexor carpi ulnaris to the dorsal wrist by rerouting
around the ulna also has been shown to provide some
degree of active supination. By improving voluntary
control of forearm rotation, hand function increases.
Wrist and finger flexion deformity is common in patients with hemiplegia. The flexor carpi ulnaris is usually
the major deforming force resulting in wrist flexion.
Transfer of the flexor carpi ulnaris to the wrist extensors
alleviates the deformity and improves wrist extension
and finger tenodesis into flexion. Simultaneous musculotendinous lengthening of the finger flexors are necessary if the extrinsic finger flexors are tight in neutral
wrist position; otherwise, the patient will develop a disabling clenched fist postoperatively. Z-lengthenings, su-

715

Orthopaedic Knowledge Update 8

Forearm, Wrist, and Hand: Pediatrics

Table 1 | House Classification for Hand Function in

Patients With Cerebral Palsy
Level

Designation

Activity Level

0
1
2

Does not use

Poor passive assist
Fair passive assist

Good passive assist

Poor active assist

Fair active assist

Good active assist

Spontaneous use, partial

Spontaneous use, complete

Does not use

Uses as stabilizing weight only
Can hold onto object placed
in hand
Can hold onto object and
stabilize it for use by other
hand
Can actively grasp object and
hold it weakly
Can actively grasp object and
stabilize it well
Can actively grasp object and
manipulate it against other
hand
Can perform bimanual activities easily and occasionally
uses the hand spontaneously
Uses hand completely independently without reference
to the other hand

(Reproduced with permission from Waters PM, Van Hest A: Spastic hemiplegia of the upper
extremity in children. Hand Clin 1998;14:119-134.)

struction, and resection of the motor branch of the ulnar

nerve. The lateral band rerouting procedure provides
both intrinsic and extrinsic rebalancing and is effective
in correcting swan neck deformities.
Patients with cerebral palsy who have disabling dynamic spasticity and fixed contractures of the wrist and
hand benefit from surgical reconstruction. Often the
more severely affected patients (House level 0 through
2) respond best to musculotendinous lengthenings, tenodesis, and joint stabilization procedures. Patients with
more functional ability (House levels 3 through 6) improve with dynamic tendon transfers and releases.

Arthrogryposis
Infants with classic arthrogryposis (amyoplasia) often
have stiffness and weakness of all joints and muscles of
the upper extremity. Elbow extension, forearm pronation, wrist palmar flexion, ulnar deviation, finger flexion,
and thumb-in-palm contractures are typical. Absence of
biceps antigravity strength is common and limits the
ability to place the hand near the face. Adaptive mechanisms are necessary for function. Children with arthrogryposis often have incomplete syndactyly of all web
spaces. Contracture of the first web space is usually
functionally significant. Patients usually have marked intrinsic muscle weakness.

Treatment
perficialis to profundus flexor tendon transfers, and
bony procedures are reserved for patients with severe
contractures and limited function (usually found in the
patient with quadriparesis).
Thumb-in-palm deformity limits dynamic pinch and
grasp and makes hygiene difficult because of severe
contractures. Static contractures are corrected with web
space Z-plasty and adductor releases. At times, the static
contractures include the flexor pollicis longus and
brevis; these muscles will need to be appropriately
lengthened or released. Dynamic rebalancing is performed with tendon transfers to weak abductors and extensors of the thumb. There are many possible donor
muscles including the palmaris longus, flexor carpi radialis, and brachioradialis. The recipient tendons include
the extensor pollicis brevis and longus and the abductor
pollicis longus. The metacarpophalangeal joint needs to
be stable postoperatively. In most patients, stability is
achieved by muscle rebalancing. On occasion, a capsulodesis or arthrodesis procedure will be needed.
Some patients with cerebral palsy have disabling
swan neck deformities. If the fingers at the proximal interphalangeal joint lock and hyperextend more than 40
and lock, limited grasp and pain may result. Multiple
surgical procedures have been advised, including flexor
digitorum superficialis tenodesis, intrinsic muscle slide,
lateral band rerouting, spiral oblique ligament recon-

716

Initial treatment involves passive range-of-motion therapy and nighttime splinting to improve joint motion and
digital strength. The condition of many children improves with growth and therapy over the first several
years of life. At the elbow, triceps V-Y lengthening and
posterior capsulectomy are performed at 18 months to 3
years of age if passive elbow flexion of approximately
90 does not occur. The wrist palmar flexion contracture
is treated with both soft-tissue and bony procedures. A
flexor carpi ulnaris release with lengthening or transfer
to the wrist extensors is performed in conjunction with a
dorsal carpal closing wedge osteotomy.
The thumb-in-palm contracture is treated with a
Z-plasty syndactyly release. Care must be taken not to
overrelease the adductor. Transfers for thumb abduction
and extension are predominantly tenodesis procedures
because of the limited strength of the donor muscles.
Children with arthrogryposis will have permanent
limited motion and strength in their arms and hands.
Because of their high level of intelligence, these children
often are quite functionally adaptive.

Forearm and Wrist Trauma

Epidemiology and Prevention Updates
In a longitudinal study from New Zealand, it was determined that 51% of boys and 40% of girls will experience a single fracture by 18 years of age. The most com-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

mon fracture sites, the wrist and forearm, accounted for
24% of all fractures. It was also found that 23% of boys
and 16% of girls had more than one fracture before age
18 years. In a population-based study from Wales, it was
noted that there was a seasonal variation in forearm and
wrist fractures in children and adolescents. The incidence of such fractures markedly increased in spring
and summer, as did the severity of fracture. Twentythree percent of patients with fractures required hospital admission in the spring and summer compared with
only 10% of patients who had fractures during the winter months. In a population-based, case control study
from Wales, a significant decrease in bone mass (measured by dual-energy x-ray absorptiometry and metacarpal morphometry) was found in children age 9 to 16
years, with wrist and forearm fractures compared with a
control group. Postural balance scores were noted to be
significantly poorer in patients with high body weight
and a history of previous forearm fracture compared
with a control group with normal body weight. The increased body mass index and poor balance appear to
contribute to the increased risk of forearm fracture in
these patients.
Participants in specific sports such as snowboarding,
horseback riding, and soccer (goal keeping) were noted
to have an increased risk of wrist fractures. In a retrospective review of skiers and snowboarders from Japan,
there was an increased incidence of upper limb injuries,
particularly elbow dislocations and wrist fractures, in
those who participated in snowboarding. In a randomized controlled study of snowboarders in Norway, it was
noted that wrist guards significantly lowered the incidence of wrist injuries without increasing the risks for
more proximal fractures or dislocations. Because of the
increased risk of head and wrist injuries in horseback
riding, and the decrease in severity and number of head
injuries with the use of protective headgear, it has been
proposed that wrist protectors may be of benefit in this
sport as well.
Ligamentous laxity correlates with the site of the
fracture. Patients with hyperlaxity are more likely to
have a supracondylar humerus fracture after a fall,
whereas children with less laxity are more likely to sustain distal radius fractures. In a study of distal radius
fractures in young soccer goalkeepers, the size of the
ball correlated with the risk of a distal radius fracture.

Forearm and Wrist Fractures

Fractures of the forearm are the most common long
bone fractures in children, accounting for 45% of all
childhood fractures and 62% of upper extremity fractures. Approximately 75% to 84% of these forearm
fractures occur in the distal third of the bone; 15% to
18% in the middle third; and less than 5% in the proximal third. Injuries to the distal radioulnar joint or carpal

American Academy of Orthopaedic Surgeons

Chapter 61 Forearm, Wrist, and Hand: Pediatrics

Figure 1 A radiograph showing intramedullary fixation of a diaphyseal radius and

ulna fracture. The ulnar pin is inserted proximally and the radial pin is inserted distally.
bones in children are much less common. Scaphoid fracture is the most common carpal bone fracture; however,
it accounts for only 0.45% of all upper extremity fractures in children.

Fractures of the Diaphyseal Radius and Ulna

Diaphyseal fractures are classified as plastic deformation, buckle or torus, greenstick or incomplete, and complete. Approximately 50% of diaphyseal fractures are
greenstick and most occur in children younger than 8
years of age. The purpose of closed reduction is to correct the malrotation first, then the malangulation.
Complete fractures are more common in preadolescent and adolescent patients and tend to be unstable;
these fractures often require internal fixation. There is
an increased incidence of loss of reduction. More recently, unstable diaphyseal forearm fractures have been
treated with intramedullary fixation. Elastic nails or Kirschner wires can be placed percutaneously after successful closed reduction. The ulnar pin is passed from proximal to distal, while the radial pin is placed from distal to
proximal with the entry site proximal to the distal radial
physis (Figure 1). Open reduction is indicated for irreducible fractures; unstable fractures, especially in the adolescent patient; displaced segmental fractures; fractures
associated with unstable Monteggia, Galeazzi, or supracondylar fractures; refracture; and for patients with a
dysvascular limb.

Fractures of the Distal Radius and Ulna

Metaphyseal Fractures
Seventy-five percent to 84% of pediatric forearm fractures involve the distal radius. Most of these radial fractures are metaphyseal and are associated with metaphyseal fractures of the ulna. Nondisplaced metaphyseal
torus fractures can be treated with brief immobilization.
Studies indicate that removal of the splint at home was
more satisfactory for families than return for doctorsupervised removal; the clinical outcome was the same.

717

Forearm, Wrist, and Hand: Pediatrics

Although closed reduction and long arm cast immobilization is indicated for fractures with greater than 10
of malalignment, several studies indicate a high incidence (up to 34%) of loss of reduction. Percutaneous
pin fixation avoids the need for rereduction of difficult
distal radius fractures but does carry the risks associated
with anesthesia, pin placement, possible infection, and
limited scarring. Two separate randomized trials of pinning compared with cast immobilization indicated that
pinning significantly reduced the risk of loss of reduction requiring remanipulaton; however, long-term clinical outcome was the same for both groups of patients.
Pinning is reasonable for fractures with initial angulation greater than 30 and displacement more than 50%
of the diameter of the radius. Remodeling of malunited
fractures in the flexion/extension plane will occur and,
to a lesser extent, remodeling will also occur in the radial deviation/ulnar deviation plane if sufficient gravity
remains. Rotational malunion will not remodel.
Percutaneous pinning is clearly indicated for fractures associated with neurovascular compromise and
significant soft-tissue swelling, and with concomitant
fractures of the elbow (floating elbow) because of the
increased risk for development of compartment syndrome in these patients. A smooth Kirschner wire is inserted obliquely from distal to proximal, starting in the
radial metaphysis just proximal to the physis. Care must
be taken to avoid the radial sensory nerve with insertion. This procedure may be done with a 5- to 10-mm incision and clearing of the pin tract. A second, crossing
Kirschner wire can be inserted from between the fourth
and fifth dorsal extensor compartments while avoiding
the extensor tendons (Figure 2).
Distal Radius Physeal Fractures
Most distal radius physeal fractures are Salter-Harris
type II injuries that occur in adolescents. Displacement
is typically dorsal with apex volar angulation. An atraumatic closed reduction and long arm cast immobilization is indicated for fractures with greater than 10 of
malangulation. The most significant complications are
associated with future growth arrest and acute neurovascular compromise. The incidence of associated
nerve injuries is approximately 8%. Studies indicate that
patients with signs or symptoms of median nerve injury
at the time of presentation would be best treated by
percutaneous pin fixation rather than cast immobilization to lessen the risk of forearm compartment syndrome, acute carpal tunnel syndrome, or median neuropathy. There is limited evidence of a significant risk of
growth arrest with the use of smooth, small diameter
pins. The pins should be removed at 3 to 4 weeks postoperatively to lessen the risk of growth arrest.
Long-term follow-up (average, 25.5 years) of 157 patients with distal radius and ulna physeal fractures who
were treated with closed reduction showed that 4.4% of

718

Orthopaedic Knowledge Update 8

Figure 2 A displaced distal radius metaphyseal fracture treated with percutaneous

fixation. Care must be taken to avoid the radial sensory nerve and, if possible, the
distal radial physis, with the radial pin. Care must be taken to avoid the extensor
tendons with the ulnar pin.
patients with distal radius fractures and 50% of patients
with distal ulna fractures had a symptomatic growth discrepancy greater than 1 cm. In addition, 38 asymptomatic
patients had growth discrepancies between 2 to 9 mm. Patients with redisplacement or late presentation should not
undergo a repeat reduction after 7 days to avoid iatrogenic physeal arrest. Late open reduction should be
avoided for the same reason. Because these fractures displace in the plane of motion of the wrist joint and are juxtaphyseal, there is tremendous potential for remodeling
of a malunion in the younger adolescent patient. If the
malunion does not remodel with growth, a dorsal opening
wedge osteotomy with bone graft and internal fixation
may be needed for the skeletally mature patient with
greater than 10 apex dorsal malangulation. Patients with
displaced physeal fractures should be examined for 1 to 2
years after the fracture occurrence to rule out a growth
disturbance. If a radial growth arrest occurs, the continued ulnar growth can lead to distal radioulnar joint incongruity, ulnocarpal impaction, and a triangular fibrocartilage tear (Figure 3).

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 61 Forearm, Wrist, and Hand: Pediatrics

Figure 3 A, Radiograph shows distal radial physeal arrest after repetitive closed reductions of a Salter-Harris type II fracture. Note the ulnar overgrowth with open physis and the
radial growth arrest. B, Clinical photograph of the patient showing the deformity resulting in ulnar-carpal impaction.

Hand Trauma
Fractures to the pediatric hand account for approximately one quarter of all childhood fractures and have
two peak periods of occurrencein adolescence (from
sport-related activities) and in infancy (from crush injuries). Most of these fractures are nondisplaced, nonphyseal injuries that do not have long-term consequences.
Physeal injuries account for up to 40% of finger fractures. A Salter-Harris type II fracture of the small finger
proximal phalanx is the most common physeal injury.
Distal tuft and phalanx fractures are the most common
fractures in infants and children up to age 8 years; proximal phalanx fractures of the small finger are most common in the 9- to 12-year age group; and fifth metacarpal
neck fractures have the highest incidence in patients age
13 to 16 years. Most pediatric hand fractures heal within
2 to 3 weeks and have excellent functional outcomes regardless of the type of treatment used. Malunions or
growth disturbances are rare. However, there is a subset
of pediatric hand fractures with comminution, severe
displacement, and intra-articular and condylar involvement, which will heal poorly if not recognized and appropriately treated. These fractures account for 12% to
20% of pediatric fractures in most studies.

Problematic Hand Fractures

Phalangeal neck fractures are usually caused by crush
injuries. As the child attempts to extract the affected
digit, the condyles become entrapped and a fracture occurs in the subcondylar region. The condylar fragment is

American Academy of Orthopaedic Surgeons

displaced in extension and often malrotates. The subcondylar fossa is obliterated, blocking interphalangeal
flexion. If not properly recognized and treated, complications of malunion and loss of motion occur. The severity of this fracture is often underappreciated in the urgent care setting. Treatment is with either closed
reduction and percutaneous pinning or by open reduction and internal fixation. If open reduction is necessary,
the collateral ligaments should not be dissected from
the distal fragment. This action increases the risk of osteonecrosis. Treatment for a late malunion includes osteotomy, subchondral fossa reconstruction, or remodeling. Remodeling of phalangeal neck malunions rarely
occurs because of the significant distance from the physis.
Osteochondral fractures in young children are often
challenging to treat. These fractures have a high risk of
nonunion, malunion, and osteonecrosis, which is particularly true in crush injuries to the middle phalanx that alter the local blood supply. The fracture is intra-articular,
generally displaced, and requires anatomic reduction for
a successful outcome. Most often, the fracture needs to
be treated aggressively with open reduction. Bone grafts
may be necessary to maintain articular congruity and
prevent collapse. Even with a well-performed open reduction, complications from osteochondral fractures can
occur in young children.
In the adolescent, treatment of intercondylar fractures is similar to that given to adults. Anatomic reduction and pin fixation is necessary to restore the joint surface and to prevent loss of reduction. Open reduction is

719

Orthopaedic Knowledge Update 8

Forearm, Wrist, and Hand: Pediatrics

appropriate for fractures that are not suitable for closed
reduction, and can be performed with a volar, midaxial,
or dorsal approach. Restoration of an anatomic joint
lessens the risk of long-term arthritis, malalignment, and
loss of motion.
The major complication of diaphyseal-level phalangeal fractures is malrotation. Frequently, children will
refuse to actively move the affected finger in the acute setting, thus preventing accurate assessment of digital alignment. However, possible tenodesis of the wrist enables the
physician to assess malrotation and should be performed
on all phalangeal and metacarpal fractures regardless of
radiographic appearance. If closed treatment is chosen, a
solitary finger should never be immobilized; it should be
secured to the adjacent digits to prevent subsequent loss
of reduction. If the fracture is malrotated and unstable, a
reduction pin or screw is necessary for stabilization. Although malrotation is uncommon, it can result in a major
complication if it is not discovered until after union takes
place. The malrotated digit impairs the function of the adjacent digits secondary to overlap, and can only be corrected with osteotomy.

Scaphoid Fractures
Distal pole scaphoid fractures in the skeletally immature patient heal readily with cast immobilization without risk of nonunion or osteonecrosis. Scaphoid waist
fractures are common in the adolescent age group. Recent studies have indicated that both adults and children
with clinical pain in the region of the scaphoid and normal radiographs may benefit from MRI, the results of
which will alter treatment choices in a high percentage
of patients. Waist fractures carry the same risks of nonunion and osteonecrosis in the child as they do in an
adult. Open reduction, bone grafting, and internal fixation should be used to treat an established nonunion in
a child. The epidemiology of scaphoid fractures continues to change; proximal pole fractures, nonunions, and
osteonecrosis now have been described in adolescents.
Treatment with distal radial vascularized bone graft has
been recommended for these unique complications. The
treatment of acute scaphoid fractures, even nondisplaced fractures, with percutaneous screw fixation is
controversial. The complication rates in patients with
acute scaphoid fractures who were treated with casting
compared with those treated with screw fixation have
been statistically equivalent.

Flexor Tendon Injuries

Tendon Lacerations
The diagnosis, surgical care, and postoperative rehabilitation of a flexor tendon injury may be more difficult in
a child than in an adult. Treatment is especially difficult
in a toddler or preschool-age child who offers limited
cooperation. The presenting digital cascade and digital

720

excursion with wrist tenodesis serve as the basis for diagnosis of a flexor tendon laceration. If doubt exits, the
wound should be explored with the patient under anesthesia. Repair of the tendon lacerations in zones I and
II requires meticulous technique with fine sutures. Postoperative cast immobilization for 4 weeks is effective.
No differences have been found in total active motion (TAM) between patients treated with early mobilization protocols and those treated with cast immobilization for 4 weeks. If cast immobilization continues
beyond 4 weeks, TAM may decrease by up to 40% by 6
weeks. There was no difference in the results in groups
from 0 to 15 years of age.
Two-stage reconstruction of unrecognized zone II
lacerations in children has poorer results than in adults,
with a higher rate of complications and a mean TAM of
only 140. Better results are achieved with supervised
rehabilitation. In a study on uncomplicated flexor pollicis longus tendon laceration repairs, long-term limited
motion (> 30) of the interphalangeal joint occurred in
one third of the patients. Short splint immobilization
had a negative effect on outcome; however, zone of injury, an early mobilization program, or concurrent digital nerve injury had no significant effect on long-term
outcome.

Amputations
The treatment of proximal, complete digital amputations with replantation in children as young as 1 year of
age is now standard. In children, the indication for replantation is more liberal than in adult patients and includes cases of multiple digit, thumb, midpalm, hand,
and distal forearm amputations, as well as single digit
amputations in zones I and II. Crush amputations
caused by compression from doors, heavy objects, or bicycle chains have a peak incidence in children 5 years of
age, whereas amputations caused by sharp objects occur
more commonly in adolescents. Digital survival rates
from replantation range from 69% to 89% in pediatric
studies. Indications that favor digital survival are amputations resulting from sharp objects, patient body weight
greater than 11 kg, more than one vein repaired, bone
shortening, interosseous wire fixation, and vein grafting
of arteries and veins. Vessel size generally exceeds 0.8
mm in digital replants in children and is not a technical
problem for the skilled microvascular surgeon. Index
and long finger replants have been more successful than
small finger replants in children. A digital survival rate
of 95% occurred in children if prompt reperfusion was
seen after arterial repair with at least one successful
venous anastomosis. Neural recovery rates far exceed
those found in adult patients; return of two-point discrimination of less than 5 mm often occurs. Tenolysis
may be necessary after tendon repair. Growth arrest or
deformity is more common if there is a crush compo-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Table 2 | Classification of Congenital Upper Limb

Malformations
I

II
III
IV
V
VI
VII
VIII

Failure of formation
A Transverse arrest
B Longitudinal arrest
Failure of differentiation (separation)
Congenital tumorous conditions
Duplication
Overgrowth
Undergrowth
Congenital constriction ring syndrome
Generalized skeletal abnormalities

nent to the amputation. Microvascular toe to thumb

transfer is a successful alternative to pollicization as
treatment for a failed replant in a young child.

Congenital Upper Extremity Malformations

The prevalence of congenital upper limb anomalies is 2
per 1,000 live births. Forty-six percent of these infants also
have another nonhand anomaly; bilateral hand anomalies
are found in more than 50% of these patients. Overall,
boys are affected more than girls by a ratio of 3:2. Advanced maternal age significantly increases the prevalence
of congenital upper extremity malformations.
Recent advances in embryology have greatly improved the understanding of human limb development.
Growth factors influence the apical ectodermal ridge
and numerous genes influence the zone of polarizing activity and polarity. Genes responsible for specific deformities such as triphalangeal thumb have been identified
on specific chromosomes. Modern ultrasound techniques also have increased the rate of prenatal diagnosis
of these disorders and may lead to parental consultation
before birth.
During embryologic growth, the limb bud appears 26
days after fertilization. The final stage of upper extremity development, separation of the ring and small digits,
is completed by day 56. Events occurring at various
stages of embryologic growth may have profound effects on the ultimate appearance of the limb.

Classification
The most commonly used classification system is that
proposed by the American Society for Surgery of the
Hand and the International Federation of Societies for
Surgery of the Hand. This system includes eight categories that are based on the proposed etiologic pathways
and are found in Table 2.

Trigger Thumb
Trigger thumb is one of the most common malformations in children. Two large studies of consecutive deliv-

American Academy of Orthopaedic Surgeons

Chapter 61 Forearm, Wrist, and Hand: Pediatrics

eries of more than 12,000 children have not found this
condition to be present at birth, supporting the belief
that it is not a true congenital condition. The child will
have a flexion deformity of the interphalangeal joint
that may not be passively correctable. A compensatory
hyperextension deformity at the metacarpophalangeal
joint may also develop. A palpable nodule at the level of
the A1 pulley of the thumb, which appears to be caused
by edema within the tendon, will also be present.
Spontaneous resolution occurs in a small number of
patients. Nonsurgical treatment has been most popular
in Asia where a combination of splinting and passive
motion has been shown to allow for correction in up to
two thirds of affected patients. When the thumb does
not show any sign of spontaneous correction, A1 pulley
release is indicated. This procedure may be performed
by either transverse incision or a chevron type incision
centered over the metacarpophalangeal joint. Longitudinal incisions over this site have been shown to adversely
affect range of motion and appearance. Long-term studies of trigger thumb release have found that although
recurrence of triggering is extremely rare, approximately 25% of the patients show some limitation in interphalangeal joint motion compared with the unaffected thumb. No long-term functional deficits have
been noted with trigger thumb release.

Trigger Finger
Based on large population studies, trigger finger, like
trigger thumb, has not been found at birth. Patients with
trigger fingers have a higher rate of spontaneous resolution and an earlier age of onset than patients with trigger thumbs. For those digits not showing spontaneous
correction, abnormalities are found well beyond the A1
pulley and may include abnormalities of the flexor digitorum sublimis tendon insertion and the A3 pulley. Surgical treatment of a trigger finger requires exploration
of the digit well beyond the A1 pulley because almost
50% of patients undergoing simple A1 pulley release
will have residual triggering.

Polydactyly
Polydactyly occurs as an isolated disorder, in association
with other malformations of the extremities, or as part
of a syndrome. It usually occurs sporadically but may be
inherited with a mainly autosomal dominant inheritance. Polydactyly is categorized as radial, central, or ulnar with the small finger the most commonly duplicated
digit.
Small finger duplication may occur as either a skin
tag or a fully developed digit including a fully formed
metacarpal bone (Figure 4). The incidence of this disorder has been reported as high as 1 in 300 births in the
African American population and 1 in 3,000 in the
white population. In the African American population,

721

Orthopaedic Knowledge Update 8

Forearm, Wrist, and Hand: Pediatrics

Table 3 | Thumb Duplication

Level

Description

I
II
III
IV
V

Bifid distal phalanx

Complete duplication distal phalanx
Complete duplication distal phalanx, bifid proximal phalanx
Complete duplication distal and proximal phalanges
Complete duplication distal and proximal phalanges, bifid
metacarpal
Complete duplication distal, proximal phalanges, and metacarpal
Triphalangeal thumb

VI
VII

Figure 4 Small finger duplication, complete.

this condition is rarely associated with any other type of
malformation, whereas up to 30% of white children
with small finger polydactyly will have underlying organ
system disorders such as gastrointestinal or genitourinary malformations. Most patients with ulnar polydactyly are treated with either ligation in the nursery or
simple excision. One recent study found a 23% complication rate following suture ligation; infection and incomplete removal were the most common complications. Patients with ulnar polydactyly involving
neurovascular structures in bone will require more sophisticated reconstruction, which may include transference of hypothenar tendons and collateral ligament reconstruction.
Radial side polydactyly involving the thumb is classified based on the level of duplication (Table 3) and
ranges from a bifid distal phalanx to complete duplication of metacarpals (Figure 5). Patients with malformations involving the thumb tips may benefit from a
combination type procedure with more proximal duplications treated by excision of the radial digit and reconstruction of the retained ulnar digit. Simple excision
produces significant complications of instability and
stiffness; formal reconstruction of the duplicated digit
including advancement of the thenar tendon and collateral ligament reconstruction should be performed. Because of the difficulty in reconstructing the ulnar collat-

722

Figure 5 Type IV thumb duplication, converging.

eral ligament, the preferred treatment is preservation of
the more ulnar digit, which in most patients is more
completely developed. Late deformities have been observed as a result of instability of the interphalangeal
joints and eccentric insertions of the flexor and/or extensor tendon. These deformities may have a radial in-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 61 Forearm, Wrist, and Hand: Pediatrics

sertion along the retained digit, allowing for a zigzag

type of deformity later in life. Although long-term studies show few functional problems with duplicate thumb
reconstruction, a large percentage of patients report
stiffness and/or instability at the interphalangeal joints.
The triphalangeal thumb falls within the category of
thumb duplication and is an autosomal dominant form
of an inherited radial abnormality (Figure 6). It may exist in combination with syndactyly. Complications associated with multiple level joint instability have led some
physicians to recommend ablation of this digit when it is
part of a radial side polydactyly. When seen in isolation,
the triphalangeal thumb is best treated by deletion of
the middle phalanx and temporary pin fixation of the
resulting interphalangeal joint.
Central polydactyly is a rare condition involving duplication of the index, long, or ring digits. These digits
often show incomplete duplication with frequent overlapping of the duplicated digits. Most patients are best
treated by deletion of these digits, which are often underdeveloped.

Radial Head Dislocation

Congenital radial head dislocation is often asymptomatic and may go undiagnosed until radiographs have
been obtained for an injury. True congenital dislocations
may have a late diagnosis because of the frequency of
bilateral involvement. Recent reports have suggested
that open reduction and ligament repair or reconstruction may successfully reduce the radial head. Longer
term studies are needed to determine if the bony
changes that include capitellar undergrowth, radial head
enlargement, and bowing may be prevented by open reduction. Radial head reduction has not been as successful in patients with either radial or ulnar dysplasia and
in patients older than 5 years of age at the time of surgery.
In patients who are not evaluated until secondary
bone changes have occurred, open reduction has not
been recommended. Radial head resection at adolescence remains a mainstay of treatment and has been
shown to improve range of motion. Regrowth of the
proximal radius has been reported after radial head resection and may cause secondary impingement and loss
of motion.

Longitudinal Epiphyseal Bracket (Delta Phalanx)

Delta phalanx may be found in isolation or within a
digit affected by other disorders such as syndactyly (Figure 7). In this condition the epiphyseal plate is oriented
both transversely and longitudinally, forming a C
shape. This abnormal epiphyseal plate may cause a significant deviation in digits that are operated on for
other conditions, and may be undetected at the time of
initial surgery. Studies have shown that this condition

American Academy of Orthopaedic Surgeons

Figure 6 Duplicate thumb, triphalangeal ulnar component.

may not be readily apparent until 18 months of age. In
patients with a deviation caused by this abnormal epiphyseal plate, a portion of the plate should be removed
and replaced by fat or other material. This technique
has been shown to successfully stop the abnormal deviation of the digit. In patients in whom treatment is delayed, treatment may require reversed wedge osteotomy
as a method to both correct deformity and arrest the abnormal epiphyseal growth.

Constriction Ring Syndrome

Constriction ring syndrome has been reported in up to 1
in 10,000 births. Although the etiology of this disorder is
not completely known, a similar disorder has been produced in laboratory animals by creating a vascular insult
late in limb development. In these experiments, the loss
of the epidermal layer causes coalescence of digits much
like that seen in an untreated burn. Involvement of the
digits ranges from mild skin dimpling to complete absence of digits. Acral syndactyly is noted with coalescence of digits at the tip combined with separation at

723

Orthopaedic Knowledge Update 8

Forearm, Wrist, and Hand: Pediatrics

Figure 7 Ring finger delta phalanx.

Figure 8 Constriction ring syndrome with web base separation.

the level of the web space (Figure 8). The underlying

structures are usually normal to the level of the digital
absence; separation of digits may be easily performed as
the proper level of the web is already established. Circumferential rings may best be released by performing
surgery on one half of the circumference at a time. On
rare occasion, this disorder may occur with significant
compromise to the digit distal to the site of the constriction ring and will require release of the ring during the
first few days of life. In most patients with ring release, a
deep fibrous layer must be excised in addition to
Z-plasty, which is usually performed at the skin level.

nism. Camptodactyly has a bimodal type of distribution,

with type I presenting in infancy and type II presenting
in adolescence. In type III, the condition is found as part
of a syndrome.
Most type I patients improve with stretching and
splinting. Patients with type II camptodactyly may improve with splinting and may have significant residual
flexion deformity, as do type III patients. Surgery for
this disorder must be undertaken with caution because
significant residual flexion deformity is usually noted.
Most authors recommend reserving surgical release for
contractures greater than 60.

Camptodactyly

Syndactyly

Camptodactyly is an isolated congenital flexion deformity of the proximal interphalangeal joint. The patients
may present with a broad spectrum of deformity and involvement. The digits have abnormalities of numerous
structures including skin, fascia, tendon sheaths, flexor
digitorum sublimis tendons, lumbrical and interosseous
muscles, bony surfaces, and the central extensor mecha-

Syndactyly is the most common congenital hand deformity occurring in approximately 1 in 2,200 births. It may
be inherited with an autosomal dominant pattern in up
to 40% of patients. Although numerous syndromes include syndactyly, the most common are Aperts syndrome and Polands syndrome. Syndactyly may be complete or incomplete based on the distal extent of the

724

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Figure 9 Complete syndactyly of the third and fourth fingers.

web (Figure 9). Syndactyly is also classified as either
simple or complex based on any component of shared
bone or nail. The geometry of most syndactylous digits
usually requires the use of skin graft for separation. A
full-thickness graft harvested from either the upper extremity or the groin provides adequate coverage. Recent
reports have promoted the separation of syndactylous
fingers without grafting by the aggressive defatting of
skin flaps. Border digits may require early release because of the disparate length of the ring and small finger
or the thumb and index finger. Complications of syndactyly release include infection, web creep, and unrecognized bony anomaly. Web creep is seen in patients undergoing surgery before 16 months of age. The
unrecognized bony anomaly of a delta phalanx also may
be undetected on radiographs until 18 months of age.
These two factors may indicate that delaying surgery
until 18 months may be advisable.

Madelungs Deformity
Madelungs deformity is a growth disturbance involving
the palmar and ulnar aspects of the distal radius at the

American Academy of Orthopaedic Surgeons

Chapter 61 Forearm, Wrist, and Hand: Pediatrics

epiphyseal plate. A physeal bar has been found that
bridges the distal metaphysis of the radius to the epiphysis along the volar aspect of the radius at the lunate
facet. Hypertrophy of palmar ligaments (including the
radiotriquetral and the short radiolunate ligaments, and
an anomalous volar ligament) has been identified. It appears that the physeal bar impedes the normal development of the distal radius causing its palmar and ulnar
deformity. The hypertrophied ligament also appears to
contribute to the deformity.
Radiographic findings include dorsal bowing of the
radius, ulnar tilting of the radius, and radial tilting of the
ulna. There is palmar tilting of the distal articular surface of the radius and a triangular appearance of the
epiphysis. A fixed pronation deformity of the distal radius is usually present.
Madelungs deformity may occur as an isolated condition or as part of a generalized dysplastic disorder associated with Leri-Weill syndrome (dyschondrosteosis),
a condition whose inheritance is autosomal dominant.
Deletions or mutations in the SHOX (short stature homeoboxcontaining) gene have now been well documented in those patients.
During the first few years of life, the ulnar portion of
the epiphysis fails to ossify, leading to a severely angulated distal articular surface. The carpus then tends to
shift toward the ulna with the lunate often overlapping
the ulnar head. Patients may seek treatment during the
preadolescent years because of the deformity. Pain is
rarely a symptom.
Numerous surgical approaches have been described.
It appears that the palmar approach to release the hypertrophied ligament combined with either epiphysiolysis or osteotomy may offer the best treatment for correcting this deformity. After epiphyseal plate closure,
procedures to correct the deformity should include correction of the articular surface angular deformity combined with ulnar shortening. Salvage type procedures include radiolunate arthrodesis combined with distal ulnar
resection.

Synostosis of the Proximal Radius and Ulna

Failure of the proximal radius and ulna to separate
leads to a fixed forearm position because of synostosis.
Most cases of synostosis are bilateral and may frequently occur in association with other organ system abnormalities. In most patients the disease is characterized
by a forearm that is in slight to moderate pronation;
minimal treatment may be required for this condition.
In patients with significant supination of the forearm, a
rotational osteotomy is recommended so that both
palms may be positioned for keyboard use. Repositioning of the forearm in patients with congenital radioulnar
synostosis may be performed using an osteotomy
through the level of synostosis or with a two-level os-

725

Orthopaedic Knowledge Update 8

Forearm, Wrist, and Hand: Pediatrics

Table 4 | Thumb Hypoplasia

Type I
Type II
Type III

A
B
Type IV
Type V

Mild underdevelopment
Thenar hypoplasia, abduction contracture, metacarpophalangeal laxity
Thenar hypoplasia, abduction contracture, metacarpophalangeal instability, extrinsic tendon hypoplasia or absence,
metacarpal underdevelopment
Stable carpometacarpal joint
Unstable carpometacarpal joint
Pouce flottant
Complete absence of thumb

mately undergo pollicization to avoid placement of a

pin through the index metacarpal.
A recent long-term assessment of the functional outcome after centralization for radial dysplasia showed that
despite surgical intervention, hand function remained
markedly abnormal, whereas upper extremity disability
was mild. Improvement of wrist alignment and increased
forearm length was not found to correlate with improved
upper extremity function. Significant radial angulation
and limited range of motion and grip strength were all
findings in this long-term follow-up study.

Thumb Hypoplasia
teotomy (one osteotomy performed at the distal third of
the radius and one at the proximal third of the ulna).
Attempts to separate the synostosis have been successful in a limited number of patients; the use of either a
pedicled or free tissue transfer to prevent reformation
of the synostosis is required.

Transverse Forearm Deficiency

Most patients with transverse failure of formation have a
unilateral, short below-elbow deficiency. Other levels of
failure of formation include transcarpal, distal forearm,
and transhumeral. Most children will have excellent adaptation to this deficiency with few functional deficits.
The use of a passive prosthesis is usually attempted
in patients with a short below-elbow deficiency. In some
patients, the forearm portion is quite short and there
may be difficulty with prosthetic fitting. In addition,
soft-tissue coverage may be inadequate to allow for
prosthetic placement. Prosthetic fitting usually begins at
4 to 6 months of age and includes a passive prosthesis to
allow for two-handed activity. Many children resist the
use of a prosthesis because of its lack of sensory feedback. More proximal levels of deficiency are rarely
treated with prosthetic usage.

Thumb hypoplasia includes the spectrum of deformities

from a slightly small thumb to complete thumb absence.
Blauth type I thumbs usually appear small but require
no reconstruction. Type II deformities have absent or
underdeveloped thenar muscles; opponensplasty is required. Type III hypoplastic thumbs are characterized
by underdevelopment of the metacarpal bones, absence
of extrinsic tendons, and thenar hypoplasia. Type IIIA
thumbs show a stable carpometacarpal joint and type
IIIB thumbs an unstable carpometacarpal joint, often
with a deficient metacarpal. Type IIIA thumbs require
significant reconstructive effort including opponensplasty, metacarpophalangeal joint stabilization, and web
space deepening. Type IIIB thumbs may best be treated
by ablation and pollicization using the index finger as is
recommended for the type IV deficiency (pouce flottant). The type V deficiency is characterized by complete absence of the thumb (Table 4). Studies of the
long-term outcome following hypoplastic thumb reconstruction did not find any benefit from early reconstruction. More than 30% of patients undergoing pollicization will require opponensplasty. Pollicization is best
performed when the patient is 2 to 3 years old because
of the increased size of the digital vessels.

Longitudinal Deficiency of the Ulna

Longitudinal Deficiency of the Radius
Radial ray deficiency is a condition occurring in 1 in
25,000 births. Radial ray deficiencies include complete
or partial absence of the radius and radial digitsmost
commonly the thumb. Radial deficiencies are usually bilateral. Radial ray deficiency has been categorized as
type I, short distal radius; type II, underdevelopment of
the entire radius; type III, partial absence of radius; and
type IV, complete absence of radius (the most common
deficiency). Treatment includes passive stretching with
either serial casting or static progressive splinting until 6
months of age at which time centralization or radialization is performed. Alternatively, soft-tissue distraction
using external fixation also may be used to provide
some level of correction before bony stabilization. Centralization is usually performed in patients who will ulti-

726

Ulnar longitudinal deficiency is one fourth as common

as radial ray deficiency and occurs in approximately 1 in
100,000 births. The ulna may show partial or complete
deficiency with varying levels of digital deletions. The elbow deformity ranges from joint hypoplasia to complete
radiohumeral synostosis. Unilateral involvement is most
common and this condition is frequently associated with
other musculoskeletal abnormalities such as scoliosis
and fibular deficiency. Surgical treatment of patients
with ulnar radial deficiency usually focuses on the hand.
Up to three fourths of hands have first web abnormalities. Syndactyly is frequently found in the digits that are
present, and also requires separation. One study found
that slightly fewer than 50% of patients with ulnar ray
deficiency also displayed some element of thenar hypoplasia.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Annotated Bibliography
Forearm and Wrist Trauma
Boyd KT, Brownson P, Hunter JB: Distal radial fractures in young goalkeepers: A case for an appropriately
sized soccer ball. Br J Sports Med 2001;35:409-411.
In a prospective, clinic-based study of young goalkeepers,
it was shown that the size of the ball had a direct effect on the
risk of a distal radius fracture.

Cannata G, DeMaio F, Mancini F, Ippolito E: Physeal

fractures of the distal radius and ulna: Long-term prognosis. J Orthop Trauma 2003;17:172-179.
One hundred sixty-three fractures of the distal radius and
ulna physes in 157 patients were reviewed at an average
follow-up of 25.5 years (14 to 46 years). Seventy-seven were
isolated fractures of the distal radius, 54 were associated with
ulnar styloid fractures, and 26 were associated with ulnar
metaphyseal fractures. Eighteen were Salter-Harris type I fractures and 139 were type II fractures. Symptomatic growth discrepancy of greater than 1 cm at skeletal maturity occurred in
4.4% of the distal radius physeal fractures and 50% of distal
ulnar physeal fractures.

Jones IE, Williams SM, Dow N, Goulding A: How many

children remain fracture-free during growth?: A longitudinal study of children and adolescents participating in
the Dunedin Multidisciplinary Health and Development
Study. Osteoporos Int 2002;13:990-995.
Results of a longitudinal study of 601 people from New
Zealand showed that approximately half (60% girls, 50%
boys) of the children in the study did not have a fracture
throughout their growth period. More than one fracture occurred in 119 children (16% girls, 23% boys). The forearm was
the most common site of fracture (24%).

Ronning R, Ronning I, Gerner T, Engebretsen L: The

efficacy of wrist protectors in preventing snowboarding
injuries. Am J Sports Med 2001;29:581-585.
Results of a randomized, prospective study of 5,029 snowboarders in Norway is presented. A significant decrease in injury rate was found with the use of wrist protectors. Beginners
and those using rental equipment were at highest risk for wrist
injury.

Wareham K, Johansen A, Stone MD, Saunders J, Jones

S, Lyons RA: Seasonal variation in the incidence of
wrist and forearm fractures and its consequences. Injury
2003;34:219-222.
A longitudinal study from Wales showed a seasonal variation in the incidence of wrist and forearm fractures in children
younger than 15 years of age. The incidence in the three winter months (5.7/1,000 per year) was half the incidence for fractures the rest of the year (10.7/1,000 per year). The fractures
that occurred during the nonwinter months were more severe
with 23% of patients with forearm and wrist fractures requiring hospitalization during the spring, summer, and fall com-

American Academy of Orthopaedic Surgeons

Chapter 61 Forearm, Wrist, and Hand: Pediatrics

pared with only 10% requiring hospitalization when fractures
occurred during the winter.

Waters PM, Bae D, Montgomery K: The surgical management of post-traumatic distal radial physeal growth
arrest in adolescents. J Pediatr Orthop 2002;22:717-724.
The authors present a case study of patients with distal radial physeal arrests with consequential ulnar overgrowth resulting in complications with ulnar-carpal impaction, distal
radial-ulnar joint incongruity, and triangular fibrocartilage
tears. Surgical planning and treatment options are outlined for
corrective osteotomies and soft-tissue repairs.

Hand Trauma
Mahabir RC, Kazemi AR, Cannon WG, et al: Pediatric
hand fractures: A review. Pediatr Emerg Care 2001;17:
153-156.
The incidence and epidemiology of hand fractures in children are discussed.

Rajesh A, Basu AK, Vaidhyanath R, Finlay D: Hand

fractures: A study of their site and type in childhood.
Clin Radiol 2001;56:667-669.
Radiographs of 280 children with hand fractures were reviewed and categorized by site of the injury and age of the patient. Distal tuft fractures were the most common in the 0- to
4-year age group; distal phalanx fractures occurred most frequently in the 5- to 8-year age group; fractures of the proximal
phalanx of the small finger were most common in the 9- to 12year age group; and fractures of the small finger metacarpal
occurred most frequently in the 13- to 16-year old age group.

Congenital Upper Extremity Malformations

Giele H, Giele C, Bover C, et al: The incidence and epidemiology of congenital upper limb anomalies: A total
population study. J Hand Surg [Am] 2001;26:628-634.
A homogeneous population study found congenital upper
extremity anomalies in approximately 2 per 1,000 live births
with nearly 50% of those affected having some other anomaly.
Failure of separation and duplications were the most common
malformations and increased in number with advanced maternal age.

Goldfarb CA, Klepps SJ, Dailey LA, et al: Functional

outcome after centralization for radius dysplasia.
J Hand Surg [Am] 2002;27:118-124.
This long-term review of patients undergoing centralization found residual radial angulation of 36, wrist arc of motion of 31, and an average ulnar length that was 54% of that
found in the uninvolved contralateral ulna. Surgery resulted in
improvement in wrist alignment but with less benefit to functional improvement than previously believed.

Greuse M, Coessens BC: Congenital syndactyly: Defatting facilitates closure without skin graft. J Hand Surg
[Am] 2001;26:589-594.

727

Forearm, Wrist, and Hand: Pediatrics

Orthopaedic Knowledge Update 8

The authors performed syndactyly separation with the use

of graft by performing extensive defatting. Skin graft was required for complex cases and reoperation was performed in 2
of 16 patients.

This long-term retrospective review found that most patients with Madelungs deformity are female (77%), with a
positive family history in 13% of those affected. Despite surgery, almost all patients had limited range of motion.

Kemnitz S, De SL: Pre-axial polydactyly: Outcome of

the surgical treatment. J Pediatr Orthop B 2002;11:79-84.

Classic Bibliography

This retrospective study of patients who underwent reconstruction of preaxial polydactyly found high levels of patient
satisfaction. Functional outcome was affected by joint stability.

Keswani SG, Johnson MP, Adzick NS, et al: In utero

limb salvage: Fetoscopic release of amniotic bands for
threatened limb amputation. J Pediatr Surg 2003;38:848851.
The use of fetoscopic laser surgery to release constriction
rings at the level of the wrist improved extremity blood flow
in utero.

McAdams TR, Moneim MS, Omer GE Jr: Long-term

follow-up of surgical release of the A (1) pulley in childhood trigger thumb. J Pediatr Orthop 2002;22:41-43.
The authors report on the results of long-term follow-up
on patients undergoing trigger thumb release. Patients with a
transverse incision had better appearance. Twenty-three percent of these patients had incomplete interphalangeal joint
motion, and 17% had metacarpophalangeal hyperextension.

Moon WN, Suh SW, Kim IC: Trigger digits in children.

J Hand Surg [Br] 2001;26:11-12.
In this study of more than 7,000 newborn infants, no instances of trigger thumb or trigger finger were found. Trigger
fingers developed earlier in life than trigger thumb and had a
higher spontaneous recovery rate.

Murase T, Tada K, Yoshida T, et al: Derotational osteotomy at the shafts of the radius and ulna for congenital
radioulnar synostosis. J Hand Surg [Am] 2003;28:133137.
The authors describe a two-level osteotomy with the radius osteotomy at the distal third and the ulnar osteotomy at
the proximal third. Pronation deformity was corrected by
manual derotation.

Rayan GM, Frey B: Ulnar polydactyly. Plast Reconstr

Surg 2001;107:1449-1454.
This study reviews the various types of ulnar polydactyly
and their treatments. Most patients (71%) had soft-tissue tags
that were treated by ligation. African American patients were
most likely to have soft-tissue duplications only, whereas white
patients were more likely to have digits with bony elements.

Schmidt-Rohlfing B, Schwobel B, Pauschert R, et al:

Madelung deformity: Clinical features, therapy and results. J Pediatr Orthop B 2001;10:344-348.

728

Bayne LG, Klug MS: Long-term review of the surgical

treatment of radial deficiencies. J Hand Surg [Am] 1987;
12:169-179.
Benson LS, Waters PM, Kamil NI, Simmons BP, Upton J
III: Camptodactly: Classification and results of nonoperative treatment. J Pediatr Orthop 1994;14:814-819.
Buck-Gramcko D: Pollicization of the index finger:
Method and results in aplasia and hypoplasia of the
thumb. J Bone Joint Surg Am 1971;53:1605-1617.
Carter PR, Ezaki M: Madelungs deformity: Surgical
correction through the anterior approach. Hand Clin
2000;16:713-721.
Cole RJ, Manske PR: Classification of ulnar deficiency
according to the thumb and first web. J Hand Surg
[Am] 1997;22:479-488.
Fitoussi F, Mazda K, Frajman JM, Jehanno P, Pennecot
GF: Repair of the flexor pollicus longus tendon in children. J Bone Joint Surg Br 2000;82:1177-1180.
Gibbons CL, Woods DA, Pailthorpe C, et al: The management of isolated distal radius fractures in children.
J Pediatr Orthop 1994;14:207-210.
Horii E, Nakamura R, Sakuma M, Miura T: Duplicated
thumb bifurcation at the metacarpophalangeal joint level: Factors affecting surgical outcome. J Hand Surg
[Am] 1997;22:671-679.
Hung L, Cheng JC, Bundoc R, et al: Thumb duplication
at the metacarpophalangeal joint: Management and a
new classification. Clin Orthop 1996;323:31-41.
James MA, McCarroll HR, Manske PR: The spectrum
of radial longitudinal deficiency: A modified classification. J Hand Surg [Am] 1999;24:1145-1155.
Mani GV, Hui PW, Cheng JC: Translation of the radius
as a predictor of outcome in distal radial fractures of
children. J Bone Joint Surg Br 1993;75:808-811.
Manske PR, Halikis MN: Surgical classification of central deficiency according to the thumb web. J Hand Surg
[Am] 1995;20:687-697.
Manske PR, Rotman MB, Dailey LA: Long-term functional results after pollicization for the congenitally deficient thumb. J Hand Surg [Am] 1992;17:1064-1072.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

McCarroll HR: Congenital anomalies: A 25 year overview. J Hand Surg [Am] 2000;25:1007-1037.
McCarty E, Mencio G, Green N: Anaesthesia and analgesia for the ambulatory management of fractures in
children. J Am Acad Orthop Surg 1999;7:81-91.
Mintzer CM, Waters PM: Surgical treatment of pediatric
scaphoid nonunions of the scaphoid. J Pediatr Orthop
1999;19:329-337.
Johnson KJ, Haigh SF, Symonds KE: MRI in the management of scaphoid fractures in skeletally immature
patients. Pediatr Radiol 2000;30:685-688.
OConnell SJ, Moore MM, Strickland JW, et al: Results
of zone I and zone II flexor tendon repairs in children.
J Hand Surg [Am] 1994;19:48-52.
Ogino T, Kato H: Clinical and experimental studies on
teratogenic mechanisms of congenital absence of digits
in longitudinal deficiencies. Cong Anom 1993;33:187196.

American Academy of Orthopaedic Surgeons

Chapter 61 Forearm, Wrist, and Hand: Pediatrics

Proctor MT, Moore DJ, Paterson JM: Redisplacement
after manipulation of distal radial fractures in children.
J Bone Joint Surg Br 1993;75:453-454.
Shoemaker S, Comstock C, Mubarak S, et al: Intramedullary kirschner wire fixation of open or unstable forearm fractures in children. J Pediatr Orthop 1999;19:329337.
Smith PJ, Grobbelaar AO: Camptodactyly: A unifying
theory and approach to surgical treatment. J Hand Surg
[Am] 1998;23:14-19.
Vransky P, Bourdelat D, Al Faour A: Flexible intramedullary pinning technique in the treatment of pediatric
fractures. J Pediatr Orthop 2000;20:23-27.
Waters PM, Kolettis GJ, Schwend R: Acute median neuropathy following physeal fractures of the distal radius.
J Pediatr Orthop 1994;14:173-177.

729

Chapter

62

Hip, Pelvis, and Femur: Pediatrics

Keith R. Gabriel, MD
Eric J. Wall, MD

Developmental Dysplasia of the Hip

Incidence
Most newborn screening studies, usually based on physical examination techniques, suggest that some degree of
hip instability can be detected in 1 in 100 to 1 in 250 babies. Actual dislocated or dislocatable hips are much less
frequent, being found in 1 to 1.5 of 1,000 live births.
Late presentation of developmental dysplasia of the hip
(DDH) is found in approximately 4 per 10,000 children.

Etiology
There is no single cause of DDH. The basic structures of
the human hip joint are well formed by the 11th fetal
week. Subsequent development of the hip requires a
continuous synergistic molding and growth of the immature femoral head and the acetabulum. Any process or
event that interrupts this interaction can result in structural abnormality and instability. Those hips that dislocate early during the course of fetal development will
have extreme anatomic abnormalities, and are called
teratologic dislocations.
The risk of DDH has been found to be 34% in identical twins, but only 3% in fraternal twins. The frequency
in siblings is approximately 6% to 7%. If one parent
and one sibling have DDH, the risk to subsequent infants rises to 36%. The genetic influence may also be
seen in comparisons of different ethnic groups, with
high rates among Lappish and very low rates among
Bantus.
Infants who have been in the breech position during
the third trimester and/or perinatally have a higher risk
of DDH, as high as 20% for those in the frank breech
position. An increased incidence is also found with conditions typically associated with intrauterine crowding,
such as oligohydramnios, congenital recurvatum or dislocation of the knee, and congenital muscular torticollis.
Postnatally, a high incidence of DDH is found in those
societies where infants are customarily strapped or
swaddled with the thighs adducted.
DDH is associated with certain neuromuscular conditions and genetic syndromes, especially those in which

American Academy of Orthopaedic Surgeons

decreased or abnormal fetal movement is a feature, such

as arthrogryposis and spina bifida. Treatment of these
hips depends as much on the nature of the overall disorder as on the dislocated hip itself. Although ligamentous
laxity has been suggested as a risk factor, DDH is not
especially associated with genetic conditions such as the
Marfan, Ehlers-Danlos, or Down syndromes, in which
generalized joint laxity is a prominent feature.
In infants who are otherwise normal, the most important risk factors for DDH are female gender, family
history, and breech positioning. In addition to routine
screening, some types of imaging studies are often recommended for these infants.

Diagnosis
Physical Examination
For newborns and neonates, the mainstay of physical diagnosis has been the palpable sensation of the hip sliding out of or into the acetabulum. Barlows test is a provocative maneuver in which the examiner attempts to
subluxate or dislocate the hip by pressing gently downward on the flexed, adducted thigh. Ortolanis test is a
reduction maneuver performed by abducting the flexed
hip while lifting gently forward under the greater trochanter. Palpable luxation or reduction of the joint constitutes a positive test. Most high-pitched clicks are
transmitted from the greater trochanteric or knee areas,
and are inconsequential.
Between 3 and 6 months of age, the soft tissues
tighten sufficiently that a reduced hip does not dislocate
with Barlows test and a dislocated hip cannot be reduced with Ortolanis test. Hip abduction in flexion will
be limited on the affected side, and the thigh will be
foreshortened. In unilateral cases, the shortening is conveniently demonstrated by the Galeazzi (or Allis) test,
in which the infant is positioned supine with the hips
and knees flexed so that the relative height of the knees
can be assessed.
After these children begin to stand, a flexion contracture usually develops at the affected hip. Increased
lumbar lordosis and pelvic obliquity are seen. The pa-

731

Hip, Pelvis, and Femur: Pediatrics

Figure 1 Standard reference lines and angles used to interpret pelvic radiographs in
DDH. The femoral ossific nucleus normally is located in the lower, inner quadrant
formed by the intersection of Hilgenreiners (horizontal) and Perkins (vertical) lines.
Shentons line is a continuous arc along the inferior border of the femoral neck and
superior margin of the obturator foramen, which is disrupted when the femoral head is
dislocated. The acetabular index measures the inclination of the acetabulum. Normal
values for a newborn are less than 30 (average 27.5). (Reproduced with permission
from Guille J, Pizzutillo P, MacEwen G: Developmental dysplasia of the hip from birth
to six months. J Am Acad Orthop Surg 2000;8:232-242.)

tients walk with gluteus medius insufficiency because of

the high position of the greater trochanter. In unilateral
cases, the limb-length inequality may either contribute
to a limp, or may be compensated by toe-walking on the
affected side. Once again, limitation of hip abduction in
flexion is a useful finding in all age groups.
Lesser degrees of dysplasia may include a functionally stable but anatomically insufficient hip. These situations are ordinarily clinically silent in younger children.
The abnormality may be discovered incidentally, but
more often presents in later life as hip pain or degenerative arthritis.
Imaging: Ultrasound, Radiography, CT, MRI, and
Arthrography
Ultrasound allows visualization of cartilaginous and
soft-tissue structures before femoral head ossification.
The static acetabular morphology, as described by Graf,
is based on two angles measured in the coronal plane. A
reference line common to both angles is first established
along the lateral wall of the ilium superior to the acetabulum. The angle is subtended by this iliac line and
a line tangential to the bony roof of the acetabulum.
The angle is subtended by the iliac line and a line tangential through the cartilaginous labrum. In general, a
normal, mature infant hip should have an angle of
greater than 60 and a angle of less than 55. The per-

732

Orthopaedic Knowledge Update 8

centage of coverage of the femoral head can also be assessed. In dynamic ultrasound assessment, the position
of the femoral head with respect to the posterior bony
wall of the acetabulum is monitored as the hip is
stressed with a modified Barlows test.
The use of ultrasound in the routine screening of all
newborns is neither cost effective nor practical. The procedure is so sensitive that routine use has resulted in
overdiagnosis, above the expected incidence of DDH.
The presence or severity of ultrasound abnormalities in
infants who have otherwise normal hip examinations
does not correlate with ultimate outcome. Routine ultrasound screening of newborns has not been shown to
reduce the prevalence of late-diagnosed dysplasia.
Ultrasound evaluation at 3 to 4 weeks after birth is
useful as an adjunct in those patients having equivocal
findings on initial nursery examination, and for those in
high-risk groups. Ultrasound can be used to check hip
stability and document reduction during treatment or at
the completion of treatment. The greatest advantage is
that serial examinations can be done without exposing
the infant to ionizing radiation.
Femoral head ossification usually begins by age 6
months, and plain radiographs then become the primary
imaging modality for DDH. Radiography also remains
useful to evaluate the distorted anatomy of teratologic
dislocations. Traditionally, several lines are projected
across the visible ossified portions of the pelvis and
proximal femur to assist in interpretation of the plain
films. Hilgenreiners, Perkins, and Shentons lines are
commonly used, as illustrated in Figure 1, to determine
the alignment of the proximal femur to the pelvis. The
acetabular index provides information about the contour of the acetabulum itself.
CT, MRI, and arthrography do not play a significant
role in the primary diagnosis of DDH. Each modality
has specific uses in treatment. CT is currently the study
of choice for assessing hip reduction in a spica cast after
closed or open treatment. Limited use of CT directly
through the acetabulum requires a minimum of radiation exposure. Sedation is seldom needed, as the patient
is immobilized in the cast and spiral acquisition techniques for CT have decreased the time required for
these studies. A method of hip reduction and spica cast
application with the infant in the MRI scanner has been
investigated, but is not yet feasible in most areas. Both
CT and MRI data can be presented as threedimensional images, and software is available to manipulate these images when planning various osteotomies.
Arthrography is unparalleled for showing dynamically
the soft-tissue impediments to reduction, and is useful
to confirm position and stability in closed or open procedures. However, it is an invasive study that requires
sedation or anesthesia.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Screening
All babies should be screened for hip dysplasia. However, the scope, timing, and methodology of neonatal
screening for DDH are areas of current controversy.
Hips dislocated at birth should be treated, with the exception of certain teratologic situations. However, 75%
to 90% of hips found to be subluxatable at birth will
spontaneously stabilize within a few weeks. A reasonable protocol is to reexamine all hips having questionable nursery examinations, and infants with recognized
risk factors for dysplasia, at 2 to 3 weeks after birth.
Hips with persistent laxity at that point should be
treated. Ultrasound may be used at week 3 or 4 as an
adjunct in questionable or high-risk situations. Hips that
are stable at the 3-week examination but show dysplastic anatomy on ultrasound should be reassessed in 2 to 3
months. If the dysplasia persists, treatment is indicated.

Treatment
The fundamental goals of treatment are the same regardless of patient age. Concentric reduction should be
obtained and maintained, with a minimum of risk to the
blood supply of the capital femoral epiphysis. The later
DDH is diagnosed, more complex interventions are
needed and the risk of complications increases.

Abduction Splinting
Infants whose hips are subluxatable, reduced but dislocatable, or dislocated but reducible can usually be
treated by splinting the hips in flexion and gentle abduction. Most of these infants will be younger than 6
months of age, although occasionally the hips of a
slightly older infant remain Ortolani positive. Sometimes abduction splinting is used to achieve reduction of
a dislocated hip in infancy. This procedure must be done
with care to avoid forced positioning, and the attempt
should be abandoned if not successful within 2 to 3
weeks.
The most popular abduction splint used in the
United States is the Pavlik harness, although it has some
limitations and potential problems. It is an active device,
in that normal muscle function is required; it is not effective in patients with paralysis or spasticity. If the anterior straps hold the hips hyperflexed, femoral nerve
palsy or inferior hip dislocation can occur. The persistent use of the Pavlik harness with the hip in a posteriorly subluxated position will result in a failure of development of the posterior wall of the acetabulum,
sometimes called Pavlik disease. Prolonged prone positioning of the infant wearing the harness should be
avoided, because the combined effect of the harness and
the weight of the torso forces the hips into maximal abduction, increasing the risk of ischemic necrosis.
Good results are also obtained with other abduction
splints such as the von Rosen splint or various Plastazote

American Academy of Orthopaedic Surgeons

Chapter 62 Hip, Pelvis, and Femur: Pediatrics

orthoses. It is critical that forced or extreme abduction be
avoided. Regardless of the splint chosen, concentric reduction must be verified within 2 to 3 weeks of initiation
of treatment. If reduction is not confirmed within that period, the splinting device should be abandoned and another method of treatment chosen. When successful, the
abduction splint should be worn on a full-time basis until
the hip is stable, and part-time wear is recommended until
acetabular remodeling is confirmed.
Closed Reduction
Most infants older than age 6 months, those whose hips
are not reducible with a simple Ortolani maneuver, and
those in whom splinting in abduction fails will need a
more formal closed or open reduction of the hip. Closed
methods are preferred up to about age 2 years, but the
reduction must be obtained and maintained without excessive force. Forcible maneuvers and forced rigid immobilization carry a high risk of ischemic necrosis.
The use of traction before closed reduction of the
hip has become less common. The goal of traction is to
gently and gradually stretch the soft tissues surrounding
the hip, so that closed reduction can be achieved without undue force. However, comparison studies have
shown that adductor tenotomy and immobilization in
the human position (hips flexed 100, abducted 45 and
neutrally rotated) can achieve comparably low rates of
ischemic necrosis.
Closed reduction should be done under general anesthesia, with an arthrogram used to verify positioning.
A concentric reduction must be obtained, with no more
than 5 mm of contrast pooling medially between the
femoral head and the acetabulum, and with no softtissue interposition. With the hips flexed, a safe zone of
abduction should be assessed: an arc of stable positioning between redislocation (toward adduction) and the
limit of comfortable, unforced abduction. If the safe
zone is unacceptably narrow, an adjunctive procedure
such as percutaneous adductor tenotomy may be indicated to release soft-tissue tension and increase the
range of available abduction. The infant is then immobilized in a spica cast in the human position (hip flexion
of about 100 and abduction of about 45) for 3 to 4
months. Part-time abduction bracing should be continued until acetabular remodeling is complete. It is not
unusual for infants treated by closed reduction to
achieve stable hips, but to require a secondary acetabular or femoral procedure to correct residual deformity.
Open Reduction
Failure to obtain a concentric closed reduction of the
hip is an indication for open reduction in all age groups.
Any hip in which extreme positioning is required to
maintain reduction should also be treated by open reduction. In general, children older than 2 years of age
will require open reduction.

733

Hip, Pelvis, and Femur: Pediatrics

Orthopaedic Knowledge Update 8

Figure 2 Redirectional osteotomies.

A, Single innominate (Salter). B, Triple innominate (Steel). C, Bernese pericapsular
(Ganz). (Reproduced with permission from
Gillingham B, Sanches A, Wenger D: Pelvic
osteotomies for the treatment of hip dysplasia in children and young adults. J Am
Acad Orthop Surg 1999;7:325-337.)
Anteromedial and medial approaches have most often been recommended for younger children (2 years
old or younger). All approaches include division of the
origin of the adductor longus. One anteromedial variation then proceeds between the pectineus medially and
the femoral neurovascular bundle laterally. Division of
the psoas tendon allows direct access to the anteromedial hip capsule. With exception of an inverted labrum,
obstacles to reduction can be easily visualized and removed. However, there is no opportunity to perform
capsulorrhaphy or osteotomy through this exposure. A
potential disadvantage is that the medial femoral circumflex artery is in the field of deep dissection, which
some believe increases the risk of ischemic necrosis.
Alternatively, after division of the origin of the adductor brevis, the pectineus may be retracted laterally
and the deep dissection can then proceed either anterior
or posterior to the adductor brevis. These two variations
are conceptually less direct than the anteromedial approach, and are not well suited for high dislocations.
Other indications and limitations are similar.
The most commonly used anterior approach for open
hip reduction is that of Smith-Peterson, generally modified by using a bikini incision parallel to the groin flexion
crease. Dissection is deepened between the tensor fascia
femoris and the sartorius, mobilizing the origins of the tensor fascia femoris from the iliac wing. Division of the origins of the rectus femoris then gives access to the anterior, superior, and lateral hip capsule. A separate
tenotomy, either percutaneous or open, of the adductor
longus is recommended. All of the impediments to reduction can be addressed via this approach. If an inverted labrum is identified, it can be released by making several
radial cuts in the tight area. The labrum should not be excised.
This approach is suitable for all age groups. The dissection preserves the medial femoral circumflex vessels.

734

Capsulorrhaphy and pelvic osteotomy can be performed

through this approach. Postoperative immobilization is
in the functional position, with approximately 15 of
flexion, 15 of abduction, and neutral rotation.
Femoral shortening at the time of open reduction
has taken the place of preoperative traction for reducing soft-tissue tension around the hip. This procedure is
virtually always necessary for children older than age 2
years, but should be considered in each patient. A separate lateral approach is necessary. The hip is reduced, a
subtrochanteric osteotomy of the femur is performed,
and the soft tissues are allowed to retract. The resulting
overlapped segment of proximal femur is resected, and
the femur is internally fixed with a plate and screws.
Varus and rotation can be adjusted through this osteotomy as needed for hip stability. It is important to avoid
excessive retroversion of the femur, especially if concomitant pelvic osteotomy is done, as this increases the
risk of posterior redislocation of the hip.
In older children, the acetabulum will be more severely dysplastic. Anterolateral coverage will be insufficient. It is frequently necessary to consider concomitant
pelvic osteotomy at the time of open reduction to ensure stability of the hip. The Salter innominate osteotomy will redirect the acetabulum to restore approximately 25 of lateral coverage and 10 of anterior
coverage (Figure 2). As indicated previously, combined
pelvic and femoral osteotomy must be done carefully to
avoid posterior instability of the joint.
Upper Age Limits
Beyond a certain age, surgical treatment is unlikely to
produce a stable, mobile, pain-free joint for the long
term. Despite obvious gait abnormalities, bilateral painfree dislocations probably should not be reduced after
age 6 or 7 years. Unilateral dislocations are generally

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 62 Hip, Pelvis, and Femur: Pediatrics

more problematic, because of limb-length inequality

and pelvic obliquity. Therefore, reduction of unilateral
dislocations is considered until adolescence.

Complications
Redislocation
Redislocation after closed reduction can usually be
treated by repeat closed reduction or open reduction with
no deleterious effect on long-term outcome. Redislocation
following open reduction is usually attributable to some
flaw in the initial procedure. Inadequate inferior capsular
release, inadequate capsulorrhaphy, and posterior instability from combined pelvic and femoral osteotomies are
common errors. Repeat surgery is almost always necessary to correct the problem, and results are generally
worse than with primary open reduction.
Ischemic Necrosis
Ischemic necrosis of the femoral head is seen with all
forms of treatment. Causes include extrinsic compression of the vasculature supplying the capital femoral epiphysis, and excessive direct pressure on the cartilaginous head. Excessive or forceful abduction, previous
failed closed treatment, and repeat surgery are associated with increased rates of ischemic necrosis. The question of whether ossification of the femoral head before
treatment might affect ischemic necrosis rates is not
fully resolved.
The diagnosis of ischemic necrosis is based on radiographic findings that include failure of appearance or
growth of the ossific nucleus 1 year after reduction,
broadening of the femoral neck 1 year after reduction,
increased density and then fragmentation of the ossified
femoral head, or residual deformity of the femoral head
and neck after ossification. Classifications of ischemic
necrosis separate partial involvement from complete necrosis, which causes progressive femoral head and neck
deformity. Treatment depends on degree of severity.

Late Dysplasia
In a growing child, the mechanically stable hip that is
reduced but dysplastic may be monitored with serial radiographs. Failure of improvement with growth is an indication for intervention, as is symptomatic dysplasia in
the adolescent. Analysis of the late dysplastic hip is facilitated by CT or MRI studies with three-dimensional
reconstruction. Femoral, pelvic, or concomitant osteotomy may be required, according to the location of the
major deformity. Pelvic osteotomies are grouped into
reconstructive and salvage types.
If it is possible to concentrically and congruently reduce a dysplastic hip, a reconstructive osteotomy may be
indicated to improve coverage of the femoral head with
the native articular cartilage of the acetabulum. Several

American Academy of Orthopaedic Surgeons

Figure 3 Reshaping osteotomy of Dega, which preserves the sciatic notch and a
variable portion of the inner table of the ileum. (Reproduced with permission from
Vitale MG, Skaggs DL: Developmental dysplasia of the hip from six months to four
years of age. J Am Acad Orthop Surg 2001;9:401-411.)
osteotomies that reorient the acetabulum without changing its shape have been described, and the choice, in
large measure, depends on patient age (Figure 2). The
single innominate osteotomy (Salter) depends on flexibility of the symphysis pubis to allow sufficient rotation
of the acetabular segment. In older children this flexibility is lost, and triple osteotomy (Steel or Tonnis) is preferred so that the acetabular segment of the pelvis can
be rotated without violating the triradiate growth potential. After closure of the triradiate cartilage, the periacetabular osteotomy (Bernese) is appropriate.
If the intrinsic shape of the acetabulum must be altered to improve congruence, reshaping osteotomies
such as the Pemberton or the Dega should be considered. These osteotomies require an open triradiate cartilage to allow hinging through the acetabulum itself. The
inner and outer tables of the ileum are divided in the
Pemberton, whereas only the outer table is cut in the
Dega (Figure 3). Both proceed into the posterior limb
of the triradiate, and preserve the posterior wall of the
ischium at the sciatic notch.
Salvage osteotomies increase the surface area available for weight bearing, and depend on fibrocartilaginous metaplasia of the interposed hip capsule to form
an articulating surface. The Chiari innominate osteotomy and various shelf acetabular augmentations are
available (Figure 4).

Developmental Coxa Vara

Developmental coxa vara has an estimated incidence of
1 in 25,000 live births. Males and females are affected
with equal frequency, there is no predilection for the
right or left side, and the condition is bilateral in approximately 50% of patients.
The underlying defect of this condition is unknown.
The most widely accepted theory regarding the progres-

735

Hip, Pelvis, and Femur: Pediatrics

Orthopaedic Knowledge Update 8

Figure 5 Hilgenreiners physeal angle is created by a line through the triradiate cartilage and its intersection with a line through the physis. The normal angle is about 25.
(Reproduced with permission from Beals RK: Coxa vara in childhood: Evaluation and
management. J Am Acad Orthop Surg 1998;6:93-99.)
Figure 4 Salvage procedures. A, Chiari. B, Shelf slotted acetabular augmentation.
(Reproduced with permission from Gillingham B, Sanchez A, Wenger D: Pelvic osteotomies for the treatment of hip dysplasia in children and young adults. J Am Acad
Orthop Surg 1999;7:325-337.)
sive deformity is that there exists an unspecified primary
ossification defect in the inferior femoral neck. Physiologic stresses of weight bearing cause fatigue failure of the
local dystrophic bone, resulting in the progressive varus
displacement. MRI and some biopsy specimens support
this theory.
Patients usually have a progressive but painless gait
abnormality during early childhood. The coxa vara creates a high-riding position of the greater trochanter (decrease or reversal of the articulotrochanteric distance)
and therefore the hip abductors are functionally weakened. The children walk with a waddling, Trendelenburg
pattern. Those with unilateral involvement have an additional component of limb-length inequality, whereas
those with bilateral coxa vara have increased lumbar
lordosis.
The high-riding greater trochanter is prominent on
physical examination. Hip range of motion is restricted
in all planes. The loss of abduction is a direct manifestation of the coxa vara. Loss of internal rotation is attributed to a progressive loss of femoral anteversion as the
femoral neck displaces. A hip flexion contracture is often present. The lower extremity length discrepancy in
patients with unilateral deformity is generally mild.
The radiographic indicator of developmental coxa
vara is a triangular metaphyseal fragment in the inferior
femoral neck, delineated by an inverted Y-shaped radiolucency. There is a decrease in the femoral neck-shaft
angle, sometimes to values below 90. The position of
the physeal plate is measured by Hilgenreiners physeal
angle, determined on the AP view as the angle between
Hilgenreiners line and the plane of the proximal femoral physis. A normal Hilgenreiners angle should be less
than 25 (Figure 5). Spontaneous healing may occur
when Hilgenreiners angle remains less than 45. A
Hilgenreiners angle of greater than 60 (increasingly

736

more vertical) is associated with progression of the coxa

vara, stress fracture and nonunion of the femoral neck,
and early degenerative arthritis of the hip. The overall
radiographic appearance of the hip is similar in some
skeletal dysplasias, so screening radiographs of other areas should be obtained for every patient on initial presentation.
Nonsurgical treatments have been unsuccessful. Surgical derotational valgus-producing osteotomy of the
proximal femur is indicated when Hilgenreiners physeal angle is greater than 60, the physeal angle is
greater than 45 and the deformity is progressing, the
femoral neck-shaft angle is less than 90, or the patient
develops a significant Trendelenburg gait. Surgical goals
include overcorrection of the femoral neck-shaft angle
to about 160, restoration of Hilgenreiners angle to
about 25, and normalization of femoral rotation. Supplemental bone grafting of the inferior femoral neck defect at the time of corrective osteotomy is not necessary.
There are differing opinions regarding the best patient
age for surgery. In general, corrective osteotomy should
be performed as soon as the criteria for intervention are
apparent.
If adequate correction into valgus has been
achieved, the triangular defect in the inferior femoral
neck closes by 6 months postoperatively. Premature closure of the proximal femoral physis occurs by 24 months
in most surgically treated hips, even though the physeal
plate is not violated during surgery. The patient must
therefore be monitored for subsequent limb-length inequality. If premature closure of the proximal femoral
physis is recognized, apophyseodesis or advancement of
the greater trochanter should be considered. Recurrence of varus is unusual if adequate valgus has been
achieved at surgery.

Slipped Capital Femoral Epiphysis

Slipped capital femoral epiphysis (SCFE) is one of the
most common hip disorders in the pubertal child. The
diagnosis can be delayed because 15% of patients have

American Academy of Orthopaedic Surgeons

Chapter 62 Hip, Pelvis, and Femur: Pediatrics

Orthopaedic Knowledge Update 8

knee pain instead of hip pain, and those patients with
hip pain are often initially believed to have a pulled
groin muscle. Historically, the condition and its treatment have been fraught with frequent complications
such as chondrolysis, osteonecrosis, slip progression,
intra-articular pin penetration, hip stiffness, and degenerative arthritis. Currently, with early diagnosis and single cannulated screw fixation, the outcome is usually excellent in stable SCFE with minimal or moderate
displacement. Osteonecrosis, which is usually seen only
in unstable SCFE, and severe slip displacement can lead
to early hip degeneration.

Etiology
The etiology of SCFE is unknown, but it is probably a
mechanical problem of increased sheer stress across a
capital femoral growth plate weakened either by rapid
growth, or by a condition such as pelvic radiation, hypothyroidism, renal failure, or growth hormone treatment. Most children with SCFE are above the 95th percentile for weight and are obese; these factors place
increased stress across the growth plate. Growth plates
show physiologic weakness during periods of rapid
growth such as the pubertal growth spurt. Endocrine
consultation or endocrine screening laboratories are not
indicated unless the patient has atypical endocrine findings. Children with SCFE who are younger than 10
years of age or 16 years and older, or whose weight is
below the 50th percentile have a high incidence of atypical SCFE and may require further workup. No specific
genetic component has been identified with SCFE.

Epidemiology
The prevalence of SCFE in the United States ranges between 2 to 10 cases per 100,000 children. Male children
outnumber females with the disease by a 3:2 ratio, and
the mean age for diagnosis is about 13.5 years for boys
and 12 years for girls, which corresponds with the age at
which the adolescent growth spurt occurs. Fifty percent
of children with SCFE are above the 95th percentile for
weight according to age. There is an increased incidence
of SCFE during the summer and fall months and it reportedly is bilateral in 17% to 50% of patients. Approximately half the children who have bilateral hip involvement are identified at the time of initial presentation.
Therefore, both hips should be examined and bilateral
radiographs obtained for every patient with a unilateral
presentation of SCFE.

Diagnosis
Studies show that approximately 85% of children have a
history of hip or proximal thigh pain and 15% have only
knee pain. Therefore, in any adolescent with knee pain,
range of motion should be assessed on the ipsilateral
hip. The patient with SCFE will usually have a paradox-

American Academy of Orthopaedic Surgeons

ical limp. When the patient bears weight on the painful

lower extremity, they will lean over that extremity during the stance phase of gait. This unusual limp pattern
helps localize the cause of pain to the hip. Conversely,
patients tend to lean away from the involved side if the
pain emanates from below the hip. Patients with SCFE
tend to externally rotate their lower extremity while
walking.
The hallmark of the clinical examination is that as
the examiner flexes the hip of a patient with SCFE, the
patients leg will externally rotate obligatorily. There is
usually complete loss of internal rotation, and any attempt at testing hip internal rotation causes pain. The
diagnosis of SCFE is confirmed with AP and frog-lateral
hip radiographs. The earliest radiographic sign of SCFE
can be growth plate widening or lucency (termed epiphysiolysis) and patients with chronic slips may develop
a radiodense blush sign on the metaphyseal side of the
capital femoral growth plate. On the AP radiograph of a
normal hip, a line drawn up along the superior border of
the femoral neck (Kleins line) should project over the
lateral edge of the femoral head. In a child with SCFE
the femoral head may not project over this line (Figure
6). Patients who have the earliest stage of SCFE may
have normal radiographs (preslip). MRI can help diagnose a preslip condition by showing metaphyseal high
signal next to the growth plate and growth plate
widening.

Classification
In the most useful classification for SCFE, the condition
is either stable or unstable. Stable slips are defined by
the mode of clinical presentation. If a patient can walk
into the examiners office, with or without crutches, they
have a stable slip. With an unstable slip, the patient has
difficulty with ambulation and usually presents on a gurney or in a wheelchair. This classification helps with
prognosis in that the most feared complication of SCFE,
osteonecrosis of the femoral head, is usually only seen
with unstable slips. Increased SCFE displacement and
increased angle of the slip relative to normal anatomy
correlate with less favorable outcomes than mild slips.

Treatment
The most popular treatment of SCFE is single screw fixation across the capital femoral growth plate performed
in situ (without reduction) in stable slips. Although single screw fixation seems to be the gold standard for
most slips, recent articles have shown slip progression
after single screw fixation. Single screw fixation using
modern fluoroscopic techniques that avoid persistent
joint penetration have virtually eliminated the risk of
chondrolysis (cartilage deterioration and hip stiffness)
that was previously a common complication of SCFE
pinning. The success and simplicity of single cannulated

737

Orthopaedic Knowledge Update 8

Hip, Pelvis, and Femur: Pediatrics

Figure 6 AP radiograph of the pelvis shows that on the left hip the lateral epiphysis
does not cross a line drawn along the superior edge of femoral neck (Kleins line). This
is consistent with a subtle SCFE on the left hip. The right femoral epiphysis does cross
this line and is normal.

screw fixation has decreased the popularity of alternative treatments of stable SCFE such as bone peg epiphysiodesis, hip spica casting, and femoral neck
osteotomy.

Technique of Cannulated Screw Fixation

Patients with SCFE are usually pinned on a fracture table that allows clear AP and lateral views of the hip. Because obesity can impair visualization of the hip with
SCFE, it is imperative to confirm visualization of the
hip joint surface on the AP and lateral radiographic
views before commencing the procedure. A radiolucent
standard operating room table can also be used instead
of the fracture table by rotating the hip between AP and
frog-lateral views; however, this is indicated only for stable hips. After the patient is on the table, AP and lateral
guidelines can be drawn on the skin surface to mark the
entry position. Large cannulated screws are the most
popular for fixation, which is usually performed in situ.
An adequate number of screw threads should be placed
into the center of the epiphysis to facilitate a stable fixation without joint penetration. After fixation is
achieved, the hip is rotated under live fluoroscopy to ascertain that the pin is not penetrating the hip joint. Pins
that are not placed centrally in the femoral head can inadvertently penetrate the joint despite two orthogonal
radiographic views showing no joint penetration. Postoperatively, patients with stable slips are allowed to bear
weight as tolerated or limited weight bearing is instituted for 6 to 8 weeks.

Controversies
Although a few recent studies have shown progressive
slippage after single screw fixation, the procedure remains the current gold standard because it provides the
least risk of inadvertent hip joint penetration and chon-

738

drolysis. Single pin fixation is also recommended for unstable slips, although some authors have stated a need
for two screws. Because unstable SCFE has a high rate
of osteonecrosis, some authors recommend early reduction and arthrotomy to reduce possible compression on
the extracapsular epiphyseal vessels. Traditionally, reduction was believed to increase the risk of osteonecrosis, hence the emphasis on in situ fixation. Frequently,
spontaneous reduction occurs when patients are placed
on the fracture table because of the weight of the proximal thigh and femur. One recent study reported an increased risk of osteonecrosis associated with reduction;
however, another recent study found no such increase
and concluded that spontaneous and gentle reduction is
safe. Prophylactic pinning of the unslipped opposite hip,
for which the risk of future slip is 25% or less, is also
controversial. After one hip has slipped, patients and
their families are much more attuned to the condition
and will seek treatment early when pain or limp develop
in the opposite hip. Most pediatric orthopaedists choose
not to perform prophylactic pinning, but recommend
that patients return to the office at the first sign of pain
or limp in the opposite hip. Because up to 57% of opposite hip slips are asymptomatic and current fixation
methods have low complication rates, recent decision
analysis reviews tend to support prophylactic pinning of
the unslipped opposite hip.

Legg-Calv-Perthes Disease
Idiopathic osteonecrosis of the femoral head in children
is termed Legg-Calv-Perthes disease. Its specific cause
is unknown, but it likely involves a temporary interruption of the blood supply to some portion of the femoral
head that can extend to the adjacent capital femoral
growth plate and the metaphysis. The condition tends to
run its initial course over a period of 3 to 5 years. In the
short term most patients recover good hip function;
however, long-term studies show that by the fifth or
sixth decade of life approximately 50% of patients with
prior Legg-Calv-Perthes disease will develop degenerative arthritis of the hip. Children who experience onset
of this disease before the age of 6 years, and those who
have healing with good hip joint congruence at maturity
have the best long-term results. The main goals of treatment in Legg-Calv-Perthes disease are to keep the
femoral head contained within the acetabulum and to
maintain motion. Treatment can range from observation
to physical therapy, casting, and femoral or pelvic osteotomy. However, the effect of treatment at altering
the natural history of Legg-Calv-Perthes disease remains controversial.

Epidemiology
Legg-Calv-Perthes disease most commonly occurs in
children age 4 to 10 years, but it has been described in

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

children age 2 years to teens. It is distinguished from
adult osteonecrosis of the hip primarily because of its
much better healing and revascularization potential. The
male to female ratio of occurrence is 5:1. The condition
tends to be bilateral in about 10% of children; however,
it usually does not present at the same time in both hips.
The differential diagnosis for simultaneous bilateral hip
osteonecrosis includes Meyers dysplasia, spondyloepiphyseal dysplasia, multiple epiphyseal dysplasia, sickle
cell disease, Gauchers disease, and hypothyroidism.

Etiology
The etiology of Legg-Calv-Perthes disease is unknown.
Thrombosis caused by abnormalities in the clotting cascade have been recently investigated as the primary etiology with an initial study showing that approximately
75% of children with this disease have abnormal clotting factors. Although the rate of clotting factor abnormalities is small, more recent studies have not validated
the high rate of inherited thrombosis and thrombophilia
in patients or animals with Legg-Calv-Perthes disease.
Animal models have shown that multiple episodes of infarction are necessary to create changes that simulate
the human Legg-Calv-Perthes disease. As a group, children with this disease tend to be of shorter stature, have
a delay in bone maturation of approximately 2 years,
have a high rate of attention deficit disorder, and have
high rates of exposure to secondhand smoke. However,
no systemic causes have been identified in children with
Legg-Calv-Perthes disease. Approximately 2% of children with transient synovitis of the hip will develop
Legg-Calv-Perthes disease; however this scenario probably results from an initial misdiagnosis and is not a
cause of the disease.

Stages/Pathogenesis
The stages of Legg-Calv-Perthes disease are based on
radiographs. During the initial stage the infarction occurs, and radiographs may remain occult for the first 3 to
6 months after this initial ischemic event. Most patients
present to a physician after radiographic alterations in
the femoral head have been noted. In the fragmentation
stage, the femoral head appears to fragment or dissolve,
either partially or totally. This indicator typically occurs
during revascularization as the infarcted bone is resorbed, leaving behind a lucent zone in the femoral head.
The third phase, termed reossification, occurs when new
bone appears. In the healing stage the femoral head
reossifies back to normal bone density; however, residual
femoral head and neck deformity including shortening
(coxa breva), widening (coxa magna), and flattening may
exist.
If the capital femoral growth plate is involved in the
process, there can be tilting of the femoral neck (coxa
valga) and relative overgrowth of the greater tro-

American Academy of Orthopaedic Surgeons

Chapter 62 Hip, Pelvis, and Femur: Pediatrics

chanter. This four-stage process typically evolves over
3 to 5 years and tends to proceed more quickly in
younger children than in older children.

Clinical Presentation
The typical child with Legg-Calv-Perthes disease has a
painless limp. The patient may report intermittent hip,
thigh, or even knee pain that is typically not severe and
does not necessitate use of a crutch. The patient may
have a Trendelenburg gait in which body weight is
shifted over the affected hip during the stance phase of
gait. This gait pattern (shifting weight away from the affected side during the stance phase of gait) helps to
identify the hip as the source of the patients pain. On
clinical examination, the patient usually exhibits joint
stiffness that is most apparent with loss of hip internal
rotation and hip abduction. Limb-length discrepancy
and leg muscle atrophy are late findings that can occur
in patients with more severe hip involvement. On clinical presentation, the differential diagnosis will include
septic arthritis, transient synovitis, proximal femoral osteomyelitis, SCFE, and hip dysplasia. Infectious etiologies can usually be ruled out through the clinical history
and laboratory studies including erythrocyte sedimentation rate and C-reactive protein levels. DDH and SCFE
are usually identified on the radiographs. Transient synovitis of the hip usually improves with antiinflammatory medications within 1 week and is typically
completely resolved within 4 weeks of presentation, versus the longer time course of Legg-Calv-Perthes disease.

Radiographic Findings
The earliest radiographic finding in Legg-Calv-Perthes
disease is an apparent joint space widening caused by
failure of the involved femoral ossific nucleus to grow
after the ischemic event.
Irregularity and increased density of the femoral head
ossification center are also early findings. A subchondral
lucent line (crescent sign) can appear in the femoral head,
and forms the basis for the Salter-Thompson classification
system. A crescent sign involving less than half the femoral head is class A, and if it involves more than half the
femoral head it is class B. The extent of fragmentation of
the ossific nucleus forms the basis of the Catterall and
Herring lateral pillar classifications. In the Catterall classification stages 1 and 2, fragmentation involves less than
half the femoral head, and in stages 3 and 4 it involves
more than half the femoral head. The Herring lateral pillar classification is based on an AP hip radiograph obtained approximately at the start of the fragmentation
phase (Figure 7). Only the lateral third of the ossific nucleus is evaluated on this view. A lateral pillar that maintains its full height is classified as group A. A lateral pillar
that partially collapses but maintains greater than 50% of

739

Orthopaedic Knowledge Update 8

Hip, Pelvis, and Femur: Pediatrics

with one obtained during maximum hip abduction. The

patients knees should be pointed forward during filming. Failure of the lateral corner of the femoral head to
slip under the edge of the acetabulum in the maximum
abduction film is evidence of hinge abduction.
Because radiographic changes may not be apparent
for 3 to 6 months after the onset of the disease, bone
scans and MRI have been investigated to provide an
earlier diagnosis, but their clinical value is not clearly established. Although rarely necessary, gadoliniumenhanced MRI has been shown to correlate well with
bone scan results in providing an early diagnosis of
Legg-Calv-Perthes disease. MRI without gadolinium
enhancement was found to be poor for the evaluation of
femoral head necrosis.

Prognosis
The long-term prognosis of Legg-Calv-Perthes disease
is improved with younger age of onset, especially with
an onset in patients younger than 6 years, and also with
less residual hip joint deformity at skeletal maturity. A
recent study suggests that the percentage of physeal involvement on MRI scan may correlate better with longterm prognosis than the percentage of femoral head involvement. The Stulberg classification is the gold standard for rating the residual femoral head deformity and
joint congruence at skeletal maturity; however, a recent
study has questioned the interobserver and intraobserver reliability of this five-tier grading scale. Lateral
epiphyseal calcification and hip subluxation over 4 mm
have been recognized as risk factors for poor prognosis.
Although most patients with Legg-Calv-Perthes disease do well during adolescence and early adult life, by
age 50 or 60 years 50% of patients develop disabling degenerative arthritis. A recent study indicates that contrary to the popular belief that girls with Legg-CalvPerthes disease have a poorer prognosis, the outcomes
for boys and girls with the disease are essentially equal.
Figure 7 Lateral pillar classification. A, Herring group A, right hip with no loss of
height in the lateral third of the epiphysis compared with the normal left hip. B, Less
than 50% collapse of the lateral pillar is shown in the right hip, rendering it a Herring
group B hip. C, A Herring group C hip with greater than 50% collapse of the lateral
pillar.

its height is group B, and a lateral pillar that maintains less

than 50% of its height is group C. The increasing grade
within each classification system tends to correlate with a
poorer prognosis.
Hinge abduction, in which a bump deformity of the
lateral femoral head prevents it from rotating under the
acetabulum, is considered a poor prognostic sign. Hinge
abduction can be seen on arthrogram taken with the patient under anesthesia, or can be inferred by comparing
an AP hip radiograph obtained in neutral abduction

740

Treatment
Initial treatment of Legg-Calv-Perthes disease is typically rest, activity restrictions, the use of nonsteroidal
anti-inflammatory drugs, and physical therapy to regain
hip motion. Bracing is no longer preferred because two
studies showed no significant effect using this treatment.
The goal of all treatments is to maintain motion of the
hip and to keep the femoral head contained within the
acetabulum. Patients who lose substantial motion, who
develop hinge abduction, or who start to subluxate or
dislocate the hip joint may require more aggressive
treatment. Surgical treatment remains highly controversial, but appears to be best suited for children older
than 8 years with moderate femoral head involvement.
Preliminary results from a large multicenter trial show
that surgical treatment is associated with better out-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

come in lateral pillar B and B/C border hips in patients
who are age 8 years and older at onset. Surgical treatment was not beneficial to lateral pillar A and C hips.
Lateral pillar classification alone does not seem to be
prognostic; however, taking into account other factors
such as age of the patient and head at risk signs (such
as calcification lateral to the epiphysis, metaphyseal lesions, lateral subluxation of the femoral head, and a horizontal growth plate) improves the prognostic value. Patients who develop severe stiffness or whose femoral
heads do not remain contained within the acetabulum
may benefit from Petrie casting with or without a softtissue release.
Femoral varus osteotomy and acetabular rotational
osteotomy can also enhance containment but should
only be performed after hip stiffness is resolved preoperatively. Salvage procedures such as a shelf arthroplasty or Chiari osteotomy or valgus femoral neck osteotomy can be considered for severe disease.
Some patients who have had Legg-Calv-Perthes
disease will develop recurrence of pain and locking or
catching of the hip during their teenage years. Osteochondritis dissecans or loose fragments of cartilage can
develop and are best investigated with MRI or CT scan.
The presence of an osteochondritis dissecans lesion in a
patient who has had Legg-Calv-Perthes disease is not
an indication for open or arthroscopic removal. Nonsurgical treatment seems to work well; the osteochondritis
dissecans lesion can end up in the cotyloid fossa where
it will cause minimal complications. Older children with
Legg-Calv-Perthes disease can also develop labral
tears that can be treated with open or arthroscopic procedures.

Annotated Bibliography
Developmental Dysplasia of the Hip
Duppe H, Danielsson LG: Screening of neonatal instability and of developmental dislocation of the hip: A
survey of 132,601 living newborn infants between 1956
and 1999. J Bone Joint Surg Br 2002;84:878-885.
This article helps to illustrate the differences in detection
and treatment rates according to the experience of screeners
and use of adjunctive ultrasound examinations.

Eberle CF: Plastazote abduction orthosis in the management of neonatal hip instability. J Pediatr Orthop 2003;
23:607-616.
One hundred thirteen consecutive newborns with hip instability on physical examination were treated with the Plastazote orthosis; only two required additional treatment and
there were no instances of ischemic necrosis.

Grudziak JS, Ward WT: Dega osteotomy for the treatment of congenital dysplasia of the hip. J Bone Joint
Surg Am 2001;83-A:845-854.

American Academy of Orthopaedic Surgeons

Chapter 62 Hip, Pelvis, and Femur: Pediatrics

This article explains and removes the confusion surrounding this osteotomy. Results for 22 patients are presented.

Guille J, Pizzutillo P, MacEwen G: Developmental dysplasia of the hip from birth to six months. J Am Acad
Orthop Surg 2000;8:232-242.
This article reviews diagnosis and management of DDH in
children from birth to age 6 months.

Holen KJ, Tegnander A, Bredland T, et al: Universal or

selective screening of the neonatal hip using ultrasound?: A prospective, randomized trial of 15,529 newborn infants. J Bone Joint Surg Br 2002;84:886-890.
The effect of screening was measured by the rate of late
detection of dysplasia in children from age 6 to 11 years. Universal ultrasound screening was not statistically better than
universal screening by physical examination, with ultrasound
reserved for questionable cases.

Konigsberg DE, Karol LA, Colby S, OBrien S: Results

of medial open reduction of the hip in infants with developmental dislocation of the hip. J Pediatr Orthop
2003;23:1-9.
Forty hips in 32 patients were reviewed at an average of
10.3 years after medial open reduction. Satisfactory results
were obtained in 75% of patients. The authors endorse the
procedure as an option for patients younger than 1 year of
age.

Lerman JA, Emans JB, Millis MB, Share J, Zurakowski

D, Kasser JR: Early failure of Pavlik harness treatment
for developmental hip dysplasia: Clinical and ultrasound
predictors. J Pediatr Orthop 2001;21:348-353.
In 26 of 137 hips, Pavlik harness treatment failed to
achieve or maintain hip reduction. Univariate risk factors for
failure were bilaterality, initial clinical examination, and initial
ultrasound percent coverage.

Luhmann SJ, Bassett GS, Gordon JE, Schootman M,

Schoenecker PL: Reduction of a dislocation of the hip
due to developmental dysplasia: Implications for the
need for future surgery. J Bone Joint Surg Am 2003;85A:239-243.
Reduction of dislocated hips before the appearance of an
ossific nucleus resulted in only a slight increase in ischemic necrosis, but delay more than doubled the need for future surgery. These authors do not recommend a delay.

Sampath JS, Deakin S, Paton RW: Splintage in developmental dysplasia of the hip: How low can we go?
J Pediatr Orthop 2003;23:352-355.
Subluxatable hips at newborn examination were reexamined. By treating only those hips with persistent instability at
2 weeks, and those with persistent anatomic dysplasia at
9 weeks, the rate of abduction splinting was lowered without
adversely affecting outcomes.

741

Hip, Pelvis, and Femur: Pediatrics

Sanchez-Sotelo J, Trousdale RT, Berry DJ, Cabanela
ME: Surgical treatment of developmental dysplasia of
the hip in adults: I. Nonarthroplasty options. J Am Acad
Orthop Surg 2002;10:321-333.
An overview of the etiology, biomechanics, and principles
of treatment of acetabular dysplasia is presented.

Vitale MG, Skaggs DL: Devlopmental dyplasia of the

hip from six months to four years of age. J Am Acad
Orthop Surg 2001;9:401-411.
This article reviews diagnosis and management of DDH in
children from ages 6 months to 4 years.

Weintrob S, Grill F: Current concepts review: Ultrasonography in developmental dysplasia of the hip.
J Bone Joint Surg Am 2000;82-A:1004-1018.
Methods, techniques, indications, and a critical analysis of
screening issues are discussed.

Developmental Coxa Vara

Beals RK: Coxa vara in childhood: Evaluation and management. J Am Acad Orthop Surg 1998;6:93-99.
This article presents an excellent and comprehensive review of coxa vara in children.

Burns KA, Stevens PM: Coxa vara: Another option for

fixation. J Pediatr Orthop B 2001;10:304-310.
The authors reported excellent fixation using a modified
veterinary plate for valgus osteotomy in 9 patients (12 hips).

Widmann RF, Hresko MT, Kasser JR, Millis MB: Wagner multiple K-wire osteosynthesis to correct coxa vara
in the young child: Experience with a versatile tailormade high angle blade plate. J Pediatr Orthop B 2001;
10:43-50.
By stacking Kirschner wires, the effect of a high angle
blade plate was achieved in patients who were otherwise too
small for conventional internal fixation devices.

Slipped Capital Femoral Epiphysis

Futami T, Suzuki S, Seto Y, Kashiwagi N: Sequential
magnetic resonance imaging in slipped capital femoral
epiphysis: Assessment of preslip in the contralateral hip.
J Pediatr Orthop B 2001;10:298-303.
MRI showed growth plate widening on T1 images as an
early sign of SCFE in patients with a preslip condition.

Kennedy JG, Hresko MT, Kasser JR, et al: Osteonecrosis of the femoral head associated with slipped capital
femoral epiphysis. J Pediatr Orthop 2001;21:189-193.
Osteonecrosis was found only in unstable SCFE in 4 of 27
patients. The magnitude of slip and the magnitude of reduction in the unstable group was not predictive of a poorer outcome.

742

Orthopaedic Knowledge Update 8

Loder RT, Greenfield ML: Clinical characteristics of
children with atypical and idiopathic slipped capital
femoral epiphysis: Description of the age-weight test
and implications for further diagnostic investigation.
J Pediatr Orthop 2001;21:481-487.
Patients with SCFE who are younger than age 10 years or
older than age 16 years and those with weight less than the
50th percentile are more likely to have an atypical (nonidiopathic) cause of SCFE such as renal failure, prior radiation
treatment, or endocrine disorder.

Schultz WR, Weinstein JN, Weinstein SL, Smith B: Prophylactic pinning of the contralateral hip in slipped capital femoral epiphysis. J Bone Joint Surg Am 2002;84-A:
1305-1314.
A decision analysis formula based on probabilities of
achieving a good long-term outcome supports prophylactic
pinning of the unslipped opposite hip.

Tokmakova KP, Stanton RP, Mason DE: Factors influencing the development of osteonecrosis in patients
treated for slipped capital femoral epiphysis. J Bone
Joint Surg Am 2003;85-A:798-801.
In this study, osteonecrosis was found only with unstable
SCFE (21 of 36 hips), and complete or partial reduction of unstable SCFE was associated with a higher rate of osteonecrosis.

Legg-Calv-Perthes Disease
Gigante C, Frizziero P: and Turra, S: Prognostic value of
Catterall and Herring classification in Legg-CalvePerthes disease: Follow-up to skeletal maturity of 32 patients. J Pediatr Orthop 2002;22:345-349.
A small study found that Catterall classification was not
prognostic but lateral epiphyseal calcification and epiphyseal
subluxation greater than 4 mm were prognostic of poor outcome. Herring classification was only prognostic when combined with patient age.

Herring J, Kim H: Browne R: Abstract: Legg-CalvePerthes disease: A multicenter trial of five treatment
methods. Pediatric Orthopaedic Society of North America Annual Meeting, Amelia Island, Florida, 2003, p26.
Preliminary results of a landmark prospective study on the
outcome of Legg-Calv-Perthes disease at maturity, showed
that only lateral pillar B hips and borderline B/C hips of patients age 8 years and older had improved outcome with surgery. Lateral pillar A and lateral pillar B hips in patients
younger than 8 years did well without treatment; there was no
treatment effect for lateral pillar C hips.

Classic Bibliography
Castelein RM, Sauter AJ, de Vierger M, et al: Natural
history of ultrasound hip abnormalities in clinically normal newborns. J Pediatr Orthop 1992;12:423-427.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chiari K: Medial displacement osteotomy of the pelvis.
Clin Orthop 1974;98:55-71.
de Sanctis N, Rondinella F: Prognostic evaluation of
Legg-Calve-Perthes disease by MRI: Part II. Pathomorphogenesis and new classification. J Pediatr Orthop
2000;20:463-470.
Ganz R, Klaue K, Vinh TS, Mast JW: A new periacetabular osteotomy for the treatment of hip dysplasias: Technique and preliminary results. Clin Orthop 1988;232:
26-36.
Goldberg MJ, Harcke TH, Hirsch A, Lehmann H, Roy
DR, Sunshine P: Clinical practice guideline: Early detection of developmental dysplasia of the hip (AC0001).
Pediatrics 2000;105:896-905.
Graf R: Fundamentals of sonographic diagnosis of infant hip dysplasia. J Pediatr Orthop 1984;4:735-740.
Guille JT, Lipton GE, Szoke G, et al: Legg-CalvePerthes disease in girls. J Bone Joint Surg Am
1998;80:1256.
Harcke HT, Kumar SJ: The role of ultrasound in the diagnosis and management of congenital dislocation and
dysplasia of the hip. J Bone Joint Surg Am 1991;73:622628.
Herring JA, Neustadt JB, Williams JJ, et al: The lateral
pillar classification of Legg-Calv-Perthes disease.
J Pediatr Orthop 1992;12:143-150.
Loder RT, Richards BS, Shapiro PS, Reznick LR, Aronson D: Acute slipped capital femoral epiphysis: The importance of physeal stability. J Bone Joint Surg Am 1993;
75:1134-1140.
Malvitz TA, Weinstein SL: Closed reduction for congenital dysplasia of the hip: Functional and radiographic results after an average of thirty years. J Bone Joint Surg
Am 1994;76:1777-1792.
McAndrew MP, Weinstein SL: A long-term follow-up of
Legg-Calve-Perthes disease. J Bone Joint Surg Am 1984;
66:860-869.
Meehan PL, Angel D, Nelson JM: The Scottish Rite abduction orthosis for the treatment of Legg-CalvePerthes disease. J Bone Joint Surg Am 1992;74:2.

American Academy of Orthopaedic Surgeons

Chapter 62 Hip, Pelvis, and Femur: Pediatrics

Neyt JG, Weinstein SL, Spratt KF, et al: Stulberg classification system for evaluation of Legg-Calve-Perthes disease: Intra-rater and inter-rater reliability. J Bone Joint
Surg Am 1999;81:1209.
Pemberton PA: Pericapsular osteotomy of the ilium for
treatment of congenital subluxation and dislocation of
the hip. J Bone Joint Surg Am 1965;47:65-86.
Reinker KA: Early diagnosis and treatment of hinge abduction in Legg-Calve-Perthes disease. J Pediatr Orthop
1996;16:3-9.
Salter RB, Dubos J-P: The first fifteen years personal
experience with innominate osteotomy in the treatment
of congenital dislocation and subluxation of the hip.
Clin Orthop 1974;98:72-103.
Salter RB, Thompson GH: Legg-Calve-Perthes disease:
The prognostic significance of the subchondral fracture
and a two-group classification of the femoral head involvement. J Bone Joint Surg Am 1984;66:479-489.
Schoenecker PL, Strecker WB: Congenital dislocation of
the hip in children: Comparison of the effects of femoral
shortening and of skeletal traction in treatment. J Bone
Joint Surg Am 1984;66:21-27.
Staheli LT, Chew DE: Slotted acetabular augmentation
in childhood and adolescence. J Pediatr Orthop 1992;12:
569-580.
Stulberg SD, Cooperman DR, Wallensten R: The natural
history of Legg-Calve-Perthes disease. J Bone Joint Surg
1981;63:1095-1108.
Tonnis D, Behrens K, Tscharani F: A midified technique
of the triple pelvic osteotomy: Early results. J Pediatr
Orthop 1981;1:241-249.
Walters R, Simon S: Joint destruction: A sequel of unrecognized pin penetration in patients with slipped capital femoral epiphysis, in The Hip. St. Louis, MO, CV
Mosby, 1980, p 145.
Weinstein SL, Ponseti IV: Congenital dislocation of the
hip: Open reduction through a medial approach. J Bone
Joint Surg Am 1979;61:119-124.

743

Chapter

63

Knee and Leg: Pediatrics

Kosmas J. Kayes, MD
David A. Spiegel, MD

Anterior Cruciate Ligament Injuries

Anterior cruciate ligament (ACL) tears in patients with
an open physis are on the rise, and optimum treatment
still remains controversial. Most of these injuries occur
during participation in sports activities. The mechanism
is a noncontact, deceleration injury with a valgus stress
component or a hyperextension injury. Patients almost
always report immediate pain and early onset of knee
effusion. Accurate history and physical examination in
the child younger than 12 years is often difficult because
children are not able to provide a good history. The examination may not confirm the diagnosis with only 50%
correlation of physical examination findings with findings on MRI or at time of arthroscopy. Most patients
will have a positive Lachman test, but a pivot shift is often difficult to elicit. Effusion and loss of motion will be
present.
Imaging should consist of plain radiographs of the
knee, including AP, lateral, tunnel, and Merchant views.
MRI is the imaging study of choice for evaluation of an
ACL tear. Studies have shown that MRI can be just as
accurate in adolescents with ACL injuries as in adults.
Specificity and sensitivity of clinical examination (done
by experienced surgeons) and MRI were shown to correlate well with findings at arthroscopy. However, in
children younger than 12 years, MRI findings were significantly lower in both sensitivity and specificity.
In children with open physis at the knee, treatment
is still controversial. Modifying sports activity or not
participating in pivoting sports until near skeletal maturity are options, but often are not acceptable to the family and patient. Bracing and physical therapy may help
but will not eliminate the possibility of recurrent episodes of instability that expose the menisci to injury.
Complete tears of the ACL are associated with a higher
rate of meniscal injury than partial tears, and patients
with recurrent episodes of subluxation are at greater
risk to eventually injure the meniscus. A teenager with
nearly closed physes, with skeletal age 14 years or older
(boys tend to be 2 years behind in skeletal maturity
compared with girls of the same chronologic age), and

American Academy of Orthopaedic Surgeons

at Tanner stage 4 to 5 can probably be considered skeletally mature and therefore receive adult treatments. If a
child has more than 2 cm of growth remaining, hamstring reconstruction should be considered to lessen the
likelihood of injury to the physes. More importantly,
careful attention should be paid to fixation to prevent
damage to the physis.
Primary repair of the ACL has never been shown to
be successful. Extra-articular reconstruction has a high
rate of failure (greater than 50%) over time and may
need to be revised with an intra-articular method when
the child is skeletally mature. Intra-articular reconstruction in the skeletally immature patient carries the risk of
growth disturbance. This possibility needs to be weighed
against the inevitable risk of further and potentially irreversible risk of meniscal damage. Transepiphyseal
techniques that avoid the physis potentially offer the solution but are technically demanding and not well studied to date. The literature has reported instances of injury to the physis in younger children after intraarticular ACL reconstruction but most of these injuries
are related to technical error in graft placement or fixation.
Partial tears of the ACL are unusual but do occur in
children younger than skeletal age 14 years with a normal Lachman test; tears involving less than 50% of the
fibers have been shown to respond well to nonsurgical
reconstruction. Many MRI studies have shown partial
tears of the ACL, however, no good data exist on which
tears to treat surgically and which to treat nonsurgically
with rehabilitation.

Meniscal Tears
Meniscal tears in children are still difficult to diagnose,
with many children not reporting a significant injury
and physical examination being less reproducible, especially in children 12 years of age or younger. MRI can
have a higher false-positive rate for diagnosing meniscal
tears because of increased vascularity of the meniscus in
children; diagnosis is improved with review by a good
pediatric skeletal radiologist. Tears of the medial menis-

745

Knee and Leg: Pediatrics

cus are more common than lateral meniscal tears if discoid menisci are excluded. ACL tears have a 50% rate
of associated meniscal tear.
Meniscal repair should be attempted in children and
adolescents to preserve meniscal tissue if possible. Suture repair is still the gold standard, with the vertical
mattress inside-out technique shown to be the strongest.
Not only will peripheral tears or red-on-white tears
heal, but new evidence suggests that tears that extend
into the white zone greater than 4 mm from the periphery had a 75% healing rate with stable inside-out repair.

Discoid Meniscus
Discoid menisci occur in 3% to 5% of the population.
They are more often lateral, with 25% occurring bilaterally. Discoid menisci are often asymptomatic but can become symptomatic at any age. Symptoms often include
pain and popping in the lateral joint line and mechanical symptoms of catching and locking. The discoid meniscus does not have the same strength characteristics as
a normal meniscus and can be more prone to tear without significant trauma. Classification by Watanabe consists of three types. In type I, the meniscus covers the
entire lateral plateau and is stable. In type II, the meniscus only covers part of the plateau and is stable. Type III
discoid menisci (Wrisberg variant) are unstable because
of absent meniscotibial attachment posteriorly.
Treatment in asymptomatic patients with no tear is
observation. For symptomatic patients, treatment consists of repair or dbridement if a tear is present and
saucerization of the meniscus. For type III unstable menisci, suture stabilization to the capsule is necessary
along with saucerization. Complete meniscectomy
should be avoided because the meniscus will not regenerate.

Orthopaedic Knowledge Update 8

tures, respectively. Treatment of type I injuries is usually
long leg casting with the knee in extension or 10 to 15
flexion for 4 to 6 weeks. Type II lesions can undergo
closed reduction with hyperextension of the knee, then
long leg casting with close follow-up to check for displacement. If the fragment does not reduce to less than
2 mm of displacement, then the same treatment as for
type III lesions should be done. Type III fractures are
typically treated with open or arthroscopic reduction of
the fragment and fixation to secure the fragment while
it heals.
The anterior horn of the medial meniscus or the transverse meniscal ligament may be trapped beneath the fragment in many patients and must be mobilized while the
fragment is reduced and then secured. Meniscal entrapment is reported in 26% of type II injuries and more than
90% of type III injuries. Several different methods of fixation have been reported, including use of Steinmann
pins, cannulated screws into the fragment stopping short
of the physis, or suture technique through fibers of the
ACL at its base and then down through the crater, exiting
out the anterior tibial epiphysis.
Controversy exists as to whether long-term instability continues, either objectively or subjectively, no matter how well the fragment is reduced. The thought is
that some ligaments undergo appreciable interstitial
failure before the fracture and that even if the fracture
is reduced, the ligament may have residual laxity. Most
reports show that even if there is some objective laxity
on clinical examination, most patients (84%) return to
their previous level of activity. A recent report on a
small number of patients demonstrates that residual laxity may not be inevitable because good results with minimal objective or subjective laxity were obtained.

Juvenile Osteochondritis Dissecans

Tibial Eminence Fracture
Avulsion of the tibial eminence occurs with a mechanism similar to an ACL tear. Usually occurring between
8 to 14 years of age, the attachment of the ACL on the
tibial eminence is avulsed off, bringing a varying amount
of bony fragment with it. Historically, these injuries occurred after a fall from a bicycle, but recently more are
reported from mishaps during skiing or while playing
football. The mechanism of injury is usually forceful hyperextension or a direct blow to the distal end of the femur with a flexed knee. A high incidence of associated
injuries (up to 40%) such as medial collateral ligament
(MCL) injuries or meniscal tears has been reported.
Meyers and McKeever classify injuries into three
types. Type I is nondisplaced, type II is elevation of the
anterior fragment up to 3 mm with the posterior hinge
intact, and type III is complete displacement of the
avulsed fragment; they occur in approximately 15%,
39%, and 45% of patients with tibial eminence frac-

746

Juvenile osteochondritis dissecans refers to an osteochondritis dissecans lesion that occurs before physeal
closure. An avascular portion of the subchondral bone
in severe cases can undergo separation from the underlying epiphyseal cancellous bone. The cartilage separates with the subchondral bone, although sometimes
there is very little bone remaining on the back side of
the cartilage. The severity ranges from minimal involvement that results in a soft cartilage bed to complete separation with production of a loose body. The etiology of
this disease is unknown but is highly associated with repetitive microtrauma and/or vascular disruption. Boys
are affected twice as often as girls, and most patients are
very active in sports, giving credence to the theory of
trauma as an etiology. The lateral aspect of the medial
femoral condyle is most often involved (75%). The lateral condyle, patella, trochlea, and tibial plateau can also
develop lesions but are affected much less often. The
posterior aspect of the lateral femoral condyle can have

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

normal variations in ossification that can mimic an osteochondritis dissecans lesion on plain radiographs.
Approximately 50% of patients describe a history of
trauma to the knee with reports of ongoing, vague knee
pain during athletic activities. The pain often improves
with rest. Mechanical symptoms of locking, catching, or
popping can occur when the condition is more advanced
and loose fragments are present. Patients may have pain
with palpation over the condyle that is involved. Effusion may or not be present.
A tunnel view is important, along with an AP and
lateral view of the knee, as many lesions are best seen
on the tunnel view. MRI can be helpful in making treatment decisions. MRI stage I lesions have articular cartilage thickening but no break in the cartilage. Stage II lesions have a break in the cartilage but no increased
signal intensity on T2-weighted images between the
bone fragment and the intact subchondral bone. Stage
III lesions have the increased signal intensity at this interface, an indication that fluid is present between the
fragment and the host bone. This finding is a sign of instability of the fragment. Stage IV lesions are loose bodies. MRI with gadolinium enhancement helps to distinguish between type II and type III lesions
Treatment in skeletally immature patients with a stable lesion depends on the symptoms. Patients with open
physes have a better prognosis for healing of the lesions.
If the patient is relatively asymptomatic, then activity
restriction alone may be enough. However, if symptoms
of pain and joint effusion are present, restricted weight
bearing and possibly immobilization may be warranted
for 6 weeks. Range-of-motion exercises should be performed out of the brace. This treatment plan can result
in healing in up to 90% of patients at 3 to 6 months. If
symptoms persist or the MRI signal does not change,
then drilling of the lesion may be indicated. This procedure helps induce vascular supply into the lesion and facilitates healing. The lesions may be drilled retrograde
through a portal while visualizing through the scope or
antegrade with fluoroscopic guidance to avoid penetrating intact cartilage.
If the lesion is unstable (type III or IV), surgical intervention is indicated. If subchondral bone is present,
the bed can be dbrided and drilled and the fragment
fixed with pins or screws. Results are variable with fixation. If there is no subchondral bone on the fragment,
the piece should be removed and the bed dbrided and
drilled. Other options include autologous chondrocyte
transplantation and osteochondral grafts. If the lesion is
smaller than 2.5 cm, dbridement of the frayed cartilage,
drilling, or microfracture may allow the lesion to heal
with fibrocartilage. In larger lesions, fixation of the fragment should be attempted only if there is subchondral
bone attached to the cartilage. Lesions on the medial
femoral condyle tend to respond the best; lesions of the

American Academy of Orthopaedic Surgeons

Chapter 63 Knee and Leg: Pediatrics

trochlea, tibia, and patella do not have as good a response to the same treatment.
With larger lesions, newer techniques of cartilage restoration can be considered. At a mean 5-year follow-up
(range, 1 to 10 years) of autologous chondrocyte transplantation, 90% of patients had good results. However,
only seven patients in this study were younger than 18
years. Mosaicplasty gave the same results at the same
length of follow-up; again, most patients who were treated
had adult-onset lesions. The outcome of these procedures
in the younger population is still unknown because results
are not separated from those found with adult-onset lesions.

Tibial Tubercle Fractures

These fractures are rare and account for approximately
2% of all epiphyseal fractures. The mechanism of injury
is usually jumping or landing during some type of sports
activity, often basketball or track. The force of the massive quadriceps contraction on takeoff or landing
against the acutely flexing knee pulls off the tubercle.
The Watson-Jones classification of these fractures includes type I (the fracture line is between the ossification centers of the tibial tubercle and the epiphysis before closure), type II (the fracture pattern courses
between the tibial tubercle and the plateau), and type
III (the fracture extends vertically behind the tibial tubercle and through the tibial plateau). A few instances
have been reported in which a tubercle avulsion fracture with an accompanying epiphyseal fracture, either
Salter-Harris II, III, or IV occurs (making a case for a
type IV tibial tubercle fracture pattern). These types of
fractures are extremely rare but can have higher potential for compartment syndrome.
Treatment of nondisplaced type I injuries is cast immobilization for 6 weeks with the leg in extension. Displaced type II and III lesions require open reduction and
screw fixation for stability. Meniscal injury can also occur
in type III fractures. Most injuries occur in older children
just before physeal closure and heal well; however, if this
injury occurs in younger children, smooth pin fixation
should be considered to avoid further damage to the proximal tibial physis. Children may have resultant recurvatum
deformity if younger than 11 years at the time of injury.

Distal Femoral Physeal Fractures

These fractures comprise less than 1% of all pediatric
fractures and about 5% to 6% of physeal fractures. The
undulation of the physis contributes to its stability. The
mechanism of injury is usually a football injury or being
hit by a motor vehicle. Concomitant ligament injury
may occur in up to 50% of patients but often is unrecognized initially. Most injuries are closed, and patients
have pain, deformity, inability to bear weight, and knee
effusion (if the fracture is intra-articular).

747

Knee and Leg: Pediatrics

Salter-Harris type II fractures are the most common
pattern and are seen in adolescents. The risk of growth
arrest is significant (between 19% to 25%) even in this
pattern. Salter-Harris type III fractures usually result
from a valgus stress and involve the medial condyle.
These fractures may be minimally displaced. SalterHarris type IV is uncommon and Salter-Harris type V is
extremely rare, with the diagnosis often made later after
deformity occurs from growth disturbance. Infants respond to remodeling and require minimal reduction
even with significant displacement. In adolescents, minimal remodeling of varus/valgus can be expected; therefore, reduction needs to be near anatomic.
Treatment of nondisplaced Salter-Harris type I and
II fractures consists of cast immobilization with close
follow-up. If reduction is required, these injuries are unstable and fixation should be considered. Smooth Steinmann pins crossed in the metaphysis are effective for
Salter-Harris type I and II fractures with a small metaphyseal fragment. If the metaphyseal fragment is large
enough, cannulated screws can be used transversely
through the metaphyseal fragment into the remaining
metaphysis, thereby avoiding the physis. Salter-Harris
type III and IV fracture patterns require anatomic reduction. If no displacement is present, simple immobilization is adequate; however, most Salter-Harris type III
and IV fractures will require open anatomic reduction
and fixation. Cannulated screws placed parallel to the
physis, which travel transversely across the epiphysis
and the metaphysis, are excellent constructs.

Proximal Tibial Physeal Fractures

Proximal tibial physeal fractures are rare injuries occurring in patients age 8 to 15 years. Half of these fractures
result from motor vehicle crashes and half from injury
during sports participation. They can also be found in
the newborn and in patients with arthrogryposis who
have undergone aggressive stretching in an attempt to
improve range of motion. The mechanism of injury can
be a direct force, flexion injury, or abduction with or
without hyperextension of the lower leg. An associated
MCL injury may be present, especially in Salter-Harris
type III fractures.
Salter-Harris type fractures account for 15% of the injuries and about half are nondisplaced. For those that are
displaced, the metaphyseal portion typically is displaced
medially or posteriorly. Salter-Harris type II fracture patterns (43%) are nondisplaced in up to two thirds of patients. When displacement does occur, the metaphyseal
fragment is usually displaced medially. Salter-Harris type
III fractures (22%) usually involve the lateral epiphysis,
and the MCL is often torn in this pattern. Salter-Harris
type IV fractures account for 27% and Salter-Harris type

748

Orthopaedic Knowledge Update 8

V account for 3% of these injuries. Type V patterns are
not usually not noticed until deformity occurs after the injury.
Treatment of Salter-Harris type I and II patterns is
casting for 6 to 8 weeks if nondisplaced and gentle
closed reduction, often under general anesthetic, if displaced. Crossed pins should be considered for stability
or cannulated screws should be used if the metaphyseal
fragment if large enough. Salter-Harris type III and IV
patterns require open reduction and internal fixation if
any displacement is present. They can be fixed with cannulated screws across the epiphysis for type III and
across both metaphysis and epiphysis for type IV. These
injuries can cause unrecognized arterial damage and
lead to compartment syndrome.

Congenital Dislocation of the Knee

Congenital knee dislocation is a spectrum of deformity
from simple positional contractures that correct spontaneously to rigid dislocation of the knee joint requiring
surgical intervention. The condition often occurs in children with myelomeningocele, arthrogryposis, or other
syndromes such as Larsens. This deformity is often associated with developmental dysplasia of the hip, clubfoot, and metatarsus adductus. It is extremely important
to look for associated hip dysplasia when a knee
dislocation is present. The deformity is recognized at
birth with a knee that is hyperextended and a foot that
is extended toward the patients head because of a
flexed hip. The pathology consists of one or all of the
following conditions: contractures of the quadriceps tendon, an absent suprapatellar pouch, tight collateral ligaments, and anterior subluxation of the hamstring tendons.
If the condition results from positioning in utero, often the deformity is mild and the knee is still fairly flexible. The knee will correct spontaneously within a few
weeks or can be assisted by gentle range-of-motion therapy to accelerate correction. Serial casting also may be
helpful if hip dysplasia is present. A Pavlik harness is
applied once knee range of motion increases enough to
allow proper fit.
A more severe contracture or true dislocation shows
minimal response to stretching and casting. This condition usually is evident within the first 3 months of treatment. Failure to gain 30 of knee flexion by 3 months is
an indication for surgery. Surgical treatment is with V-Y
quadricepsplasty or Z-lengthening of the tendon, release of the anterior joint capsule, transposition posteriorly of the hamstring tendons, and mobilization of the
collateral ligaments. Ninety degrees of knee flexion
should be achieved and then the knee should be placed
in a cast in 45 to 60 flexion for 3 to 4 weeks.

American Academy of Orthopaedic Surgeons

Chapter 63 Knee and Leg: Pediatrics

Orthopaedic Knowledge Update 8

Congenital Dislocation of the Patella

A true congenital dislocation of the patella is rare. This
entity must be distinguished from a recurrent dislocation that occurs later in life. In congenital dislocation of
the patella, the patella is hypoplastic or absent, and the
femoral trochlea is often flattened. The lateral retinaculum is tight, and the patella is completely dislocated laterally. The patella is often adherent to the iliotibial band
and often irreducible over the condyle. Genu valgus is
often present, as is a flexion contracture of the knee.
Surgical intervention is required and involves extensive
lateral retinacular release, often to the greater trochanter, medial plication, and either hamstring tenodesis
or transfer of one half of the patellar tendon.

Nail-Patella Syndrome
Nail-patella syndrome is rare, involving a tetrad of orthopaedic manifestations: (1) nail dysplasia; (2) hypoplastic (or absent) and often dislocated patellae; (3) iliac
horns; and (4) elbow dysplasia, mostly consisting of radial head dislocation. The syndrome is of autosomal
dominant inheritance with great variability in the phenotype. Foot abnormalities may also accompany the diagnosis, as well as upper and lower extremity tendon
contractures. The foot deformities include metatarsus
adductus, pes planus, equinus, clubfoot, vertical talus,
and calcaneovalgus. Many patients have nephropathy
and glaucoma.
The orthopaedic implications primarily involve the
patellar dislocations. Most patients will require early
surgical intervention to stabilize the extensor mechanism and prevent long-term complications. Surgical
treatment is with lateral release (extensive), medial plication, and hemipatellar tendon transfer or hamstring
tenodesis. The nail problems are usually well tolerated.
Surgery for radial head dislocation is usually not needed
because patients usually will not have significant complications. The iliac horns are asymptomatic.

Patellar Fracture
Fractures of the patella are uncommon in children
younger than 16 years. Diagnosis in the younger patient
is difficult and often delayed. The mechanism of injury
is by direct impact or contraction of the quadriceps
muscle. Osteochondral fractures are more likely in the
younger age group, especially after a dislocation event.
Avulsion fractures are classified by location. Superior
pole avulsion is least common; medial avulsion is often
caused by dislocation of the patella and is the most
common, followed by inferior pole avulsion. Lateral
avulsion must be distinguished from the congenital variant bipartite patella. Often a larger fragment is fractured off than appears on radiographs because of the
large cartilaginous portion.

American Academy of Orthopaedic Surgeons

Treatment options vary but generally follow the

same guidelines as for adult fractures. Nondisplaced
fractures can be treated in a cast. For displacement
greater than 3 mm, open reduction and fixation is recommended. Small fracture pieces may be excised and
screw fixation of a larger fragment should be considered.

Evaluation of Genu Varum in Infants and

Children
Bowing of the lower extremities in infants and children
is a common reason for referral to an orthopaedic surgeon. Infantile genu varum is physiologic in most instances. Pathologic causes include infantile Blounts disease, skeletal dysplasias, rachitic syndromes (nutritional
or metabolic), posttraumatic or postinfectious deformities, and rarely, focal fibrocartilagenous dysplasia.
Historical features that elevate the index of suspicion for infantile Blounts disease include obesity, early
ambulation, race (African American), and rarely, a positive family history. The deformity is localized to the
proximal tibia, in contrast to physiologic bowing, where
the deformity is usually more evenly distributed
throughout the diaphyseal region. A varus thrust also
should raise the index of suspicion.
A host of radiographic parameters have been investigated to screen for the development of infantile
Blounts disease (Figure 1). The tibial metaphysealdiaphyseal angle (TMDA) is most frequently used for
this purpose. The TMDA has been shown in several
studies to be effective (< 5% error) for angles less than
9 (physiologic) or greater than 16 (Blounts); however,
a significant percentage of cases may be indeterminate
(9 to 16).
In an attempt to improve the accuracy of radiographic screening, several other measurements have
been evaluated for their sensitivity, specificity, and predictive value. Both the percentage of tibial deformity
(tibial varus/total limb varus) and the ratio of the femoral metaphyseal-diaphyseal angle to the TMDA may be
more sensitive and more specific than the TMDA alone.
The epiphyseal-metaphyseal angle (EMA) may also be
helpful because patients 1 to 3 years of age with a
TMDA of 10 or greater and an EMA of greater than
20 are at increased risk for Blounts disease. Serial radiographs are often required to establish the diagnosis,
and a comprehensive approach to screening should include both standardized measurements and knowledge
of the sequential radiographic changes described by
Langenskild.

Blounts Disease
Infantile Blounts disease is a progressive varus deformity of the proximal tibia. The etiology is most likely
multifactorial, and the pathophysiology involves a dis-

749

Knee and Leg: Pediatrics

Orthopaedic Knowledge Update 8

Figure 1 Radiographic indices used in the evaluation of lower extremity bowing in infants and children. A, Tibiofemoral angle (mechanical). This is the angle in between a line
drawn from the center of the hip to the center of the knee and a line from the center of the knee to the center of the ankle. B, TMDA. The TMDA has been used most commonly
and is measured as the angle between a line drawn though the most distal aspect of the medial and lateral beaks of the proximal tibia and a line perpendicular to the anatomic
axis (most reproducible if drawn along the lateral tibial cortex). C, Femoral metaphyseal-diaphyseal angle. This angle is measured in between the anatomic axis of the femur and
a line drawn perpendicular to a line parallel to the distal femoral physis. D, Epiphyseal-metaphyseal angle. This represents the angle between a line drawn through the proximal
tibial physis (parallel to the base of the epiphyseal ossification center) and a line connecting the midpoint of the base of the epiphyseal ossification center with the most distal
point on he medial beak of the tibia. E, Percent deformity of the tibia ([%DT] = tibial varus [TV]/total limb varus). The %DT is calculated as the degree of tibial varus (angle
between the mechanical axis of the tibia and a line parallel to the distal femoral condyles) divided by the total limb varus (tibial varus + femoral varus [FV]). Femoral varus
represents the angle between the mechanical axis of the femur and the line parallel to the distal condyles of the femur.
turbance of growth in the posterior and medial regions
of the proximal tibia, most likely caused by mechanical
overload in a genetically susceptible individual. The sequence of radiographic changes that evolves over the
first decade of life has been described by Langenskild.
Plain radiographs may suggest a depression in the medial tibial plateau in later stages of the disease; however,
arthrography (or MRI) often reveals that this region is
occupied by unossified cartilage.
The best outcomes are associated with early diagnosis and unloading of the medial joint by either bracing
or osteotomy. Although somewhat controversial, a kneeankle-foot orthosis may be indicated for patients with
stage I and II disease who are younger than 3 years. The
device should be worn at least during weight-bearing
hours, and desirable features include a drop lock and a
valgus producing strap.
Valgus osteotomy is indicated for patients with progressive deformity. Overcorrection is essential, and the

750

distal fragment should be translated laterally. An intraoperative arthrogram facilitates visualization of the tibial joint surface. The lowest risk of recurrence is seen in
those with Langenskild III disease who are younger
than 4 years. Unfortunately, recurrence remains common even after a well-executed osteotomy.
Although MRI is not routinely indicated, it may be
helpful in documenting the presence, size, and location
of a physeal bar in later stages of the disease. Treatment
of a physeal bar must be individualized. In addition to
angular correction, either a resection of the bar or a lateral hemiepiphysiodesis may be required. Limb lengths
should be followed closely, and either lengthening or
contralateral epiphysiodesis may be indicated.
Adolescent or late-onset Blounts disease also involves a progressive varus deformity of the proximal
tibia and is diagnosed in children older than 8 years.
Most patients are obese, and mechanical overload has
been implicated. A common radiographic feature is wid-

American Academy of Orthopaedic Surgeons

Chapter 63 Knee and Leg: Pediatrics

Orthopaedic Knowledge Update 8

Table 1 | Clinical Findings in the Ipsilateral Extremity That

May Be Observed in Patients With Longitudinal Deficiency
of the Tibia and Fibula

Figure 2 Common osseous anatomy of fibular (A) and tibial (B) deficiency.

ening and irregularity of the medial proximal tibial physis, which is consistent with a chronic compressive injury. This multiplanar deformity includes varus,
procurvatum, and internal torsion. Treatment is by osteotomy or hemiepiphysiodesis, and suggested advantages of external fixation include the ability to address
all components of the deformity, to minimize neurovascular complications by enabling a gradual correction, to
make adjustments in alignment postoperatively based
on standing radiographs, and to enable simultaneous
limb lengthening. Hemiepiphysiodesis may be considered for milder deformities in patients with sufficient
growth remaining. A coexisting deformity of the distal
femur may require either hemiepiphysiodesis or osteotomy.

Longitudinal Deficiency of the Fibula

Given the spectrum of coexisting abnormalities associated with longitudinal deficiency of the fibula (Figure 2
and Table 1), postaxial hypoplasia is a term that has
been suggested as more comprehensive in describing
this condition. The treatment plan must address limb
length inequality, foot and ankle deformity, and coexisting abnormalities such as valgus at the knee and ankle.
Growth inhibition varies from 5% to 30% and the fe-

American Academy of Orthopaedic Surgeons

Fibular Deficiency

Tibial Deficiency

Femur
Shortening
External rotation
Lateral condylar hypoplasia
PFFD
Coxa vara
Knee
Absent or attenuated cruciate
ligaments
Genu valgum
Patellar dysplasia and/or
subluxation
Tibia
Shortened
Anteromedial bowing
Fibula
Hypoplastic or absent

Femur
Hypoplasia
PFFD

Ankle
Valgus
Ball and socket
Foot
Equinovalgus
Tarsal coalition
Longitudinal deficiency (lateral)
Fused metatarsals

Knee
Absent cruciates
Dysplastic or absent patella

Tibia
Deficient or absent
Fibula
Angulation (varus, posterior)
Proximal instability/dislocation
Ankle
Fibulocalcaneotalar articulation
Foot
Equinovarus
Tarsal coalition
Longitudinal deficiency (medial)

PFFD = proximal femoral focal deficiency

mur contributes to limb shortening in up to 50% of patients.

Classification schemes have traditionally focused on
the degree of fibular deficiency and the predicted limb
length inequality; however, both the treatment and
prognosis also depend on deformity of the foot and ankle. Fibular morphology does not seem to correlate with
either the percentage of growth inhibition or the degree
of foot deformity; therefore, treatment recommendations should not be based solely on the degree of fibular
deficiency. Congenital lower extremity deficiencies are
associated with a constant percentage of growth inhibition; thus, the discrepancy at maturity can be estimated
at an early age. The position of the foot of the shorter
extremity relative to the longer extremity also may be
used to estimate the percentage of shortening at skeletal
maturity (proximal one third 26%, middle one third
approximately 16% to 25%, distal one third 15%).
The treatment should be individualized. Patients
with a functional (or reconstructable) foot and a milder

751

Knee and Leg: Pediatrics

Orthopaedic Knowledge Update 8

Figure 3 The Kalamchi and Dawe classification for longitudinal deficiency of the tibia. A, Type I represents complete absence of the tibia. B, Type II deficiencies are associated
with absence of the distal tibia. C, Type III deficiencies have a distal tibial deficiency (hypoplasia) with a diastasis of the distal tibiofibular articulation.

projected discrepancy may be treated by foot reconstruction and an epiphysiodesis, a single lengthening
procedure, or both. A foot with three or more rays can
potentially be treated by reconstruction rather than amputation. Those extremities with a nonfunctional foot
and/or a large projected limb-length discrepancy (up to
25 cm) are most commonly treated by amputation
(Syme or Boyd). In general, most patients with complete absence of the fibula require amputation. The indications for amputation versus limb salvage remain nebulous for the subset of patients with a functional (or
reconstructable) foot and a large projected discrepancy.
Although a staged reconstruction including two or three
lengthening procedures is technically feasible, it remains
to be determined whether limb function and patient satisfaction will be enhanced compared with early amputation. The psychosocial impact of multiple reconstructive
procedures during childhood and adolescence must also
be considered.
Coexisting deformities must also be addressed to
maximize outcome. The severity of ankle valgus correlates with the degree of fibular deficiency. There is typically a wedge-shaped distal tibial epiphysis and a tarsal
coalition. Milder degrees of ankle valgus may be treated
by hemiepiphysiodesis, whereas techniques to restore
the lateral buttress in patients with greater deformity include fibular lengthening or distal tibial osteotomy. Valgus deformity at the knee results from hypoplasia of the
lateral femoral condyle, and treatment options include a
distal femoral osteotomy or distal femoral hemiepiphysiodesis (open versus physeal stapling). Anteromedial
bowing of the tibia is common but typically only requires an osteotomy in patients with complete absence
of the fibula.

752

Longitudinal Deficiency of the Tibia

A spectrum of limb abnormalities is seen in patients
with partial or complete absence of the tibia. Clinical
concerns include instability at the knee and the ankle,
an equinovarus foot deformity with or without the loss
of medial rays, and limb-length discrepancy. The classification scheme presented by Kalamchi and Dawe is perhaps most practical (Figure 3). The condition occurs
sporadically; however, both autosomal dominant and
autosomal recessive forms of transmission have been
documented. In contrast to fibular deficiencies, associated congenital anomalies in the musculoskeletal system
(most often in the hand) and the viscera (congenital
heart disease, testicular anomalies) have been reported
in patients with tibial deficiencies.
Complete absence of the tibia is usually treated by
through knee amputation. A small subset of patients
with a strong quadriceps muscle and no knee flexion
contracture may be candidates for fibular centralization
(Brown procedure). Because a significant subset of
these patients may subsequently require an amputation
for progressive flexion contracture, most physicians
have abandoned this procedure. When the proximal
tibia has ossified, stability is achieved by proximal tibiofibular synostosis, and either amputation (Syme or
Boyd) or foot salvage is also performed depending on
the local anatomy. Two or more limb lengthenings may
be required if the foot is retained, and the patients family must be prepared for the effects on the patient of
multiple procedures throughout childhood and adolescence.
The third major type of deformity involves hypoplasia of the distal tibia associated with a diastasis between
the distal tibia and fibula, for which the most common

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

treatment has been early Syme or Boyd amputation.
Limbs with lesser degrees of anticipated discrepancy
and a functional (or reconstructable) foot may be considered for salvage, and the choice of procedure is based
on the local anatomy. Options for ankle stabilization include arthrodesis of either the distal tibia or fibula to either the talus or the calcaneus. Talectomy may be required to facilitate closure of the diastasis, and
synostosis of the distal tibia to the fibula may be required to maintain stability. An osteotomy of the distal
tibia and fibula also may be required to correct angular
deformity, and limb-length discrepancy will need to be
treated by one or more lengthenings and/or epiphysiodesis.

Congenital Pseudarthrosis of the Tibia

and/or Fibula
The etiology of congenital pseudarthrosis of the tibia remains unknown, and the pathophysiology appears to involve a localized deregulation of bone modeling and
growth. The condition is associated with neurofibromatosis in approximately 50% of patients; however, only
10% of patients with neurofibromatosis will develop
pseudarthrosis of the tibia. In contrast to anterolateral
bowing of the tibia, which is associated with a prepseudarthrosis or frank pseudarthrosis of the tibia, posteromedial bowing is generally benign and improves with
time. However, patients with posteromedial bowing will
require either an epiphysiodesis or a limb lengthening
to treat limb-length discrepancy.
The goal of treatment is not only to obtain union,
but also to prevent progressive deformity/refracture, to
treat coexisting limb-length discrepancy and ankle valgus, and to maximize function of the foot and ankle.
Follow-up beyond skeletal maturity is required to determine the outcome of a given treatment method. There is
evidence to suggest that delaying surgery may increase
the chances of achieving union and that results of treatment may be better in patients presenting for surgery at
a later age (older than 5 years).
For patients with anterolateral bowing and a prepseudarthrosis, the goal is to prevent or delay fracture
by full-time immobilization in a clamshell ankle-foot
orthosis or similar device. Prophylactic bypass grafting
has not been widely accepted. The primary goal for an
established pseudarthrosis is to achieve union. Principles of surgery include resecting all abnormal tissue,
providing a stable mechanical environment, and grafting
with autogenous bone. The role of electrical stimulation
as an adjunct has yet to be determined.
Resection and grafting with intramedullary fixation
has been associated with variable rates of union (28%
to 80%); concomitant fibular surgery may improve the
outcome. In the absence of fibular involvement, osteotomy is usually required to facilitate apposition of the

American Academy of Orthopaedic Surgeons

Chapter 63 Knee and Leg: Pediatrics

tibial fragments. A coexisting pseudarthrosis of the fibula worsens the prognosis and should be treated by resection and grafting as well. Intramedullary fixation of
the fibula should be considered. Ankle valgus may be
treated by medial distal tibial hemiepiphysiodesis or distal tibiofibular synostosis because a distal tibial osteotomy may fail to unite.
There has been considerable interest in the use of
external fixation and distraction osteogenesis. Following
resection of the pathologic tissue, acute shortening and
compression at the pseudarthrosis site may be combined
with bone transport via a proximal tibial corticotomy.
Excellent rates of union have been reported; however,
refracture remains a concern. The technique is most appropriate for children older than 5 years. Intramedullary
fixation may help to protect against refracture once
union has been achieved.
Free vascularized fibular grafting results in union
rates of 92% to 95%; however, additional procedures
are commonly required for delayed union at the distal
transfer site, for refracture or stress fractures, for ankle
valgus, and for limb-length discrepancy. Pseudarthrosis
may recur above or below the site of transfer. Donor
site morbidity (weakness, ankle valgus) remains a concern, and restoring fibular continuity by grafting of the
periosteal envelope, or by distal tibiofibular fusion,
should be considered. Ipsilateral fibular transfer may be
considered in those with no evidence of fibular involvement.
Amputation is most commonly performed for resistant pseudarthroses following multiple attempts to
achieve union or when union has been achieved but
function is impaired by chronic pain or stiffness. The
Syme or Boyd techniques are preferred in children,
whereas below-knee amputation may be considered in
skeletally mature patients.
Isolated congenital pseudarthrosis of the fibula is extremely rare. The lesion is typically in the distal third,
and the goals of treatment are to restore fibular continuity and to prevent or treat ankle valgus. Although observation or splinting may be an option in the absence
of a progressive ankle valgus, an attempt to restore fibular continuity by resection, bone grafting, and intramedullary fixation is generally recommended. In younger
children, valgus may improve with growth if union is
achieved. Arthrodesis of the distal fibula to the distal
tibia may be performed as an alternative to resection
and grafting or following failed attempts to achieve
union at the pseudarthrosis. The procedure will prevent
the progression of ankle valgus in older children and adolescents but will not result in correction of valgus. The
most common treatment of an established valgus deformity is either distal tibial hemiepiphysiodesis or osteotomy.

753

Knee and Leg: Pediatrics

Annotated Bibliography
Anterior Cruciate Ligament Injuries
Anderson AF: Transepiphyseal replacement of the anterior cruciate ligament in skeletally immature patients.
J Bone Joint Surg Am 2003;85-A:1255-1263.
In this small study, skeletally immature patients with ACL
tears were treated with transepiphyseal ACL reconstruction,
thereby avoiding crossing the physis with the grafts. The technique is demanding but safe and provided effective treatment
in this population.

Kocher MS, Micheli LJ, Zurakowski D, Luke A: Partial

tears of the anterior cruciate ligament in children and
adolescents. Am J Sports Med 2002;30:697-703.
Partial ACL tears diagnosed by MRI and arthroscopically
confirmed were studied in 45 skeletally mature and 17 skeletally immature patients. Patients who had good outcomes with
nonreconstructive treatment were younger than skeletal age
14 years, had normal or near-normal results on the Lachman
and pivot shift tests, and had 50% or less predominantly posterolateral tears. Patients not meeting those criteria required reconstructive treatment for a good result.

Meniscal Tears
Noyes FR, Barber-Westin SD: Arthroscopic repair of
meniscal tears extending into the avascular zone in patients younger than twenty years of age. Am J Sports
Med 2002;30:589-600.
In this prospective study, 71 meniscal tears extending into
the avascular zone in patients younger than 20 years were examined at an average follow-up of 18 months postoperatively.
The repairs were all made with an inside-out vertical divergent
suture technique. Seventy-five percent of these patients had
good results with no evidence of ongoing symptoms in the tibiofemoral compartment.

Tibial Eminence Fracture

Owens BD, Crane GK, Plante T, Busconi BD: Treatment
of type III tibial intercondylar eminence fractures in
skeletally immature athletes. Am J Orthop 2003;32:103105.
In this study, 12 patients with type III tibial eminence fractures had good results with absorbable suture repair. These
patients had 1 mm or less residual laxity at an average
follow-up of 36 months. Five of the 12 patients had associated
injuries, including two MCL injuries, two lateral meniscal
tears, one medial meniscal tear, and one patellar tendon avulsion.

Shepley RW: Arthroscopic treatment of type III tibial

spine fractures using absorbable fixation. Orthopedics
2004;27:767-769.
Five patients (average age, 15 years) were treated with absorbable polydioxanone pins for type III tibial spine fractures.

754

Orthopaedic Knowledge Update 8

No patient had any objective instability on KT-1000 testing
and no functional limitations in high-demand sporting activities.

Juvenile Osteochondritis Dissecans

Peterson L, Minas T, Brittberg M, Lindahl A: Treatment
of osteochondritis dissecans of the knee with autologous
chondrocyte transplantation. J Bone Joint Surg Am
2003;85-A(suppl 2):17-24.
Fifty-eight knees underwent autologous chondrocyte
transplantation for osteochondritis dissecans lesions (average
of 5.7 cm2). Mean age at the time of surgery was 26.4 years
and mean follow-up was 5.6 years. The population was mixed
regarding lesion size, location, and time of onset (skeletally
mature versus immature). Ninety-one percent of patients had
good or excellent results at follow-up based on clinical evaluation and self-assessment questionnaires.

Nail-Patella Syndrome
Beguiristain JL, De Rada PD, Barriga A: Nail-patella
syndrome: Long-term evolution. J Pediatr Orthop B
2003;12:13-16.
Eight patients with knee pain caused by patellar instability
were reviewed. Five patients were treated surgically and three
were not treated.

Evaluation of Genu Varum in Infants and Children

Bowen RE, Dorey FJ, Moseley CF: Relative tibial and
femoral varus as a predictor of progression of varus deformities of the lower limbs in young children. J Pediatr
Orthop 2002;22:105-111.
Progression of varus deformity was evaluated with six radiographic measurements. The percentage of tibial deformity
(tibial varus/total limb varus) was more sensitive and more
specific than the metaphyseal-diaphyseal angle. All patients
with both a deformity of the tibia greater than 50% and a
TMDA greater than 16 experienced progression of deformity.

Davids JR, Blackhurst DW, Allen BL Jr: Radiographic

evaluation of bowed legs in children. J Pediatr Orthop
2001;21:257-263.
The sensitivity, specificity, and positive predictive value of
the mechanical axis, the TMDA, and the EMA were evaluated
in children 1 to 3 years of age. Those with a TMDA of 10 or
greater and an EMA of greater than 20 are at increased risk
for Blounts disease.

McCarthy JJ, Betz RR, Kim A, et al: Early radiographic

differentiation of infantile tibia vara from physiologic
bowing using the femoral-tibial ratio. J Pediatr Orthop
2001;21:545-548.
A femoral-tibial ratio (FTR = femoral metaphysealdiaphyseal angle/tibial metaphyseal-diaphyseal angle) of 0.7 or
less or 1.4 or greater was associated with a less than 5% error
risk. The FTR had a greater sensitivity and specificity versus
the TMDA alone, with less rotational effect.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Blounts Disease
Chotigavanichaya C, Salinas G, Green T, et al: Recurrence of varus deformity after proximal tibial osteotomy
in Blount disease: Long-term follow-up. J Pediatr Orthop 2002;22:638-641.
The recurrence rate in three groups was evaluated (patients younger than 4 years [46%], patients older than 4 years
with crossed pins [94%], and patients older than 4 years with
external fixation [72%]). Early osteotomy and postoperative
alignment in valgus decreased the risk of recurrence.

Craig JG, van Holsbeeck M, Zaltz I: The utility of MR

in assessing Blount disease. Skeletal Radiol 2002;31:208213.
In patients with infantile Blounts disease, ossification of
the medial tibial epiphysis is delayed, and they may have widening and depression of the medial physis, metaphyseal cartilaginous intrusions, edema, lateral physeal widening, femoral
osteochondral injury, hypertrophy of the medial meniscus, and
focal physeal arrest.

Longitudinal Deficiency of the Fibula

Stanitski DF, Stanitski CL: Fibular hemimelia: A new
classification system. J Pediatr Orthop 2003;23:30-34.
This classification evaluates fibular morphology (I = nearly
normal, II = small fibula, III = absent fibula), tibiotalar joint
and distal tibial epiphyseal morphology (H = horizontal, V =
valgus, S = spherical), the presence of a tarsal coalition, and
the number of rays in the foot.

Longitudinal Deficiency of the Tibia

Alekberov C, Shevtsov VI, Karatosun V, et al: Treatment
of tibia vara by the Ilizarov method. Clin Orthop 2003;
409:199-208.
Feldman DS, Madan SS, Koval KJ, et al: Correction of
tibia vara with six-axis deformity analysis and the Taylor
spatial frame. J Pediatr Orthop 2003;23:387-391.

Chapter 63 Knee and Leg: Pediatrics

Grafting and intramedullary fixation was most successful
when concomitant fibular surgery was performed (osteotomy
or resection, with or without intramedullary fixation). Outcome did not correlate with the initial radiographic appearance or age at surgery. Fibular involvement correlates with
valgus deformity at the ankle.

Kim HW, Weinstein SL: Intramedullary fixation and

bone grafting for congenital pseudarthrosis of the tibia.
Clin Orthop 2002;405:250-257.
Five of 12 pseudarthroses united after one procedure
(three refractures), and 4 of the other 7 eventually healed with
multiple procedures. A poor prognosis was associated with distal pseudarthroses and concomitant pseudarthrosis of the fibula.

Ng BK: Saleh M: Fibula pseudarthrosis revisited treatment with Ilizarov apparatus: Case report and review of
the literature. J Pediatr Orthop B 2001;10:234-237.
Yang KY, Lee EH: Isolated congenital pseudarthrosis of
the fibula. J Pediatr Orthop B 2002;11:298-301.
These two articles review four patients with the rare disorder of pseudarthrosis of the fibula. Treatment priorities are to
restore union and to prevent or treat valgus at the ankle.

Classic Bibliography
Achterman C, Kalamchi A: Congenital deficiency of the
fibula. J Bone Joint Surg Br 1979;61:133-137.
Anderson DJ, Schoeneker PL, Sheridan JJ, et al: Use of
an intramedullary rod for the treatment of congenital
pseudarthrosis of the tibia. J Bone Joint Surg Am 1992;
74:161-168.
Boyd HB: Pathology and natural history of congenital
pseudarthrosis of the tibia. Clin Orthop 1982;166:5-13.

In these two studies the advantages of external fixation for

the treatment of infantile and adolescent Blounts disease are
shown in 91 patients. Multiplanar correction can be achieved
gradually, and alignment can be adjusted based on standing radiographs. Limb-length discrepancy can be addressed simultaneously.

Feldman MD, Schoenecker PL: Use of the metaphysealdiaphyseal angle in the evaluation of bowed legs. J Bone
Joint Surg Am 1993;75:1602-1609.

Tokmakova K, Riddle EC, Kumar SJ: Type IV congenital deficiency of the tibia. J Pediatr Orthop 2003;23:649653.

Johnston CE II: Infantile Tibia Vara. Clin Orthop 1990;

255:13-23.

Both reconstruction and amputation are considered for

this rare deformity, and patients with a functional (or reconstructable) foot and a lesser degree of projected limb-length
discrepancy may benefit from reconstruction of the ankle mortise and subsequent limb lengthening.

Congenital Pseudarthrosis of the Tibia and/or Fibula

Johnston CE II: Congenital pseudarthrosis of the tibia:
Results of technical variations in the Charnley-Williams
procedure. J Bone Joint Surg Am 2002;84-A:1799-1810.

American Academy of Orthopaedic Surgeons

Greene WB: Infantile Tibia Vara. J Bone Joint Surg Am

1993;75:130-143.

Jones D, Barnes J, Lloyd-Roberts GC: Congenital aplasia and dysplasia of the tibia with intact fibula: Classification and management. J Bone Joint Surg Br 1978;60:
31-39.
Kalamchi A, Dawe RV: Congenital deficiency of the
tibia. J Bone Joint Surg Br 1985;67:581-584.
Schoeneker PL, Capelli AM, Millar EA, et al: Congenital longitudinal deficiency of the tibia. J Bone Joint Surg
Am 1989;71:278-287.

755

Chapter

64

Ankle and Foot: Pediatrics

Vincent S. Mosca, MD
William Hennrikus, MD

Clubfoot (Talipes Equinovarus)

Idiopathic clubfoot occurs in approximately 1 of every
1,000 live births, affects males twice as often as females,
and is bilateral in approximately 50% of cases. Recent
genetic research supports the hypothesis of a multifactorial pattern of inheritance in which genetics plays a central role. Environmental factors, such as maternal smoking and amniocentesis in the first trimester, may
modulate the genetic expression of the disorder. Clubfoot can be identified on fetal ultrasound as early as the
12th week of gestation. The sensitivity and specificity
are high, making prenatal counseling about treatment
and prognosis an appropriate and beneficial activity.
The clubfoot is characterized by four basic deformities: (1) cavus (plantar flexion of the forefoot on the
hindfoot, also described as pronation of the forefoot on
the hindfoot); (2) adductus of the forefoot on the midfoot/hindfoot; (3) varus (or inversion) of the subtalar
joint complex; and (4) equinus of the foot at the ankle.
These deformities are not passively correctable. There is
a single (occasionally double) posterior ankle skin
crease and a deep transverse skin crease across the midfoot. In a clubfoot, the talus is smaller than normal, and
the talar neck is deviated in a plantar medial direction.
The axes of the anterior and middle facets of the calcaneus create a more acute angle than in the normal foot,
with the anterior facet oriented inward. This creates a
corresponding varus deformity of the distal end of the
calcaneus, resulting in a medial orientation of the calcaneocuboid joint. There also may be varying degrees of
medial subluxation at that joint.
A complete physical examination of the child is indicated to rule out a neurogenic or syndromic cause for
the deformity. There is no reported increased risk for
developmental dysplasia of the hip in children with
clubfoot; however, examination of the hips in a newborn
is always an important part of the musculoskeletal
screening examination. The diagnosis of clubfoot in a
newborn can and should be based solely on clinical findings. If radiographs are obtained, either before, during,
or after initiation of treatment, they should be taken as

American Academy of Orthopaedic Surgeons

stress dorsiflexion AP and lateral views. There is no indication for routine radiographs or ultrasound imaging
of the childs hips.
The goal of treatment is a plantigrade foot with good
joint mobility that is functional, painless, stable over
time, and free of calluses. Initial treatment is nonsurgical. More than 50 years ago, Kite and Ponseti independently proposed significantly different manipulation and
casting techniques for clubfoot deformity correction
with each reporting extremely high success rates. For
unknown reasons, neither technique came into widespread use and the past five decades have been marked
by a proliferation of extensive and radical clubfoot surgeries. Short- and intermediate-term follow-up studies
on these radical surgical procedures revealed a high percentage of painful, stiff, and deformed feet with the
need for additional surgery in 5% to 50% of cases.
The Ponseti method has recently been reintroduced,
based on scientific support of the efficacy of this
method that was documented in a long-term follow-up
study that was published in 1995. That study showed little difference in appearance, comfort, and function in
25- to 45-year-old patients with clubfoot who were
treated as infants, compared with age-matched controls.
With the Ponseti method, the clubfoot is manipulated for 1 to 2 minutes before application of an aboveknee cast. Cavus is corrected first, by dorsiflexing the
first metatarsal against a fulcrum that is the dorsolateral
aspect of the head of the talus. In subsequent weekly
manipulations and above-knee cast applications, the adductus and varus are corrected by abducting the forefoot against the same fulcrum. Cavus, adductus, and
varus are slightly overcorrected and the foot is externally rotated 60 to 70 in relation to the thigh after four
to seven casts have been worn. Percutaneous tenotomy
of the Achilles tendon is performed in approximately
90% of feet. The final above-knee cast is worn for 3
weeks. The Achilles tendon reforms in a lengthened
state during that time. Straight-last shoes externally rotated on an abduction bar are then worn full-time for 3
months and at night for up to 3 years. At least 95% of
clubfoot deformities can be corrected without the need

757

Ankle and Foot: Pediatrics

for extensive surgery if the treatment is begun soon after birth. To be successful, this method must be studied
in detail, mastered, and should not be modified.
Up to 40% of clubfeet will require lateral transfer of
the tibialis anterior to the lateral cuneiform between the
ages of 2 and 5 years to correct a dynamic muscle imbalance (supination deformity) that is inherent in the clubfoot and not related to the initial method of deformity
correction.

Metatarsus Adductus
Metatarsus adductus is characterized by medial deviation of the forefoot on the midfoot with neutral or slight
valgus alignment of the hindfoot. An important pathogenic factor may be a developmental abnormality that
results in a trapezoid shape of the medial cuneiform
with medial orientation of the first metatarsal-medial
cuneiform joint. Metatarsus varus refers to a similar deformity in which the metatarsals are adducted as well as
supinated. A skewfoot combines adduction and plantar
flexion of the forefoot, with moderate to severe valgus
deformity of the hindfoot. The literature is inconsistent
with these definitions.
Metatarsus adductus can be classified according to
the degree of deformity as mild, moderate, or severe using the heel bisector line. A second classification system,
based on flexibility, is prognostic. Flexible metatarsus
adductus deformities can be easily abducted beyond
straight alignment. Partly flexible metatarsus adductus
deformities correct to a straight foot alignment with
passive abduction, and rigid feet do not straighten manually. Flexible metatarsus adductus, which accounts for
90% to 95% of all deformities, corrects spontaneously
in the first 3 to 5 years of life. Partly flexible and rigid
feet benefit from serial manipulation and casting in infants younger than 1 year of age. The lateral pressure
point is at the calcaneocuboid joint, not the head of the
talus as in clubfoot casting. The subtalar joint is held in
slight inversion and the ankle in slight plantar flexion to
prevent inadvertent eversion of the subtalar joint. Three
to four weekly manipulations and long leg cast applications are needed to slightly overcorrect the deformity. A
holding device, such as a reverse or straight-last shoe,
should be used for several months thereafter to prevent
recurrence of deformity. Good results can be expected
at long-term follow-up without a need for surgery, even
when there is mild to moderate residual deformity. Severe, rigid deformity in the older child may cause pain,
callus formation, and shoe-fitting problems for which
surgery would be indicated.
Tarsometatarsal capsulotomies and osteotomies at
the base of the metatarsals are associated with significant complications. An opening wedge osteotomy of the
trapezoid-shaped medial cuneiform offers treatment at

758

Orthopaedic Knowledge Update 8

the site of the deformity and is associated with few risks
or complications. It may be necessary to perform a concurrent closing wedge osteotomy of the cuboid or osteotomies at the base of the lesser metatarsals.

Congenital Vertical Talus

Congential vertical talus is a dorsolateral dislocation of
the talonavicular joint associated with extreme plantar
flexion of the talus. The foot has a rocker-bottom appearance, and the head of the talus is prominent and
palpable on the plantar-medial aspect of the midfoot.
The hindfoot is plantar flexed. There is a dorsiflexion
contracture of the forefoot on the hindfoot that prevents the creation of a longitudinal arch by passive manipulation. It is reported that the deformity occurs in association with syndromes such as arthrogryposis,
myelomeningocele, and sacral agenesis in more than
50% of patients. Congential vertical talus is bilateral in
50% of affected children. The primary differential diagnosis is positional calcaneovalgus deformity, a condition
that resolves completely in most patients. It is differentiated clinically by flexibility of the midfoot but a dorsiflexion contracture of the ankle joint. A lateral radiograph of a congenital vertical talus in maximum
dorsiflexion shows fixed plantar flexion of the talus. A
lateral radiograph in maximum plantar flexion shows an
irreducible dislocation of the talonavicular joint and
confirms the diagnosis.
Surgery is required for deformity correction in
nearly all feet with congenital vertical talus. Preliminary
serial plantar flexion casting is used to stretch the dorsal
tendons, skin, and neurovascular structures of the foot.
Reduction of the talonavicular joint dislocation is desired, but rarely achieved with casting.
In the past, treatment has been surgical reconstruction between ages 6 and 18 months using a single stage
circumferential release. Surgical reconstruction through
a dorsal approach to the talonavicular joint, with concurrent Achilles tendon lengthening, has more recently
been shown to result in excellent and lasting deformity
correction, with less scar formation and shorter surgical
time. Naviculectomy, by shortening the elongated medial column of the foot, is an effective surgery for correcting recurrent or untreated congenital vertical talus
deformity in the older child. This procedure preserves
some motion in the resultant pseudo-subtalar joint complex. Subtalar and triple arthrodeses have also been recommended in the older age group, but it is technically
challenging to correct severe valgus deformity of the
hindfoot with either procedure, and these surgeries lead
to stress transfer to adjacent joints. Subtalar arthrodesis
has also been associated with progressive overcorrection
of the deformity in some feet.

American Academy of Orthopaedic Surgeons

Chapter 64 Ankle and Foot: Pediatrics

Orthopaedic Knowledge Update 8

Tarsal Coalition
Tarsal coalition is a fibrous, cartilaginous, or bony connection between two or more tarsal bones that results
from a congenital failure of differentiation and segmentation of primitive mesenchyme. It affects at least 1% to
2% of the general population and is most commonly
seen as an autosomal dominant condition with nearly
full penetrance. Talocalcaneal and calcaneonavicular coalitions occur with about equal frequency, are usually bilateral, and together account for nearly all coalitions.
There may be more than one tarsal coalition in the
same foot.
Maturation of the coalition coincides with the development of progressive valgus deformity of the hindfoot
and restriction of subtalar motion, all of which are more
severe in a foot with a talocalcaneal coalition. For an individual with a tarsal coalition who develops pain, the
onset of symptoms may coincide with bony transformation of a previously cartilaginous coalition. This generally occurs between 8 and 12 years of age for those children with calcaneonavicular coalitions, and between 12
and 16 years of age for those with talocalcaneal coalitions. The onset of vague, aching, activity-related pain in
the sinus tarsi area or along the medial aspect of the
hindfoot is often insidious. The peroneal tendons appear
to be in spasm and can develop a late contracture. The
exact etiology and the anatomic location of the pain and
spasm are debated. Examination of the foot reveals
stiffness or rigidity of an everted subtalar joint. The
hindfoot remains in valgus alignment even with toe
standing.
Calcaneonavicular coalitions are best seen on an oblique radiograph of the foot. Talocalcaneal coalitions are
best seen on coronal plane images of a CT scan. A CT
scan should be obtained for feet with calcaneonavicular
coalitions because a talocalcaneal coalition may coexist.
MRI is helpful in identifying coalitions that are still in
the fibrous stage and not visualized on radiographs or
CT scans, but this should not be a first-line study. A
bone scan can be helpful in identifying other possible
etiologies for a rigid flatfoot with an atypical presentation of pain (osteoid osteoma, infection, fracture).
Treatment is indicated for symptomatic tarsal coalitions. Nonsurgical treatment such as cast immobilization, soft orthotic devices, and anti-inflammatory medications should be used initially. Surgical treatment is
indicated if pain recurs after initially successful nonsurgical treatment. Good long-term pain relief can be expected in most patients after resection of calcaneonavicular coalitions with interposition of the extensor
digitorum brevis. An osteologic study has shown at least
three anatomic variations for calcaneonavicular coalitions. Poor results may correlate with the exact pathologic anatomy of the coalition. The short-term results of
resection and soft-tissue interposition for talocalcaneal

American Academy of Orthopaedic Surgeons

coalitions have been good, but the results deteriorate

rapidly with time and long-term studies are not available. Unsatisfactory results with resection have been reported in feet in which the ratio of the surface area of
the coalition to the surface area of the posterior facet
was greater than 50% (as determined by CT scan) and
in which there was significant valgus deformity of the
hindfoot and narrowing of the posterior facet. The independent influence of the size of the coalition was not
determined in that or any other study. Some authors
recommend resection of a symptomatic talocalcaneal
coalition regardless of its size if the foot is not fixed in
severe valgus alignment. The presence of a talar beak
does not necessarily indicate the presence of degenerative arthrosis and is not, by itself, a contraindication for
resection. Documented degenerative arthrosis and persistence or recurrence of pain following coalition resection are considered by some to be indications for triple
arthrodesis, though supportive data are lacking. Severe
symptomatic valgus deformity of the hindfoot associated with a talocalcaneal coalition can be corrected and
pain relieved with a calcaneal osteotomy with or without concurrent resection of the coalition.

Fractures
Intra-articular ankle fractures are uncommon but problematic in children. Four general fracture types are medial malleolar fractures, Tillaux fractures, triplane fractures, and pilon fractures. Controversial aspects in the
treatment of these injuries include the indications for a
CT scan and how much displacement indicates reduction and fixation. Complications stemming from these
injuries center on growth arrest and articular incongruity.
Medial malleolar fractures occur in young children
with an average age of injury of 8 years. The mechanism
of injury is supination and inversion of the foot resulting
in a Salter-Harris type III or IV injury. Three radiographic views of the ankle are recommended: AP (Figure 1), lateral, and oblique. CT is not needed in most instances. Any displacement of a physeal fracture of the
medial malleolus requires reduction and possible fixation. These injuries occur in younger children with many
years of growth remaining; therefore, anatomic reduction is needed to minimize angular deformity and shortening. Treatment strategies involve closed reduction and
pinning or cannulated screw fixation. If open reduction
and internal fixation is needed, minimal stripping of the
periosteum is recommended to prevent iatrogenic
growth arrest (Figure 2). When treating a Salter-Harris
type IV injury, the Thurston-Holland fragment can be
safely removed to visualize the physis. Visualization of
the joint can be accomplished by a small anterior arthrotomy. Fixation should be performed parallel to
rather than across the physis. Screws provide compres-

759

Ankle and Foot: Pediatrics

Orthopaedic Knowledge Update 8

Figure 1 AP radiograph of the ankle showing a displaced medial malleolar physeal

fracture and a displaced distal fibular physeal fracture.

Figure 2 AP radiograph following closed reduction of the fibular fracture and open
reduction and internal fixation of the medial malleolar fracture.

sion and may therefore be superior to Kirschner wires.

Osteotomy and physeal bar resection may be needed if
growth arrest occurs. Long-term follow-up shows that
osteoarthritis results in about 10% of these injuries.
Tillaux fractures occur in older children (average
age, 13 years). The mechanism of injury is supination
and external rotation resulting in the avulsion of the
anterior-lateral distal tibial physis by the pull of the
anterior-inferior tibia-fibula ligament. The maturing distal tibial physis closes centrally, then medially, then laterally. The open lateral physis is vulnerable to injury in
this age group. Three radiographic views are needed. In
one third of patients, the Tillaux fracture is only seen on
the oblique view. The articular surface displacement
needs to be reduced to less than 2 mm to minimize the
development of osteoarthritis. Articular congruity is the
major concern; physeal arrest is not a concern because
the patient is close to skeletal maturity. CT is commonly
used and appears to be better than radiographs in determining if the fracture is displaced greater than 2 mm

(Figure 3). Treatment strategies include attempted

closed reduction by internal rotation, supination, and direct pressure; percutaneous manipulation with a Steinmann pin; and open reduction and internal fixation using a small anterior arthrotomy. Compression screw
fixation can cross the physis because the remaining
growth potential is minimal. Long-term follow-up studies are needed.
Triplane fractures occur in children nearing skeletal
maturity (average age, 14 years). The mechanism of injury is supination and external rotation. The number of
fracture fragments depends on which part of the physis
is closed at the time of injury. The fracture occurs in
three planes: the coronal planethe lateral radiograph
shows a Salter-Harris type II fracture; transverse
planeepiphysiolysis through the anterior and lateral
physis; or sagittal planethe AP radiograph typically
shows a Salter-Harris type III fracture. The extent of the
injury is often underappreciated on plain films; therefore, CT is recommended in most cases. CT defines the

760

American Academy of Orthopaedic Surgeons

Chapter 64 Ankle and Foot: Pediatrics

Orthopaedic Knowledge Update 8

the two most common reasons for a patient with a plantar puncture wound to be referred for orthopaedic care.
Pseudomonas septic arthritis or osteomyelitis should be
suspected in any patient with swelling and foot pain that
occurs after stepping on a nail while wearing a sneaker.
Pseudomonas bacteria commonly grow in the foam rubber in the sole of the sneaker. Pseudomonas have an affinity to invade the cartilaginous joint surfaces and physes. Therefore, in infection stemming from a nail
puncture wound through a sneaker, open surgical dbridement and a course of intravenous antibiotics are recommended. Consultation with an infectious disease specialist should be considered in problematic cases. In
patients with suspected retained foreign bodies, radiographs are indicated to detect radiodense objects such
as needle fragments and some types of glass; ultrasound
can be used to detect radiolucent glass fragments, toothpicks, and splinters. Not all patients require wound exploration and a search for a foreign body. A 3-week period of short leg casting results in the extrusion of
radiolucent foreign bodies in some instances. Wound exploration, when deemed necessary, is best done in the
operating theater with adequate anesthesia, magnification, and fluoroscopy. A tetanus update is recommended
for any patient with a plantar puncture wound.
Figure 3 Reformatted CT showing the extent of displacement of a Tillaux fracture.
number of fracture fragments, defines the displacement,
and helps in planning surgical incisions. Articular congruity, not physeal arrest, is the major concern with
triplane fractures because the patient has almost
reached skeletal maturity. The goal of treatment is to restore the articular displacement to less than 2 mm to
minimize long-term osteoarthritis. Treatment strategies
include closed reduction, closed reduction and fixation,
and open reduction and internal fixation. Fixation can
cross the physis, and screws are recommended to provide compression. The 2-mm limit of displacement may
not be the only factor in long-term outcome. Injury to
the distal tibiofibular joint and damage to the articular
cartilage at the time of injury may lead to long-term
symptoms despite an adequate reduction. Pilon fractures are discussed in chapter 39.

Puncture Wounds
Most puncture wounds of the foot are treated by emergency department and primary care physicians and do
not require orthopaedic care. One study defined the
natural history of plantar puncture wounds by following
63 patients prospectively. Treatment in the emergency
department included surface cleaning alone. Five infections (8%) and two retained foreign bodies (3%) occurred. An infection and/or a retained foreign body are

American Academy of Orthopaedic Surgeons

Apophysitis/Osteochondrosis
Severs disease (apophysitis of the calcaneus) is a common cause of heel pain in the child and adolescent.
Kohlers disease and Freibergs infraction are less common causes of midfoot pain and metatarsal pain in this
age group.

Severs Disease
Severs disease is a traction apophysitis at the insertion
of the Achilles tendon. The patient reports heel pain
that worsens during participation in running sports.
Physical examination shows tenderness at the insertion
of the Achilles tendon on the calcaneus. In addition,
most patients have decreased ankle dorsiflexion compared with the uninvolved side and pain with forced
dorsiflexion of the ankle. The pain associated with Severs disease increases with activity and decreases with
rest. Radiographs are normal and rule out conditions
such as calcaneal bone cyst or calcaneal stress fracture
that can have similar symptoms.
Symptomatic treatment consisting of rest, Achilles
tendon stretching, ice, nonsteroidal anti-inflammatory
drugs, and modification of running activity is recommended. In addition, a 1-cm heel cushion may be helpful to decrease traction on the apophysis and reduce
symptoms. If the pain persists despite the previously
outlined treatment, immobilization in a short leg walking cast for 1 to 2 months is recommended. Unfortunately, an undulating course with possible recurrences of

761

Orthopaedic Knowledge Update 8

Ankle and Foot: Pediatrics

pain is common, and patient education about the natural history of Severs disease is key. The problem abates
at skeletal maturity when the apophysis closes. No longterm complications have been reported.

osteotomy or shortening osteotomy. Metatarsal head resection is not recommended because this procedure may
transfer pressure and pain to an adjacent metatarsal
head.

Kohlers Disease

Adolescent Bunion

Kohlers disease is osteonecrosis of the tarsal navicular.

Radiographs show flattening sclerosis and irregularities.
The disorder is more common in boys and is bilateral in
about 20% of patients. It is believed that Kohlers disease stems from repetitive microtrauma to the midfoot
with weight bearing. A possible genetic predisposition
has been suggested. The patient with Kohlers disease
limps and bears weight on the lateral side of the foot.
Physical examination shows tenderness and occasional
swelling at the medial midfoot. Treatment consists of
foot support. Rest combined with a soft arch support is
recommended for mild symptoms. A short leg walking
cast for 1 to 2 months followed by a soft arch support is
recommended for more severe symptoms. Casting appears to hasten the resolution of symptoms compared
with orthotic devices alone. However, the duration of
time for radiographic bone restoration averages
8 months despite the treatment selected. Clinical deformity such as flatfoot does not occur.

Adolescent bunion is a prominence of the medial aspect

of the first metatarsal head associated with a valgus deformity of the hallux. Weight-bearing AP and lateral radiographs of the feet confirm the clinical diagnosis. Normal radiographic angles between the first and second
metatarsals (intermetatarsal angle) are from 6 to 10
and between the first metatarsal and the proximal phalanx (hallux valgus angle) are from 10 to 20. A bunion
in an adolescent is present if the intermetatarsal angle
exceeds 10 and the hallux valgus angle exceeds 20. The
etiology of adolescent bunion is multifactorial. The patient usually comes for treatment because of the cosmetic deformity and is rarely symptomatic. Physical examination shows abnormal angulation and mobility of
the first metatarsocuneiform joint, medial deviation of
the first metatarsal, and a congruous first metatarsophalangeal joint. The results of surgery in an adolescent
with a bunion are unpredictable with a 50% recurrence
rate. Associated pes planus, a tight Achilles tendon, and
a long great toe result in the highest recurrence rate.
Therefore, surgery should be delayed until skeletal maturity if possible. Suggested treatment options until the
patient reaches skeletal maturity include a shoe with a
wide, rounded toe box, a total contact insole with a
fixed toe separator, and daily exercises combined with a
nighttime thermoplastic splint. Surgery may be considered in patients who have exhausted nonsurgical treatment options and still experience pain that interferes
with normal activities. No single surgical technique is
applicable in all patients. Proximal metatarsal osteotomy should be used with caution if the proximal physis
is open; otherwise, growth arrest and shortening may occur. Shortening can result in transfer metatarsalgia.

Freibergs Infraction
Freibergs infraction is osteonecrosis of one of the metatarsal heads. The etiology may involve repetitive microtrauma disrupting the blood supply. The condition
occurs most often in adolescent female athletes. The
longest metatarsal, usually the second, is typically affected. Radiographs show metatarsal head irregularity,
enlargement, flattening, and sclerosis. In patients with
advanced etiology, joint space narrowing and osteochondritis dissecans may develop. Patients with Freibergs infraction report forefoot pain. Physical examination
shows local swelling and tenderness over the metatarsal
head with occasional stiffness of the metatarsophalangeal joint. Most patients can be treated symptomatically with activity modification, nonsteroidal antiinflammatory drugs, a metatarsal bar shoe insert, or a
change in shoe wear to relieve the weight-bearing stress
on the involved metatarsal head. A short leg walking
cast followed by a metatarsal bar is sometimes required
for more severe cases. Usually, the disease runs its
course and the metatarsal head reconstitutes in about 2
years. In the minority of patients in whom conservative
care is unsuccessful, surgery is indicated. Surgery should
be performed with caution because the reported series
are small and without comparison groups. Surgical options include metatarsophalangeal arthrotomy and removal of loose bodies, drilling the metatarsal head, subchondral bone grafting, interposition arthroplasty using
the extensor digitorum longus tendon, and dorsiflexion

762

Annotated Bibliography
Clubfoot (Talipes Equinovarus)
Dietz F: The genetics of idiopathic clubfoot. Clin Orthop
2002;401:39-48.
An excellent and comprehensive review of this important
aspect of clubfoot is presented.

Herzenberg JE, Radler C, Bor N: Ponseti versus traditional methods of casting for idiopathic clubfoot.
J Pediatr Orthop 2002;22:517-521.
This is the first report from outside of Iowa comparing the
Ponseti method for clubfoot deformity correction with traditional casting followed by extensive posteromedial release surgery. The authors results with 34 clubfeet treated with the
Ponseti method matched the results reported by Ponseti de-

American Academy of Orthopaedic Surgeons

Chapter 64 Ankle and Foot: Pediatrics

Orthopaedic Knowledge Update 8

cades ago and recently, and revealed superiority over the traditional method.

Ippolito E, Farsetti P, Caterini R, Tudisco C: Long-term

comparative results in patients with congenital clubfoot
treated with two different protocols. J Bone Joint Surg
Am 2003;85-A:1286-1294.
Forty-seven clubfeet treated with extensive posteromedial
release (after an average of 16 abduction/pronation casts)
were compared with 49 clubfeet treated with an open Achilles
Z-lengthening and posterior ankle capsulotomy (after an average of six Ponseti-type supination/abduction casts). Follow-up
at 25 years for group one and 19 years for group two showed
significantly better results in the second group, although not as
good as the results reported by Cooper and Dietz using the
traditional Ponseti technique with percutaneous Achilles tenotomy. The authors recommend treating clubfeet according to
the method of Ponseti.

Kuo KN, Hennigan SP, Hastings ME: Anterior tibial tendon transfer in residual dynamic clubfoot deformity.
J Pediatr Orthop 2001;21:35-41.
Forty-two full anterior tibial tendon transfers to the lateral
cuneiform were compared with 29 split anterior tibial tendon
transfers to the cuboid as treatment of dynamic supination deformity of previously operated clubfeet. There was no significant difference in results between the two techniques.

Pirani S, Zeznik L, Hodges D: Magnetic resonance imaging study of the congenital clubfoot treated with the
Ponseti method. J Pediatr Orthop 2001;21:719-726.
MRI was performed on infant clubfeet undergoing manipulation and cast treatment according to the Ponseti method at
the beginning, middle, and end of treatment. Correction of abnormal relationships between tarsal bones was documented.
Additionally, the abnormal shapes of the individual tarsal osteochondral anlages were corrected. This effect can be accounted for by changes in mechanical loading of these fastgrowing tissues.

Skelly AC, Holt VL, Mosca VS, Alderman BW: Talipes

equinovarus and maternal smoking: A population-based
case-control study in Washington state. Teratology 2002;
66:91-100.
This article presents an epidemiology study showing an increased risk of having a child with clubfoot if the mother
smoked during pregnancy. There was a dose response with increased risk correlated with the number of cigarettes smoked.

Tredwell SJ, Wilson D, Wilmink MA, et al: Review of the

effect of early amniocentesis on foot deformity in the
neonate. J Pediatr Orthop 2001;21:636-641.
Clubfoot was identified in 1.63% of live births in mothers
who had undergone early amniocentesis (11 to approximately
12 gestational weeks/day) versus 0.12% in those who underwent midtrimester amniocentesis (15 to approximately 16 ges-

American Academy of Orthopaedic Surgeons

tational weeks/day). The authors hypothesize that decreased

amniotic fluid in the early group, during the time of maximum
foot growth velocity, leads to decreased fetal movement resulting in joint and limb deformities.

Congenital Vertical Talus

Mazzocca AD, Thomson JD, Deluca PA, Romness MJ:
Comparison of the posterior approach versus the dorsal
approach in the treatment on congenital vertical talus.
J Pediatr Orthop 2001;21:212-217.
Twenty-five consecutive congenital vertical tali that were
treated with a traditional posterior release operation (average
age, 18 months; range, 6 to 67 months) were compared with
eight consecutive feet treated with deformity correction via
the dorsal approach (average age, 26.5 months; range, 6 to 65
months). The dorsal approach group required 30% less surgical time, had better clinical scores, and fewer complications including no recurrences (versus 32%) and no cases of osteonecrosis of the talus (versus 48%).

Tarsal Coalition
Cooperman DR, Janke BE, Gilmore A, Latimer BM,
Brinker MR, Thompson GH: A three-dimensional study
of calcaneonavicular tarsal coalitions. J Pediatr Orthop
2001;21:648-651.
Thirty-seven presumed calcaneonavicular tarsal coalitions
from the Todd Osteological Collection in Cleveland were studied. The anterior facet of the calcaneus was completely spared
by the coalition in 8 specimens, partially replaced by the navicular portion of the coalition in 7, and completely replaced by
the navicular portion of the coalition in 22. It is hypothesized
that this variable coalition anatomy is a potential cause for
poor results after resection, particularly if the resection creates
joint instability.

Fractures
Barmada A, Gaynor T, Mubarak SJ: Premature physeal
closure following distal tibia physeal fractures. J Pediatr
Orthop 2003;23:733-739.
A residual physeal gap of more than 3 mm following treatment of a distal tibial physeal fracture was associated with
high rates of premature physeal closure. The authors suggest
that open reduction and removal of entrapped periosteum
may be beneficial.

Horn BD, Crisci K, Krug M, et al: Radiologic evaluation

of juvenile Tillaux fractures of the distal tibia. J Pediatr
Orthop 2001;21:162-164.
CT is the preferred imaging modality for detecting more
than 2 mm of displacement in juvenile Tillaux fractures.

Leetum DT, Ireland ML: Arthroscopically assisted reduction and fixation of a juvenile Tillaux fracture.
Arthroscopy 2002;18:427-429.
Arthroscopic visualization assisted with the anatomic reduction of the articular fragment, obviating the need for ankle
arthrotomy, is discussed.

763

Ankle and Foot: Pediatrics

Orthopaedic Knowledge Update 8

Leets M, Davidson D, McCaffrey M: The adolescent pilon fracture: management and outcome. J Pediatr
Orthop 2001;21:20-26.

Clark M, DAmbrosia R, Ferguson A: Congenital vertical talus: Treatment by open reduction and navicular excision. J Bone Joint Surg Am 1977;59:816-824.

Pilon fractures in adolescents result from high-energy

trauma and are associated with a high complication rate. Results were good to excellent in only 63% of cases. The authors
suggest a new classification system for adolescent pilon fractures.

Cooper D, Dietz F: Treatment of idiopathic clubfoot: A

thirty-year follow-up note. J Bone Joint Surg Am 1995;
77:1477-1489.

Marsh JL, Buckwalter J, Gelberman R, et al: Articular

fractures: does anatomic reduction really change the result? J Bone Joint Surg Am 2002;84-A:1259-1271.
The authors suggest that there is no convincing evidence
that improved articular reduction leads to a better outcome or
a lower risk of osteoarthritis. Rather, the authors suggest that
damage to the articular cartilage at the time of injury is the
most important factor in outcome.

Seifert J, Matthes G, Hinz P, et al: Role of MRI in the diagnosis of distal tibia fractures in adolescents. J Pediatr
Orthop 2003;23:727-732.
The authors report that MRI provided anatomic detail and
information about the joint surfaces superior to plain film radiographs.

Coughlin MJ, Bordelon RL, Johnson KA, et al: Evaluation and treatment of juvenile hallux valgus. Contemp
Orthop 1990;21:169-203.
Duncan RD, Fixsen JA: Congenital convex pes valgus.
J Bone Joint Surg Br 1999;81:250-254.
Ezra E, Hayek S, Gilai AN, Khermosh O, Wientroub S:
Tibialis anterior tendon transfer for residual dynamic
supination deformity in treated clubfeet. J Pediatr Orthop B 2000;9:207-211.
Farsetti P, Weinstein SL, Ponseti IV: The long-term functional and radiographic outcomes of untreated and nonoperatively treated metatarsus adductus. J Bone Joint
Surg Am 1994;76:257-265.

Apophysitis/Osteochondrosis

Freiberg AH: The so-called infarction of the second

metatarsal bone. J Bone Joint Surg 1926;8:257.

Tang SF, Chen CP, Pan JL, et al: The effects of a new
foot-toe orthosis in treating painful hallux valgus. Arch
Phys Med Rehabil 2002;83:1792-1795.

Grosoia JA: Juvenile hallux valgus: A conservative approach to treatment. J Bone Joint Surg Am 1992;74:
1367-1374.

The authors report that using a total contact insole with a

fixed toe separator reduced pain and improved the hallux valgus angle without causing skin ulcers or blisters.

Tsirikos AI, Riddle EC, Kruse R: Bilateral Kohlers disease in identical twins. Clin Orthop 2003;409:195-198.
Bilateral Kohlers disease in identical twins suggests that a
genetic predisposition to the disorder may exist.

Volpon JB, de Carvalo Filho G: Calcaneal apophysitis: A

quantitative radiographic evaluation of the secondary
ossification center. Arch Orthop Trauma Surg 2002;122:
338-341.
A radiographic study of patients with Severs disease compared with normal control patients is presented. The sclerotic
aspect of the calcaneus apophysis is a normal finding and
should not be used to establish the diagnosis of Severs disease.

Classic Bibliography
Bleck E: Metatarsus adductus: Classification and relationship to outcomes of treatment. J Pediatr Orthop
1983;3:2-9.
Caterini R, Farsetti P, Ippolitio E: Long term follow up
of physeal injury to the ankle. Foot Ankle 1991;11:372383.

764

Honein MA, Paulozzi LJ, Moore CA: Family history,

maternal smoking, and clubfoot: An indication of a
gene-environment interaction. Am J Epidemiol 2000;
152:658-665.
Huurman WW, Bhuller GS: Nonoperative treatment of
retained radio lucent foreign bodies in lower limbs.
J Pediatr Orthop 1982;2:506-508.
Ippolitoo E, Ricciardi Pollini PT, Falez F: Kohlers disease of the tarsal navicular: Long-term follow up of 12
cases. J Pediatr Orthop 1984;4:416-417.
Katcherian DA: Treatment of Freibergs disease. Orthop
Clin North Am 1994;25:69-81.
Kling TF, Bright RW, Hensinger RN: Distal tibial physeal fractures in children that may require open reduction. J Bone Joint Surg Am 1984;66:647-657.
Kumar S, Guille J, Lee M, Couto J: Osseous and nonosseous coalition of the middle facet of the talocalcaneal joint. J Bone Joint Surg Am 1992;74:529-535.
Laaveg S, Ponseti I: Long-term results of treatment of
congenital club foot. J Bone Joint Surg Am 1980;62:2331.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 64 Ankle and Foot: Pediatrics

Leonard MA: The inheritance of tarsal coalition and its

relationship to spastic flat foot. J Bone Joint Surg Br
1974;56:520-526.

Rebbeck T, Dietz F, Murray J, Buetow K: A single-gene

explanation for the probability of having idiopathic talipes equinovarus. Am J Hum Genet 1993;53:1051-1063.

McHale KA, Lenhart MK: Treatment of residual clubfoot deformity, the bean-shaped foot, by open wedge
medial cuneiform osteotomy and closing wedge cuboid
osteotomy: Clinical review and cadaver correlations.
J Pediatr Orthop 1991;11:374-381.

Schwab RA, Powers RD: Conservative therapy for plantar puncture wounds. J Emerg Med 1995;13:291-295.

Micheli LJ, Ireland ML: Prevention and management of

calcaneal apophysitis in children: An overuse syndrome.
J Pediatr Orthop 1987;7:34-38.

Seimon L: Surgical correction of congenital vertical talus under the age of 2 years. J Pediatr Orthop 1987;7:
405-411.
Sever JW: Apophysitis of the os calcis. NY Med J
1912;95:1025.

Pennycook A, Makaower R, ODonnell AM: Puncture

wounds of the foot: Can infective complications be
avoided? J R Soc Med 1994;87:581-583.

Wilde P, Torode I, Dickens D, Cole W: Resection for

symptomatic talocalcaneal coalition. J Bone Joint Surg
Br 1994;76:797-801.

Ponseti I: Congenital Clubfoot. Oxford, England, Oxford

University Press, 1996.

Williams GA, Cowell HR: Kohlers disease of the tarsal

navicular. Clin Orthop 1981;158:53-58.

Ponseti IV, Becker JR: Congenital metatarsus adductus:

The results of treatment. J Bone Joint Surg Am 1966;48:
702-711.

American Academy of Orthopaedic Surgeons

765

Chapter

65

Pediatric Spine Trauma

William C. Warner, Jr, MD
Gregory A. Mencio, MD

Introduction
Cervical spine injuries are uncommon in children and
usually are associated with motor vehicle crashes,
pedestrian-vehicle accidents, or falls in young children.
In older children, sports injuries, diving accidents, and
gunshot injuries are the most common causes. An
awareness of the unique aspects of the pediatric cervical
spine and an understanding of its growth and development are necessary for correct diagnosis and proper
treatment. Normal physes may be mistaken for fractures, resulting in overtreatment, and certain fractures
that occur through open physes may be undertreated.
The atlas develops from three ossification centers
(Figure 1). The posterior arches fuse by 3 to 4 years of
age, and the neurocentral synchondrosis between the lateral masses and the body fuse at approximately 7 years
of age. The odontoid process is separated from the body
of the axis by a synchondrosis, which usually is fused by
6 to 7 years of age. This synchondrosis appears as a cork
in a bottle on an open mouth odontoid radiograph. The
lower cervical vertebrae also are composed of three primary ossification centers, one for the body and two for
the neural arches. The neural arches fuse posteriorly by
3 years of age, and the neurocentral synchondrosis fuses
with the body between 3 and 6 years of age. The vertebral bodies are wedge-shaped until 7 years of age, and
then gradually become rectangular.
Upper cervical spine injuries are more common in
children between birth and 8 years of age. After 8 years
of age, the injury patterns become more like those in
adults, with the lower cervical spine more frequently involved. Factors contributing to the increased frequency
of upper cervical spine injuries in the young child include the relatively horizontal facets, the large head size
relative to trunk size, muscle weakness, and the increased physiologic motion of the neck in children.
In young children, diagnosis of a cervical spine injury may be difficult; repeated examinations and a high
index of suspicion often are needed. Upper cervical
spine injuries are frequent in young children with facial
trauma (fractures) and head trauma. Any pain or persis-

American Academy of Orthopaedic Surgeons

tent tenderness, paraspinal muscle spasms, limitation of

motion, or persistent torticollis should alert the examiner to a possible cervical spine injury. An adequate
neurologic examination is difficult in a frightened child,
and frequent examinations may be required to reliably
determine neurologic status.
Initial radiographs should include cross-table lateral,
AP, and odontoid open-mouth views. It is mandatory that
the cervicothoracic junction be visible on the plain radiographs. On the lateral radiographs, four lines should be
drawn corresponding to the anterior vertebral bodies, the
posterior vertebral bodies, the inside of the lamina (spinolaminar line), and the tips of the spinous processes (Figure 2). All four of these lines should follow a smooth, even
contour.The articular facets should be parallel and the interspinous ligaments balanced.The retropharyngeal space
should be less than 7 mm, and the retrotracheal space
should be less than 14 mm in children.An atlanto-dens interval of 4 to 5 mm is normal in young children; in adults
and adolescents, this interval should be 3 mm.The atlantodens interval is increased in young children because a significant portion of the dens is cartilaginous and not visible on plain radiographs. This situation also gives the
appearance of overriding of the atlas on the unossified odontoid on extension lateral radiographic views. Oblique
radiographs show details of the facet joints and pedicles
and are useful for determining if a fracture or fracturedislocation is present. Flexion and extension radiographs
should be obtained only in an awake and cooperative
child under the supervision of a physician; these radiographs may be inappropriate in a very young child or an
obtunded patient. Because of the increased physiologic
motion in young children, pseudosubluxation of the second cervical vertebra on the third or the third cervical vertebra on the fourth may be present, most commonly in
children 1 to 7 years of age. Swischuks line is helpful in
differentiating this phenomenon from true injury.This line
is drawn along the posterior arch (spinolaminar line) of
C1 to C3 and should pass within 1.5 mm of the posterior
arch of C2 (Figure 3).
CT scans with three-dimensional reconstruction views
may be helpful in identifying fractures of the upper cer-

767

Pediatric Spine Trauma

Orthopaedic Knowledge Update 8

Figure 1 A, Ossification centers of the atlas. B, Ossification centers for the axis.
(Reproduced from Copley LA, Dormans JP: Cervical spine disorders in infants and
children. J Am Acad Orthop Surg 1998;6:205.)
vical spine (base of the skull, C1 or C2 vertebra) and in
evaluating atlantoaxial rotatory subluxation. MRI is especially useful for ruling out cervical spine injuries in patients who are obtunded or have a closed head injury and
may be difficult to evaluate because of their associated injuries. In a recent study, MRI was able to clear the cervical spine in intubated, obtunded, and uncooperative
children with suspected cervical spine injuries. MRI also
was useful in documenting or ruling out injuries suggested
by plain radiographs and CT scans. MRI confirmed the
plain radiography diagnosis in 66% of patients and altered
the diagnosis in 34%.
Adequate immobilization of the cervical spine is difficult in children. Because commercial cervical collars often do not fit properly, they do not provide adequate immobilization. Sandbags can be placed on each side of the
head to prevent motion. Spine boards used for children
should be modified to accommodate the large size of the
head in relationship to the trunk. An occipital recess or a
split mattress technique should be used to prevent unwanted flexion of the cervical spine (Figure 4). A halo ring
and vest can be used for immobilization of the cervical
spine in children, but an increased complication rate has
been reported in children compared with adults. CT scanning can help in pin placement to avoid cranial sutures
and thin areas of the skull. Eight to 12 pins with low insertional torques of 1 to 5 inch-lb are used in children.The
vest often must be custom fitted to avoid motion in the
vest portion while the head is fixed in the halo portion of
the orthosis.

Specific Cervical Spine Injuries

Atlanto-Occipital Dislocation
In the past, atlanto-occipital dislocation usually was a
fatal injury, but with current emergency medical care

768

Figure 2 Normal relationships in the lateral aspect of the cervical spine. 1 = spinous
processes, 2 = spinolaminar line, 3 = posterior vertebral body line, and 4 = anterior
vertebral body line. (Reproduced from Copley LA, Dormans JP: Cervical spine disorders in infants and children. J Am Acad Orthop Surg 1998; 6:205.)

and increased awareness of this injury, more children

are surviving atlanto-occipital dislocations. Dislocation
of the atlanto-occipital joint is caused by a sudden deceleration injury, such as a motor vehicle or pedestrianvehicle accident. The childs head is thrown forward on
the relatively fixed trunk, causing sudden craniovertebral separation. Because the atlanto-occipital joint has
little inherent bony stability and most of its stability is
provided by its ligamentous attachments, most atlantooccipital dislocations are unstable and require surgical
stabilization. The diagnosis of this condition may be difficult but is suggested by the mechanism of injury and
the significant amount of anterior soft-tissue swelling
visible on lateral radiographs.
The three most reliable radiographic findings to assist in the diagnosis of atlanto-occipital dislocation are
(1) the Wackenheim line, (2) the Powers ratio, and (3)
the occipital condylar distance. The Wackenheim line is
drawn along the clivus and should intersect tangentially
the tip of the odontoid (Figure 5). An anterior or posterior shift of this line indicates an anterior or posterior
displacement of the occiput on the atlas. The Powers ratio is determined by drawing a line from the basion to
the posterior arch of the atlas and a second line from
the opisthion to the anterior arch of the atlas (Figure 6).
A ratio of more than 1.0 or less than 0.55 represents a

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 65 Pediatric Spine Trauma

Figure 4 Spine boards used for transportation of young children should be modified
to include either an occipital recess (top figure) or a mattress pad (bottom figure) to
accommodate the relatively large head. (Reproduced from Dormans JP: Evaluation of
children with suspected cervical spine injury. Instr Course Lect 2002;51:403.)

Figure 3 The spinolaminar (Swischuks) line is used to differentiate pseudosubluxation from true injury. (Reproduced from Copley LA, Dormans JP: Cervical spine disorders in infants and children. J Am Acad Orthop Surg 1998;6:205.)

disruption of the atlanto-occipital joint. An occipital

condyle facet distance of more than 5 mm from the occipital condyle to the C1 facet also represents a disruption of the atlanto-occipital joint. MRI also is useful for
documenting soft-tissue injury associated with atlantooccipital dislocation.
Atlanto-occipital dislocation should be stabilized
surgically with a posterior occiput to C1 or C2 fusion.
Because of the instability of this injury, the preoperative
use of a halo or traction may be contraindicated.

Fractures of the Ring of C1

Fractures of the ring of C1 are uncommon injuries in
both children and adults. The mechanism of injury is an
axial load to the head; the force is transmitted through
the occipital condyles to the lateral masses of C1. In
adults, the ring usually breaks in two places, but in children the open synchondrosis of C1 allows a single fracture of the ring and a greenstick fracture through the
synchondrosis. Widening of the lateral masses of more
than 7 mm beyond the borders of the axis on an AP radiograph indicates an injury to the transverse ligament.
In children, avulsion of the ligament from its attachments is more likely than a true rupture of the transverse ligament. Nonsurgical treatment is recommended
for most patients with this injury.

Odontoid Fractures
Odontoid fractures are one of the most common cervical spine fractures in children. Most are associated with
head trauma from a motor vehicle crash or a fall from a
height, although odontoid fracture can occur with trivial
head trauma. In children, this fracture most often occurs

American Academy of Orthopaedic Surgeons

Figure 5 Wackenheim clivus-canal line is drawn along the clivus into the cervical
spinal canal and should pass just posterior to the tip of the odontoid. (Reproduced
with permission from Menezes AH, Ryken TC: Craniovertebral junction abnormalities, in
Weinstein SL (ed): The Pediatric Spine: Principles and Practice. New York, NY, Raven,
1994.)
through the synchondrosis of C2 distally at the base of
the odontoid and appears on radiographs as a physeal
(Salter-Harris type I) injury. The fracture usually is apparent on plain lateral radiographs, which show the anterior displacement of the odontoid. If the fracture
through the synchondrosis has spontaneously reduced, it
appears as a nondisplaced Salter-Harris type I fracture.
CT and MRI may be necessary to fully delineate the injury. Most odontoid fractures in children heal uneventfully and complications are rare. Closed reduction is obtained by extension or slight hyperextension of the
neck. At least 50% apposition should be obtained (com-

769

Pediatric Spine Trauma

Orthopaedic Knowledge Update 8

Acute rupture of the tranverse ligament is rare, reported to occur in less than 10% of pediatric cervical
spine injuries; avulsion of the attachment of the transverse ligament to C1 is more common. The transverse
ligament is the primary stabilizer of an intact odontoid
against forward displacement. The normal distance from
the anterior cortex of the dens to the posterior cortex of
the anterior ring of C1 is 4.5 mm in children and a distance of more than this, measured on a lateral radiograph, suggests disruption of the transverse ligament.
CT is useful to show avulsion of the transverse ligament
from the ring of C1. For acute injuries, reduction in extension is recommended, followed by surgical stabilization of C1 and C2 and immobilization for 8 to 12 weeks
in a Minerva cast, halo brace, or cervical orthosis.

(most common and benign); type II, unilateral facet subluxation with 3 to 5 mm of anterior displacement; type
III, bilateral anterior facet displacement of more than 5
mm; type IV, posterior displacement of the atlas (Figure
7). Types III and IV are rare, but neurologic involvement may be present or instantaneous death can occur;
these types must be treated with great care.
Children with acute atlantoaxial rotatory subluxation usually report neck pain and headaches and hold
the head tilted and rotated to one side, resisting any efforts to move the head. If the deformity becomes fixed,
the pain subsides but the torticollis and decreased range
of motion persist.
Radiographic evaluation may be difficult because of
the position of the head. AP and open-mouth odontoid
views should be taken with the shoulders flat and the
head in as neutral position as possible. Lateral masses
that have rotated forward appear wider and closer to
the midline, whereas the opposite lateral mass appears
narrower and farther away from the midline. On the lateral view, the lateral facet appears anterior and usually
wedge-shaped rather than the normal oval shape. Flexion and extension views can be used to exclude instability. CT scanning is useful to show superimposition of C1
on C2 in a rotated position and to determine the degree
and amount of malrotation. Three-dimensional CT scans
are helpful to identify rotatory subluxation. MRI is of
little value unless neurologic findings are present.
Treatment depends on the duration of symptoms.
Many patients probably never receive medical treatment because symptoms are mild and the subluxation
reduces spontaneously over a few days. If rotatory subluxation has been present for a week or less, a soft collar, anti-inflammatory drugs, and an exercise program
are sufficient. If symptoms persist after a week of this
treatment, head halter traction should be initiated, either in the hospital or at home; muscle relaxants and analgesics may be needed. If the subluxation is present for
longer than a month, halo traction can be used. If reduction cannot be obtained or maintained, if signs of instability or neurologic deficits are present, or if the deformity has been present for more than 3 months, posterior
arthrodesis is recommended to relieve muscle spasms
associated with the malrotation and produce normal
head appearance.

Atlantoaxial Rotatory Subluxation

Hangmans Fracture

Atlantoaxial rotatory subluxation is a common cause of

childhood torticollis and is most often caused by trauma
or infection. The most common cause is an upper respiratory infection (Grisels syndrome), but subluxation
can occur after a retropharyngeal abscess, tonsillectomy,
pharyngoplasty, or trivial trauma. Atlantoaxial rotatory
subluxation is classified into four types: type I, unilateral
facet subluxation with an intact transverse ligament

Bilateral spondylolisthesis of C2, or hangmans fractures, are caused by forced hyperextension and are most
frequent in children younger than the age of 2 years,
probably because of the disproportionately large head,
poor muscle control, and hypermobility present in this
age group. The possibility of child abuse must be considered. Radiographs show a lucency anterior to the pedicles of the axis, usually with some forward subluxation

Figure 6 The Powers ratio is determined by drawing a line from the basion (B) to the
posterior arch of the atlas (C) and a second line from the opisthion (O) to the anterior
arch of the atlas (A). The length of line BC is divided by the length of the line OA. A ratio
of more than 1 is diagnostic of anterior atlanto-occipital translation and a ratio of less
than 0.55 indicates posterior translation. (Reproduced with permission from Parfenchuck TA, Bertrand SL, Powers MJ, et al: Posterior occipitoatlantal hypermobiliy in
Down syndrome: An analysis of 199 patients. J Pediatr Orthop 1994; 304.)
plete reduction of the translation is not necessary) before immobilization in a Minerva or halo cast or custom
orthosis for 6 to 8 weeks. Manipulation under anesthesia
or open reduction and internal fixation rarely are required.

Traumatic Ligamentous Dislocation

770

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 65 Pediatric Spine Trauma

Figure 7 Fielding and Hawkins classification of atlantoaxial rotatory displacement showing four types of rotatory fixation. A, Type I, no anterior displacement and odontoid acting
as the pivot. B, Type II, anterior displacement of 3 to 5 mm and one lateral articular process acting as the pivot. C, Type III, anterior displacement of more than 5 mm.
D, Type IV, posterior displacement. (Reproduced with permission from Dormans JP: Evaluation of children with suspected cervical spine injury. Instr Course Lect 2002;51:403.)
of C2 on C3. This injury must be differentiated from a
persistent synchondrosis of the axis. Treatment is symptomatic, with immobilization in a Minerva cast, halo, or
cervical orthosis for 8 to 12 weeks. If union does not occur, posterior or anterior arthrodesis can be done to stabilize the fracture.

Subaxial Injuries
Fractures and dislocations involving C3 through C7 are
rare in children and infants. Because these injuries occur
most frequently in older children and adolescents and
have fracture patterns similar to those in adults, they
generally can be treated as in adults. Atlantoaxial screws
and lateral mass plates have been used successfully for
fixation of unstable fractures of the cervical spine in
children. Image-guided techniques make accurate placement of these implants easier in a childs small vertebrae.

Pediatric Halo Use

Halo vest immobilization is being used with increasing
frequency in children with cervical spine injuries. It affords superior immobilization to a rigid cervical collar
and is easier to apply and more versatile than a Minerva
cast. It permits access for skin and wound care while
avoiding the skin problems (maceration, ulceration) typically associated with both hard collars and casts. However, complication rates as high as 68% have been reported with pediatric halo use. The most common
problems are pin site infections; however, pin perforation and brain abscesses have also been reported. The
thickness of the skull in children is decreased and, in
children younger than age 6 years, it has been suggested
that CT of the skull to measure calvarial thickness can
be helpful in determining optimal sites for pin placement.
In children older than 6 years, the standard adult
halo construct using four pins (two anterolaterally, two
posterolaterally) inserted at standard torques of 6- to 8-

American Academy of Orthopaedic Surgeons

inch-lb is generally successful. In younger children, more

pins (up to 12) placed with lower insertional torques (2to 4-inch-lb) have been advocated (Figure 8). Standard
pediatric halo rings fit most children, but infants and
toddlers usually require custom sizing. Although standard pediatric halo vests are available, custom vests or
body casts generally provide superior fit and immobilization.

Spinal Cord Injury Without Radiographic

Abnormality
The possibility of spinal cord injury without radiographic abnormality (SCIWORA) should be considered
in children, particularly in patients younger than 8 years.
SCIWORA is defined as spinal cord injury in a patient
in whom there is no visible fracture on plain radiographs or CT scan. MRI may be diagnostic in showing
spinal cord edema or hemorrhage, soft-tissue or ligamentous injury, or apophyseal end plate or disk disruption, but is completely normal in approximately 25% of
patients. SCIWORA is the cause of paralysis in approximately 20% to 30% of children with injuries of the spinal cord. Involvement of the cervical spine has been
found to be slightly more common than other levels in
most studies.
Potential mechanisms of SCIWORA include hyperextension of the cervical spine, which can cause compression of the spinal cord by the ligamentum flavum
followed by flexion, which can cause longitudinal traction; transient subluxation without gross failure; or unrecognized cartilaginous end plate failure (Salter-Harris
type I fracture). Ligamentous laxity, hypermobility of
the spine, and immature spinal vasculature are thought
to be contributing factors. Regardless of the specific
mechanism, injury to the spinal cord in this syndrome
occurs because of the variable elasticity of the elements
of the immature spinal column. Experimentally, it has
been shown that the bone, cartilage, and soft tissue in
the spinal column can stretch about 2 inches without
disruption but that the spinal cord ruptures after a 0.25-

771

Pediatric Spine Trauma

Orthopaedic Knowledge Update 8

spine to prevent recurrent injury. Immobilization with a
rigid cervical collar for 2 to 3 months is usually adequate treatment of SCIWORA. There have been no reports of recurrent spinal cord injury when the cervical
spine has been immobilized in this manner. Surgery is
occasionally necessary for unstable injury patterns. The
prevalence of scoliosis following infantile paralysis is
more than 90% for patients with quadriplegia and 50%
for patients with paraplegia. Long-term follow-up to
monitor for spinal deformity is necessary.

Thoracic and Lumbar Fractures

Figure 8 Photograph of a 2-year-old patient showing a halo construct with a total of

10 pins placed with lower insertional torques (2- to 4-inch-lb).
inch displacement. Spinal cord injury occurs when deformation of the musculoskeletal structures of the spinal
column exceeds the physiologic limits of the spinal cord.
Neurologic injury may be complete or incomplete.
Partial spinal cord syndromes reported in SCIWORA
include Brown-Squard, anterior, and central cord syndromes, as well as mixed patterns of injury. Incomplete
neurologic injuries have a good prognosis for recovery,
whereas complete injuries carry a dismal prognosis. Approximately 50% of patients have delayed onset of neurologic symptoms or late neurologic deterioration after
an initially less severe degree of injury.
SCIWORA may also occur in the thoracolumbar
spine in association with high-energy thoracic or abdominal trauma. The mechanisms of injury include vascular insult to the watershed area of the spine associated with profound/prolonged hypotension, distraction
mechanism in the seatbelt-restrained patient, or hyperextension mechanism following a crush injury as most
often occurs when a child is rolled over by a car while in
the prone position, resulting in the spine collapsing into
the chest cavity.
Prognosis following SCIWORA is correlated to MRI
findings, if any are present, and to the severity of neurologic injury. Children younger than 10 years are more
likely to have permanent paralysis than older children, reflecting differences in the types of injuries that occur in
these two age groups. SCIWORA in younger children is
usually the result of higher-energy trauma such as a motor vehicle crash, whereas in older children the mechanism
of injury is more likely to be the result of a lower-energy
event such as sports-related trauma or a fall.
Effective treatment requires careful evaluation of
the cervical spine to exclude osseous or cartilaginous injury or mechanical instability, and stabilization of the

772

Thoracic, lumbar, and sacral fractures are relatively uncommon in children. Most of these injuries are caused
by motor vehicle crashes or falls. The most common injuries are compression fractures and flexion-distraction
injuries. In infants and young children, nonaccidental
trauma (child abuse) may be a cause of significant spinal trauma. Avulsion fractures of the spinous processes,
fractures of the pars or pedicles, or compression fractures of multiple vertebral bodies are the most common
patterns of injury that usually occur from severe shaking
or battering. These injuries may be associated with other
signs of child abuse, including fractures of the skull, ribs,
or long bones and cutaneous lesions. Apophyseal end
plate fractures or slipped apophyses are injuries that are
unique to older children and teenagers whose symptoms
mimic disk herniation.
Compression fractures are caused by a combination
of hyperflexion and axial compression. Because the disk
in children is stronger than cancellous bone, the vertebral body is the first structure in the spinal column to
fail. It is common for children to sustain multiple compression fractures. Compression rarely exceeds more
than 20% of the vertebral body. When loss of vertebral
body height exceeds 50%, the possibility of injury to the
posterior column of the spine should be considered and
is best evaluated with CT. Most of these fractures are
treated nonsurgically with rest, analgesics, and bracing.
Surgical stabilization may be indicated if there is posterior column involvement and instability.
Flexion-distraction injuries (seat belt injuries) occur in
the upper lumbar spine in children wearing a lap belt.
With sudden deceleration, the belt slides up on the abdomen where it acts as a fulcrum. As the spine rotates
around this axis it fails in tension, resulting primarily in
disruption of the posterior column with variable patterns
of extension into the middle and anterior column. Four
patterns of injury have been described. Type A is a bony
disruption of the posterior elements extending to a variable degree into the middle column. Type B is an avulsion
of the spinous process with facet joint disruption or fracture and extension into the vertebral apophysis. Type C is
a disruption of the interspinous ligament with a fracture
of the pars interarticularis extending into the body. Type

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

D is a posterior ligamentous disruption with laminar fracture and disruption of the vertebral apophysis. Because of
the transverse plane of orientation of this group of injuries, abnormalities may be missed by thick-section CT and
may not be detected even with complementary thin sections, unless sagittal reconstructions are included. A lateral radiograph showing widening of the interspinous
space is the most helpful study in diagnosing this fracture,
although increased distance between the spinous processes may occasionally be seen on the AP radiograph.
MRI may be the single best imaging modality because it
can accurately identify soft-tissue and disk injury as well
as having predictive value in determining spinal cord and
neural injury.
Approximately two thirds of patients have intraabdominal injuries including ruptures of internal organs
and mesenteric tears, which may be life-threatening if
not diagnosed and treated appropriately. Neurologic injury is unusual. Lap belt injuries with mostly bony involvement and kyphosis less than 20 can be treated
with hyperextension casting. Those with posterior ligamentous disruption and soft-tissue injury require surgical stabilization with compression instrumentation and
posterior arthrodesis.
Fracture of the vertebral end plate (slipped vertebral
apophysis) usually occurs in teenagers and is characterized by traumatic disruption of the vertebral ring apophysis and disk into the spinal canal. The clinical symptoms are essentially the same as a herniated disk.
Patients may have muscle weakness, sensory changes,
absent reflexes, and root tension signs. This injury most
commonly involves the caudal end plate of L4, but may
occur at any level in the lumbar spine. The injuries may
be purely cartilaginous with herniation of the apophysis
and disk or osseous with fractures of the cortical and
cancellous rim of the vertebral body. This injury usually
cannot be identified on plain radiographs. CT or MRI is
needed to make the diagnosis. Treatment is removal of
the bony and cartilaginous fragments and usually requires more extensive exposure (bilateral laminotomies)
than simple diskectomy.
Fracture-dislocations of the spine are unstable injuries that usually occur at the thoracolumbar junction
and often are associated with neurologic deficits. They
are rare injuries in children that require surgical stabilization and fusion. Burst fractures are also rare injuries
in children that result from axial compression and typically occur at the thoracolumbar junction or in the lumbar spine. The need for surgical treatment is determined
by the stability of the fracture and the presence of neurologic deficits. Nonsurgical treatment is a viable option
in neurologically intact children, although most will develop a progressive angular deformity during the first
year after the fracture. It has been shown that surgical
stabilization prevents kyphotic deformity and decreases
the length of hospitalization. Instrumentation should in-

American Academy of Orthopaedic Surgeons

Chapter 65 Pediatric Spine Trauma

clude two levels above and below the fracture. In children with complete neurologic deficits, longer constructs
provide better stability and may prevent subsequent
paralytic spinal deformity from occurring.

Annotated Bibliography
Specific Cervical Spine Injuries
Flynn JM, Closkey RF, Mahboubi S, Dormans JP: Role
of magnetic resonance imaging in the assessment of pediatric cervical spine injuries. J Pediatr Orthop 2002;22:
573-577.
In this study of 74 children, MRI confirmed the plain radiography diagnosis in 66% and altered the diagnosis in 34%.
MRI is valuable in the evaluation of potential cervical spine
injury, especially in obtunded children or those with equivocal
plain radiographs.

Kenter K: Worley G, Griffin T, Fitch RD: Pediatric traumatic atlanto-occipital dislocation: Five cases and a review. J Pediatr Orthop 2001;21:585-589.
Of five children with traumatic atlanto-occipital dislocation, the three survivors had posterior occipitovertebral fusions. The diagnosis was missed initially in three children. The
authors recommend detailed measurements of the initial cervical spine radiographs in pediatric patients at risk for traumatic atlanto-occipital dislocation.

Lustrin ES, Karakas SP, Ortiz AO, et al: Pediatric cervical spine: Normal anatomy, variants, and trauma.
Radiographics 2003;23:539-560.
Knowledge of the normal embryologic development and
anatomy of the cervical spine is important to avoid mistaking
synchondroses for fractures and to correctly interpret imaging
studies. Familiarity with mechanisms of injury and appropriate
imaging modalities also aids in the correct interpretation of radiographs of the pediatric cervical spine.

Thoracic and Lumbar Fractures

Clark P, Letts M: Trauma to the thoracic and lumbar
spine in the adolescent. Can J Surg 2001;44:337-345.
This article describes thoracolumbar fractures in adolescents. The treatment of these injuries follows many of the
same principles as spinal fractures in adults. Nonsurgical treatment is used more frequently because there is less spinal instability and better tolerance of bed rest and spinal immobilization in this population.

Lalonde F, Letts M, Yang JP, Thomas K: An analysis of

burst fractures of the spine in adolescents. Am J Orthop
2001;30:115-120.
This article describes the results of treatment in 11 children (average age 14.4 years) with burst fractures of the spine;
6 children were treated with posterior spinal fusion and instrumentation. Results showed that (1) mild progressive angular
deformity developed at the site of the fracture; (2) spinal instrumentation and fusion prevented kyphotic deformity and

773

Pediatric Spine Trauma

decreased the length of hospitalization without contributing to
further spinal cord injury; and (3) nonsurgical treatment was a
viable option in neurologically intact children, but progressive
angular deformity occurred during the first year after the fracture.

Reddy SP, Junewick JJ, Backstrom JW: Distribution of

spinal fractures in children: does age, mechanism of injury, or gender play a significant role? Pediatr Radiol
2003;33:776-781.
Of the 2,614 pediatric patients referred to a trauma center
over a 5-year period, 84 sustained vertebral fracture and 50
had neurologic injury without radiographic abnormality. A total of 164 fractures were identified. The thoracic region (T2T10) was most commonly injured, accounting for 47 fractures
(28.7%), followed by the lumbar region (L2-L5) with 38 fractures (23.2%), the midcervical region with 31 fractures
(18.9%), the thoracolumbar junction with 24 fractures
(14.6%), the cervicothoracic junction with 13 fractures (7.9%),
and the cervicocranium with 11 fractures (6.7%). There was no
relationship to gender or mechanism of injury.

Sledge JB, Allred D, Hyman J: Use of magnetic resonance imaging in evaluating injuries to the pediatric
thoracolumbar spine. J Pediatr Orthop 2001;21:288-293.
This study is a retrospective review of 19 children with
thoracolumbar fractures associated with neurologic deficits
from three level 1 trauma centers. The authors conclude that
MRI is the imaging modality of choice for these fractures because it can accurately classify injury to bones and ligaments
and because the cord patterns as determined by MRI have
predictive value of neurologic status.

Classic Bibliography
Akbarnia BA: Pediatric spine fractures. Orthop Clin
North Am 1999;30:521-536.
Banerian KG, Wang AM, Samberg LC, Kerr HH, Wesolowski DP: Association of vertebral end plate fracture
with pediatric lumbar intervertebral disk herniation:
Value of CT and MR imaging. Radiology 1990;177:763765.
Donahue DJ, Muhlbauer MS, Kaufman RA, Warner
WC, Sandford RA: Childhood survival of atlantooccipi-

774

Orthopaedic Knowledge Update 8

tal dislocation: Underdiagnosis, recognition, treatment,
and review of the literature. Pediatr Neurosurg 1994;21:
105-111.
Dormans JP, Criscitiello AA, Drummond DS, Davidson
RS: Complications in children managed with immobilization in a halo vest. J Bone Joint Surg Am 1995;77:
1370-1373.
Finch GD, Barnes MJ: Major cervical spine injuries in
children and adolescents. J Pediatr Orthop 1998;18:811814.
Glass RB, Sivit CJ, Sturm PF, Bulas DI, Eichelberger
MR: Lumbar spine injury in a pediatric population: Difficulties with computed tomographic diagnosis.
J Trauma 1994;37:815-819.
Harris JH, Carson GC, Wagner LK: Radiologic diagnosis of traumatic occipitovertebral dissociation: 1. Normal
occipitovertebral relationships on lateral radiographs of
supine subjects. AJR Am J Roentgenol 1994;162:881-886.
Herzenberg JE, Hensinger RN, Dedrick DK, Phillips
WA: Emergency transport and positioning of young
children who have an injury of the cervical spine. J Bone
Joint Surg Am 1989;71:15-22.
Judd DB, Liem LK, Petermann G: Pediatric atlas fracture: a case of fracture through a synchondrosis and review of the literature. Neurosurgery 2000;46:991-994.
Mubarak SJ, Camp JF, Vueltich W, et al: Halo application in the infant. J Pediatr Orthop 1989;9:612-614.
Odent T, Langlais J, Glorion C, et al: Fractures of the
odontoid process: a report of 15 cases in children
younger than 6 years. J Pediatr Orthop 1999;19:51-54.
Pouliquen JC, Kassis B, Glorion C, Langlais J: Vertebral
growth after thoracic or lumbar fracture of the spine in
children. J Pediatr Orthop 1997;17:115-120.
Subach BR, McLaughlin MR, Albright AL, Pollack IF:
Current management of pediatric atlantoaxial rotatory
subluxation. Spine 1998;23:2174-2179.

American Academy of Orthopaedic Surgeons

Chapter

66

Pediatric Spinal Deformity

Daniel J. Sucato, MD
B. Stephens Richards III, MD

Idiopathic Scoliosis
Idiopathic scoliosis is the most common type of scoliosis
and, as its name implies, there is no known definitive
etiology for this condition. It is defined as a lateral curvature of the spine with a Cobb angle of 10 or greater
and axial plane rotation. The sagittal plane usually demonstrates hypokyphosis in the thoracic spine, junctional
kyphosis between two structural curves, and segmental
hypolordosis of the lumbar spine when a structural
curve is present. Classification of idiopathic scoliosis is
usually defined according to the age of the patient at
the time of curve development as follows: infantile
(from birth to age 3 years), juvenile (from age 3 years to
10 years), and adolescent (from age 10 years to 18
years). The age classifications, although somewhat arbitrary, allow the surgeon to characterize curves and assist
in treatment algorithms from the outset.

Infantile Idiopathic Scoliosis

Infantile idiopathic scoliosis is uncommon, and its etiology is not determined. However, patients who present
with this presumed diagnosis require more careful evaluation because an underlying cause is more often found
with infantile idiopathic scoliosis than juvenile and adolescent scoliosis. A recent multicenter study demonstrated that 21.7% of patients who presented with infantile idiopathic scoliosis measuring greater than or equal
to 20 had neural axis abnormalities, and 80% of these
patients required neurosurgical intervention. The authors of this study recommended that an MRI scan be
obtained at the time of presentation for any patient with
presumed infantile idiopathic scoliosis whose curve
measures 20 or greater.
Infantile idiopathic scoliosis is the least common
type of idiopathic scoliosis. It is more commonly seen in
males when compared with juvenile or adolescent idiopathic scoliosis (1:1 male to female ratio), and left
curves are more often reported.
Spontaneous correction of infantile idiopathic scoliosis may occur, and can be predicted based on the position of the rib relative to the vertebra and the rib-

American Academy of Orthopaedic Surgeons

vertebral angle difference (RVAD) as described by

Metha. If the rib overlaps the vertebral body at the apex
of the curve (phase 2 rib), the curve is likely to progress,
whereas a phase 1 rib pattern (no overlap of the rib) requires measurement of the RVAD to assess for curve
progression. A patient with a phase 1 rib and an RVAD
greater than 20 has a significant risk for curve progression. Careful observation is warranted in patients with
these curves, and treatment with an orthotic device is indicated for those curves that have shown progression or
are greater than 30. For the young patient with a large
curve or those patients who do not tolerate an orthotic
device, a Risser cast is appropriate to loosen up the
spine and allow for improved wear of the orthotic device.
Surgical treatment should be delayed as long as possible, primarily because of the concern with creating a
small chest and its detrimental effect on pulmonary function. Although there are no definitive indications for surgical treatment, progression despite orthotic treatment
or curves greater than 50 are generally accepted indications. A variety of surgical treatment strategies have
been described, with the most predictable outcome resulting from an anterior and posterior fusion. Posterior
instrumentation is generally recommended, and it is possible in even small patients because of the newer and
smaller instrumentations that are currently available. A
5-year follow-up study of 13 patients demonstrated that
convex epiphysiodesis combined with a Luque trolley resulted in overall good results, and instrumented growth
was 32% of what was expected. Newer instrumentation
using dual posterior rods and connectors in the middle
of the construct to allow intermittent lengthenings have
demonstrated favorable early results. This approach appears to limit the typically high complication rate observed in the traditional growing rod scenario. Longer
follow-up is required for these patients.

Juvenile Idiopathic Scoliosis

Juvenile idiopathic scoliosis occurs in patients between
the ages of 3 to 10 years; approximately 15% of all pa-

775

Pediatric Spinal Deformity

tients with idiopathic scoliosis have juvenile idiopathic
scoliosis. Females are more commonly affected than
males, and this disparity increases with age. As with adolescent idiopathic scoliosis, right thoracic curves are
more common.
The natural history of juvenile idiopathic scoliosis is
one of steady progression until age 10 years, when
curves generally progress fairly rapidly. Unlike infantile
idiopathic scoliosis, 95% of all patients with juvenile idiopathic scoliosis have progressive curves, whereas only
5% resolve. The thoracic curves tend to progress more
commonly and require fusion more often. Curve patterns may change with extension of the primary curve
or a development of secondary structural curves.
As with those with infantile idiopathic scoliosis, patients with juvenile idiopathic scoliosis should have MRI
studies to evaluate the neural axis because of a 20% to
25% incidence of abnormalities. These abnormalities often require surgical treatment; therefore, it has been
recommended that MRI be done at the time of the initial evaluation.
The nonsurgical treatment of juvenile idiopathic
scoliosis is similar to that for adolescent idiopathic scoliosis. Bracing is used to treat patients with curves that
have progressed to 30 or for those with curves exceeding 20 that have demonstrated 5 of progression. Surgical management is also similar to that for patients with
adolescent idiopathic scoliosis, although the crankshaft
phenomenon assumes greater importance in the skeletally immature patient.

Adolescent Idiopathic Scoliosis

Adolescent idiopathic scoliosis is relatively common,
with a reported incidence of 2% to 3% for curves between 10 and 20, and an incidence of 0.3% for curves
greater than 30. There is an equal incidence of smaller
curves among males and females, whereas the female to
male ratio of curves greater than 30 is 10:1.
The natural history of adolescent idiopathic scoliosis
has been well studied, and it is generally accepted that
curves progress in two scenarios: (1) continued spine
growth and (2) large curve magnitude despite the completion of spine growth. Continued spine growth is assessed using an array of both clinical and radiographic
parameters. Clinical parameters include female menarcheal status and height measurements performed at
each clinic visit to determine growth velocity. Peak
growth velocity is approximately 10 cm per year and occurs just before the onset of menses in females. Radiographic parameters that indicate skeletal immaturity include an open triradiate cartilage, and a Risser grade 0
to 1. Following the completion of growth, curve progression is more likely for patients with thoracic curves exceeding 45 to 50 and thoracolumbar/lumbar curves exceeding 40, especially when coronal imbalance is

776

Orthopaedic Knowledge Update 8

present. The rate of progression in these curves is approximately 1 per year.
Patients with adolescent idiopathic scoliosis should
have a complete and organized clinical evaluation to
confirm the diagnosis, rule out neural axis abnormalities,
and assist in determining treatment. A careful inspection of the skin is necessary to rule out lesions such as
caf-au-lait spots (neurofibromatosis) and cutaneous
manifestations of dysraphism. The extremities, especially
the feet, should be inspected to ensure that no deformities (cavovarus feet indicating neural axis abnormalities)
are present. Observation of gait (including heel and toe
walking) will allow the surgeon to obtain a general
sense of the patients strength and coordination. The
neurologic examination should include an assessment of
motor function in all muscle groups, a sensory examination, and deep tendon reflexes. Abdominal reflexes that
are symmetric confirm a normal neural axis; when abdominal reflexes are asymmetric, the surgeon should obtain an MRI scan. Examination of the patients back includes an assessment of balance, trunk imbalance, waist
asymmetry, pelvic tilt, and limb-length discrepancy.
Shoulder height differences are important, especially
when deciding whether an upper thoracic curve requires
fusion. The Adams forward bend test is performed to
analyze axial plane deformity, which is an indication
that the curve is structural. Asymmetric forward bending of the patient or tenderness of the spine to palpation
may indicate a neural axis abnormality.
Initial radiographic examination should include PA
and lateral radiographs of the entire spine (lower cervical spine down to the hips). The PA radiograph should
be used to evaluate the spine and determine the Cobb
angle for all curves. Skeletal maturity should be assessed
using the Risser sign and the maturity of the triradiate
cartilage (open or closed). The PA radiograph should always be assessed to ensure that two pedicles are present
for each vertebra, no signs of congenital vertebra exist,
and that there is no evidence for other abnormalities
such as neurofibromatosis (penciling of the ribs, endosteal scalloping, or significantly wedged vertebra). An
assessment of spinal balance can be made by dropping a
C7 plumb line and comparing it with the center sacral
vertical line. Structural characteristics of the proximal
thoracic curve include T1 tilt, Cobb angle of greater
than 30, Nash-Moe apical rotation of II or greater, and
a transitional vertebra between the upper and middle
thoracic curves of T6 or lower. Recently, the clavicle angle (the angle subtended by the intersection of a horizontal line and the line that is tangential to the highest
points of each clavicle) has been shown to be the best
predictor of postoperative shoulder balance. The lateral
radiograph can be used to assess sagittal balance, which
is variable, and to ensure that thoracic hypokyphosis is
present. The rib heads of the thoracic curve should demonstrate apical lordosis in patients with adolescent idio-

American Academy of Orthopaedic Surgeons

Chapter 66 Pediatric Spinal Deformity

Orthopaedic Knowledge Update 8

pathic scoliosis, and its absence should warrant investigation of the neural axis with MRI. A recent study
demonstrated that 97% of patients with adolescent idiopathic scoliosis and a normal MRI study had apical lordosis, whereas this sign was absent in 75% of patients
with scoliosis and a syringomyelia. The presence of a
spondylolysis and/or spondylolisthesis should always be
evaluated at the lumbosacral junction and is present in
1% to 2% of patients with adolescent idiopathic scoliosis. Preoperative radiographic assessment should include
supine best-effort bend films to the right and left.
MRI should not be used as a screening tool and is
indicated when there is an atypical curve pattern, absence of thoracic lordosis, atypical pain, neurologic abnormalities or foot deformities, or extremely rapid curve
progression.
The King-Moe classification has been traditionally
used for determining curve types in patients with adolescent idiopathic scoliosis. This is a five-part classification that was developed to describe thoracic curve patterns to help guide the surgeon when implanting
Harrington instrumentation. As with many orthopaedic
classifications, it has fair interobserver and intraobserver
reliability. The newer classification system described by
Lenke and associates provides a more comprehensive
evaluation of both the PA and lateral radiographs and
also uses best-effort supine bend films. The following
three components are defined in this classification system: curve types (1 through 6), lumbar modifiers (A, B,
or C), and thoracic kyphosis (negative, N = normal, or
positive). Although more comprehensive, this classification system has 42 possible combinations and therefore
introduces more complexity into the curve evaluations.
When each of the three components were evaluated individually, the reliability was similar to that of the KingMoe classification. As with any classification system for
adolescent idiopathic scoliosis, it should always be used
in conjunction with a careful clinical evaluation of the
patient.
Nonsurgical treatment includes observation and
management with an orthotic device. Observation is indicated for any curve less than 45 in a skeletally mature
patient without significant clinical deformity. In the skeletally immature patient, observation of curves less than
25 is warranted. Bracing is used for the skeletally immature patient (Risser grades 0, 1, or 2) with a curve between 30 and 45 or an initial curve between 20 and
25 that has demonstrated 5 of progression. Although
numerous braces are available today, the Boston and
Charleston braces are most commonly used. The Boston
brace can be used to treat all curve patterns. Successful
treatment depends on the amount of time the brace is
worn, as was reported in a recent study in which a compliance monitor was incorporated into the brace. The
Charleston (nighttime) brace has been shown to be ef-

American Academy of Orthopaedic Surgeons

fective in treating thoracolumbar/lumbar curves that

measure between 25 and 35.

Surgical Treatment
Generally, the indications for surgical treatment of patients with adolescent idiopathic scoliosis are thoracic
curves greater than 45 in the skeletally immature patient or greater than 50 in the skeletally mature patient. Because thoracolumbar/lumbar curves are more
likely to progress despite a smaller curve magnitude,
surgical intervention is indicated for patients with
curves greater than 40 to 45, especially when there is
significant rotation and/or translation. The first goal of
surgery is to prevent curve progression with spinal arthrodesis. The second goal of surgery is to safely improve the three-dimensional deformity. Surgical planning depends on the radiographic and clinical deformity
present and the skeletal maturity of the patient. Fusion
levels depend on the surgical approach used and a careful assessment of the radiographs and clinical deformity.
The standard, more traditional posterior approach can
be used for all curve patterns, and it is best for double
or triple curves. The anterior approach is more commonly used for thoracolumbar/lumbar curve patterns
because removal of the disk assists in achieving improvement in coronal plane deformity, and lumbar lordosis can be restored. The anterior approach can also be
used for thoracic curve patterns, especially when hypokyphosis is present. In this instance, correction is
achieved with convex compression, which produces kyphosis. Use of anterior instrumentation requires close
attention to screw length and direction (in the vertebral
body) because of the proximity of the aorta to the left
side of the spine in patients with right thoracic scoliosis.
For those undergoing posterior instrumentation and
fusion, an anterior diskectomy and fusion should also be
performed in those who are skeletally immature (Risser
grade 0 with open triradiate cartilage) and in those patients who have very large (> 80) and stiff (< 50% flexibility index) curves.
Advances in spinal instrumentation have improved
the correction of scoliotic deformities. Studies have
demonstrated improvement and maintenance of deformity when pedicle screws are used in the thoracolumbar/lumbar spine (Figure 1). The use of pedicle
screws in the thoracic spine also improves curve correction when compared with hooks, and initial reports
demonstrate safe placement (Figure 2). However, anatomic studies demonstrate that it may be challenging to
place thoracic pedicle screws, especially on the concavity of the curve, because of the narrow width of the
pedicle and the proximity of the aorta laterally and spinal cord medially. Improvement in lateral and posterior
translation of the thoracic spine may not be significantly
improved when compared with segmental hook and

777

Pediatric Spinal Deformity

Orthopaedic Knowledge Update 8

Figure 1 Preoperative AP (A) and lateral (B) and 3-year postoperative AP (C) and lateral (D) radiographs of a 14-year-old girl who underwent posterior spinal fusion and
instrumentation with a combination of hooks, sublaminar wires, and pedicle screw fixation. Restoration of coronal and sagittal balance is seen in the postoperative radiographs,
with excellent correction of the lumbar curve.

Figure 2 Preoperative (A and B) and 2-year postoperative (C and D) radiographs of a 14-year-old boy after posterior spinal fusion and instrumentation using pedicle screw
fixation alone.
wire fixation; however, the improvement in the ability to
correct axial rotation may prove to be its greatest advantage. Confirmation of screw placement is more difficult in the thoracic spine because radiographic visualization is obscured by the ribs and soft tissues. The results
of using electromyographic stimulation of screws to confirm intrapedicle placement is not as reliable as lumbar
screw stimulation.
The thoracoscopic approach to perform an anterior
release appears as effective as open thoracotomy and
has minimized the incisions required for anterior access
to the spine. A recent study demonstrated that an ante-

778

rior thoracoscopic release/fusion performed with the patient in the prone position is very effective and better
tolerated when compared with a thoracoscopic release
with the patient in the lateral position. Thoracoscopic
instrumentation for single thoracic curves achieves correction comparable to open anterior or posterior instrumentation (Figure 3). Although less scarring provides
excellent cosmetic improvement, the duration of thoracoscopic surgery continues to be significantly longer
when compared with more conventional approaches.
Despite recent advancements in spinal instrumentation and techniques, the ultimate goal of surgical treat-

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 66 Pediatric Spinal Deformity

Figure 3 Preoperative (A and B) and 3-year postoperative (C and D) radiographs of a 14-year-old girl after a thoracoscopic anterior spinal fusion and instrumentation.
ment is to achieve solid fusion while minimizing complications. Autogenous iliac crest bone continues to be the
gold standard to promote fusion in adolescent idiopathic
scoliosis. Neurologic monitoring using somatosensoryevoked potentials and/or motor-evoked potentials is now
the standard of care. A 50% decrease in amplitude and/
or an increase in latency of 10% are generally considered
thresholds for concern for neurologic injury when assessing patients using somatosensory-evoked potentials. Critical threshold values for motor-evoked potentials are not
as clear and are dependent on the mode of stimulation.
The Stagnara wake-up test, which is used when neurologic injury is suspected, is the gold standard for neurologic
assessment. Acute complications from the surgical treatment of adolescent idiopathic scoliosis are relatively rare.
However, the need for revision following a posterior spinal fusion can be as high as 19%, with revision for lateonset surgical pain from prominent hardware occurring
in 8% of patients. Delayed infections can occur up to 3 to
4 years postoperatively and can present as a small draining wound or fluctuance, accompanied by low-grade fevers and a mildly elevated erythrocyte sedimentation
rate. Treatment consists of removal of the instrumentation and primary closure, followed by oral administration
of antibiotics. Intraoperative cultures usually grow Staphylococcus epidermidis or Propionibacterium acnes, an
organism that requires culture incubation up to 2 weeks.

consistent, and comparable in reliability and validity to

the Medical Outcomes Study Short Form-36 Health Survey Questionnaire (SF-36). The Scoliosis Research
Society-22 Patient Outcome Instrument (SRS-22) is
shorter and more focused on health issues related to
scoliosis than is the SF-36. Following surgical treatment
for adolescent idiopathic scoliosis, outcomes using the
SRS-22 have shown significant improvement from the
preoperative status in the domains of pain, general selfimage, function from back condition, and level of activity. Improvement in coronal Cobb angle correction does
not correlate with improved SRS-22 scores.
One study reported that minimum 20-year follow-up
for patients who had posterior surgery with Harrington
instrumentation for idiopathic scoliosis demonstrated
similar back function when these patients were compared with an age-matched control group from the general population. Those who underwent surgery had a
greater likelihood of having back pain (78% versus
58%) and lumbar pain (65% versus 47%). In a 50-year
follow-up study of untreated patients compared with a
matched control group, the authors reported a greater
incidence of shortness of breath (22% versus 15%),
which was associated with a Cobb angle of greater than
80. Despite a greater likelihood of chronic back pain
(61% versus 35%), patients with scoliosis were productive and functioning at a high level.

Scoliosis Research Society-22 Patient Outcome

Instrument

Congenital Spinal Deformities

This health-related quality-of-life patient questionnaire

was developed to correlate patients perceptions and
satisfaction with their scoliotic deformities. Studies have
shown this instrument to be simple to use, internally

American Academy of Orthopaedic Surgeons

Congenital vertebral abnormalities lead to a variety of

spinal deformities, including scoliosis, kyphosis, or a
combination of the two. Its cause remains unknown. Genetic abnormalities or any other traumatic or teratologic
type of maternal insults during pregnancy are rarely Ki-

779

Pediatric Spinal Deformity

dentified in patients with congenital vertebral abnormalities. Recently, congenital scoliosis has been documented in two monozygotic twins, a finding that
contradicts most of the findings in the congenital deformity literature. A potential increase in exposure to
chemical fumes and carbon monoxide in mothers of
children with congenital spine deformities has also been
reported.
Spinal dysraphism, which includes numerous abnormalities such as diastematomyelia, syringomyelia, diplomyelia, Arnold-Chiari malformations, intraspinal tumors,
and tethering of the spinal cord, is consistently found in
30% to 40% of patients with congenital spinal deformities. MRI, which best identifies these dysraphic abnormalities, has been recently recommended for all patients
with congenital spinal deformity as part of the initial
evaluation, even in the absence of clinical findings. The
mere presence of a potentially tethering intraspinal lesion may be sufficient reason for prophylactic surgical
treatment to address the lesion before the development
of any neural dysfunction. This approach, in the absence
of neurologic findings, remains controversial.

Congenital Scoliosis
The variety of vertebral anomalies found in congenital
scoliosis makes its natural history uncertain. The two basic types, defects of vertebral formation and defects of
vertebral segmentation, may occur separately or in combination. In 80% of patients with congenital scoliosis,
the anomalies can be classified into one of the two
types.
Defects of vertebral formation may be partial or
complete. True hemivertebrae result from the complete
failure of formation on one side and cause the formation of laterally based wedges consisting of half the vertebral body, a single pedicle, and hemilamina. When
present in the thoracic spine, hemivertebrae are usually
accompanied by extra ribs. When located at the lumbosacral junction, a significant obliquity between the
spine and pelvis can result and is usually accompanied
above by a long compensatory scoliosis. This lumbosacral deformity is best treated surgically (usually with
hemivertebrectomy) at an early age before the compensatory curve becomes fixed.
Defects of segmentation result in an osseous bridge
between two or more vertebrae, either unilaterally or
involving the entire segment. The combination of unilateral failure of segmentation and contralateral hemivertebra carries the worst prognosis in congenital scoliosis
because it produces the most severe and rapidly progressive deformity. Curves of this kind located in the
thoracolumbar spine can be expected to exceed 50 by
the age of 2 years. Without treatment, patients with thoracolumbar, midthoracic, or lumbar curves experience
severe deformity at an early age. Rib fusions that ac-

780

Orthopaedic Knowledge Update 8

company unilateral segmentation defects on the concavity of the curve may adversely affect thoracic growth, resulting in severe limitations in pulmonary function and
growth. This rare condition is known as thoracic insufficiency syndrome, which is defined as the inability of the
thorax to support normal respiration or lung growth.
Left untreated, progressive deterioration occurs and can
result in death at an early age. Methods to expand the
thoracic cage are being investigated, with the goals being to provide an acute increase in the thoracic volume
with stabilization of any flail chest wall defects and to
maintain these improvements as the patient grows.
In patients with severe congenital scoliosis, plain radiographs may not provide sufficient information regarding vertebral abnormalities. Should surgical intervention be necessary, a CT scan with reformatted threedimensional reconstruction provides excellent detail for
the understanding of the deformity. These reconstructions are extremely valuable in the preoperative planning for severe deformities. MRI of the spine also must
be done for all patients with congenital scoliosis who
are undergoing surgical intervention to assess intracanal
abnormalities (30% to 40% prevalence). Currently, reformatted images from MRI can provide a clear picture
of the canal contents despite the severe threedimensional deformity associated with some congenital
scoliosis patients.
Although bracing has no beneficial effect on congenital curves, it may help to control long flexible compensatory curves below the congenital component. The primary goal of surgical intervention is to stop further
progression. Partial correction that can be obtained
safely is an added benefit. Even if the curves are relatively small (< 40), once curve progression has been
confirmed, surgical intervention should be undertaken.
Surgery should be performed prior to the development
of notable deformity, a concept different from that in idiopathic scoliosis. Various surgical approaches are used
depending on patient maturity, deformity location, and
type of deformity. These approaches include anterior
and posterior convex hemiepiphysiodesis, anterior and
posterior spinal fusion, posterior fusion with or without
instrumentation, hemivertebra excision, and spine osteotomies. Expansion thoracoplasty, a new approach to
surgery for congenital scoliosis, is indicated for very
young patients who either have or will potentially have
thoracic insufficiency syndrome. It consists of lengthening the concave hemithorax with rib distraction by
means of a prosthetic rib distractor (Figure 4). This results in increased growth of both the concave and convex sides of the curve, thus allowing additional volume
for growth of the underlying lungs. This technique, currently undergoing clinical trials, appears very promising.
In patients with severe congenital spinal deformities,
slow gradual correction has recently been reported to
be successful in some individuals using preoperative

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 66 Pediatric Spinal Deformity

Figure 4 A, Radiograph of a 4-month-old girl with congenital scoliosis, fused ribs, and a unilateral unsegmented bar. B, Radiograph of the same patient at age 11 years. After
several expansion thoracoplasty procedures, the patients hemithorax is increased in size beyond that expected with no intervention.
halo traction for 6 to 12 weeks. If this method is used,
very close monitoring for any neurologic change (numbness, tingling, and weakness) is essential. When partial
correction is obtained or a plateau has been reached,
the spine is stabilized by instrumentation and fusion.
Few patients with congenital scoliosis secondary to a
hemivertebra need to have the hemivertebra excised.
The main indication for hemivertebra excision is a fixed
decompensation in a patient in whom adequate alignment cannot be achieved through other procedures
(usually involving a hemivertebra of the fourth or fifth
lumbar level). Although a combined anterior and posterior resection has been the standard procedure for
hemivertebra excision, several recent studies report success with excision through a posterior approach only,
with correction maintained using transpedicular instrumentation.

Congenital Kyphosis
Congenital kyphosis represents an abrupt posterior angulation of the spine resulting from a localized congenital malformation of one or more vertebrae. This deformity is caused either by defects of formation (type 1),
defects of segmentation (type 2), or a combination of

American Academy of Orthopaedic Surgeons

the two. In contrast to congenital scoliosis, failure of segmentation in congenital kyphosis is less common and
produces much less deformity than failure of formation.
In type 1 kyphosis, there is a partial (or complete) deficiency of the vertebral body, but the posterior elements
remain present. With growth, a relentless progression in
the kyphosis occurs, leading to anterior impingement on
the spinal cord. When this type of deformity is diagnosed, plans for surgical intervention should begin immediately because of the risk of neurologic deficits.
In type 2 kyphosis, the anterior portion of two or
more adjacent vertebral bodies are fused, which leads to
a deformity that is less progressive, produces less deformity, and has a much lower risk of paraplegia than that
seen in patients with type 1 kyphosis.
When imaging congenital kyphosis, MRI will provide the clearest picture of the spinal cord and, in very
young patients, the clearest picture of the vertebral bodies. Cord compression may be evident on MRI before
any clinical neurologic deficits. Three-dimensional imaging of the spine using CT scan reconstructions is useful
for the evaluation of the vertebral anomalies, especially
in the older child. Both tests should be obtained before
any surgical intervention.

781

Pediatric Spinal Deformity

Because nonsurgical treatment has no beneficial effect on congenital kyphosis, the use of an orthotic device is inappropriate. In patients with type 1 kyphosis,
surgical intervention should be considered, even in the
infant, with the primary goal being prevention of future
paraplegia. If the deformity is recognized in patients
younger than age 3 to 5 years and before the kyphosis
exceeds 50, simple posterior fusion without instrumentation may be considered. A hyperextension cast is used
postoperatively for 4 to 6 months followed by a thoracolumbosacral orthosis for another 6 months. This approach may allow for some growth anteriorly in the abnormal region of the spine, which, over time, may result
in progressive improvement in the localized kyphosis.
Alternatively, in patients with kyphosis that exceeds 50,
an anterior release with strut graft must accompany the
posterior fusion. In the older child or the adult, combined anterior and posterior arthrodesis is mandatory.

Rotatory Dislocation
Segmental spinal dysgenesis, congenital dislocation of
the spine, and congenital vertebral displacement of the
spine are conditions that create the most severe localized kyphosis of the spine and lead to a neurologic deficit in 50% to 60% of patients. These conditions can be
difficult to differentiate from one another. The deformities include severe kyphosis; anterior, posterior, or lateral subluxation of the spine; and scoliosis in association
with a severely stenotic spinal canal. The treatment requires combined anterior and posterior spinal fusion because posterior fusion alone is insufficient to achieve
solid arthrodesis in patients with these types of congenital instabilities. Exploration and augmentation of the
posterior fusion mass should be considered because of a
high occurrence of pseudarthrosis. No sudden correction
should be attempted in older patients with severe angular kyphosis and progressive neurologic deficit. Function
must be favored over cosmetic appearance. Neurosurgical decompression should be used only for patients with
a proven recent and progressive neurologic deficit.

Spondylolysis and Spondylolisthesis

Spondylolysis is a defect of one or both of the pars interarticularis in the posterior element of a lower lumbar
vertebra. Although it generally results from repetitive
stress to the pars interarticularis, it can occur abruptly.
Seen in 4% to 6% of the general population, spondylolysis has a predilection in those whose activities involve
increased hyperextension of the lumbar spine. The condition primarily affects L5 (in 87% to 95% of patients)
and less frequently L4 (in up to 10%) and L3 (in up to
3%). Spondylolisthesis represents a forward slippage of
all or part of one vertebra on another and in children is
nearly always located between L5 and the sacrum.

782

Orthopaedic Knowledge Update 8

Patients usually present with pain localized to the
low back that is usually aggravated by extension activities and relieved by rest. Radiation of the pain into the
buttocks or posterior thighs is uncommon and a neurologic deficit is rare. Unless the slippage is high grade, the
back alignment appears normal and no localized tenderness is present. Hamstring tightness may be present and
can lead to a shortened stride length. In patients with a
high-grade slip, the buttocks may appear flattened, a
step-off may be palpable between the area of the unaffected lumbar spine and loose posterior elements, the
torso may appear short, and an olisthetic scoliosis may
be present. L5 or S1 nerve root compression symptoms
are rare.
Spondylolisthesis is always evident on an upright lateral radiograph of the lumbosacral junction, but spondylolysis may be difficult to visualize. If suspicion of
spondylolysis is high, oblique lumbar radiographs
should demonstrate sclerosis, elongation, or a distinct
defect in the pars interarticularis. If these radiographs
are not conclusive, single photon emission CT may demonstrate increased uptake in patients with recent
spondylolysis or prefractured stressed regions, but may
be normal in patients with established spondylolysis. CT
will definitively demonstrate the occult fracture that
may not be evident on oblique radiographs. MRI is not
needed for diagnosis.
The two most common types of spondylolysis seen
in children are isthmic and dysplastic spondylolysis. Isthmic spondylolysis is more common and represents a fatigue fracture of the pars interarticularis. The dysplastic
type occurs only at L5-S1 and results from congenital
dysplasia of the L5-S1 facet joints. Patients with this
type of spondylolysis have an elongated pars interarticularis and are more prone to developing neurologic
symptoms and deformity during growth.
Traditionally, radiographic sagittal descriptions of
spondylolisthesis have involved the Meyerding classification (grades I through IV) and the slip angle. The
Meyerding classification measures the forward translation of L5 on the sacrum. The slip angle measures the
sagittal rotation of L5 on the sacrum and reflects the localized kyphosis at this junction, increasing in highergrade deformities (Meyerding grade IV or spondyloptosis). Pelvic incidence, a newer radiographic
measurement that assesses the sacral anatomy and its
relationship to the pelvis and spine balance, appears to
be more predictive of spondylolisthesis progression. It
describes the obliquity of the sacrum to the pelvis and is
measured on the lateral radiograph by the angle between the line perpendicular to the middle of the sacral
plate and the line joining the middle of the sacral plate
to the center of the acetabular axis (femoral heads)
(Figure 5). This fixed angle is significantly larger in
those patients with higher-grade deformities (when
compared with those with low-grade slips and control

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

subjects). Pelvic incidence has a strong correlation with
the Meyerding classification; therefore, it may be used
early to predict the ultimate severity of spondylolisthesis in early adolescence.
The risk of progression in spondylolisthesis is greatest during the adolescent growth spurt, especially for
patients with dysplastic spondylolysis. Findings, such as a
vertical, dome-shaped sacrum, a trapezoidal L5, and a
kyphotic slip angle increase the risk for progression.
Asymptomatic patients with low-grade spondylolisthesis do not require treatment or activity restrictions.
Symptomatic patients will need to be counseled regarding their disorder and temporarily curtail participation
in athletic activities. Ultimately, those participating in
high-risk sports are five times more likely to have an
unfavorable clinical outcome. An acute spondylolysis
may heal following the use of a thermoplastic lumbosacral orthosis worn full-time for 6 months, particularly if
the spondylolysis is unilateral. If the pars interarticularis
defect is chronic, a brace can be used until the child is
asymptomatic. Exercises that include hamstring stretching, pelvic tilts, and abdominal strengthening are begun
when the patient is pain free. Athletic activities may
then be resumed with or without a low-profile antilordotic brace. Should the symptoms recur after a return to
higher levels of activity, the following two options need
to be discussed: (1) discontinue the activity that produces the symptoms or (2) proceed with surgical treatment to either repair the lytic defect or eliminate movement at the spondylolytic (or spondylolisthetic) segment
by a single-level fusion.
Surgical treatment options include repair of a
spondylolytic defect, single-level (or two-level) fusions,
and reduction of higher-grade spondylolisthesis deformities. The high-grade deformities are uncommon (1%
of patients), yet much of the recent literature focuses on
the treatment of these deformities. Options that have
successfully been described in the recent literature include partial reduction, decompression, and posterior
lumbosacral fixation; additional anterior column support; and even in situ posterolateral fusion followed by
cast immobilization. Partial (or complete) reduction and
instrumentation should be undertaken only by experienced spine surgeons (see chapter 45 for more in-depth
information regarding the treatment of adult patients
with spondylolisthesis).
Direct repair of spondylolytic defects in patients
without spondylolisthesis is usually reserved for those
patients with defects at L4 or L3. Defects that occur at
L5 can be effectively treated by single-level in situ
L5-S1 fusion. If a direct repair is considered, preoperative MRI should be done to rule out adjacent disk pathology. A variety of techniques have been described, including screws across the pars defects, compression
wires from transverse process to the spinous process,
pedicle screws with wires around the spinous process,

American Academy of Orthopaedic Surgeons

Chapter 66 Pediatric Spinal Deformity

Figure 5 Illustration showing how pelvic incidence in patients with spondylolisthesis

is determined by the angle between the line perpendicular to the middle of the sacral
plate and the line joining the middle of the sacral plate to the center of the acetabular
axis. Pelvic incidence is significantly larger in patients with higher-grade deformities.

and pedicle screws with hooks/rods over the lamina.

When a patient with spondylolysis at L5-S1 or lowgrade spondylolisthesis requires surgical treatment, a bilateral posterolateral L5-S1 in situ fusion remains the
gold standard, and it should be done early in patients
with a dysplastic lesion because of the high propensity
for progression. Patients with higher-grade spondylolisthesis may require fusion extending to L4. Postoperative
brace immobilization may be used for comfort. These
treatment options will successfully manage most adolescent patients.

Scheuermanns Kyphosis
Scheuermanns disease represents an exaggerated structural kyphosis involving the thoracic spine. The primary
report of poor posture in the adolescent is commonly
accompanied by a dull, aching, midscapular, nonradiating discomfort. Physical examination demonstrates an
increased, inflexible thoracic kyphosis, which is most evident during forward bending. A compensatory lumbar
hyperlordosis is common, but it remains debatable
whether this leads to an increased incidence of spondy-

783

Pediatric Spinal Deformity

lolysis. Mild scoliosis is present in 15% of patients, but
the neurologic examination is almost always normal.
Radiographic features of Scheuermanns disease include kyphosis exceeding 45, with the apex usually in
the middle to lower thoracic spine. If the exaggerated
kyphosis is mild, then other accompanying features
must be recognized to differentiate Scheuermanns disease from postural kyphosis. These include anterior
wedging of three or more adjacent vertebrae in the apical region, end plate irregularities, and narrowing of disk
spaces. Schmorls nodes (herniation of the disk into the
vertebral end plate) are occasionally present. Hyperextension lateral radiographs are also helpful in differentiating Scheuermanns disease from postural kyphosis because the apical region in patients with Scheuermanns
disease will remain relatively inflexible. Preoperative
MRI can be used to rule out disk herniation if necessary.
Nonsurgical treatment consists of exercises and bracing. Although exercises can help alleviate thoracic and
lumbar discomfort, they will not result in improvement
of the kyphosis. Bracing has been reported to be effective in improving moderate deformity, but lack of patient compliance is a primary limitation.
Surgical indications include large deformities
(> 70), curve progression, persistent pain despite nonsurgical treatment, and, most importantly, genuine cosmetic concerns of the patient. Over the past 15 years,
anterior release followed by posterior instrumentation
and fusion has been most commonly used. However, recent reports documenting the success of posterior fusion
alone, particularly with the use of threaded 4.8-mm
compression rods, may obviate the need for additional
anterior surgery. Progressive localized kyphosis above
or below the fusion can be avoided if the implant extends from the second thoracic vertebra proximally to
one level beyond the first lordotic lumbar disk space inferiorly and if the correction is not excessive (> 50%).

Annotated Bibliography
Idiopathic Scoliosis
Asher M, Min Lai S, Burton D, Manna B: The reliability
and concurrent validity of the Scoliosis Research
Society-22 patient questionnaire for idiopathic scoliosis.
Spine 2003;28:63-69.

Orthopaedic Knowledge Update 8

Danielsson A: Back pain and function 23 years after fusion for adolescent idiopathic scoliosis: A case control
study. Part II. Spine 2003;28:E373-E383.
This is a study of 142 patients who had posterior spinal fusion and instrumentation using Harrington rods at a minimum
20-year follow-up; these patients were compared with 100 ageand sex-matched control subjects. Patients with scoliosis had
significantly more degenerative disk changes and lumbar pain
than control subjects (65% versus 47%, respectively). However, only 25% of patients who underwent fusion had daily
pain, and there were no differences in back function or general health-related quality of life. Patients who underwent fusion also reported pain over their iliac crest incisions when it
was made separately.

Dobbs M, Lenke LG, Szymanski DA, et al: Prevalence

of neural axis abnormalities in patients with infantile idiopathic scoliosis. J Bone Joint Surg Am 2002;84:22302234.
This multicenter study analyzed 46 consecutive patients
between 1992 and 2000 with infantile idiopathic scoliosis and a
curve magnitude greater than or equal to 20. Ten of the 46
patients (21%) had evidence of neural axis abnormalities on
MRI. Eight of these 10 patients had neurosurgical treatment
of these abnormalities.

Kuklo TR, Lenke LG, Graham EJ, et al: Correlation of

radiographic, clinical, and patient assessment of shoulder balance following fusion versus nonfusion of the
proximal thoracic curve in adolescent idiopathic scoliosis. Spine 2002;27:2013-2020.
In this retrospective review of 112 patients who underwent
surgical treatment for a double thoracic curve pattern, the authors evaluated T1 tilt, clavicle angle, coracoid height difference, trapezius length, first rib-clavicle height difference, shoulder asymmetry as measured by soft-tissue shadows on
radiographs, and the translation of the structural curves. They
reported that the clavicle angle was the best preoperative predictor of postoperative shoulder balance.

Lenke LG, Betz RR, Harms J, et al: Adolescent idiopathic scoliosis: A new classification to determine extent
of spinal arthrodesis. J Bone Joint Surg Am 2001;83:
1169-1181.

Asher M, Min Lai S, Burton D, Manna B: Scoliosis Research Society-22 patient questionnaire: Responsiveness
to change associated with surgical treatment. Spine
2003;28:70-73.

This study introduced the new scoliosis classification,

which includes the following three components: curve type,
lumbar spine modifier, and a sagittal thoracic modifier. The
authors who developed the new system and seven randomly
selected surgeons from the Scoliosis Research Society then
tested its reliability. The interobserver and intraobserver cap
of values for the components of the classification were analyzed and demonstrated good interobserver and intraobserver
reliability.

These two studies confirm the reliability and validity of

the SRS-22 questionnaire when compared with the SF-36. The
authors conclude that the SRS-22 is responsive to changes in
the postoperative period.

Liljenqvist UR, Allkemper T, Hackenberg L, Link TM,

Steinbeck J, Halm HF: Analysis of vertebral morphology in idiopathic scoliosis with use of magnetic reso-

784

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

nance imaging and multiplanar reconstruction. J Bone
Joint Surg Am 2002;84:359-368.
The authors analyzed the morphology of the thoracic pedicles in 307 vertebrae using MRI in multiplanar reconstruction.
Maximum intravertebral deformity at the apex of the curve
was noted, with transverse endosteal width of the apical pedicles between 2.3 to 3.2 mm on the concavity and 3.9 to 4.4 mm
on the convexity.

Merola AA, Haher TR, Brkaric M, et al: A multi-center

study of the outcomes of the surgical treatment of adolescent idiopathic scoliosis using the Scoliosis Research
Society (SRS) Outcome Instrument. Spine 2002;27:20462051.
This multicenter study measured outcomes following surgical treatment of adolescent idiopathic scoliosis in 242 patients. Outcome categories included improvement from preoperative pain, general self-image, function from back condition,
and level of activity. Overall, the patients were satisfied with
the results of surgery. The authors concluded that preoperative
pain is typically present in patients with adolescent idiopathic
scoliosis, and it improves significantly following surgical treatment.

Newton PO, Betz R, Clements DH, et al: Anterior thoracoscopic instrumentation: A matched comparison to
anterior open instrumentation and posterior open instrumentation. 70th Annual Meeting Proceedings. Rosemont, IL, American Academy of Orthopaedic Surgeons,
2003.
This multicenter study compared three treatment approaches (thoracoscopic, open anterior, and posterior) for patients with right thoracic curves. The radiographic and functional outcomes were similar for the three approaches;
however, the patients who had thoracoscopic anterior instrumentation and fusion had longer surgical times.

Ouellet JA, LaPlaza J, Erickson M, Birch JG, Burke S,

Browne R: Sagittal plane deformity in the thoracic
spine: A clue to the presence of syringomyelia as a
cause of scoliosis. Spine 2003;28:2147-2151.
Thirty patients with scoliosis and documented evidence of
syringomyelia were compared with 54 patients with adolescent
idiopathic scoliosis and a normal MRI. The authors analyzed
the lateral radiographs for the presence or absence of Dickson
apical lordosis. Apical lordosis was seen in 97% of patients
with adolescent idiopathic scoliosis and normal MRI scan,
whereas only 25% of patients with syringomyelia-associated
scoliosis had apical lordosis.

Pratt R, Webb JK, Burwell RG, Cummings SL: Luque

trolley and convex epiphysiodesis in the management of
infantile and juvenile idiopathic scoliosis. Spine 1999;24:
1538-1547.
The authors reported 5-year follow-up of patients in whom
either a Luque trolley alone or a Luque trolley together with a
convex epiphysiodesis was used to treat progressive infantile

American Academy of Orthopaedic Surgeons

Chapter 66 Pediatric Spinal Deformity

or juvenile idiopathic scoliosis. The authors concluded that
convex epiphysiodesis together with Luque trolley instrumentation may slow down or improve progressive infantile or juvenile idiopathic curves.

Richards BS: Delayed infections following posterior spinal instrumentation for the treatment of idiopathic
scoliosis. J Bone Joint Surg Am 1995;77:524-529.
Ten patients (average age, 25 months) with delayed treatment of deep wound infections were observed after undergoing posterior instrumentation for adolescent idiopathic scoliosis. The authors reported that the patients usually had
drainage from the wound, fluctuance, and mildly elevated
erythrocyte sedimentation rates and were treated using instrumentation removal, primary wound closure, and short-term
administration of antibiotics. The authors also discuss the importance of longer culture incubation to identify the infectious
organisms.

Richards BS, Sucato DJ, Konigsberg DE, Ouellet JA:

Comparison of reliability between the Lenke and King
classification systems for adolescent idiopathic scoliosis
using radiographs that were not premeasured. Spine
2003;28:1148-1157.
Four surgeons analyzed 50 radiographs that had not been
premeasured and assigned a classification using both the King
and the Lenke classification. The intraobserver and interobserver reliability of the King classification were 83.5% and
68.0%, respectively. These values were similar when only the
Lenke curve types were analyzed; however, when the complete classifications were assigned (curve type, lumbar modifier, and thoracic sagittal modifier), there was fair intraobserver and interobserver reliability (65.0% and 55.5%,
respectively).

Sucato DJ, Duchene C: The position of the aorta relative to the spine: A comparison of patients with and
without idiopathic scoliosis. J Bone Joint Surg Am 2003;
85:1461-1469.
Axial T1-weighted MRI scans of the thoracic and lumbar
spine were compared for normal control subjects and patients
with right idiopathic scoliosis. The aorta was positioned more
laterally and posteriorly to the vertebral bodies in patients
with idiopathic scoliosis, which was in line with a well-placed
vertebral body screw and would be in jeopardy of a laterally
misplaced left pedicle screw.

Sucato DJ, Elerson E: A comparison between the prone

and lateral position for performing a thoracoscopic anterior release and fusion for pediatric spinal deformity.
Spine 2003;28:2176-2180.
The technique of an anterior thoracoscopic release using a
regular endotracheal tube is described, with ventilation of
both lungs with lower tidal volumes. The patient is positioned
prone. When compared with patients who underwent this procedure in the lateral position (double lumen endotracheal
tube for single-lung ventilation), the anesthesia preparation

785

Pediatric Spinal Deformity

time and the delay between the anterior and posterior procedure were both shorter and there were fewer respiratory complications (in 0 versus 14.8% of patients).

Suk SI, Kim WJ, Lee SM, Kim JH, Chung ER: Thoracic
pedicle screw fixation in spinal deformities: Are they really safe? Spine 2001;26:2049-2057.
This study analyzed 462 patients who had 4,604 thoracic
screws placed to treat spinal deformity. Neurologic complications directly related to the screws occurred in four patients
(0.8%), one of whom had transient paraparesis and three had
dural tears. The authors concluded that thoracic pedicle screw
fixation is safe when treating spinal deformity.

Weinstein SL, Dolan LA, Spratt KF, Peterson KK,

Spoonamore MJ, Ponseti IV: Health and function of patients with untreated idiopathic scoliosis. JAMA 2003;
289:559-567.
In this prospective natural history study, 117 patients with
untreated scoliosis were compared with 62 age- and sexmatched control subjects. The minimum follow-up was 50
years, and multiple parameters were evaluated. The probability of survival was similar between the two groups, however,
the incidence of shortness of breath and chronic back pain
were greater in patients with scoliosis.

Congenital Spinal Deformities

Basu PS, Elsebaie H, Noordeen MH: Congenital spinal
deformity: A comprehensive assessment at presentation.
Spine 2002;27:2255-2259.
A series of 126 consecutive patients with congenital spinal
deformity were assessed for incidence of intraspinal anomaly.
This incidence was found in 37% of patients (26% of patients
had cardiac defects and 21% had urogenital anomalies). The
authors concluded that MRI and echocardiography should be
an essential part of the evaluation of patients with congenital
spinal deformity.

Campbell RM, Hell-Vocke AK: Growth of the thoracic

spine in congenital scoliosis after expansion thoracoplasty. J Bone Joint Surg Am 2003;85:409-420.
Expansion thoracoplasty consists of osteotomizing fused
ribs on the concavity of the spine followed by expansion of the
chest cage using a vertical, expandable prosthetic titanium rib
implant. The authors reported that longitudinal growth of the
spine was achieved using this technique, likely providing additional volume for growth of the underlying lungs.

Campbell RM, Smith MD, Mayes TC, et al: The characteristics of thoracic insufficiency syndrome associated
with fused ribs and congenital scoliosis. J Bone Joint
Surg Am 2003;85-A:399-408.
This landmark article introduces and defines thoracic insufficiency syndrome, the inability of the thorax to support

786

Orthopaedic Knowledge Update 8

normal respiration or lung growth, which is caused by a rare
condition of multiple fused ribs and congenital scoliosis.

Kim YJ, Otsuka NY, Flynn JM, et al: Surgical treatment

of congenital kyphosis. Spine 2001;26:2251-2257.
In this study, 26 patients were retrospectively reviewed.
The authors found a low rate of pseudarthrosis even without
routine augmentation of the fusion mass if instrumentation
was used. They report that although gradual correction of kyphosis occurs with growth in patients younger than 3 years of
age with type II and type III deformities after posterior fusion,
it appears to be unpredictable.

Klemme WR, Polly DW, Orchowske JR: Hemivertebral

excision for congenital scoliosis in very young children.
J Pediatr Orthop 2001;21:761-764.
Six children (average age, 19 months) underwent anteriorposterior hemivertebra excision. Correction was maintained
with plaster immobilization for 3 months. The authors reported
that excellent improvements in the curves were obtained and
maintained at a minimum 2-year follow-up.

McMaster MJ, Singh H: The surgical management of

congenital kyphosis and kyphoscoliosis. Spine 2001;26:
2146-2154.
In this study, 65 patients with congenital kyphosis or kyphoscoliosis were treated with spine arthrodesis. The authors
concluded that all patients with type I or type III congenital
kyphosis or kyphoscoliosis should be treated using posterior
arthrodesis before age 5 years and before the kyphosis exceeds 50. If the kyphosis does not reduce to less than 50, an
anterior release and arthrodesis using strut grafting is needed
before posterior arthrodesis can be done.

Ruf M, Harms J: Hemivertebra resection by a posterior

approach: Innovative operative technique and first results. Spine 2002;27:1116-1123.
In this retrospective study, 21 consecutive patients with
congenital scoliosis were treated with hemivertebra resection
using a posterior approach only with transpedicular instrumentation. Early surgery is recommended to avert severe local
deformities, to prevent secondary structural changes, and to
avert extensive fusions.

Sink EL, Karol LA, Sanders J, et al: Efficacy of perioperative halo-gravity traction in the treatment of severe
scoliosis in children. J Pediatr Orthop 2001;21:519-524.
Perioperative halo traction was used in 19 patients, including those with congenital scoliosis. The technique improved balance and frontal and sagittal alignment. No neurologic complications occurred.

Sturm PF, Chung R, Bormze SR: Hemivertebra in

monozygotic twins. Spine 2001;26:1389-1391.
This is a report on two monozygotic female twins with thoracic hemivertebrae that led to scoliosis.

American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8

Chapter 66 Pediatric Spinal Deformity

Suh SW, Sarwark JF, Vora A, et al: Evaluating congenital spine deformities for intraspinal anomalies with magnetic resonance imaging. J Pediatr Orthop 2001;21:525531.

Papagelopoulos PJ, Klassen RA, Peterson HA, et al:

Surgical treatment of Scheuermanns disease with segmental compression instrumentation. Clin Orthop 2001;
386:139-149.

In this study, 41 children with congenital spinal deformities

underwent MRI. Evidence of intraspinal anomalies were visible for 31% of the patients, including tethered cord, syringomyelia, and diastematomyelia. The authors recommend MRI
as part of the initial evaluation, even in the absence of clinical
findings.

Twenty-one patients with kyphotic deformities of 50 or

greater underwent posterior compression instrumentation.
Seven patients also had anterior release. The authors concluded that the use of the posterior procedure by itself provided significant correction, thereby avoiding the development
of any secondary deformity in most patients.

Spondylolysis and Spondylolisthesis

Grzegorzewski A, Kumar SJ: In situ posterolateral spine
arthrodesis for grades III, IV, and V spondylolisthesis in
children and adolescents. J Pediatr Orthop 2000;20:506511.
This study reports on 21 patients who underwent in situ
posterolateral L4-S1 fusions to treat severe spondylolisthesis,
followed by pantaloon cast for 4 months. The authors report
satisfactory results using this technique.

Hanson DS, Bridwell KH, Rhee JM, Lenke LG: Correlation of pelvic incidence with low and high-grade isthmic spondylolisthesis. Spine 2002;27:2026-2029.
In this study, pelvic incidence, a fixed angle in an individual, was reported to be significantly higher in patients with
low-grade and high-grade isthmic spondylolisthesis when compared with control subjects and correlated significantly with
the Meyerding grades of severity.

Lenke LG, Bridwell KH: Evaluation and surgical treatment of high-grade isthmic dysplastic spondylolisthesis.
Instr Course Lect 2003;52:525-532.
The authors reported that high-grade isthmic dysplastic
spondylolisthesis should be treated surgically with appropriate
central and foraminal decompressions at the L5-S1 level, followed by lumbosacral fusion. Partial reduction (to improve
the slip angle) provides less risk to the L5 nerve root than
complete reduction. Anterior and posterior fusion at L5-S1
appears to provide the best long-term results.

Scheuermanns Kyphosis
Johnston CE, Sucato DJ, Elerson E: Correction of adolescent hyperkyphosis with posterior-only threaded rod
compression instrumentation. 38th Annual Scoliosis Research Society Meeting Manual. Quebec, Canada, Scoliosis Research Society, 2003, p 121.
In this study, threaded 4.8-mm posterior compression rods
were used to treat 14 patients with thoracic kyphosis (average
kyphosis, 78.6 preoperatively). Anterior release was not performed. Correction to 40 was maintained 2.5 years postoperatively. The authors concluded that anterior spinal fusion is not
necessary when kyphosis is corrected using this technique.

American Academy of Orthopaedic Surgeons

Poolman RW, Been HD, Ubags LH: Clinical outcome

and radiographic results after operative treatment of
Scheuermanns disease. Eur Spine J 2002;11:561-569.
In this study, 23 patients underwent combined anterior and
posterior fusion of their kyphotic deformities. On extended
follow-up, thoracic kyphosis significantly increased, which was
thought to be caused primarily by removal of the posterior implant. This occurred despite solid fusions being shown at the
time of implant removal. Use of the SRS-22 questionnaire
showed only fair outcomes after surgical intervention, leading
the authors to question the indications for surgery.

Stotts AK, Smith JT, Santora SD, et al: Measurement of

spinal kyphosis: Implications for the management of
Scheuermanns kyphosis. Spine 2002;27:2143-2146.
In this study, a broad range of intraobserver and interobserver differences occur in the measurement of thoracic kyphosis in patients with Scheuermanns disease.

Classic Bibliography
Blount WP, Schmidt AC: The Milwaukee brace in the
treatment of scoliosis. J Bone Joint Surg 1957;39:693.
Cook S, Asher M, Lai S-M, Shobe J: Reoperation after
primary posterior instrumentation and fusion for idiopathic scoliosis: Toward defining later operative site
pain of unknown cause. Spine 2000;25:463-468.
Metha MH: The rib-vertebra angle in the early diagnosis
between resolving and progressive infantile scoliosis.
J Bone Joint Surg Br 1972;54:230-243.
Nachemson AL, Peterson L-E: Effectiveness of treatment with a brace in girls who have adolescent idiopathic scoliosis: A prospective, controlled study based
on data from the Brace Study of the Scoliosis Research
Society. J Bone Joint Surg Am 1995;77:815-822.
Weinstein SL, Ponseti IV: Curve progression in idiopathic scoliosis. J Bone Joint Surg Am 1983;65:447-455.

787