AANEM PRACTICE TOPIC

UTILITY OF ELECTRODIAGNOSTIC TESTING IN EVALUATING PATIENTS

WITH LUMBOSACRAL RADICULOPATHY: AN EVIDENCE-BASED REVIEW
S. CHARLES CHO, MD,1 MARK A. FERRANTE, MD,2 KERRY H. LEVIN, MD,3
ROBERT L. HARMON, MD, MS,4 and YUEN T. SO, MD, PhD1
1
Stanford University School of Medicine, Stanford, California, USA
2
Ohio State University School of Medicine, Columbus, Ohio, USA
3
Cleveland Clinic, Cleveland, Ohio, USA
4
Pain Management and Rehabilitation Specialists, PC and Medical College of Georgia, Augusta, Georgia, USA
Accepted 11 April 2010

ABSTRACT: This is an evidence-based review of electrodiag- nostic role. A correct diagnosis of lumbosacral
nostic (EDX) testing of patients with suspected lumbosacral ra- radiculopathy is important for implementation of
diculopathy to determine its utility in diagnosis and prognosis.
Literature searches were performed to identify articles applying timely and appropriate treatments.
EDX techniques to patients with suspected lumbosacral radicu- The primary difficulty in this systematic review
lopathy. From the 355 articles initially discovered, 119 articles was the lack of an established reference standard
describing nerve conduction studies, electromyography (EMG),
or evoked potentials in adequate detail were reviewed further. for the diagnosis of lumbosacral radiculopathy.
Fifty-three studies met inclusion criteria and were graded using This lack of a standard makes comparison of
predetermined criteria for classification of evidence for diagnos- reported sensitivity and specificity impossible. Struc-
tic studies. Two class II, 7 class III, and 34 class IV studies
described the diagnostic use of EDX. One class II and three tural abnormalities by imaging modalities had high
class III articles described H-reflexes with acceptable statistical rates of false positives,1–4 and occasional false nega-
significance for use in the diagnosis and confirmation of sus- tives.5,6 Surgical visualization also did not consistently
pected S1 lumbosacral radiculopathy. Two class II and two
class III studies demonstrated a range of sensitivities for use of translate into clinical symptoms or physiological
muscle paraspinal mapping. Two class II studies demonstrated changes.5 In addition, the clinical presentation itself
the utility of peripheral myotomal limb electromyography in ra- may be inconclusive. For example, injuries affecting
diculopathies.
Muscle Nerve 42: 276–282, 2010 the sacroiliac and zygapophysial joints, ligaments,
muscles, and the peripheral disk annulus may cause
The diagnosis of lumbosacral radiculopathy is referred pain suggestive of radiculopathy.7–9
based on clinical history and examination, imaging The Lumbosacral Radiculopathy Task Force was
studies, and electrodiagnostic (EDX) testing. Con- charged by the American Association of Neuromus-
verging lines of evidence localize the lesion, estab- cular and Electrodiagnostic Medicine (AANEM) to
lish the diagnostic certainty, and assist with prog- perform a systematic review of the available litera-
nostication. In those cases with negative imaging ture to evaluate the utility of EDX studies in the di-
findings or atypical clinical presentations, EDX is agnosis of lumbosacral radiculopathies. Task force
especially useful; whereas, in those with abnormal members were selected on the basis of their EDX
imaging findings, it serves a complementary diag- and methodological expertise. This practice param-
eter describes the results and recommendations of
Abbreviations: AANEM, American Association of Neurology and Electro- this systematic review.
diagnostic Medicine; EDX, electrodiagnostic; EMG, electromyographic;
MEP, motor evoked potential; MeSH, medical subject heading; NPV, neg-
ative predictive value; PM, paraspinal mapping; PPV, positive predictive
value; SEP, somatosensory evoked potential METHODS
Key words: electrophysiology; herniated disk; lumbosacral radiculopathy;
lumbosacral radiculitis; low back pain; root compression; sciatica A search of the United States National Library of
This report is provided as an educational service of the AANEM. It is Medicine’s MEDLINE database was performed in
based on an assessment of the current scientific and clinical literature. It is
not intended to include all possible methods of care for a particular clinical October 2003 for articles in English pertaining to
problem, or all legitimate criteria for choosing to use a specific procedure. humans, using the medical subject headings
Neither is it intended to exclude any reasonable alternative methodologies.
This statement is not intended to address all possible uses of, or issues (MeSHs) of lumbosacral radiculopathy, radiculitis,
regarding, the evaluation of lumbosacral radiculopathy, and in no way low back pain, sciatica, herniated disk, root com-
reflects upon the usefulness of electrodiagnostic studies in those areas
not addressed. The AANEM recognizes that specific patient care pression (limited to the lumbosacral region), elec-
decisions are the prerogative of the patient and his/her physician and are trophysiology, neurophysiology, nerve conduction
based on all of the circumstances involved. These guidelines are not a
substitute for the experience and judgment of a physician. This review studies, electromyography, and spinal nerve root
was not written with the intent that it be used as a basis for compression. A second search was conducted in
reimbursement decisions. This report was developed by committees of
the AANEM. It did not undergo additional peer review by Muscle & Nerve. October 2005, a third in January 2006, and a
Correspondence to: J. Vavricek, AANEM, 2621 Superior Drive NW, fourth in August 2006, using the expanded search
Rochester, MN 55901; e-mail: [email protected]
criteria based on the original search, additional
V
C 2010 Wiley Periodicals, Inc.

Published online 15 July 2010 in Wiley InterScience (www.interscience.

MeSH terms from the final articles, bibliographies,
wiley.com). DOI 10.1002/mus.21759 relevant textbooks, and relevant articles presented
276 EDX in Lumbosacral Radiculopathy MUSCLE & NERVE August 2010
by the AANEM members and committees. All the In the literature review, the authors did not
abstracts were reviewed by at least two authors, and find any consensus for a reference case definition
articles that described EDX in the diagnosis or of lumbosacral radiculopathy. Case definitions con-
prognosis of lumbosacral radiculopathy were sidered potentially acceptable included an appro-
obtained and evaluated in further detail. priate combination of clinical findings, imaging
studies demonstrating a structural lesion and surgi-
Inclusion Criteria. The author panel limited the cal confirmation of root pathology, and a consen-
scope of the review to standard electrophysiologi- sus diagnosis based on unanimous independent
cal methods, including electromyography (EMG), clinical opinions of a team consisting of a physia-
paraspinal mapping (PM), nerve conductions trist, a radiologist, and a neurosurgeon.13
(including H-reflex and F-waves, and motor evoked
RESULTS
potentials with nerve root stimulation), and soma-
Nine articles providing Class II or Class III evidence
tosensory evoked potentials (SEPs).10,11 Articles
were identified.13–21 Three articles studied the use of
that applied thermography, magnetic stimulation,
F-wave latencies13,14,17 and evoked potentials,14,16,21
and composite mathematical calculations in the di-
and four articles addressed H-reflexes13,14,17,20 and
agnosis of radiculopathy were excluded. Case
EMG.13,15,18,19 Four studies applied the EDX test in
reports and uncontrolled case series were also
a blinded evaluation.13,15,18 The outcome assess-
excluded. The author panel limited the review to
ments in the remaining articles were not blinded,
published studies that met at least four of the fol-
but were classified as Class III as the EDX measures
lowing eight criteria of high-quality studies:
were objective (i.e., F-wave, H-reflex, and SEP stud-
ies, or EMG interpretation limited to the presence
1. Prospective cohort survey or case–control study.
or absence of fibrillations potentials and positive
2. Masking of EDX results when applying the refer-
sharp waves). Class IV studies are not considered fur-
ence standard.
ther in this review.
3. Criteria for patient selection clearly defined.
4. Valid reference standard for the diagnosis of Study Characteristics. The study design and
lumbosacral radiculopathy, including clinical, cohort assembly method varied. Two studies were
imaging, or surgical confirmation. prospective, masked, and controlled13,15; five stud-
5. The EDX procedure described in enough detail ies were prospective and case-controlled13,14,18,20;
or referenced to a prior publication to allow and three studies were retrospective and case-con-
duplication of the technique. trolled.16,17,21 All of these studies, except two,13,18
6. The reference values and criteria for interpret- have a narrow spectrum of patients who had lum-
ing the results adhered to the accepted stand- bosacral radiculopathy, both clinically and by vari-
ards of EDX practice. ous other tests. Conversely, the two studies that
7. For papers applying nerve conduction and SEP employed a broad spectrum of patients included
techniques, criteria for abnormal results were asymptomatic patients and those with low back
defined adequately in statistical terms using data pain who did not have spinal stenosis. The number
from a reference population. of patients with lumbosacral radiculopathy in each
8. For papers applying nerve conduction and SEP study ranged from 16 to 206. Mean ages ranged
techniques, limb temperature is reported. from 18 to 80 years, although one study did not
state the inclusive ages.20
Three hundred fifty-five articles met the broad In two Class II prospective, controlled studies,
search criteria. One hundred nineteen of these the EDX physician was blinded to the clinical diag-
articles applied EDX testing in the diagnosis of lum- nosis.13,15 In one study, subjects were designated
bosacral radiculopathy, and 53 of these 119 articles normal control, low back pain without spinal ste-
fulfill at least four of the criteria. For these 53 stud- nosis, or spinal stenosis.13 The reference standard
ies, the panel abstracted citation information, ele- was an independent consensus agreement of the
ments relevant to broad application (inclusion crite- physiatrist, neuroradiologist, and neurosurgeon for
ria, source, and spectrum of patients), and quality the presence of spinal stenosis. However, the defi-
of evidence (study design, comparison/control nition of spinal stenosis was not provided, and no
group, objectiveness of the outcome variables, and attempt was made to correlate the EMG findings
presence or absence of masking). The quality of evi- to the level or the side of radiologic or clinical ab-
dence provided by each article was rated by at least normality. Another Class II study utilized data
two members of the author panel, using a previ- from Knutsson’s 1961 article to evaluate the lum-
ously published classification scheme for diagnostic bosacral EMG screen.15 The clinical data were pre-
studies adopted by the AANEM and the American sented in adequate detail to permit this reformula-
Academy of Neurology.12 tion. The data compared EMG, myelogram, and
EDX in Lumbosacral Radiculopathy MUSCLE & NERVE August 2010 277
clinical symptoms to the intraoperative findings of tainty of lumbosacral radiculopathy. Imaging was
a narrow spectrum of 206 patients who had sciat- classified as normal and abnormal with abnormal-
ica. One hundred eighty-five patients had intrao- ities classified as possible, probable, single level, or
perative confirmation of root compromise. The multilevel. Extremity EMG was given in mean and
study group included subjects with sciatica and standard deviation scores of abnormal spontaneous
intraoperative abnormalities, whereas the control activity, motor unit potential changes, and compos-
group consisted of patients with sciatica and no ite EMG rather than raw data. Comparing only the
abnormal intraoperative findings. The EMG assess- certain (and excluding the possible and probable)
ments were blinded, performed prospectively, and single-level abnormality imaging findings to PM,
compared with an objective ‘‘gold standard.’’ the sensitivity was 63%, specificity 92%, positive
The spectrum of clinical presentations varied. predictive value (PPV) 87%, and negative predic-
Two studies mandated dermatomal sensory deficits tive value (NPV) 75%; the values were 71%, 92%,
and an L5 or S1 pattern of weakness (dorsiflexion 83%, and 85%, respectively, for the multilevel
or plantarflexion),14,20 whereas another allowed imaging abnormalities. Compared with imaging,
clinical symptoms to include or exclude sensory the false-positive rate was 8% (1 of 13) and the
paresthesias in a dermatomal distribution.16 Clini- false-negative rate was 33% (6 of 18). The false-
cal presentations included back pain, pain radia- negative rate decreased to 5% with the addition of
tion, reflex abnormalities, strength deficits, and lower extremity EMG. All 7 patients with an iso-
ambulation ability in both the radiculopathy lated S1 radiculopathy had normal PM studies.
cohort and the group with low back pain that did In another Class III study, PM was compared
not have a radiculopathy.13 Other symptoms were with imaging, extremity EMG, physical examina-
sciatica over a certain amount of time, and 60% of tion, and combined EMG/imaging as separate ref-
patients had a reduced Achilles tendon reflex, erence standards.18 Of note, the authors calculated
17% had a reduced patellar reflex, 61% had weak- different values than what were recorded, and they
ness of great toe extension, 30% had reduced sensa- presented those calculations. Using imaging as the
tion, 40% had muscular atrophy, and 95% had posi- reference standard the sensitivity of PM was 66%
tive straight leg raises.15 Most of the studies used and the specificity was 92%; using extremity EMG
radicular pain as an initial screen13,14,16,18,20,21; how- as the reference standard, these values were 50%
ever, one article did not explicitly state the clinical and 85%, respectively; using physical examination
features of presentation.17 Four articles included as the gold standard, they were 54% and 69%,
diminished ankle jerks and positive straight leg respectively; and using combined EMG/imaging as
raises as part of the clinical criteria.14–16,20 Three the gold standard, they were 58% and 90%,
studies did not detail the clinical examination.17,18 respectively.
Two studies explicitly excluded patients with A Class II study documented PM of four or
previous surgery.16,18 Another two studies implied more muscles. Abnormalities defined as membrane
exclusion because surgery was part of the reference instability on one side demonstrated sensitivity of
standard, but the authors did not explicitly state 30.4% and specificity of 100%, meeting statistical
the surgery exclusion.14,17 Although the spectrum significance compared with control subjects (P ¼
of disease and the associated presentations varied, 0.004).13 Additional EMG examination in an over-
the common feature among the prospective lapping myotomal distribution in limb muscles
patient assembly was radiating lower back pain. (tensor fascia lata, vastus medialis, tibialis anterior,
Four groups included surgical observations as part extensor hallucis longus, medial gastrocnemius)
of the ‘‘gold standard.’’14,15,17,19 Others used some had sensitivity of 47.8% and specificity of 87.5%
form of EDX,14,16,19,20 and most implemented vari- (P ¼ 0.008), when only fibrillation potentials were
ous imaging studies.13,16,18,21 A recent study imple- accepted as abnormal. When a combination of any
mented a unanimous consensus among the physia- needle EMG abnormalities was used (fibrillation
trist, neuroradiologist, and neurosurgeon as the potential, typical motor unit potential amplitude,
case definition of spinal stenosis.13 number of polyphasic motor unit potential, motor
unit potential recruitment), the sensitivity increased
EMG/Paraspinal Mapping. Several studies investi- to 79.2%, and specificity fell to 50% (P ¼ 0.035).
gated the use of limb myotomal EMG and PM, In another Class II study,15 a four- and five-mus-
using abnormal spontaneous activities as the defi- cle extremity EMG screen that included PM had
nition of abnormalities (Table 1). A Class III study sensitivity in the range of 89–92%. Four- and five-
compared PM with imaging (either computed to- muscle screens without PM were not as sensitive
mography or magnetic resonance imaging) in 43 (77–90%). Specificity could not be calculated
patients and extremity EMG in 110 patients.19 The because the data did not include false-positive or
patient groups were classified by the degree of cer- true-negative values.
278 EDX in Lumbosacral Radiculopathy MUSCLE & NERVE August 2010
 One Class II13 and two Class III stud-

NPV

66.7

61.8

58.5
H-Reflex.

33

Dx, diagnosis; EMG, electromyography; PM, paraspinal mapping; MRI, magnetic resonance imaging; NCS, nerve conduction study; NPV, negative predictive value; PPV, positive predictive value; SEP, somatosensory
14,20
ies investigated the tibial nerve H-reflex in the
diagnosis of lumbosacral radiculopathy (Table 2).

78.6
PPV

100
The diagnostic sensitivity and specificity varied

66

92
Test screening yield
widely. One study13,14 noted 100% sensitivity and
specificity, whereas others13,14,20 reported sensitivity
Specificity

of 51%, specificity of 91%, PPV of 64%, and NPV

87.5

100
37

92
of 84% in the S1 radiculopathy group. The sensi-
tivity dropped to 6% in the L5 group. The Class II
study13 did not specify the root level of abnormal-
Sensitivity

ity or describe stimulating or recording techniques.

89
92
77
90
66.7

45.8

29.2
62

Using H-wave absence (ipsilateral or bilateral) or

asymmetry as markers of abnormality, sensitivities
were 36.4% and 18.2%, and specificities were
SEP, F-wave, H-reflex, NCS

91.3% and 100%, respectively.

Surgical, secondary other test
Table 1. Design characteristics and outcomes in EMG studies for lumbosacral radiculopathy.

‘‘Gold Standard’’ Dx

F-Wave Studies. Two Class III studies used pero-

neal nerve (L5 radiculopathy) and posterior tibial
Surgical visualization
Surgical visualization

History, exam, MRI;

History, exam, MRI;

3 consensus Dxs

3 consensus Dxs

nerve (S1 radiculopathy) F-wave latency and inter-

side latency differences as markers of abnormality
and compared these with normal values (Ta-
Imaging

ble 3).14,17 One used the abductor digiti minimi

for posterior tibial and extensor digitorum brevis
for peroneal recording.17 The sensitivity was 65%
using the peroneal nerve, and 56% using the tibial
subjects, 30 pain, but
asymptomatic control

60; 30 control subjects,

60 subjects; 30 control

nerve. The data did not allow the specificity, PPV,

30 pain but no MRI
114 consecutive, 35

and NPV to be calculated. The other Class III

Cohort size

study did not state the muscles used and reported

206
206
16

a combined peroneal/tibial cumulative sensitivity

subjects

no MRI

of 25%, specificity of 62%, PPV of 57%, and NPV

of 29% in detecting lumbosacral radiculopathies.14
A Class II study13 mentioned peroneal F studies,
but did not report tibial studies. Markers of abnor-
Prospec/retro

mality included F-wave absence (ipsilateral or bilat-

Prospective

Prospective
Prospective
Prospective

Prospective
Prospective

eral) or asymmetry. Asymmetry was not defined.

Sensitivity was 4.8%, and specificity was 95.5%.

Motor Evoked Potential with Root Stimulation. One

Class III study21 tested motor evoked potentials
Blind
Yes

Yes
Yes
Yes

Yes

Yes

(MEPs) stimulating with surface electrodes over

the midline of the lumbosacral region and record-
ing from the tibialis anterior (L5) and soleus (S1)
3 (9 control
subjects)

muscles. The MEP latency was considered abnor-

Class

2
2
3

mal when it was 3 standard deviations greater than

the mean of normal control subjects, or when an
interside latency difference exceeded 0.8 ms. The
EMG without PM

L5 MEP latencies were 72% sensitive and 100%

EMG/fibrillation
EMG with PM

any muscle

specific in detecting lumbar radiculopathies, with a

Test

PPV of 100% and an NPV of 83%. Similarly, the

EMG/PM

EMG/PM

S1 latencies were 66.7% sensitive and 100% spe-

EMG

cific in detecting sacral radiculopathies, with a PPV

of 100% and an NPV of 74%.
Dillingham (2000)
Dillingham (2000)

Dermatomal/Segmental SEP. There were two Class

evoked potential.
Albeck (2000)

III studies using dermatomal SEP (Table 4).14,16

Author (year)

Haig (1997)

Haig (2005)

Haig (2005)

One study used a site of stimulation 6 cm above

the medial malleolus for L4, the medial side of the
second metatarsal bone for L5, and the lateral side
EDX in Lumbosacral Radiculopathy MUSCLE & NERVE August 2010 279
 Table 2. Design characteristics and outcomes in nerve conduction studies for lumbosacral radiculopathy.
Test screening yield
Author Prospec/ Cohort
(year) Test Class Blind Retro size ‘‘Gold Standard’’ Dx Sensitivity Specificity PPV NPV
Albeck H-reflex (S1 3 (9 control Yes Prospective 16 Surgical, secondary 100 100 100 100
(2000) radiculopathy) subjects) other test SEP,
F-wave, H-reflex,
NCS, EMG
Haig H-reflex 2 (30 control Yes Prospective 60 History, exam, MRI, 33.3 91.7 80 57.9
(2005) subjects, 30 pain 3 consensus Dxs
but no MRI)
Marin H-reflex 3 (53 normal No Prospective 17 L5, EMG, F-wave, SEP 50 91
(1995) control subjects) 18 S1

Dx, diagnosis; EMG, electromyography; PM, paraspinal mapping; MRI, magnetic resonance imaging; NCS, nerve conduction study; NPV, negative predic-
tive value; PPV, positive predictive value; SEP, somatosensory evoked potential.

of the fifth metatarsal bone for S1.14 Of the 5 mal abnormality, and 10% of the patients with a
patients with herniation and an abnormal SEP, 1 confirmed S1 radiculopathy had an L5 dermato-
had an S1 lesion coupled to an L5 SEP abnormal- mal abnormality.
ity, 1 had an L5 lesion coupled to an L4 SEP ab-
normality, and the other 3 showed concordance CONCLUSIONS
with the actual localization. The sensitivity was
29%, specificity 67%, PPV 63%, and NPV 33%. In 1: In patients with suspected lumbosacral radicu-
the other study,14,16 the sural and superficial pero- lopathy, the following EDX studies probably aid
neal nerves were stimulated first, and then a site at the clinical diagnosis:
the superomedial margin of the first metatarsopha- (a) Peripheral limb EMG (Class II evidence,
langeal joint was employed for L5. A site at the Level B recommendation).
midsection of the fifth metatarsal bone on the lat- (b) PM with needle EMG in lumbar radiculopa-
eral foot was used for S1 dermatomal SEP. The thy (Class II evidence, Level B
L5 dermatomal SEP was 50% sensitive and 97.7% recommendation).
specific at both the 90% and 95% confidence (c) H-reflex in S1 radiculopathy (Class II and
intervals (CIs). The S1 dermatomal SEP was 20% III evidence, Level C recommendation).
sensitive and 97% specific at 90% CI, and 10% 2: Evidence suggests a low sensitivity of peroneal
sensitive and 97.9% specific at 95% CI. The super- and posterior tibial F-waves (Class II and III evi-
ficial peroneal SEP showed the best sensitivity for dence, Level C recommendation).
L5 radiculopathy of 70% and 60%, respectively, at 3: There is inadequate evidence to reach a conclu-
90% and 95% CI. Of note, there was a high false- sion on the utility of the following EDX studies:
negative rate. Twenty percent of the patients with (a) Dermatomal/segmental SEP of the L5 or S1
a confirmed L5 radiculopathy had an S1 dermato- dermatomes (Class III evidence, Level C

Table 3. Design characteristics and outcomes in nerve conduction studies for lumbosacral radiculopathy.
Test screening yield
Author Prospec/ Cohort ‘‘Gold
(year) Test Class Blind retro size standard’’ Dx Sensitivity Specificity PPV NPV
Albeck F-waves 3 (9 control Yes Prospective 16 Surgical, secondary 25 62 57 29
(2000) subjects) other test SEP,
F-wave, H-reflex,
NCS, EMG
Eisen F-wave 3 No Prospective 60 normal, Myelography; 6 50
(1977) latency 25 subjects surgically proven
Haig F-waves 2 (30 control Yes Prospective 60 History, exam, MRI; 4.2 95.8 50 50
(2005) absent subjects, 3 consensus Dxs
(either or 30 pain but
both sides) no MRI)

Dx, diagnosis; EMG, electromyography; MRI, magnetic resonance imaging; NCS, nerve conduction study; NPV, negative predictive value; PPV, positive
predictive value; SEP, somatosensory evoked potential.

280 EDX in Lumbosacral Radiculopathy MUSCLE & NERVE August 2010

 Table 4. Design characteristics and outcomes in somatosensory evoked potential studies for lumbosacral radiculopathy.
Test screening yield
Author Prospec/ Cohort ‘‘Gold
(year) Test Class Blind retro size standard’’ Dx Sensitivity Specificity PPV NPV
Albeck Dermatomal 3 (9 control Yes Prospective 16 Surgical, secondary 31 67 62 31
(2000) SEP subjects) other test SEP,
F wave, H reflex,
NCS, EMG
Dumitru Segmental 3 No Prospective 20; 43 ‘‘well defined’’ L5 20 95
(1996) SEP sural control or S1 Hx, exam,
stimulation subjects imaging, EDX
Dumitru Segmental 3 No Prospective 20; 43 ‘‘well defined’’ L5 70 95
(1996) SEP peroneal control or S1 Hx, exam,
stimulation subjects imaging, EDX
Tabaraud MEP responses 3 No Prospective 45; 25 L5 radiculopathy 72 100 100 83.3
(1989) prolonged control
response subjects
latency
Tabaraud MEP responses 3 No Prospective 45; 25 S1 radiculopathy 66.7 100 100 73.5
(1989) prolonged control
response subjects
latency

Dx, diagnosis; EDX, electrodiagnostic; EMG, electromyography; Hx, history; MEP, motor evoked potential; NCS, nerve conduction study; NPV, negative
predictive value; PPV, positive predictive value; SEP, somatosensory evoked potential.

recommendation). diagnosing suspected lumbosacral radiculopathy.

(b) PM with needle EMG in sacral radiculopa- These elements are recommended for diagnostic
thy (one small Class II study, Level U). accuracy:
(c) MEP with root stimulation in making an in-
dependent diagnosis of lumbosacral radicu- 1. The study design should be a prospective cohort
lopathy (Class III evidence, Level U). survey.
2. The cohort should include a wide spectrum of
patients with alternative diagnoses, allowing the
DISCUSSION EDX to independently assess and determine the
The available evidence is limited by the lack of a presence of radiculopathy.
universally accepted case definition of lumbosacral 3. A consensus-based reference standard (gold
radiculopathy. Furthermore, none of the studies standard) of lumbosacral radiculopathy must be
presented methodological data in adequate detail developed for research purposes.
to allow unbiased comparison of an objective refer- 4. Studies should employ a wide spectrum of
ence standard to the EDX diagnosis. Also, with the patients with alternative diagnoses and with vary-
exception of one study,13 a narrow spectrum of ing degrees of severity of radiculopathy.
patients were used. 5. All patients enrolled should complete the EDX.
In the setting of suspected lumbosacral radicu- The presence of radiculopathy should be meas-
lopathy, EDX testing is often applied as part of an ured, and the level and side documented.
evaluation that also includes clinical history, physi- 6. An investigator who is unaware of the EDX
cal examination, and imaging study. It would be results should determine the final diagnosis.
useful to determine the independent contribution 7. Studies may be performed to assess the utility
of EDX to the diagnosis. The studies reviewed did of EDX in combination with or in isolation
not address the optimal combination or sequence from alternative diagnostic studies and the
of testing in the overall evaluation process. Finally, sequence of studies and study combinations that
none of the reviewed studies addressed the utility provide the highest yield for identifying lumbo-
of EDX in prognosticating outcome or response to sacral radiculopathies.
treatments.

RECOMMENDATIONS FOR FUTURE RESEARCH DEFINITIONS FOR STRENGTH OF EVIDENCE

Future studies should eliminate potential biases Class I: Evidence provided by a prospective study
and provide data sufficient to determine the inde- in a broad spectrum of persons with the suspected
pendent contribution of EDX techniques used in condition, using a reference (gold) standard for
EDX in Lumbosacral Radiculopathy MUSCLE & NERVE August 2010 281
case definition, where the test is applied in a 2. Wiesel SW, Tsourmas N, Feffer HL, Citrin CM, Patronas N. A study
of computer-assisted tomography. I. The incidence of positive CAT
blinded evaluation, and enabling the assessment of scans in an asymptomatic group of patients. Spine 1984;9:549–551.
appropriate tests of diagnostic accuracy. All 3. Jensen MC, Brant-Zawadzki MN, Obuchowski N, Modic MT, Malka-
sian D, Ross JS. Magnetic resonance imaging of the lumbar spine in
patients undergoing the diagnostic test have the people without back pain. N Engl J Med 1994;331:69–73.
presence or absence of the disease determined. 4. Weishaupt D, Zanetti M, Hodler J, Boos N. MR imaging of the
Class II: Evidence provided by a prospective lumbar spine: prevalence of intervertebral disk extrusion and seques-
tration, nerve root compression, end plate abnormalities, and osteo-
study of a narrow spectrum of patients with the arthritis of the facet joints in asymptomatic volunteers. Radiology
1998;209:661–666.
suspected condition, or a well-designed retrospec- 5. Olmarker K, Rydevik B. Pathophysiology of sciatica. Orthop Clin N
tive study of a broad spectrum of persons with an Am 1991;22:223–234.
established condition (by gold standard) compared 6. Marshall LL, Trethewie ER, Curtain CC. Chemical radiculitis. A clini-
cal, physiological and immunological study. Clin Orthop Rel Res
with a broad spectrum of controls, where testing is 1977;124:61–67.
applied in a blinded evaluation, and enabling the 7. Schwarzer AC, Aprill CN, Bogduk N. The sacroiliac joint in chronic
low back pain. Spine 1995;20:31–33.
assessment of appropriate tests of diagnostic 8. Schwarzer AC, Aprill CN, Derby R, Fortin J, Kine G, Bogduk N. Clin-
accuracy. ical features of patients with pain stemming from the lumbar zyg-
Class III: Evidence provided by a retrospective apophysial joints. Is the lumbar facet syndrome a clinical entity?
Spine 1994;19:1132–1137.
study where either persons with the established 9. Park ES, Park CI, Kim AY, Park MK. Relationship between electromy-
condition or controls are of a narrow spectrum, ography and computed tomography in the evaluation of low back
pain. Yonsei Med J 1993;34:84–89.
and where the reference standard, if not objective, 10. So YT. Guidelines in electrodiagnostic medicine. Practice parameter
is applied by someone other than the person who for needle electromyographic evaluation of patients with suspected
cervical radiculopathy. Muscle Nerve 1999;8(suppl):S209–S221.
performed the test. 11. Marciniak C, Armon C, Wilson J, Miller R. Practice parameter: utility
Class IV: Any design where test is not applied of electrodiagnostic techniques in evaluating patients with suspected
peroneal neuropathy: an evidence-based review. Muscle Nerve 2005;
in an independent evaluation OR evidence pro- 31:520–527.
vided by expert opinion alone or in descriptive 12. Edlund WG, Yuen S, Franklin G. Clinical practice guideline process
manual. St. Paul, MN: American Academy of Neurology; 2004.
case series (without control subjects). 13. Haig AJ, Tong HC, Yamakawa KS, Quint DJ, Hoff JT, Chiodo A,
et al. The sensitivity and specificity of electrodiagnostic testing for
Rating System for Strength of Recommendations the clinical syndrome of lumbar spinal stenosis. Spine 2005;30:
2667–2676.
• Level A: Established as effective, ineffective, or 14. Albeck MJ, Taher G, Lauritzen M, Trojaborg W. Diagnostic value of
electrophysiological tests in patients with sciatica. Acta Neurol Scand
harmful (or established as useful/predictive or 2000;101:249–254.
15. Dillingham TR, Dasher KJ. The lumbosacral electromyographic
not useful/predictive) for the given condition in screen: revisiting a classic paper. Clin Neurophysiol 2000;111:2219–
the specified population. 2222.
• Level B: Probably effective (or probably useful/ 16. Dumitru D, Dreyfuss P. Dermatomal/segmental somatosensory
evoked potential evaluation of L5/S1 unilateral/unilevel radiculopa-
predictive or not useful/predictive) for the given thies. Muscle Nerve 1996;19:442–449.
condition in the specified population. 17. Eisen A, Schomer D, Melmed C. An electrophysiological method for
examining lumbosacral root compression. Can J Neurol Sci 1977;4:
• Level C: Possibly effective (or possibly useful/ 117–123.
predictive or not useful/predictive) for the given 18. Haig AJ. Clinical experience with paraspinal mapping. II: A simpli-
fied technique that eliminates three-fourths of needle insertions.
condition in the specified population. Arch Phys Med Rehabil 1997;78:1185–1190.
• Level U: Data inadequate or conflicting. Given 19. Haig AJ. Clinical experience with paraspinal mapping. I: Neurophysi-
ology of the paraspinal muscles in various spinal disorders. Arch
current knowledge, the test is unproven. Phys Med Rehabil 1997;78:1177–1184.
20. Marin R, Dillingham TR, Chang A, Belandres PV. Extensor digito-
rum brevis reflex in normals and patients with radiculopathies. Mus-
REFERENCES cle Nerve 1995;18:52–59.
1. Boden SD, Davis DO, Dina TS, Patronas NJ, Wiesel SW. Abnormal 21. Tabaraud F, Hugon J, Chazot F, Moreau JJ, Vidal J, Vallat JM, et al.
magnetic-resonance scans of the lumbar spine in asymptomatic sub- Motor evoked responses after lumbar spinal stimulation in patients
jects. A prospective investigation. J Bone Joint Surg [Am] 1990;72: with L5 or S1 radicular involvement. Electroencephalogr Clin Neuro-
403–408. physiol 1989;72:334–339.

This guideline is greater than 5 years old. Every five years, an interim literature search is performed and the guideline reviewed. While new studies
have been published since this guideline was last reviewed, the Practice Issue Review Panel Committee of the AANEM has determined that these
studies are not sufficient to mandate a revision of this guideline at the present time. The information contained in this guideline and the recommendations
offered are still relevant to current practice.
Reaffirmation approved by the Practice Issue Review Panel: September, 2017.

282 EDX in Lumbosacral Radiculopathy MUSCLE & NERVE August 2010