research article
research article
ABSTRACT
Purpose: The objective of this study is to analyze the immediate effects on neck pain and active cervical range of motion
after a single cervical high-velocity low-amplitude (HVLA) manipulation or a control mobilization procedure in
mechanical neck pain subjects. In addition, we assessed the possible correlation between neck pain and neck mobility.
Methods: Seventy patients with mechanical neck pain (25 males and 45 females, aged 20-55 years) participated in this
study. The lateral gliding test was used to establish the presence of an intervertebral joint dysfunction at the C3 through C4
or C4 through C5 levels. Subjects were divided randomly into either an experimental group, which received an HVLA
thrust, or a control group, which received a manual mobilization procedure. The outcome measures were active cervical
range of motion and neck pain at rest assessed pretreatment and 5 minutes posttreatment by an assessor blinded to the
treatment allocation of the patient. Intragroup and intergroup comparisons were made with parametric tests. Within-group
effect sizes were calculated using Cohen’s d coefficient.
Results: Within-group changes showed a significant improvement in neck pain at rest and mobility after application
of the manipulation ( P b .001). The control group also showed a significant improvement in neck pain at rest
( P b .01), flexion ( P b .01), extension ( P b .05), and both lateral flexions ( P b .01), but not in rotation. Pre-post
effect sizes were large for all the outcomes in the experimental group (d N 1), but were small to medium in the
control mobilization group (0.2 b d b 0.6). The intergroup comparison showed that the experimental group obtained a
greater improvement than the control group in all the outcome measures ( P b .001). Decreased neck pain and
increased range of motion were negatively associated for all cervical motions: the greater the increase in neck
mobility, the less the pain at rest.
Conclusions: Our results suggest that a single cervical HVLA manipulation was more effective in reducing neck pain at
rest and in increasing active cervical range of motion than a control mobilization procedure in subjects suffering from
mechanical neck pain. (J Manipulative Physiol Ther 2006;29:511-517)
Key Indexing Terms: Neck Pain; Manipulation, Spinal; Range of Motion, Articular; Manual Therapy
a
Escuela de Osteopatı́a de Madrid, Madrid, Spain. Submit requests for reprints to: César Fernández de las Peñas,
b
Escuela de Osteopatı́a de Madrid, Madrid, Spain; and Facultad de Ciencias de la Salud, Universidad Rey Juan Carlos,
Department of Physical Therapy, Occupational Therapy, Physical Avenida de Atenas s/n, 28922 Alcorcón, Madrid, Spain
Medicine and Rehabilitation, Universidad Rey Juan Carlos, (e-mails: [email protected], [email protected]).
Alcorcón, Spain. Paper submitted December 1, 2005; in revised form January 5,
c
Escula de Osteopatı́a de Madrid, Madrid, Spain. 2006; accepted April 20, 2006.
d
Escula de Osteopatı́a de Madrid, Madrid, Spain. 0161-4754/$32.00
e
Escula de Osteopatı́a de Madrid, Madrid, Spain; and Depart- Copyright D 2006 by National University of Health Sciences.
ment of Physical Therapy, Universidad de Sevilla, Sevilla, Spain. doi:10.1016/j.jmpt.2006.06.022
511
512 Martı́nez-Segura et al Journal of Manipulative and Physiological Therapeutics
Immediate Effects on Pain and Range of Motion of Spinal Manipulation September 2006
M
echanical neck pain affects 45% to 54% of the
general population at some time during their
lives1 and can result in severe pain and disability.2
The exact pathology of mechanical neck pain is not clearly
understood, but it is purported to be related to various
anatomic structures, including ligaments, muscles, zygapo-
physial joints, uncovertebral joints, intervertebral disks, or
neural tissues. Bogduk and Aprill3 hypothesized that one of
the most common causes of neck pain is related to
mechanical dysfunction of the cervical spine. Intervertebral
joint dysfunctions are defined as a reduction of mobility of a
cervical segment4 and, if identified on clinical examination, Fig 1. High-velocity low-amplitude procedure used in this study.
is often the focus of treatment of mobilization/manipula-
tion.5 It has been found that spinal manipulative therapy is pain and were referred by their primary care physician to a
one of the most used therapies in the management of private clinic of physical therapy and osteopathy in Alicante,
mechanical neck disorders.5,6 Spain, from February to September 2005 were recruited to
The aim of mechanical neck pain treatment is to reduce participate in this study. For the purpose of this study,
the pain and restore normal function of the cervical spine. mechanical neck pain was defined as generalized neck and/or
Previous studies have demonstrated that spinal manipulative shoulder pain with mechanical characteristics, including
therapy is effective in reducing pressure pain threshold7 and
symptoms provoked by maintained neck postures, by neck
increasing cervical range of motion8 - 10 in patients present-
movement, or by palpation of the cervical muscles. Inclusion
ing with mechanical neck pain. Another study has reported requirements for patients to be participants were (1) patients
that no lasting changes in passive cervical range of motion suffering from mechanical neck pain of at least 1-month
occur after spinal manipulation.11 Cassidy et al8 compared duration and (2) clinical presentation of intervertebral joint
the immediate effects of spinal manipulative therapy, with dysfunction at C3 through C4 or C4 through C5 levels
muscle energy techniques as the mobilization procedure, on diagnosed by the lateral gliding test of the cervical spine.13
pain and range of motion in the cervical spine. These Patients were excluded if they exhibited any of the following:
authors found that both treatments were equally effective in
(1) any contraindication to manipulation, (2) diagnosis of
improving cervical range of motion, but spinal manipulation
fibromyalgia syndrome,14 (3) previous history of a whiplash
was more effective than muscle energy techniques in injury, (4) history of cervical spine surgery, (5) diagnosis of
decreasing pain.9 Pikula12 also analyzed the immediate cervical radiculopathy or myelopathy determined by their
effects of spinal manipulative therapy, compared with primary care physician, (6) having undergone spinal manip-
detuned ultrasound therapy, on neck pain and cervical range ulative therapy within the past month before the study, (7)
of motion. This study found greater decreased neck pain and exhibiting a positive result in extension-rotation test15 or, (8)
increased cervical range of motion in those patients who younger than 18 years. The clinical history for each patient
received spinal manipulation compared with those receiving
was solicited from their primary care physician to assess the
placebo therapy (detuned ultrasound therapy).12
presence of any exclusion criteria or bred flagsQ (eg, infection,
Although Cassidy et al8 and Pikula12 analyzed the osteoporosis). A radiograph of the cervical spine was
immediate effects on pain and range of motion of spinal solicited for all patients older than 30 years to assess the
manipulative therapy in the cervical spine, these studies did possibility of degenerative cervical alteration.
not include a control group receiving a manual mobilization This study was supervised by the Escuela de Osteopatı́a
procedure. The aim of this randomized controlled study was de Madrid, Spain. It was approved by the ethical commit-
to analyze the immediate effects on neck pain and active tee in clinical research of the Escuela de Osteopatı́a de
cervical range of motion after a single cervical high-velocity
Madrid. All subjects signed the required consent before
low-amplitude (HVLA) manipulation or a control mobi-
beginning the study.
lization procedure in patients suffering from mechanical
neck pain. In addition, we assessed the possible correlation
between the decrease in neck pain and the increase in neck Cervical HVLA Technique
mobility on each cervical motion. A cervical manipulation directed at the dysfunctional
level was applied by an experienced therapist as described
below. The patient was supine with the cervical spine in a
METHODS neutral position. The index finger of the therapist applied
Subjects contact over the posterior lateral aspect of the articular pillar
Seventy subjects (25 males and 45 females; 20-55 years at the dysfunctional side of the identified vertebra (assessed
old; mean, 37; SD, 10) who presented with mechanical neck by the lateral gliding test). The therapist’s other hand
Journal of Manipulative and Physiological Therapeutics Martı́nez-Segura et al 513
Volume 29, Number 7 Immediate Effects on Pain and Range of Motion of Spinal Manipulation
cradled the patient’s head. Gentle ipsilateral side flexion and Table 1. Characteristics of each group at the beginning of
contralateral rotation were introduced from the restricted the study
side until slight tension was palpated in the tissues at the
Control Experimental
contact point. An HVLA thrust was directed upward and group group P
medially in the direction of the patient’s contralateral eye
(Fig 1).16 A specific cracking or popping sound, indicating No. of subjects 37 34
joint cavitation, accompanied all manipulations. Sex (male/female) 13/24 13/21 .7
Age (mean F SD) 39 F 10 35 F 10 .2
Length of neck pain 4.5 F 4.6 4F 3.4 .6
Control Group Neck pain at rest 5.5 F 1.5 5.7 F 1.5 .4
Cervical flexion 43 F 9 45 F 7 .2
This group received a manual mobilization procedure. For
Cervical extension 55 F 7 57 F 9 .2
the procedure, the patient was supine with the cervical spine Left lateral flexion 35 F 7 37 F 6 .2
in a neutral position. The index finger of the therapist applied Right lateral flexion 33 F 6 34 F 7 .4
contact over the lateral aspect of the patient’s neck. The Left rotation 56 F 7 57 F 10 .8
therapist’s other hand cradled the patient’s head. Gentle Right rotation 53 F 6 55 F 9 .3
cervical side flexion and contralateral rotation were intro- Scores are expressed as means F SD.
duced to a certain degree. That position was held for
30 seconds without any additional btensionQ and without
additional HVLA thrust. The side of the manual contact Procedure
was randomized. Subjects underwent a screening process by means of the
lateral gliding test for the cervical spine13 to establish the
presence of an intervertebral joint dysfunction at C3 through
Outcome Measures
C4 or C4 through C5 levels, which was done by clinician 1
The outcome measures of this study were the active who had more than 5 years of experience in manual joint
cervical range of motion and neck pain at rest. These assessment. After that, pretreatment measurements were
outcomes were assessed by an examiner blinded to the assessed by clinician 2. First, subjects reported their neck
treatment allocation of the subject. pain at rest on the VAS, and second, active cervical range of
A cervical goniometric device (Performance Attainment motion was taken. After pretreatment measurements, sub-
Associates, St Paul, Minn) was used for active cervical jects were divided randomly into 2 groups using a table of
range of motion. The cervical goniometer has been shown to random numbers: the experimental group that received the
be a reliable method of measurement,17,18 showing an HVLA thrust and the control group that received the manual
intratester reliability ranging from 0.7 to 0.9 and an mobilization procedure. The appropriate technique was
intertester reliability ranging from 0.8 to 0.87.19,20 More- applied by clinician 1 who was blinded to pretreatment
over, goniometric measurements have proved to be highly data. Postintervention data were assessed 5 minutes after the
correlated with radiographic measurements of cervical application of each intervention by clinician 2 who was
flexion and extension (r = 0.97, P b .001).21 blinded to the treatment allocation of the subject, in the
Neck mobility was assessed in a relaxed sitting position. same way as in the pretreatment assessment.
It was recorded as half cycles, namely, movements in a
single direction, that is, flexion or extension and right or left.
For this purpose, all subjects were asked again to sit Statistical Analysis
comfortably on the chair with both feet flat on the floor, hips Data were analyzed with the SPSS package, version 13.0
and knees positioned at 908 angles, and buttocks positioned (SPSS, Inc, Chicago, Ill). Mean and SDs of the values were
against the back of the chair. Then, the goniometer was calculated for each variable. A normal distribution of
placed at the top of their head. Once the goniometer was set quantitative data was assessed by means of the Kolmo-
in the neutral position, they were asked to move the head as gorov-Smirnov test ( P N .05). Baseline features were
far as possible in a standard form: forward (flexion), compared between groups using the independent t tests for
backward (extension), right lateral flexion, left lateral continuous data and v 2 tests of independence for categorical
flexion, right rotation, and left rotation. Two measurements data. Within-group differences were assessed with the paired
were recorded for each type of movement, and the mean Student t test. Within-group effect sizes were calculated
was used in further statistical analysis. using Cohen’s d coefficient.23 An effect size greater than
Neck pain at rest was assessed with a visual analogue 0.8 was considered large, around 0.5 moderate, and less than
scale (VAS). The patient placed a vertical mark on a 0.2 small.23 Intergroup comparisons between both study
continuous 10-cm line to indicate his/her pain, ranging from groups were analyzed with the unpaired Student t test. The
no pain or discomfort (0) to the worst pain he/she could Student t test was also used to assess sex differences in the
possibly feel (10). The reliability and validity of the VAS as improvement in both neck mobility and neck pain at rest
a measure of pain has been established previously.22 within each group and to assess possible differences in the
514 Martı́nez-Segura et al Journal of Manipulative and Physiological Therapeutics
Immediate Effects on Pain and Range of Motion of Spinal Manipulation September 2006
Table 2. Within pre-post values and effect sizes (Cohen’s d) of each group for each outcome measure
Within-group Within-group
Pre-int Post-int Pre-post P Cohen’s d Pre-int Post-int Pre-post P Cohen’s d
Neck pain at rest 5.7 (1.5) 2.2 (1.5) 3.5 (1.2) b.001 2.9 5.5 (1.7) 5.1 (1.9) 0.4 (0.6) b.01 0.6
Cervical flexion 45 (7) 52 (7) 7 (5) b.001 1.4 43 (9) 44 (9) 1.5 (2.5) b.01 0.6
Cervical extension 57 (9) 65 (9) 8 (7) b.001 1.2 55 (7) 56 (8) 1.4 (3.3) b.05 0.4
Left lateral flexion 37 (6) 42 (6) 5 (4) b.001 1.2 35 (7) 36 (6) 0.8 (1.5) b.01 0.5
Right lateral flexion 34 (7) 39 (7) 5 (4) b.001 1.2 33 (6) 33 (6) 0.8 (1.6) b.01 0.5
Left rotation 57 (10) 66 (9) 9 (5) b.001 1.8 56 (7) 56 (6) 0.3 (0.8) NS 0.4
Right rotation 55 (9) 65 (8) 10 (5) b.001 2 53 (6) 53 (6) 0.4 (1.5) NS 0.3
Table 3. Intergroup comparison of the improvement (pre-post scores) between both groups
Outcome measures Pre-post scores of experimental group Pre-post scores of control group F a ( P score of Levene) P
Neck pain at rest 3.5 (3.9-3.1) 0.4 (0.5-0.2) 13.3 (0.001) b.001
Cervical flexion 7 (9-5) 1.5 (2.4-0.7) 17.3 (0.001) b.001
Cervical extension 8 (11-6) 1.4 (2.6-0.4) 28.6 (0.001) b.001
Left lateral flexion 5 (6-3) 0.8 (1.3-0.3) 32.3 (0.001) b.001
Right lateral flexion 5 (6-4) 0.8 (1.4-0.3) 15.3 (0.001) b.001
Left rotation 9 (11-8) 0.3 (0.6-0) 36.3 (0.001) b.001
Right rotation 10 (12-8) 0.4 (0.9-0) 17.2 (0.001) b.001
Data are expressed as mean (95% confidence interval) unless otherwise indicated.
P values come from independent (unpaired) samples Student t test analysis.
a
Levene test to assess the homogeneity of the variance.
improvement in both outcomes depending on the side of the Within-group changes showed a significant improvement
manipulative procedure. Finally, the Pearson correlation test in both neck pain at rest and neck mobility after application
(r) was used to analyze the association between the of the cervical HVLA manipulation ( P b .001). The control
improvement in both neck pain at rest and neck mobility group also showed a significant improvement in neck pain
for each cervical motion. The statistical analysis was at rest ( P b .01), flexion ( P b .01), extension ( P b .05), and
conducted at a 95% confidence level. A P value less than both lateral flexions ( P b .01), but not in any rotation
.05 was considered as statistically significant. motion. Pre-post effect sizes were large for all the outcomes
in the experimental group (d N 1), but small to medium in
the control mobilization group (0.2 b d b 0.6). Within each
group, no significant differences in the improvement on
RESULTS both outcomes were found between male and females.
Thirty-four subjects, including 13 men and 21 women Within the experimental group, changes on each outcome
between the ages of 20 and 55 years (mean age, 35 F were not significantly different whether the HVLA manip-
10 years), formed the experimental group, whereas 37 ulation was applied either to the left or right side ( P N .2).
subjects, 13 men and 24 women between the ages of 21 and Table 2 summarizes the within-group pre-post scores and
55 years (mean age, 39 F 10 years), formed the control the within-group effect sizes of each group.
mobilization group. No significant differences were found The intergroup comparison (unpaired t test analysis)
for sex ( P = .7), age ( P = .2), neck pain at rest ( P = .4), or between the improvement (pre-post scores) in both groups
cervical range of motion ( P N .1) between groups, so it showed that the experimental group obtained a greater
could therefore be assumed that both groups were com- improvement than the mobilization group in all the outcome
parable in all respects at the start of the study. Baseline data measures ( P b .001). Table 3 details the intergroup
of each group are given in Table 1. comparison of pre-post scores between both groups.
Journal of Manipulative and Physiological Therapeutics Martı́nez-Segura et al 515
Volume 29, Number 7 Immediate Effects on Pain and Range of Motion of Spinal Manipulation
Finally, the Pearson correlation test showed a negative joint produced a significant immediate amelioration of
association between the improvement in neck pain at rest passive atlantoaxial rotation asymmetry regardless of
and the improvement on each cervical range of motion: whether the HVLA technique was applied unilaterally, either
flexion (r = 0.6, P b .001), extension (r = 0.6, P b .001), toward or away the restricted rotation, or bilaterally. More-
left lateral flexion (r = 0.65, P b .05), right lateral flexion over, we also found that the increase in cervical range of
(r = 0.7, P b .001), left rotation (r = 0.7, P b .001), and motion after the manipulative procedure did not depend on
right rotation (r = 0.65, P b .001). The greater the increase the side of the manipulation. Therefore, it could be that
in neck mobility, the lesser the neck pain at rest, that is, the HVLA thrust has inherent qualities that can alter the cervical
greater the improvement in neck pain. biomechanics, independently of the side and direction of the
thrust. It is possible that experienced symptomatic improve-
ment after HVLA thrust also influences the range of motion
DISCUSSION improvement. It that way, it is possible that the effects of pain
The present study demonstrated that a single cervical modulation rather than direct range of motion effects can led
HVLA manipulation was more effective in reducing neck to the changes in active range of motion.
pain at rest and in increasing active cervical range of motion The neurophysiologic mechanisms by which spinal
than a control mobilization procedure in subjects suffering manipulative therapy is effective in reducing pain are not
from mechanical neck pain. Furthermore, the effect size in the completely understood. One possible mechanism can be that
manipulative group was large, suggesting a strong clinical the mechanical stimulation of joint capsule proprioceptors
effect, whereas the effect size of the control group was small. and muscle spindles, caused by the spinal manipulation,
The present study also demonstrated that decreased neck pain may induce a reflex inhibition of pain, reflex muscle
and increased range of motion were negatively associated. relaxation, and improve mobility.26,27 Pickar28 demonstrated
Cassidy et al8 also found a relationship between a decrease in that spinal manipulation modifies the discharge of groups I
neck pain and an increase in cervical range of motion, but and II (proprioceptive) afferent. Another mechanism might
only significant for both rotation motions. Because of the be that the afferent bombardment from joint and myofascial
small sample size of the Cassidy et al study, there was a receptors provoked by the manipulative procedure can
greater probability of creating a type II error. produce presynaptic inhibition of segmental pain pathways
Our results are in agreement with previous studies. and possibly activation of the endogenous opiate sys-
Cassidy et al8 reported that spinal manipulation in the tem.29,30 After an extensive review of neurophysiologic
cervical spine was equally effective in improving cervical effects of spinal manipulative therapy, Pickar31 concluded
range of motion, but more effective in reducing neck pain, that more than 1 mechanism likely explains the effects of
than a muscle energy technique. Pikula12 also found greater spinal manipulation. However, we cannot completely
decreased neck pain and increased cervical range of motion exclude a placebo effect by pure fact to put the therapists’
with a cervical manipulation than with detuned ultrasound hand on the symptomatic area.
therapy. Vernon et al7 reported that cervical manipulation Our study has several limitations. First, we only
produced significantly higher increases in pressure pain examined the short-term effects of spinal manipulative
threshold of tender points surrounding a cervical dysfunction therapy directed at the cervical spine. The fact that statisti-
in subjects with mechanical neck pain. Fernández-de-las- cally significant changes occurred after spinal manipulation
Peñas et al24 have recently demonstrated that a supine provides impetus for future research in this area. Therefore,
cervical rotation manipulation resulted in increased inter- further studies are needed to examine long-term effects of
segmental motion at the dysfunctional side of a cervical cervical manipulation. Moreover, Cleland et al32 have
vertebra as measured with plain film radiographs during recently demonstrated that thoracic manipulation was more
contralateral cervical side flexion. Previous and current effective in reducing neck pain than a sham manual
findings suggest that spinal manipulative therapy is more procedure. Because thoracic manipulation was also effective
effective in reducing pressure pain threshold and increasing in reducing neck pain, it is plausible that spinal manipulative
cervical range of motion than control mobilization procedure, therapy directed at the thoracic spine also provokes an
muscle energy techniques,9 or detuned ultrasound therapy.12 increase in cervical range of motion. That hypothesis needs
It has been purported that intervertebral joint dysfunctions to be tested in futures studies. Second, we cannot say that
are characterized by a reduction of mobility of a spinal our subjects were truly blinded because patients could know
segment, and that spinal manipulation can potentially affect that they had been allocated to receive high velocity-low
the mobility of the joint, resulting in alterations of the amplitude thrust (joint cavitation) or control mobilization
kinematic behavior of the spine.4 If treatment is precise, the procedure (nontissue tension). The third limitation was the
spinal manipulative procedure should affect the mobility of sample size. To definitely establish a cause-and-effect
the hypomobile joint and lead to an increased range of relationship between spinal manipulative therapy and
motion at that particular segment.24 However, Clements decreased neck pain and increased range of motion, our
et al25 found that HVLA manipulation of the atlantoaxial findings must be confirmed in a large number of subjects.
516 Martı́nez-Segura et al Journal of Manipulative and Physiological Therapeutics
Immediate Effects on Pain and Range of Motion of Spinal Manipulation September 2006
28. Pickar JG. An in vivo preparation for investigating neural 31. Pickar JG. Neurophysiological effects of spinal manipulation.
responses to controlled loading of a lumbar vertebra in the Spine J 2002;2:357 - 71.
anesthetized cat. J Neurosci Methods 1999;89:87 - 96. 32. Cleland JA, Childs JD, McRae M, Palmer JA, Stowell T.
29. Will T. The biochemical basis of manipulation. J Manipulative Immediate effects of thoracic manipulation in patients with
Physiol Ther 1978;1:155 - 9. neck pain: a randomized clinical trial. Man Ther 2005;10:
30. Haldeman S. Pain physiology as a neurological model for 127 - 35.
manipulation. Man Med 1981;19:5 - 11.