Content uploaded by Rochenda Ann Howard
Author content
All content in this area was uploaded by Rochenda Ann Howard on Feb 04, 2015
Content may be subject to copyright.
Pilates-Based Therapeutic Exercise: Effect
on Subjects With Nonspecific Chronic Low
Back Pain and Functional Disability: A
Randomized Controlled Trial
Rochenda Rydeard, PT, MSc
1
Andrew Leger, PT, PhD
2
Drew Smith, PhD
3
Study Design: A randomized controlled trial, pretest-posttest design, with a 3-, 6-, and 12-month
follow-up.
Objectives: To investigate the efficacy of a therapeutic exercise approach in a population with
chronic low back pain (LBP).
Background: Therapeutic approaches developed from the Pilates method are becoming increas-
ingly popular; however, there have been no reports on their efficacy.
Methods and Measures: Thirty-nine physically active subjects between 20 and 55 years old with
chronic LBP were randomly assigned to 1 of 2 groups. The specific-exercise-training group
participated in a 4-week program consisting of training on specialized (Pilates) exercise
equipment, while the control group received the usual care, defined as consultation with a
physician and other specialists and healthcare professionals, as necessary. Treatment sessions were
designed to train the activation of specific muscles thought to stabilize the lumbar-pelvic region.
Functional disability outcomes were measured with The Roland Morris Disability Questionnaire
(RMQ/RMDQ-HK) and average pain intensity using a 101-point numerical rating scale.
Results: There was a significantly lower level of functional disability (P= .023) and average pain
intensity (P= .002) in the specific-exercise-training group than in the control group following the
treatment intervention period. The posttest adjusted mean in functional disability level in the
specific-exercise-training group was 2.0 (95% CI, 1.3 to 2.7) RMQ/RMDQ-HK points compared to
a posttest adjusted mean in the control group of 3.2 (95% CI, 2.5 to 4.0) RMQ/RMDQ-HK points.
The posttest adjusted mean in pain intensity in the specific-exercise-training group was 18.3 (95%
CI, 11.8 to 24.8), as compared to 33.9 (95% CI, 26.9 to 41.0) in the control group. Improved
disability scores in the specific-exercise-training group were maintained for up to 12 months
following treatment intervention.
Conclusions: The individuals in the specific-exercise-training group reported a significant decrease
in LBP and disability, which was maintained over a 12-month follow-up period. Treatment with a
1
Graduate student (at time of study), School of Rehabilitation Therapy, Queen’s University, Kingston,
Ontario, Canada.
2
Educational Developer, Centre for Teaching and Learning, Queen’s University, Canada; Assistant
Professor (at time of study), School of Rehabilitation Therapy, Queen’s University, Kingston, Ontario,
Canada.
3
Senior Lecturer, Department of Sport and Exercise Science, The University of Auckland, Auckland, New
Zealand; Assistant Professor (at time of study), Department of Rehabilitation Sciences, The Hong Kong
Polytechnic University, Hung Hom, Hong Kong SAR, China.
The protocol for this study was approved by The Faculty of Health Science Research Ethics Board,
Queens University, Canada and Hong Kong Polytechnic University Human Subjects Ethics Subcommittee,
Hong Kong Special Administrative Region.
Address corresondence to A. B. Leger, Centre for Teaching and Learning, Queen’s University, Kingston,
Ontario K7L 3N6, Canada. E-mail: AL7@post.queensu.ca
modified Pilates-based approach was more effi-
cacious than usual care in a population with
chronic, unresolved LBP. J Orthop Sports Phys
Ther 2006;36(7):472-484. doi:10.2519/jospt.
2006.2144
Key Words: exercise rehabilitation,
lumbar spine muscle recruitment,
stabilization exercises
Low back pain (LBP)
represents the most
common cause of dis-
ability in persons un-
der 45 years of age.
1
Spinal disorders represent at least
40% of the compensated disorders
treated by physiotherapists, and
70% of these spinal disorders in-
volve the lumbar spine.
39
The effectiveness of therapeutic
exercise in the treatment of
chronic LBP is currently under
review.
22,28,39,43,44
General condi-
tioning programs to train strength
and endurance of the spine mus-
culature have been shown to re-
duce pain intensity and disability
28
and to be useful in the treatment
of nonspecific chronic LBP
22,44
and ‘‘activity-related spinal disor-
ders.’’
39
Much of the literature
examining chronic LBP and exer-
cise interventions study a popula-
tion whose pain and disability
472 Journal of Orthopaedic & Sports Physical Therapy
manifest in ‘‘deconditioning syndrome,’’ as described
by Mayer.
22
It follows that exercise-training programs
directed generally at muscle strength, endurance, and
reconditioning are appropriate. However, physical
deconditioning may not be the limiting factor to
recovery for many patients seeking treatment for
chronic low back disorders; it is now accepted that
muscle dysfunction in chronic LBP may not simply be
a problem of muscle strength or endurance. Instead,
the problem may be one of altered neuromuscular
control mechanisms affecting muscular stability of the
trunk and movement efficiency.
5,15,17,29
Bergmark
3
classifies spinal stability as being comprised of 2
(muscular) mechanisms: local mechanisms, whereby
deep, local muscles act to control movement at the
intervertebral segment, and global mechanisms,
whereby muscles control movement of the spine
generally and at multiple segments. Effective control
of both mechanisms is necessary for efficient stabiliza-
tion of the spine and alterations in neuromuscular
control and the loss of normal patterns of spinal
motion will cause pain.
26
Further, Edgerton et al
5
suggest that pain and pathology result in changes in
neural input to motor neuron pools affecting muscle
activation. A general reconditioning approach,
through its lack of specificity, may reinforce abnormal
muscle recruitment and perpetuate compensatory
strategies that may have developed as a result of
neuromuscular adaptation over time.
Consequently, specifically designed therapeutic ex-
ercise approaches that enhance spinal stability and
modulate neuromuscular control in the presence of
chronic LBP have been embraced by physiotherapists.
Techniques evaluated in the literature to date address
muscle activation directed at the intervertebral seg-
ment via the cocontraction of the deep abdominals
and the paraspinals, enhancing stabilization at the
lumbar spine segments.
25
Recent evidence supports
the effectiveness of such a specific-exercise approach
in a chronic LBP population with a diagnosis of
spondylolysis or spondylolisthesis.
25,44
O’Sullivan and
colleagues
25
found a significant reduction in pain
intensity (P= .0006) and functional disability levels
(P= .0001) in a group who received specific exercise
with maintenance of effect over a 30-month follow-
up. No significant changes were seen in a control
group receiving usual care.
The Pilates Method
37
is an exercise method popu-
lar for decades in dance training and the dance
medicine community. The Pilates Method is a unique
approach to training in mind-body awareness and
control of movement and posture. Specialized appa-
ratus provides an opportunity to train a variety of
movement patterns and postures. The neuromuscular
demands of traditional Pilates methods can be quite
high and therefore a modification of this method is
necessary for application to physiotherapeutic inter-
ventions. The technique used in this study was
adapted but consistent with traditional Pilates tech-
niques, focusing on postural symmetry and controlled
movement. Particular emphasis, however, was placed
on specific muscle activation strategies thought to
stabilize the lumbar-pelvic region.
A modified Pilates approach to improve posture
and control movement can thus be supported within
a theoretical context of neuromuscular control and
builds upon the concept of stability about a local
spinal segment. Global stability mechanisms to con-
trol the lumbar-pelvic region are then incorporated
into this background of segmental lumbar control.
This may be achieved by incorporating specific activa-
tion of the gluteal muscles to stabilize the lumbar-
pelvic region while performing hip extension.
34,35
Stability of the trunk is thereby accomplished by
using an overlapping of stabilization strategies. Fur-
ther, the use of Pilates apparatus to train stabilization
strategies during movement may enhance the effect
of a relatively more static mat exercise. This may be
important for retention of treatment effects and
transfer to everyday movement and functional activi-
ties.
The function of the hip extensors and the gluteus
maximus is thought to be central to stability and
control in the lumbar-pelvic region.
14,18
The gluteus
maximus muscle may play an important role in
lumbar-pelvic mechanics and load transfer from the
lumbar spine to the pelvis and lower extremities.
46
Altered recruitment of the lumbar-pelvic musculature
and dysfunction of the gluteus maximus muscle is
reported in LBP conditions.
4,24,47
Janda
14,15
describes
a characteristic ‘‘pseudoparesis’’ of the gluteus
maximus in LBP, characterized by hypotonia and a
delay in activation. A concomitant imbalance in the
functional length or recruitment of the hamstrings
and/or superficial lumbar erector spinae relative to
the gluteus maximus has been associated with
LBP.
2,13-15,24,36
The resultant pull of muscle forces
may impact adversely on neuromuscular control of
hip extension and the ability of the trunk to stabilize
effectively against the demands of loading during
activities such as walking.
Evidence supporting the effect on pain and disabil-
ity from specific exercise approaches applied to
nonspecific chronic LBP has not been confirmed.
Similarly, no studies to date have examined the
effects of a specific exercise training approach that
emphasizes lumbar-pelvic stability and the function of
the gluteus maximus. As well, no clinical research
activities have been reported to date that elucidate
the efficacy of an exercise intervention based on the
Pilates Method in the treatment of chronic LBP.
Therefore, the objectives of this study were to
investigate the efficacy of a specific-exercise interven-
tion based on the Pilates Method and emphasizing
specific-activation strategies of the gluteus maximus
J Orthop Sports Phys Ther • Volume 36 • Number 7 • July 2006 473
RESEARCH REPORT
and the effect on LBP intensity and functional
disability in an identified population with nonspecific
chronic LBP.
METHODS
Subjects
Fifty-five subjects were recruited over a 4-month
period through notices posted to private and public
physicians’ and physiotherapists’ offices, notices
posted to local sports clubs and Universities, and by
advertisement in an English-language newspaper (Fig-
ure 1). The subjects were recruited from a popula-
tion of physically active adults between 20 and 55
years of age, living in Hong Kong, with longstanding,
persistent LBP (with or without leg pain) of greater
than 6 weeks duration or recurring LBP (with at least
2 painful incidences per year) of sufficient intensity
to restrict functional activity in some manner. ‘‘Physi-
cally active’’ was defined as participation in a mini-
mum of three 30-minute sessions per week of activity
requiring a moderate effort in keeping with estab-
lished guidelines for physical fitness.
6,42
This criterion
ensured that findings cannot simply be attributed to
disuse secondary to deconditioning and low activity
levels. Similarly, this criterion may have helped to
eliminate those LBP conditions that might have
responded to general physical conditioning exercise
without requiring more specific neuromuscular train-
ing. Subsequent clinical screening was performed by
an independent physiotherapist evaluator to test for
evidence of neuromuscular dysfunction and fulfill-
ment of the following criteria: (a) strength recording
of grade 4 or less out of 5 on manual muscle testing
of the gluteus maximus
14,18
;(b) altered recruitment
of the gluteus maximus muscle as determined clini-
cally by visual and manual inspection during a prone
leg extension test.
13,36,38
Prospective subjects were excluded from the study
if they were pregnant, had a past history of spinal
surgery or spinal fracture, were diagnosed with in-
flammatory joint disease, systemic metabolic disorder,
rheumatic disease, or chronic pain syndrome, showed
evidence of overt neurological compromise or acute
inflammatory process, or had difficulty understanding
written or spoken English. Thirty-nine subjects met
the selection criteria for inclusion in the study and 16
subjects were excluded. Testing was performed at the
Hong Kong Polytechnic University. The clinical inter-
vention was conducted at a private physiotherapy
clinic in Hong Kong, which specialized in Pilates-
based interventions. All subjects signed informed
consent forms upon entry to the trial and the rights
of the subjects were protected. The Faculty of Health
Science Research Ethics Board, Queens University,
Kingston, Ontario and the Hong Kong Polytechnic
University Human Subjects Ethics Subcommittee,
Hong Kong Special Administrative Region, granted
ethics approval for the study. Subject characteristics
are reported in Table 1.
Study Design
The study consisted of 2 parts, a pretest-posttest
control group design in part 1 (the primary study)
and in part 2 (a follow-up on the exercise interven-
tion group over a 1-year period). Simple randomiza-
tion was performed at entr y to the trial after
eligibility was determined. Randomization was admin-
istered by independent office staff. Subjects randomly
pulled a card from a box of concealed premarked
cards to obtain assignment to either the specific-
exercise-training group (SETG) or control group
(CG) without specific exercise training (Figure 1).
The subjects had no preconceived expectations for
treatment because at the time of the study the Pilates
method was not commonly known. The subjects were
advised that the study was designed to evaluate the
effectiveness of this specific-exercise intervention in
the treatment of LBP.
Intervention
Those in the CG group received no specific exer-
cise training and continued with usual care, defined
as consultation with a physician and other specialists
and healthcare professionals as necessary. They were
not restricted from seeking any other treatment if
they so wished. Subjects were instructed to continue
to do what they were previously doing, including
regular physical activity. For ethical reasons the CG
had the option to receive, free of charge, the
specific-exercise-training program 4 weeks later, after
collection of posttreatment intervention outcome
data from the main study. During the main study
period, although subjects were aware they could
receive the exercise intervention, they were not
familiar with the technique or aware of any details of
what the treatment entailed.
The SETG received a treatment protocol consisting
of training on specialized (Pilates) exercise apparatus
in the clinic for three 1-hour sessions per week, and
training in a 15-minute home program performed 6
days per week for 4 weeks. Treatment was provided at
no charge to the subjects. The apparatus used in the
clinic consisted of a floor mat and a Pilates Reformer
with standing platform and jump-board attachments
(Balanced Body, Sacramento, CA). The Pilates Re-
former is made of a sliding horizontal platform
within a box-like frame, upon which the subject sits,
stands, kneels, or reclines. Varying resistance to
movement is provided via light springs attached to
the moving platform and through a simple pulley
system. The subject moves against the low external
resistance offered by the springs.
474 J Orthop Sports Phys Ther • Volume 36 • Number 7 • July 2006
Figure 1. Subject flow during the study.
J Orthop Sports Phys Ther • Volume 36 • Number 7 • July 2006 475
RESEARCH REPORT
TABLE 1. Subject characteristics and demographic data for the specific exercise-training group (SETG) and the control group (CG).
Subject Characteristics CG SETG PValue
Gender P= .77*
Male 8 6
Female 13 12
Age (y) 34 (8) 37 (9) P= .34
†
Height (cm) 171(10) 169 (8) P= .65
†
Mass (kg) 69 (15) 68 (14) P= .92
†
Duration of symptoms (y) Median, 9 (range, 1-20) Median, 5.5 (range, 0.5-27.0) P= .25
‡
Nature of condition P= .58*
Chronic 16 15
Recurrent 5 3
Area of symptoms P= .27*
Low back pain (LBP) 11 9
LBP and leg pain above knee 3 6
LBP and leg pain below knee 7 3
Previous physiotherapy treatment 19 16 P= .87*
Included exercise therapy 15 14 P= .65*
Functional disability (mean ± SD) 4.2 ± 3.6 3.1 ± 2.5 P= .14
†
Pain intensity score (mean ± SD) 30.4 ± 17.6 23.0 ± 17.7 P= .56
†
* Chi-Square test.
†
Unpaired ttest.
‡
Mann-Whitney Utest.
The clinic treatment protocol was provided in an
individualized manner by 1 of 2 experienced physio-
therapists trained in the treatment protocol and
blinded to the results of testing. The standardized,
progressive treatment protocol addressed targeted
muscle activation strategies throughout a variety of
movement patterns involving hip extension. The
subject was required to consciously recruit specific
muscles—the deep anterolateral abdominals (with
coactivation of the pelvic floor and lumbar
multifidus), followed by activation of the gluteus
maximus muscles. Static postures were initially
trained (Figure 2), followed by training a variety of
movement patterns to stress the lumbar-pelvic region
and involving hip extension (Figure 3). The training
was progressed on the Pilates Reformer (Figure 4)
over the 4-week period as tolerated. Initially move-
ments were practiced using weight-bearing patterns in
supine, with the lumbar spine in the neutral position.
Gradually more upright postures and controlled
movement of the lumbar-pelvic region out of neutral
posture were incorporated. Prescribed movements
were performed slowly, smoothly, and without pain.
Individualized facilitation strategies were provided by
the physiotherapist to correct technique, control
speed, assist appropriate muscle activation, or modify
the exercise or the progression to suit the subjects’
needs. Facilitation strategies included mental and
visual imagery, manual or verbal cueing, and demon-
stration.
The home treatment protocol consisted of 2 parts:
(1) floor exercises to specifically activate the deep
anterolateral abdominals and local stability syner-
gists
31
and the gluteus maximus muscle by moving
the leg in a manner similar to that utilized on the
apparatus (Figures 2 and 3) and (2) skill drills in
which difficult tasks were broken down into move-
ment components and practiced in isolation incorpo-
rating correct abdominal and gluteal control.
Compliance with the home exercise program was
encouraged and was self-monitored on a log sheet.
The treating physiotherapist kept clinical notes docu-
menting details of the treatment program and the
subject’s progress.
Outcome Measure Testing
Data collection monitored both pain intensity and
functional status and included 2 self-report question-
naires administered by the research assistant, an
independent physiotherapist investigator blinded to
group assignment. In part 1 all subjects were tested
on baseline measures at entry to the study (pretest-
ing). Retesting of both groups was done at the end of
the 4-week treatment intervention period (posttest-
ing). In part 2, disability measures for retention of
treatment effect were collected from the SETG using
a questionnaire mailed to the subjects over a 12-
month follow-up period.
Pain Intensity Outcome The NRS-101, a 101-point
numerical rating scale, was used to measure subjec-
tive pain intensity.
16
The subject rated his or her
perceived pain level between 0 and 100, with 0
representing no pain and 100 representing pain as
bad as it could be. The test protocol asked subjects to
verbally state the number that best described the
average amount of pain they had experienced in the
past week. The number noted by the subject was
recorded on the subject’s record card and used for
data analysis.
476 J Orthop Sports Phys Ther • Volume 36 • Number 7 • July 2006
Figure 2. The subject was required to consciously recruit specific
muscles: the deep anterolateral abdominals (with coactivation of the
pelvic floor and lumbar multifidus), followed by activation of the
gluteus muscle to control a static posture.
Functional Disability Outcome Functional disability was
evaluated with the RMQ
32,33
and the RMDQ-HK.
11
This tool is a self-administered questionnaire listing
activities that can be compromised by LBP. The RMQ
(English version)/RMDQ-HK (Chinese version vali-
dated in a Hong Kong Chinese population) is a
self-administered questionnaire consisting of 24 items
to measure disability secondary to LBP. The scores
range from 0 to 24, with 0 representing no disability
and a score of 24 indicating severe disability.
Statistical Analysis
Statistical analysis was performed using SPSS soft-
ware. Results were considered statistically significant if
the Pvalue was less than .05
Part 1 Subject characteristics, such as height and
body mass, were compared between groups prior to
the treatment intervention using unpaired ttests.
Gender distribution, nature of condition, area of
symptoms, previous physiotherapy treatment, and the
inclusion exercise therapy were compared with a
nonparametric statistics. Duration of symptoms
(years), however, was analyzed with a nonparametric
test, considering the positively skewed nature of the
data. Outcome measures following the 4-week treat-
ment intervention period were compared between
the 2 groups using an analysis of covariance accord-
ing to the general linear model, with group (2 levels:
CG and SETG) as main factor, prestest measurements
as a covariate, and posttest measurements as depen-
dent variable.
Part 2 RMQ/RMDQ-HK data were collected for the
SETG immediately after and at 3, 6, and 12 months
following the treatment intervention period. As
follow-up information was not available for some
participants, a sensitivity analysis with 4 intention-to-
treat analyses was conducted to evaluate the retention
of treatment effect. First, missing data of all random-
ized subjects were handled with the ‘‘last observation
carried forward’’ (LOCF) imputation method and
analyzed with a repeated-measures ANOVA on the
different periods that data were collected, followed by
post hoc analyses using Fisher’s least significant
difference test. To verify the robustness of the conclu-
sions of the analysis, 3 intention-to-treat analyses were
conducted with 3 alternative approaches. The
ANOVAs were carried out for the subjects with a
complete data set only, and then the worst-case value
was imputed to the missing data, and finally the
best-case value. Post hoc analyses were once again
conducted using the Fisher least significant difference
test.
RESULTS
Subjects
Subjects were recruited over a 4-month period. The
treatment intervention took place over a 4-week
period from the time of randomization. Analysis
indicated no significant difference between the
groups regarding baseline characteristics (Table 1).
The study sample represented a population of
longstanding LBP conditions with the median dura-
tion of symptoms in the CG (n = 18) and SETG (n =
J Orthop Sports Phys Ther • Volume 36 • Number 7 • July 2006 477
RESEARCH REPORT
Figure 3. Static posture training was followed by training the control
of a variety of movement patterns involving hip extension.
21) at 9 years (range, 1-20 years) and 5.5 years
(range, 0.5-27 years), respectively. Twenty percent of
the entire subject group described their LBP condi-
tion to be one of a recurring nature and 80% of the
subjects described their condition to be of a chronic,
persistent nature. Ninety percent of all subjects had
received previous physiotherapy intervention and
74% of those interventions had included exercise
therapy.
All subjects in the main study completed the
4-week treatment intervention according to the study
protocol and questionnaire data were collected for
analysis from all subjects in the main study. Response
rate to posttreatment intervention questionnaires in
part 2 was 86% at 3 months, 57% at 6 months, and
62% at 12 months (Figure 1).
Treatment Efficacy (Part 1)
After adjusting for measurements at pretest, there
was a significant reduction in average pain intensity
(P= .002) and in functional disability (P= .023) in
the SETG following the treatment intervention pe-
riod (Table 2). No significant interactions between
pretreatment and group were detected for either test.
The means and adjusted means and standard errors
of the mean for the SETG and CG pretreatment and
posttreatment intervention period are depicted
graphically in Figures 5 and 6. The posttest adjusted
mean in the SETG was 2.0 (95% CI, 1.3 to 2.7)
RMQ/RMDQ-HK points compared to a posttest ad-
justed mean in the CG of 3.2 (95% CI, 2.5 to 4.0)
RMQ/RMDQ-HK points. The posttest adjusted mean
in pain intensity in the SETG was 18.3 (95% CI, 11.8
to 24.8) NRS points, as compared to 33.9 (95% CI,
26.9 to 41.0) NRS points in the control group.
Retention of Treatment Effect (Part 2)
Retention of treatment effects was examined in the
SETG for RMQ/ RMDQ-HK data collected at 3, 6,
and 12 months following the completion of the main
study (Table 3).
The first intention-to-treat analysis, using LOCF,
revealed significant improvements in RMQ/
RMDQ-HK scores over the 12-month period (P⬍.01)
(Figure 7). Post hoc analysis showed a significant
difference between pretreatment and posttreatment,
and pretreatment and the 3 follow-up periods. Data
for the posttreatment differed from the 3- and
6-month periods, but not from the 12-month period.
However, no differences were found among the 3
follow-up periods. The analysis suggests that treat-
ment effects were not only retained over time but
that the functional disability score decreased further
following the completion of the treatment interven-
tion up until 3 months. From 3 months to 12
months, the effects seem to be maintained, although
these results should be considered with caution,
knowing the lack of significant difference between
posttreatment results and the 12-month follow-up.
The other 3 intention-to-treat analyses supported,
for the most part, the results found. Significant
improvements were found for the group of subjects
with a complete data set (P⬍.01) and for the data set
with best-case value imputed to the missing data
(P⬍.01). Post hoc tests also showed similar results,
478 J Orthop Sports Phys Ther • Volume 36 • Number 7 • July 2006
Figure 4. Training progressed on the Pilates Reformer initially involving weight-bearing movements in a lumbar-pelvic neutral position. If
tolerated, more non–weight-bearing movements and control of postures out of neutral were introduced.
with the exception that the best-case method did not
find a difference between the posttreatment and the
12-month period. When the worst-case value was
imputed to the missing data, the results from the
analysis did not show a difference between the
periods (P= .12).
DISCUSSION
The main finding of this study was that a program
of specific exercise directed at retraining
neuromuscular control, provided by a physiotherapist,
and based on the Pilates method was more effica-
J Orthop Sports Phys Ther • Volume 36 • Number 7 • July 2006 479
RESEARCH REPORT
TABLE 2. Pretest means (SEM) and adjusted posttest means for functional disability and pain intensity for the specific-exercise-training
group and the control group.
Control Group (n = 18)
Specific-Exercise-Training
group (n = 21)
Outcome Measures Pretreatment Posttreatment Pretreatment Posttreatment PValue*
Functional disability 4.2 (0.8) 3.2 (0.4) 3.1 (0.6) 2.0 (0.3) .023
Pain intensity score 30.4 (4.2) 33.9 (3.5) 23.0 (3.9) 18.3 (3.2) .002
* Comparison of posttest scores using pretest scores as covariate.
cious in reducing pain intensity and functional dis-
ability levels when compared to usual care. In this
study there were significant improvements in mean
average pain intensity and in functional disability
levels in the SETG following treatment that were not
apparent in the CG. Both groups completed the
study and compliance was high, with 100% atten-
dance at scheduled clinic appointments. Although
compliance with the home protocol was not mea-
sured, it was monitored on a log, and a verbal report
from the treating physiotherapist reported good com-
pliance in general. The ability to exercise without
pain, the opportunity to practice at home, and the
quick changes in pain and carryover to function
relative to previous chronic condition were reported
as important motivating factors for compliance.
The results of this study are in agreement with the
conclusion of a systematic review of the literature
45
and the findings by O’Sullivan et al
25
and Lindgren
et al.
20
These investigators found that a training
approach that followed the principles of segmental
stabilization and neuromuscular control was effective
in reducing pain
20,25
and disability
25
in a group of
individuals with chronic LBP related specifically to
radiological instability. The current study demon-
strates that an exercise training approach similarly
addressing neuromuscular control mechanisms is ef-
fective in decreasing pain and improving function in
an identified group with nonspecific chronic LBP
when compared to a control group. The subjects in
the O’Sullivan et al
25
study were trained in stabiliza-
tion exercises designed to enhance local muscular
stability of the intervertebral segment. Although the
theoretical rationale of training muscle activation and
control was similar in both studies, the training
approach in the current study necessarily differed
and addressed different components of
neuromuscular dysfunction. The subjects in the cur-
rent study did not demonstrate a primary segmental
instability but clinically appeared to display problems
in control of the muscles thought to stabilize the
lumbar-pelvis region during hip extension and load-
ing.
14,38,46
Therefore, in both the home and the
clinic protocol, both local and global stabilization
strategies were employed. The transversely oriented
abdominal muscles, the lumbar multifidi, the dia-
phragm, and the muscles of the pelvic floor have all
been shown to be important for local stabiliza-
tion.
7,8,10,12
Specific activation of the gluteal muscles
was emphasized in this study to assist with global
stability of the lumbar-pelvis-hip region during move-
ment. The Pilates Reformer was used to train a
variety of functional movement patterns involving hip
extension.
In the current study it is not known whether the
treatment effect found was due to training the local
stability system or from training more general stabili-
zation strategies. Similarly, it is impossible to disassoci-
ate the effect of the clinic intervention from the
effects of the home intervention.
The subjects in the study reflected a chronic LBP
population with a mean duration of symptoms of 8.2
years. Despite chronic symptoms, this group was able
to participate in some form of physical activity at a
frequency of 3 times per week. Mean initial RMQ/
RMDQ-HK scores for both groups fell within the
lowest strata (0-9) identified by Stratford et al,
40
suggesting a relatively low level of reported disability.
Despite the apparently low disability and moderate
activity levels, all subjects continued to report func-
tional restrictions not resolved with previous interven-
tions. It is speculated that this functional limitation
was the motivation to continue to seek treatment. All
subjects had received treatment in the past and 90%
had received previous physiotherapy treatment(s),
74% of which had included exercise therapy. Most of
the subjects had seen more than 1 medical specialist
over the years and were continuing to seek treatment.
There were no differences between the 2 groups in
any of these characteristics. This population may
represent a significant subgroup of patients with
chronic LBP who seek ongoing treatment in the
clinic setting and it may be argued that the needs of
this group are not adequately met by traditional
interventions, identifying a gap in physical medicine
service deliver y.
It may be argued that a more useful indicator of
outcome may be the clinical significance of changes
in disability that are identified in a population with
low initial RMQ/RMDQ-HK scores. The minimal
clinically important difference (MCID) represents the
change in function that is important to an individual
patient,
4
and it is a function of the initial RMQ
scores. Stratford and colleagues
40
have determined
480 J Orthop Sports Phys Ther • Volume 36 • Number 7 • July 2006
Figure 5. Average pain intensity scores for specific-exercise-training
group (SETG) (n = 21) and control group (CG) (n = 18) previous to
(pre) and immediately following (post) the treatment intervention
period. Values are means (pre) and adjusted means (post) and
standard error of the means. *Significant difference of post scores,
using pre scores as covariate.
Figure 6. Functional disability scores for specific-exercise-training
group (SETG) (n = 21) and control group (CG) (n = 18) previous to
(pre) and immediately following (post) the treatment intervention
period. Values are means (pre) and adjusted means (post) and
standard error of the means. *Significant difference of post scores,
using pre scores as covariate.
that a change of 1 to 2 RMQ points is reflective of
clinically important changes in subjects whose initial
RMQ scores fall within the lowest strata. These results
may reflect clinically important changes in functional
disability in the group who received specific exercise
training in contrast to the group who did not. This
lends further support that treatment with specific
Pilates-based exercise was more effective than usual
care in attaining clinically meaningful changes in
functional capacity in our group of subjects. Further-
more, although the changes in functional disability
are small when examining a group whose pretreat-
ment and posttreatment scores fall within the lowest
strata, it may also be important that no subjects
reported an increase in disability throughout the
study period.
The subjects in this study fall under the broad
classification of nonspecific LBP, although the inclu-
sion criteria were strictly controlled. Classification
into more homogenous groups of LBP diagnosis with
an intervention tailored according to the needs of
the particular group is thought likely to enhance
treatment efficacy.
19
Roland and Morris
32,33
suggest
that if an intervention is applied indiscriminately to
all patients with LBP, it is unlikely that any major
effect from treatment will be discerned. Due to the
natural history of LBP it was felt necessar y to control
for some of its variability and the tight inclusion
criteria attempted to facilitate homogeneity. All sub-
jects reported unresolved chronic LBP, were physi-
cally active, and showed evidence on clinical
tests
13,18,38
of altered performance in the muscles
about the lumbar-pelvic region, including the gluteus
maximus muscle. It was felt that subjects meeting the
criteria would be more likely to respond to this
treatment approach. Therefore it follows that a limita-
tion of this study may be that results cannot be
extrapolated with confidence to chronic LBP condi-
tions outside of the selection criteria for this study,
and especially individuals with acute or more dis-
abling LBP. The high percentage of subjects receiving
recent care from a health professional, the extent of
past physical treatments, and the prior level of
exposure to physiotherapy and exercise interventions
suggest a response to the intervention itself as op-
posed to a placebo effect. Additionally, given the
chronic nature of the subjects’ conditions, it is not
likely that the results in the SETG are due to the
passage of time.
Similarly, the results in the SETG cannot simply be
explained by the introduction of physical training, as
only physically active subjects were selected for the
study. Subjects in both groups continued with general
physical training or advice as prescribed by the
independent evaluator and their attending practi-
tioner throughout the duration of the study.
Finally, all of the subjects in the CG were instructed
that they would have the opportunity to receive
treatment after a 4-week period.
0
10
20
30
40
50
60
Pre Pos t
Ave rag e Pain Inte ns it y (0 t o 100 NRS p oi nt s)
(0 = no pai n, 100 = severe pai n)
CG
SETG
*
0
1
2
3
4
5
6
7
8
9
Pre Po st
Functional Disability (0 to 24 RMQ/RMDQ-HK points)
(0 = no disabilty, 24 = severe disability)
CG
SETG
*
J Orthop Sports Phys Ther • Volume 36 • Number 7 • July 2006 481
RESEARCH REPORT
TABLE 3. Retention of treatment effects for functional disability for the specific-exercise-training group previous to (prettreatment), im-
mediately following (posttreatment), 3, 6, and 12 months after the treatment intervention for the data collected and with the last obser-
vation carried forward intention-to-treat analysis. Values are in means (SEM).
Pretreatment Posttreatment 3 Months 6 Months 12 Months
Data collected 3.1 (0.6) 1.7 (0.4) 0.9 (0.4) 0. 4 (0.2) 0.9 (0.6)
n 2121181213
Last observation carried forward 3.1 (0.6) 1.7 (0.4) 1.0 (0.4) 1.0 (0.4) 1.1 (0.4)
n 2121212121
Figure 7. Functional disability scores for specific-exercise-training group (SETG) previous to (pre) and immediately following (post) the
treatment intervention and at 3-, 6-, and 12-month follow-up, using the last observation carried forward intention-to-treat analysis. Values
are means and standard error of the means (n = 21).
To measure for retention of treatment effect,
disability measures were collected from the SETG by
questionnaire over a 12-month period following
completion of the treatment. Due to the noncompli-
ance of some subjects, the data were analyzed using a
sensitivity analysis contrasting 4 intention-to-treat
analyses. First the missing data were replaced by the
LOCF value. Analysis of these data indicated lower
mean functional disability levels were relatively well
maintained over the 12-month follow-up period. The
biggest changes occurred from the pretreatment to
the 3-month follow-up. The disability levels in the
treatment group were negligible by the 6-month
period and any further changes may not have been of
clinical importance.
To evaluate the robustness of these results, the
analysis was conducted again but with 3 different
methods of handling the missing data. Results were
the same with the sample of subjects with a complete
data set as well as with the best-case value as imputing
method. Results did not show a significant difference,
however, when using the worst-case value. Consider-
ing this last analysis and the substantial loss of data,
the conclusions drawn from the results should be
considered with caution.
The lower response rates of 57% for the 6-month
and 62% for the 12-month follow-ups may confound
the strength of any findings beyond the 3-month
follow-up, depending on the reason for dropout from
this part of the study. Three subjects were lost
482 J Orthop Sports Phys Ther • Volume 36 • Number 7 • July 2006
following posttesting and 1 subject was lost at the
12-month follow-up. Some of the subjects who did
not respond at 6 months did at 12 months and vice
versa. An analysis of the group of subjects who did
not respond to either or both of the 6- and 12-month
follow-ups were shown to have responded similarly to
the program on all outcomes compared to the group
who provided all follow-up data. These findings
would suggest that factors other than a difference in
status or treatment response acted to influence par-
ticipation throughout the follow-up period in this
group.
Ability and concomitant disability are relative to
individual expectations of function. Also important
were unsolicited comments on the questionnaires
returned from subjects whose RMQ/RMDQ-HK
scores throughout the follow-up period were 0 out of
24. These subjects described a progressive ability to
return to activities that had been previously stopped
secondary to low back problems that could not be
reflected in the RMQ/RMDQ-HK scores. The RMQ/
RMDQ-HK may not be sensitive enough to pick up
these changes in ability despite the significance for
normal function in this population. An alternative,
more sensitive measure for this population that has
similar measurement properties to the RMQ/
RMDQ-HK is not currently available. Such a measure
may be useful in demonstrating change in this
subgroup of patients with chronic LBP who continue
to seek care in physiotherapy.
Clinical Implications
This group of patients may represent individuals
seen in physiotherapy clinics and who seek ongoing
treatment from healthcare practitioners after subse-
quent return to normal daily activities despite
chronic LBP. The return to normal activities is often
with some restrictions and limitations and accompa-
nied with longstanding or recurrent pain. It is this
lack of full recovery and ongoing pain that appears
to prompt this clientele to seek ongoing treatment. It
would appear that a treatment to re-educate
neuromuscular control strategies with a structured
approach in the clinic environment and comple-
mented with incorporation into a subject-specific
functional task is beneficial in the restoration of
functional ability levels acceptable to the patient. This
specialized exercise approach can be adjunctive to, or
may follow other treatment at some point in the
continuum of rehabilitation interventions.
In summary, an exercise approach that targets
selective muscle recruitment and neuromuscular re-
training of stabilization strategies may have a signifi-
cant effect on pain and disability in a population of
active subjects with chronic LBP, with long-lasting
effects. A modification of the Pilates method may
provide a useful and cost effective treatment modality
in the management of such conditions and merits
further study.
CONCLUSIONS
The results of the study support the hypothesis that
an exercise therapy approach based on the Pilates
method and directed at neuromuscular control
mechanisms was efficacious in the treatment of a
group of individuals with nonspecific chronic LBP. A
4-week treatment intervention was more efficacious
than usual care in reducing average pain intensity
and functional disability levels, changes were main-
tained over a 12-month period.
ACKNOWLEDGMENTS
The authors would like to thank Professor Lo Sing
Kai, Deakin University and Karine Charpentier,
Queen’s University, for statistical advice.
REFERENCES
1. Acute Low Back Problems in Adults. Clinical Practice
Guideline No. 14. AHCPR Publication No. 95-0642.
Rockville, MD: Agency for Health Care and Policy
Research, Public Health Service, US Department of
Health and Human Services; 1994.
2. Arendt-Nielsen L, Graven-Nielsen T, Svarrer H, Svens-
son P. The influence of low back pain on muscle
activity and coordination during gait: a clinical and
experimental study. Pain. 1996;64:231-240.
3. Bergmark A. Stability of the lumbar spine. A study in
mechanical engineering. Acta Orthop Scand Suppl.
1989;230:1-54.
4. Binkley JM. Measurement of functional status, progress,
and outcome in orthopaedic clinical practice. Orthop
Div Rev. 1998;September/October:7-17.
5. Edgerton VR, Wolf SL, Levendowski DJ, Roy RR.
Theoretical basis for patterning EMG amplitudes to
assess muscle dysfunction. Med Sci Sports Exerc.
1996;28:744-751.
6. Health Canada and the Canadian Society for Exercise
Physiology. Canada’s physical activity guide to healthy
active living: Health Canada and the Canadian Society
for Exercise Physiology. Ottawa, Ontario, Canada:
Health Canada and the Canadian Society for Exercise
Physiology; 1998.
7. Hodges PW, Gandevia SC. Activation of the human
diaphragm during a repetitive postural task. J Physiol.
2000;522 Pt 1:165-175.
8. Hodges PW, Richardson CA. Altered trunk muscle
recruitment in people with low back pain with upper
limb movement at different speeds. Arch Phys Med
Rehabil. 1999;80:1005-1012.
9. Hodges PW, Richardson CA. Feedforward contraction
of transversus abdominis is not influenced by the
direction of arm movement. Exp Brain Res.
1997;114:362-370.
10. Hodges PW, Richardson CA. Inefficient muscular stabi-
lization of the lumbar spine associated with low back
pain. A motor control evaluation of transversus
abdominis. Spine. 1996;21:2640-2650.
J Orthop Sports Phys Ther • Volume 36 • Number 7 • July 2006 483
RESEARCH REPORT
11. Hong Kong Hospital Authority. Examining reliability,
validity and responsiveness of Hong Kong version of
Roland-Morris Disability Questionnaire (RMDQ-HK) as
an outcome measure of patients with low back pain: a
multicentre study. Hong Kong, China: Working Group
on Low Back Pain Outcomes Assessment, Coordinating
Committee for Physiotherapists, Hong Kong Hospital
Authority; 1999.
12. Hungerford B, Gilleard W, Hodges P. Evidence of
altered lumbopelvic muscle recruitment in the presence
of sacroiliac joint pain. Spine. 2003;28:1593-1600.
13. Janda V. Evaluation of muscular imbalance. In:
Liebenson C, ed. Rehabillitation of the Spine: A Practi-
tioner’s Manual. Baltimore, MD: Lippincott, Williams &
Wilkins; 1996:
14. Janda V. Muscle weakness and inhibition
(pseudoparesis) in back pain syndromes. In: Grieve GP,
ed. Modern Manual Therapy of the Vertebral Column.
New York, NY: Churchill-Livingstone; 1986.
15. Janda V. Muscles, central nervous motor regulation and
back problems. In: Korr M, ed. The Neurobiologic
Mechanisms in Manipulative Therapy. New York, NY:
Plenum Press; 1978.
16. Jensen MP, Karoly P, Braver S. The measurement of
clinical pain intensity: a comparison of six methods.
Pain. 1986;27:117-126.
17. Jull GA, Janda V. Muscles and motor control in low
back pain: assessment and management. In: Twomey
LT, ed. Physical Therapy of the Low Back. New York,
NY: Churchill-Livingstone; 1987.
18. Kendall FP, Kendall EM, Provance PG. Muscles: Testing
and Function. 4th ed. Baltimore, MD: Williams &
Wilkins; 1993.
19. Leboeuf-Yde C, Lauritsen JM, Lauritzen T. Why has the
search for causes of low back pain largely been
nonconclusive? Spine. 1997;22:877-881.
20. Lindgren KA, Sihvonen T, Leino E, Pitkanen M, Man-
ninen H. Exercise therapy effects on functional radio-
graphic findings and segmental electromyographic
activity in lumbar spine instability. Arch Phys Med
Rehabil. 1993;74:933-939.
21. Lo SK, Lin CF, Phillips BA. On the use and presentation
of analysis of variance in rehabilitation research. Hong
Kong Physiother J. 1998;16:7-12.
22. Maher C, Latimer J, Refshauge K. Prescription of activity
for low back pain: What works? Aust J Physiother.
1999;45:121-132.
23. Mayer TG, Gatchel RJ. Functional Restoration for Spinal
Disorders: The Sports Medicine Approach. Philadelphia,
PA: Lea & Febiger; 1988.
24. McGill S, Grenier S, Bluhm M, Preuss R, Brown S,
Russell C. Previous history of LBP with work loss is
related to lingering deficits in biomechanical, physi-
ological, personal, psychosocial and motor control
characteristics. Ergonomics. 2003;46:731-746.
25. O’Sullivan PB, Phyty GD, Twomey LT, Allison GT.
Evaluation of specific stabilizing exercise in the treat-
ment of chronic low back pain with radiologic diagno-
sis of spondylolysis or spondylolisthesis. Spine.
1997;22:2959-2967.
26. Panjabi MM. Clinical spinal instability and low back
pain. J Electromyogr Kinesiol. 2003;13:371-379.
27. Pool-Goudzwaard AL, Vleeming A, Stoeckart R,
Snijders CJ, Mens JM. Insufficient lumbopelvic stability:
a clinical, anatomical and biomechanical approach to
’a-specific’ low back pain. Man Ther. 1998;3:12-20.
28. Rainville J, Hartigan C, Martinez E, Limke J, Jouve C,
Finno M. Exercise as a treatment for chronic low back
pain. Spine J. 2004;4:106-115.
29. Richardson CA, Jull GA. Concepts of rehabilitation for
spinal stability. In: Boyling JD, Palastanga N, eds.
Grieve’s Modern Manual Therapy of the Vertebral
Column. Edinburgh, UK: Churchill-Livingstone;
1994:705-720.
30. Richardson CA, Jull GA. Muscle control-pain control.
What exercises would you prescribe? Man Ther.
1995;1:2-10.
31. Richardson CA, Jull GA, Hodges P, Hides J. Therapeutic
Exercise for Spinal Segmental Stabilization in Low Back
Pain. London, UK: Churchill Livingstone; 1999.
32. Roland M, Morris R. A study of the natural history of
back pain. Part I: development of a reliable and
sensitive measure of disability in low-back pain. Spine.
1983;8:141-144.
33. Roland M, Morris R. A study of the natural history of
back pain. Part II: development guidelines for trials of
treatment in primary care. Spine. 1983;8:144-150.
34. Rydeard RA. Die rehabilitation eines elite-triathleten.
Trainingstherapie nach dem konzeptder Biokinetik
Exercice Technique: ein Fallbericht (Teil 1).
Physiotherapie Fisioterapia. 2000;4:12-16.
35. Rydeard RA. Die rehabilitation eines elite-triathleten.
Trainingstherapie nach dem konzeptder Biokinetik
Exercice Technique: ein Fallbericht (Teil 2).
Physiotherapie Fisioterapia. 2000;5:12-16.
36. Sahrmann SA. Posture and muscle imbalance: faulty
lumbar pelvic alignment and associated musculo-
skeletal pain syndromes. Orthop Div Rev. 1992;Nov-
Dec:13-20.
37. Siler B. The Pilates Body. New York, NY: Random
House; 2000.
38. Singer K. Suggestions from the clinic: a new
nusculoskeletal assessment in a student population.
J Orthop Sports Phys Ther. 1986;8:34-41.
39. Spitzer WO. Scientific approach to the assessment and
management of activity-related spinal disorders. A
monograph for clinicians. Spine. 1987;12:S1-S59.
40. Stratford PW, Binkley JM, Riddle DL, Guyatt GH.
Sensitivity to change of the Roland-Morris Back Pain
Questionnaire: part 1. Phys Ther. 1998;78:1186-1196.
41. Tsang RCC. Measurement properties of the Chinese
version of the Roland-Morris Disability Questionnaire.
Hong Kong Physiother J. 2004;22:40-49.
42. US Department of Health and Human Services
(USDHHS), Centers for Disease Control and Prevention.
Physical activity and health: a report of the Surgeon
General. Altanta, GA: USDHHS; 1996.
43. van Tulder M, Malmivaara A, Esmail R, Koes B. Exercise
therapy for low back pain: a systematic review within
the framework of the cochrane collaboration back
review group. Spine. 2000;25:2784-2796.
44. van Tulder MW, Koes BW, Bouter LM. Conservative
treatment of acute and chronic nonspecific low back
pain. A systematic review of randomized controlled
trials of the most common interventions. Spine.
1997;22:2128-2156.
45. van Tulder MW, Malmivaara A, Esmail R, Koes BW.
Exercise therapy for low back pain. Cochrane Database
Syst Rev. 2000;CD000335.
46. Vleeming A, Pool-Goudzwaard AL, Stoeckart R, van
Wingerden JP, Snijders CJ. The posterior layer of the
thoracolumbar fascia. Its function in load transfer from
spine to legs. Spine. 1995;20:753-758.
47. Vogt L, Pfeifer K, Banzer W. Neuromuscular control of
walking with chronic low-back pain. Man Ther.
2003;8:21-28.
484 J Orthop Sports Phys Ther • Volume 36 • Number 7 • July 2006
