Pilates for low back pain: a systematic review
Complementary Therapies in Clinical Practice
Pilates for low back pain: A systematic review
Paul Posadzki ,, Pawel Lizis , Magdalena Hagner-Derengowska
a Complementary Medicine, Peninsula Medical School, 25 Victoria Park Road, Exeter, Devon EX2 4NT, UKb Institute of Physiotherapy, Saint Cross Physiotherapy College, Kielce, Polandc Rehabilitation Clinic, Department of Health Sciences, Collegium Medicum, Nicolas Copernicus University, Bydgoszcz, Poland
Objective: The aim of this paper is to systematically review all controlled clinical trials of Pilates to treat
Data sources: A systematic review of nine databases (Cochrane Central Register of Controlled Trials,
MEDLINE, EMBASE, CINAHL, AMED, PsycINFO, Pedro, Rehadat, Rehab Trials) was conducted and the
reference lists of all the papers were checked for further relevant publications until May 2010.
Study selection: A ﬁrst selection was performed by means of title and abstract. A second selection wasmade by means of predeﬁned inclusion criteria: randomized controlled clinical trials testing Pilates inpatients of any age or sex with low back pain.
Data extraction: Data relating to changes in body function, quality of life and pain from the includedstudies were independently extracted by the reviewers on a standardized form. Study quality wasassessed using the Oxford scale.
Data synthesis: Four eligible randomized controlled clinical trials (n ¼ 4) involving Pilates for themanagement of low back pain were included. The methodological quality of the RCTs was relatively low,varying from 1e4 on the Oxford scale. All studies were heterogeneous in terms of population of patients,control groups, inclusion and exclusion criteria, and outcome measures making a meta-analysis notfeasible. Although there is some evidence supporting the effectiveness of Pilates in the management oflow back pain, no deﬁnite conclusions can be drawn except that further research is needed with largersamples and using clearer deﬁnitions of the standard care and comparable outcome measures.
Conclusions: There is a wide diversity in research investigating the clinical and cost-effectiveness ofPilates in patients with low back pain.
experience inappropriate neuromuscular adaptations to maintainand/or preserve functions such as walking, running, or other activi-
Chronic low back pain (LBP) is one of the commonest musculo-
ties.Potentially effective therapies for this disorder are appropriate
skeletal problems in modern societand is a highly prevalent and
education, i.e. Alexander Technique and cognitive-behavioural
very expensive health dilemmaThe aetiology of this disability is
therapy; other alternative modalities such as hypnosis, biofeedback,
complex and multidimensional; however, physical and (partially)
relaxation, massage, spinal manipulation and traction treatme
psychosocial occupational factors seem to play an important aetio-
and spinal stabilization exercises.Some researchers suggest that
logical rolLBP is deﬁned as pain localized between the twelfth rib
weakened muscles such as the transversus abdominis (TA) and
and the inferior gluteal folds, with or without leg pain and in 90% of
multiﬁdius (MF) may be responsible for decreased spinal stability and
cases is non-speciﬁc.Other researchers conclude that is best
consequently the onset of LBPThe Pilates method strengthens
deﬁned as a low level continuous or essentially continuous lumbar,
these muscles and hence may be an effective modality for LBThe
sacral or lumbosacral spinal pain that is punctuated by exacerbations
Pilates method was originally developed by Joseph Pilates during the
of pain, each of which is characterized as ‘acute’. Patients who expe-
First World War and since then it has brought new insight to lower
rience this disability are limited in their daily living activities and may
back rehabilitation methods.Pilates’ initial concept mixed elementsof gymnastics, martial arts, yoga and dance, focusing on the rela-tionship between the body and mental discipline.The goal of Pilatestraining is to improve general body
strength and posture, and to coordinate movement with the breath.
1744-3881/$ e see front matter Ó 2010 Published by Elsevier Ltd.
P. Posadzki et al. / Complementary Therapies in Clinical Practice 17 (2011) 85e89
To date, scientiﬁc evidence of the effectiveness of the Pilates method
further evaluation of which 4, involving 228 patients, were eligible
for the treatment of LBP is rather anecdotaand therefore the aim of
for inclusion (see diagram). Reasons for exclusion included no
this review is to systematically assess evidence of its efﬁcacy in the
speciﬁc outcome measure, trials including patients with conditions
treatment of low back pain from all controlled clinical trials.
other than LBP such as ﬁbromyalgia syndrome, and trials thatincluded healthy individuals only. For example, Culligan et al.and
Sekendiz et al’studies were excluded because they did notinclude LBP patients; Kloubec’research was excluded as it
Literature searches were performed to identify all controlled
considered healthy individuals only; Merrithew’s was
clinical trials of Pilates as treatment for LBP. The following
excluded as it was not a randomized controlled trial; Da Fonseca et
databases were used: Cochrane Central Register of Controlled
al’sstudy was excluded because pain was not the primary
Trials, MEDLINE, EMBASE, CINAHL, AMED, PsycINFO, Pedro,
outcome measure; and Cairns et and Hides et were
Rehadat, Rehab Trials and web pages such as
excluded as their work was concerned with core stability exercises
using the search terms ‘low back pain’, ‘chronic’, ‘discogenic low
back pain’, ‘non-speciﬁc low back pain’, ‘Pilates’, ‘rehabilitation’and ‘physiotherapy’ to identify all relevant published articles onthe subject. The reference lists of the papers initially identiﬁed
were scanned for further relevant literature. No language barrierswere imposed.
Four studies meeting the criteria mentioned above were
included.They originated from the UK,the and
Canada.Pilates was used in all four studies. LBP patient pop-ulations were heterogeneous, and the descriptions of pain included
All retrieved data including uncontrolled trials, case studies,
chronic pain, discogenic pain and non-speciﬁc back pain. Control
pre-clinical and observational studies were reviewed for safety
groups were standard or usual Back School,drug therapy
information. However, only randomized controlled clinical trials
testing Pilates in patients of any age or sex with low back pain were
Gladwell et al. conducted an RCT to evaluate the effect of
included. Studies in any language published in peer-reviewed
modiﬁed Pilates on 49 active individuals with chronic LBP.Study
scientiﬁc journals between 1980 and 2010 were considered eligible.
participants were randomly allocated to a control or a Pilates group.
Main outcome measures included (1) the visual analogue scale
3.1. Methodological quality of the studies
(VAS), (2) subjective improvement of symptoms, (3) an assessmentof back-speciﬁc functional status. They report that Pilates was
The quality of the studies was assessed using the ﬁve-point
found to be superior to the controls and improvements were seen
Oxford which has good inter-examiner reliabilityThis
in this group’s general health, sports functioning, ﬂexibility and
scale assesses methodological quality such as randomization and
proprioception, and they experienced less pain. There are some
blinding procedures, descriptions of withdrawal and dropout rates,
limitations to this study: ﬁrstly, the randomization is not clearly
using a scale from 0 (poorest) to 5 (highest). Points were awarded as
described; secondly, it is single-blinded trial only; thirdly, no
follows: study described as randomized, 1 point; appropriate
intention to treat analysis was performed; and fourthly, the sample
randomization method, 1 additional point; inappropriate random-
was rather heterogeneous. The strengths of this trial include the
ization method, deduct 1 point; patient blinded to intervention
good-quality statistical methods used and the adequate description
(patient blinding was assumed where the control intervention was
of loss to follow-up rate. We gave this study a score of 3.
indistinguishable from the treatment group), 1 point; evaluator
Donzelli et al. conducted a randomized controlled trial with 53
blinded to intervention, 1 point; description of withdrawals and
patients with non-speciﬁc LBPPatients entered either a Pilates
dropouts,1 point. Clinical trials scoring 4 or 5 points were considered
therapy or a Back School treatment group. They used the Oswestry
Low Back Pain Disability Scale (OLBPDQ) and VAS and at six monthsno signiﬁcant differences were found between the groups. None-
3.2. Data extraction and quality assessment
theless, the Pilates method group showed better compliance andsubjective response to treatment. Although this study is described as
Initial screening of the abstracts of the studies was performed by
randomized, there is no speciﬁcation at all of how the randomization
two authors (PP, PL) independently. If it was not clear from the
and allocation to groups was performed. The blinding procedure is
abstract whether or not the study should be included in the review,
not adequately described, neither does the study elaborate on
the full text of the article was assessed. All articles included were
whether the statistical analysis was masked. Intention to treat
read in full. Data relating to sample size, diagnosis, gender of
analysis is not mentioned either. Another possible limitation of this
patients in the samples used, their previous incidences of LBP,
study is the lack of statistical analysis between the intervention and
therapeutic intervention and control, treatment time, primary (and
control groups, with frequency tables and distribution of variables
secondary) outcome measures and results were extracted by the
presented only. This study was given a score of 1; however, the
ﬁrst author and validated by the second. The third reviewer (MHD)
dropout (loss to follow-up) rate and eligibility criteria were sufﬁ-
further validated data using a predeﬁned standardized form. The
ciently described and the group was relatively homogeneous.
authors met to come to a consensus and discrepancies were solved
Rydeard et conducted a randomized controlled trial to
investigate the efﬁcacy of the Pilates approach with 39 chronic lowback pain (LBP) patients. Patients were randomly assigned to the
Pilates group while the control group received ‘the usual care’. Theyused Roland-Morris Disability Questionnaire (RMDQ) and NRS-101,
The search strategy generated a total of 199 references, of which
a 101-point numerical rating scale assessing pain intensity. There
51 were considered potentially relevant. We did not locate any
was a signiﬁcantly lower level of functional disability and average
unpublished trials, nor relevant papers published in any language
pain intensity in the speciﬁc exercise training group than in the
other than English. A total of 11 clinical trials were retrieved for
control group following the treatment intervention period.
P. Posadzki et al. / Complementary Therapies in Clinical Practice 17 (2011) 85e89
Obejctive: To systematically review the literature on efficacy of Pilates in LBP.
Study selection: - RCTs and CCTs published between 1980 and 2010 - All LBP patients - Pilates therapy; any task improving physical recovery
Two reviewers performed the independent search
Data Sources:Computer search -Clinicaltrials.gov -Cochrane Register of Controlled Trials- Medline -CINAHL - PsycINFO - EMBASE -AMED - Pedro - Rehadat - RehabTrials - Reference check
Included I Class Articles:Outcome: 4 articles
Two independent reviewers rated the selected studies with the JADAD score
This is a relatively well-designed trial with the randomization
and wore a lumbar cryobrace for 15 min before going to bed at night
clearly speciﬁed. The inclusion and exclusion criteria were pre-
while the control group received drugs and a lumbar cryobrace only.
determined and the group was homogeneous. The loss to follow-up
Outcome measures included RMDQ, numeric pain rating score,
ratio is adequately established. Reasonably strong statistical
patient satisfaction score, measured forward ﬂexion, use of drugs,
procedures were used and intention to treat analyses was con-
time off work and rate of symptom recurrence. The authors report
ducted. However, this study is partially blinded as only the phys-
that at the 12-month follow-up 70% of the therapeutic group reported
iotherapists were blinded to the results of testing and the relatively
over 50% pain reduction and good or better patient satisfaction
low response rate of 57% at 6 months may confound the strength of
compared to 33% in the controls (P ¼.001). We gave this study a score
the ﬁndings. We gave this study a score of 4.
of 2 for several reasons. Firstly, it lacks explicit description of the
Vad et alconducted a prospective randomized study on 87
randomization process. The statistics used are mainly descriptive. No
patients with discogenic LBP to determine the efﬁcacy of the Back Rx
intention to treat analysis was performed, nor are dropouts sufﬁ-
programme, which comprises elements of physical therapy, rehabil-
ciently described. The study also lacks blinding procedures: neither
itation, yoga and Pilates. The treatment group also received drugs,
therapists nor assessors were blinded to the intervention and analysis
namely celecoxib and hydrocodone with acetaminophen as needed,
respectively. Although the authors deﬁne a successful outcome as
P. Posadzki et al. / Complementary Therapies in Clinical Practice 17 (2011) 85e89
Table 1Summary of clinical trials of Pilates for Low Back Pain.
Visual Analogue Scale (VAS) and subjective
ﬂexibility, proprioception,and a decrease in pain
Patient satisfaction score Forward ﬂexiontest Use of drugs Time off workSymptom recurrence
‘greater than 50% pain reduction with good or better patient satis-
maximus muscle using the Lovett test and by visual observation of
faction’, there is no reference for this claim. The study does have some
this muscle maximizes the risk of bias in recruitment to the study.
methodological strengths such as clearly speciﬁed inclusion and
There is considerable inconsistency across the studies regarding
exclusion criteria that may compensate for possible biases.
assessment of LBP at the baseline, with Donzelli et alusing the
Generally, the populations of patients with chronic non-speciﬁc
Lasegue test, which is known for its limited diagnostic accuracy
LBP were well deﬁOnly Vad et deﬁne patients as
and low speciﬁcityThis may have resulted in an increased
‘discogenic with LBP’. Similarly, homogeneous outcome measures
number of false positives in recruitment to their study. Ideally
were applied. For instance, all the studies measured pain and func-
there should be one standardized assessment to determine
tional disability, two using and two using
inclusion and exclusion criteria in future research. Van der Windt
Additional measurements across the studies included the use of
et al. suggest that a combination of SLR and imaging can increase
drugs, symptom recurrence, time off worand quality of life and
speciﬁcity and sensitivity in LBP diagnosis in primary health care
physiological indicators such as the Stork test and the Sit-and-Reach
settings.There is wide variation in terms of the onset and
test, both for physical ﬁtness, in Gladwell et al’s trial.On the other
duration of LBP in the samples. Gladwell et alconsider chronic
hand, several inconsistencies were noticed. First of all, with respect
LBP patients to be those who have had symptoms for more than
to intervention there is no uniform physical/functional assessment
twelve weeks; Donzelli et althree months; Rydeard et al.six
nor eligibility criteria at the baseline of all the studies. For example,
weeks and Vad et al.,‘at least three months. The authors also
to assess and conﬁrm the existence of LBP, patients in the Vad et al.
include vague and undeﬁned categories such as ‘regular physical
studyshould have documented evidence of disk pathology (e.g.
activity’ (Gladwell et al.and ‘sufﬁcient intensity to restrict
protrusion) as indicated by magnetic resonance imaging. Rydeard
functional activity in some manner’ when recruiting LBP patients
et al.measured the relative strength of the gluteus maximus muscle
(Rydeard et alThere is little homogeneity in terms of control
using the Lovett test and by visual observation to conﬁrm the
group intervention, varying between ‘routine’ or ‘standard inter-
presence of LBP. Donzelli et al.established inclusion criteria for
vention’ (Rydeard et al.)the Back School Program (Donzelli
patients with a negative Laseque sign for a straight leg raise, and
studier a; no intervention (Gladwell et al.and drug therapy
Wasserman tests. Interestingly, Gladwell and colleagues looked at
(Vad et Clearer deﬁnitions of ‘usual’ or ‘standard’ care are
the use of drugs and pain in the lower back at the baseline.Similar
needed. Rydeard et al. deﬁne usual care as ‘consultation with
discrepancies were observed in terms of the intervention itself:
a physician and other specialists and health care professionals as
Gladwell et al.describe modiﬁed Pilates; Vad et al.the Back Rx
necessary’,but standard care may also include analgesics,
program (which includes Pilates), Donzelli, Pilates Cova Techand
leaﬂets,or electrotherapy and general exercise.It is impor-
Rydeard et al. mention specialized Pilates exercise equipment
tant to emphasize that standard care for LBP patients needs to be
manufactured by Pilates Reformer, Balance Body Sacramento CA.
redeﬁned if included in future research.
Overall, all four studies lack formal power and sample size
calculation and employed relatively small samples and therefore alllack generalizability, meaning that results cannot be extrapolated
This review explores the clinical effectiveness of Pilates in LPB
to the wider LBP population. Furthermore, the limited sample sizes
patients. Very few studies have been identiﬁed that investigate
in these trials did not allow us to perform a meta-analysis.
the effectiveness of the Pilates method for the treatment of LBP
The potential limitation of this review is that the authors did not
patients. There was one good study (Rydeard et al., 2006in
include such key words as pelvic ﬂoor, core stability, core strength,
which the risk of bias was relatively low. Reasonably high
transversus abdominis and multiﬁdius in their search strategy, since
homogeneity is observed across all studies in terms of outcome
strengthening core stability and the above-mentioned muscles is one
measures. All the reviewed papers focus on functional disability
of the purposes of the Pilates method. However, if the authors had
and pain. Nevertheless, this review indicates that there is
included these key words this approach could be criticized for limiting
heterogeneity at various levels including methodology, physical
the Pilates method to strengthening the pelvic ﬂoor muscles only.
examination, population, the intervention itself and the outcome
Also, from the methodological standpoint comparing and contrasting
measures. For instance, the duration of the interventions ranged
Pilates as an intervention with trials that focus on core stability per se
from six weeks (Gladwell et al. stud) to twelve months (Vad
only was not feasible, as Pilates focuses on a more global approach
et alDespite the fact that Rydeard et al.conducted the
including coordination, endurance, ﬂexibility, cognitive processes and
highest-quality clinical trial of those included in this review, the
self-awareness. Some researchers have written that the important
subjectivity inherent in measuring the strength of the gluteus
elements of improving back pain, including biological, educational,
P. Posadzki et al. / Complementary Therapies in Clinical Practice 17 (2011) 85e89
and psychological aspects, are encompassed within the principles of
15. Latey P. Updating the principles of the Pilates method. J Bodyw Mov Ther
Pilates training and that therefore this method should be regarded as
16. Sekendiz B, Altun O, Korkusuz F, Akın S. Effects of Pilates exercise on trunk
strength, endurance and ﬂexibility in sedentary adult females. J Bodyw MovTher; 2007; Nov;(4):318e26.
17. Latey P. 2001. The Pilates method: history and philosophy. J Bodyw Mov Ther;
18. Altan L, Korkmaz N, Bingol U, Gunay B. Effect of pilates training on people
Although some of the authors of the reviewed studies conclude
with ﬁbromyalgia syndrome: a pilot study. Arch Phys Med Rehabil 2009Dec;90(12):1983e8.
that Pilates yielded better therapeutic results than usual or standard
19. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJM, Gavaghan GJ, et al.
carethe ﬁndings of this review suggest that the evidence avail-
Assessing the quality of reports of randomized clinical trial: is blinding
able for its clinical effectiveness is inconclusive. This systematic review
necessary? Control Clin Trials 1996;17:1e12.
20. Clark O, Castro AA, Filho JV, Djubelgovic B. Interrater agreement of Jadad’s
shows that the evidence base for Pilates method remains scarce and
therefore larger and better-designed clinical trials are needed.
21. Kloubec JA. Pilates for improvement of muscle endurance, ﬂexibility, balance,
and posture. J Strength Cond Res 2010 Mar;24(3):661e7.
22. Merrithew M. Pilates for pain management. Rehab Manag 2009 Mar;22(2)
23. da Fonseca JL, Magini M, de Freitas TH. Laboratory gait analysis in patients with
low back pain before and after a pilates intervention. J Sport Rehabil 2009
1. Egle UT, Nickel R. Chronic low back pain as a somatoform pain disorder.
24. Donzelli S, Di Domenica E, Cova AM, Galletti R, Giunta N. Two different tech-
2. Furlan AD, Imamura M, Dryden T, Irvin E. Massage for low-back pain. Cochrane
niques in the rehabilitation treatment of low back pain: a randomized
Database Syst Rev 2008 Oct 8;(4):CD001929.
controlled trial. Eura Medicophys 2006 Sep;42(3):205e10.
3. Don AS, Carragee E. A brief overview of evidence-informed management of
25. Gladwell V, Head S, Haggar M, Beneke R. Does a program of pilates improve
chronic low back pain with surgery. Spine J 2008 Jan-Feb;8(1):258e65.
chronic non-Speciﬁ c low back pain? J Sport Rehabil 2006;15:338e50.
4. Donelson R. Is your client’s back pain ‘rapidly reversible’? improving low back
26. Rydeard R, Leger A, Smith D. Pilates-based therapeutic exercise: effect on
care at its foundation. Prof Case Manag 2008 Mar-Apr;13(2):87e96.
subjects with nonspeciﬁc chronic low back pain and functional disability:
5. Pradhan BB. Evidence-informed management of chronic low back pain with
a randomized controlled trial. J Orthop Sports Phys Ther 2006 Jul;36(7):472e84.
watchful waiting. Spine J 2008 Jan-Feb;8(1):253e7.
27. Vad VB, Bhat AL, Tarabichi Y. The role of the Back Rx exercise program in
6. Malanga G, Wolff E. Evidence-informed management of chronic low back pain
diskogenic low back pain: a prospective randomized trial. Arch Phys Med
with trigger point injections. Spine J 2008 Jan-Feb;8(1):243e52.
7. Seidler A, Liebers F, Latza U. Prevention of low back pain at work. Bundesge-
28. Devillé WL, van der Windt DA, Dzaferagic A, Bezemer PD, Bouter LM. The test
sundheitsblatt Gesundheitsforschung Gesundheitsschutz 2008 Mar;51(3):322e33.
of Lasègue: systematic review of the accuracy in diagnosing herniated discs.
8. Hammill RR, Beazell JR, Hart JM. Neuromuscular consequences of low back pain
and core dysfunction. Clin Sports Med 2008 Jul;27(3):449e62.
29. van der Windt DA, Simons E, Riphagen II , Ammendolia C, Verhagen AP,
9. Hebert J, Koppenhaver S, Fritz J, Parent E. Clinical prediction for success of inter-
Laslett M, et al. Physical examination for lumbar radiculopathy due to disc
ventions for managing low back pain. Clin Sports Med 2008 Jul;27(3):463e79.
herniation in patients with low-back pain. Cochrane Database Syst Rev 2010 Feb
10. Little P, Lewith G, Webley F, Evans M, Beattie A, Middleton K, et al. Randomised
controlled trial of Alexander technique lessons, exercise, and massage (ATEAM)
30. Koumantakis GA, Watson PJ, Oldham JA. Trunk muscle stabilization training
for chronic and recurrent back pain. BMJ; 2008:337. a884.
plus general exercise versus general exercise only: randomized controlled trial
11. Cairns MC, Foster NE, Wright C. Randomized controlled trial of speciﬁc spinal
of patients with recurrent low back pain. Phys Ther 2005 Mar;85(3):209e25.
stabilization exercises and conventional physiotherapy for recurrent low back
31. Norris C, Matthews M. The role of an integrated back stability program in
pain. Spine (Phila Pa 1976) 2006 Sep 1;31(19):E670e81.
patients with chronic low back pain. Complement Ther Clin Pract 2008 Nov;14
12. Hodges PW, Richardson CA. Inefﬁcient muscular stabilization of the lumbar
spine associated with low back pain. A motor control evaluation of transverses
32. Mohseni-Bandpei MA, Rahmani N, Behtash H, Karimloo M. The effect of
abdominis. Spine 1996;21(Suppl. 22):2640e50.
pelvic ﬂoor muscle exercise on women with chronic non-speciﬁc low back
13. Hides JA, Jull GA, Richardson CA. Long-term effects of speciﬁc stabilizing
exercises for ﬁrst-episode low back pain. Spine 2001 Jun 1;26(11):E243e8.
33. Simmonds MJ, Dreisinger TE. Lower back pain syndrome. In: Durstine JL,
14. Culligan PJ, Scherer J, Dyer K, Priestley JL, Guingon-White G, Delvecchio D, et al.
Moore GE, editors. ACSM’s exercise management for Persons with chronic
A randomized clinical trial comparing pelvic ﬂoor muscle training to a Pilates
Diseases and Disabilities. 2nd ed. Champaign, Ill: Human Kinetics; 2003.
exercise program for improving pelvic muscle strength. Int Urogynecol J Pelvic
Floor Dysfunct 2010 Apr;21(4):401e8 [Epub 2010 Jan 22].
NIH Public Access Author Manuscript Oral Dis . Author manuscript; available in PMC 2008 February 28. Oral Dis . 2007 November ; 13(6): 508–512. Oral biofilms, periodontitis, and pulmonary infections S Paju 1,2 and FA Scannapieco 3 1Institute of Dentistry, University of Helsinki, Helsinki, Finland 2Department of Oral and MaxillofacialDiseases, Helsinki University Central Hospital,
Food and Chemical Toxicology 40 (2002) 1243–1255Center for Occupational and Health Psychology, School of Psychology, Cardiﬀ University, UKThe literature suggests that the following eﬀects on behavior of adult humans may occur when individuals consume moderateamounts of caﬀeine. (1) Caﬀeine increases alertness and reduces fatigue. This may be especially important in low arousal sit