Вы находитесь на странице: 1из 8

Journal of Evaluation in Clinical Practice ISSN 1356-1294

The treatment effect of exercise programmes for chronic low back pain
jep_1174 484..491

Caroline Smith PhD1 and Karen Grimmer-Somers PhD2


1 2

Senior Researcher, Centre for Allied Health Evidence, University of South Australia, Adelaide, Australia Director, Centre for Allied Health Evidence, University of South Australia, Adelaide, Australia

Keywords effect size, exercise intervention, long-term effectiveness, low back pain Correspondence Professor Karen Grimmer-Somers Centre for Allied Health Evidence University of South Australia City East Campus North Terrace Adelaide 5000 Australia E-mail: Karen.grimmer-somers@unisa.edu.au Accepted for publication: 18 November 2008 doi:10.1111/j.1365-2753.2009.01174.x

Abstract
Rationale This paper summarizes evidence of long-term effectiveness of physiotherapy exercise therapy for chronic low back pain (LBP). Methods A literature search was undertaken for experimental studies (20012007), which reported any post-intervention (follow-up) outcomes. Studies were critically appraised using the PEDro instrument. Comparative statistics were calculated, relative to the type of follow-up outcome data. Results Fifteen moderate quality trials were included [mean PEDro score 7.7, SD 1.3 (range 510)]. Nine trials reported pain scales, and six reported LBP reoccurrence. Trials which reported on pain scales at 6-month follow-up found signicant differences in favour of exercise [standardized mean differences -0.57, 95%CI -0.75 to -0.39 (555 participants)]. At 12-month follow-up, a small pain scale benet from exercise persisted [standardized mean differences -0.25, 95%CI -0.44 to 0.06 (434 participants)]. There was unconvincing evidence of exercise effectiveness on pain scales after this time. Three of the four trials which reported dichotomous outcomes at 6-month follow-up demonstrated large clinical benets of exercise (relative risk reduction of reoccurrence 45246%, absolute risk reduction of reoccurrence 3642 for every 100 patients; and number needed to treat approximating 3, to prevent one patient suffering a LBP recurrence). The effect of exercise on LBP reoccurrence was variably reported beyond 6 months. Conclusion Exercise programmes are effective for chronic LBP up to 6 months after treatment cessation, evidenced by pain score reduction and reoccurrence rates. The way in which follow-up data are reported assists clinical interpretation of research ndings.

Introduction
A commitment to evidence-based practice requires clinicians to identify, interpret and apply research ndings to individual patient requirements [1]. Thus it is important for researchers to report experimental research outcomes in terms of meaningful effects, for ready interpretation by clinicians [2,3]. Experimental study outcomes are reported in dichotomous and/or equal interval forms at different measurement points. Dichotomous data express outcomes such as disease reoccurrences, while continuous data quantify outcomes, for instance intensity of pain. Experimental research outcomes are most commonly measured at the end of the intervention period. Outcomes can be measured again later (at longer-term follow-up) to determine whether short-term treatment effects are sustained. With interval data such as pain scales, the treatment effect is calculated as the difference between the amount of pain scale change comparing intervention and control groups. This is usually
484

expressed as the weighted mean difference and 95% condence intervals (CIs). Reoccurrences are usually reported in dichotomous form (Occurred/Not) and thus require different comparative statistics. The experimental event rate (EER) measures the percentage of experimental subjects reporting a reoccurrence, while the control event rate (CER) similarly describes the control subjects reporting a reoccurrence. Three comparative measures of effectiveness can then be applied: The absolute risk reduction (ARR) is the difference in reoccurrence rates in the experimental and control groups (subtracting CER from EER). The relative risk reduction (RRR) is the proportional reduction in reoccurrence rates in the experimental and control group subjects (difference between the EER and CER, divided by the CER). The number needed to treat (NNT) expresses the number of patients who need to be treated in order to prevent one reoccurrence, calculated as 1/ARR [4].

2010 Blackwell Publishing Ltd, Journal of Evaluation in Clinical Practice 16 (2010) 484491

C. Smith and K. Grimmer-Somers

Treatment effect of LBP exercise

This paper summarizes recently published literature which reports any outcome data on long-term effectiveness of physiotherapy exercise programmes for chronic low back pain (CLBP). CLBP is dened by its duration, commonly considered as three or more months. CLBP has high lifetime prevalence [5], and high likelihood of negative outcomes including work disability, high medical costs and time lost from work [6,7]. At any one time, approximately 15% of adults report suffering LBP [8]. Approximately one in three adults will experience yearly LBP recurrences [5,8]. Physiotherapy-administered exercise programmes for CLBP are reportedly effective in the short term. There is evidence of enhanced effectiveness if exercise is combined with methods to increase individual compliance [9,10].

outcome data were dichotomous, relative risks (RR), EER, CER, RRR, ARR and/or NNT were calculated at each follow-up period, per study.

Results
Study characteristics
We identied 19 trials which met the inclusion and exclusion criteria, of which 15 were subsequently considered for this review. The remaining trials did not meet the prescribed denition of CLBP [2932]. Table 1 reports characteristics of the included trials, including a summary of intervention and comparison treatments, the methodological quality score, the follow-up periods and outcomes assessed at follow-up. There was considerable variability in the exercise interventions, and the measures of outcome, with no homogenous studies. Overall the trials were of moderate methodological quality [average PEDro score 7.7, SD 1.3 (range 510)]. Common methodological aws were lack of blinding of subjects and therapists, lack of information on intention to treat analysis, and high rates of drop-out and loss to follow-up. Twelve trials reported follow-up at 12 months. Thirteen trials reported follow-up data within the 12-month period (<6 months 53%, 611 months 60%). Reporting on follow-up outcomes beyond 12 months was less common, with only three trials reporting outcomes at 2 years or longer. Nine studies reported interval outcome data using a range of measures [e.g. pain visual analogue scales, SF36 pain domain, Oswestry LBP disability questionnaire] (Table 2). Data could be pooled from six studies, which reported on pain visual analogue scales at 6-month follow-up. There was a signicant overall reduction in pain in the exercise intervention group compared with the control [SMD -0.57, 95%CI -0.75 to -0.39 (555 participants)]. At 12-month follow-up, a small sustained benet from exercise was found between groups [SMD -0.25, 95%CI -0.44 to 0.06 (434 participants)]. Three studies reported on outcomes beyond 12 months; however, there was insufcient evidence of a sustained effect from exercise [SMD -0.03, 95%CI -0.29 to 0.07 (500 subjects)]. The effect of exercise on dichotomous pain-related outcomes for LBP was reported in six trials, which demonstrated an overall benet of exercise at 6 months (RR 1.74, 95%CI 1.33 to 2.27, 296 subjects) (Table 3). These benets were demonstrated by: RRR of reoccurrence between 45% and 246%; ARR of reoccurrence between 36 and 42 for every 100 subjects, and NNT of 3 to avoid a recurrence of LBP. Of the seven studies reporting dichotomous pain outcomes at 12 months, the overall benet from participating in exercise programmes was not sustained (RR 1.18, 95%CI 0.81 to 1.72, 472 subjects) (see Table 3). One of these studies [19] also reported on 24-month follow-up and demonstrated a sustained benet of exercise (RR 10.61, 95%CI 4.27 to 26.57, 56 subjects). The RRR calculated from data in this paper indicated a 265% reduction in pain from exercise, compared with the control. The ARR suggested that for every 100 subjects provided with exercise intervention, 53 poor outcomes would be avoided, while the NNT
485

Methods
Literature search
A comprehensive search was undertaken of peer-reviewed literature published from 2001 to April 2007, using the databases of Academic search elite, Allied and Complementary Medicine, AMI, AUSThealth, BioMed Central Gateway, Blackwell Synergy, CINAHL, Cochrane library, Digital Dissertations, Health source consumer edition, Health source nursing/academic edition, HighWire Press, Ingenta Connect, MEDLINE, NLM Gateway, PEDro, Social services abstracts, PsycArticles, PsycInfo, PubMed and Web of Science. The search strategy used combinations of keywords: (1) CLBP (lasting 3+ months duration); (2) exercise; (3) physiotherapy; and (4) physical therapy. We considered all randomized controlled trials that included adults with CLBP, and any outcome related to CLBP measured at any follow-up period after cessation of the intervention. Studies were excluded if they related to acute LBP; did not use a control group; did not use a parallel study design, and did not report a physiotherapy exercise programme as the primary intervention. We assessed the methodological quality of each included study using the 11-item PEDro scale [11], which is based on the Delphi list developed by Verhagen and colleagues [12]. The PEDro scale assesses the internal and external validity of experimental studies.

Data extraction
Data were extracted from each included study on the nature of exercise intervention, sample size in treatment arms, duration of follow-up periods, and any CLBP-related outcome measure at follow-up. Outcome measures typically comprised interval data such as pain scales or dichotomous data such as reoccurrences of LBP, return to work or measures of function. A purpose-built MS Excel workbook was developed for data extraction.

Statistical analysis
Per-study weighted mean differences and 95%CIs were calculated from available interval data, if this was not already reported. Where there was more than one study with similar data measured at the same follow-up time period, pooled effect sizes were calculated using Cochrane REVMAN software Version 5 [13], and reported as standardized mean differences (SMD). Where the

2010 Blackwell Publishing Ltd

Treatment effect of LBP exercise

C. Smith and K. Grimmer-Somers

Table 1 Characteristics of included trials PEDro score 9 Length of follow-up 12 months Outcome measures Disability function, pain, quality of life

Author Cairns et al. [14]

Inclusion criteria LBP (note current mean duration of episode 7 months)

Intervention Therapy group: conventional physiotherapy plus spinal stabilization exercise (endurance training for deep abdominal and back extensor muscles Control group: Conventional physiotherapy consisting of general active exercise and manual therapy Therapy group: Chiropractic exion distraction procedure. Slow controlled manual traction Comparison group: Active exercise programme, consisting of strength exercises, exibility and cardiovascular exercises. Aim to strengthen the muscles surrounding the spine and increase trunk exibility Therapy group: Orthosis plus kinesitherapy. Standardized protocol of 12 appointments, 3 times a week for 4 weeks. Exercises included diaphragm breathing, proprioceptive trunk exercises with the aim of maintaining neutral zone at level of lumbar lordosis, gluteal and ischiocrural stretching exercises performed in an uploaded way, contraction exercises of the lumbar stabilising muscles Comparison group: Orthosis only consisting of a cloth band with splints of the CAMP brand, prescribed 90 days of continuous use Therapy group: Spinal stabilization surgery Comparison group: Intensive rehabilitation. Physiotherapist led. Daily exercises individually tailored and paced to increase repetitions and duration. Exercise consisted of; stretching of major muscle groups, spinal exibility, muscle strengthening, spinal stabilization and cardio-vascular work. Daily sessions of hydrotherapy were available Therapy: Motivation and exercise. Motivation: counselling and information strategies, reinforcement agreements and oral agreements, maintenance of exercise diary and display of treatment contract at home. Exercise consisted of individual, sub-maximal gradually increased exercise. Aim to improve spinal mobility, trunk and lower limb length, force, endurance and co-ordination Comparison group: Exercise consisted of individual, sub-maximal gradually increased exercise. Aim to improve spinal mobility, trunk and lower limb length, force, endurance and co-ordination Therapy group A: Spinal stabilization (10-week course retraining of the transverse abdominis, pelvic oor and diaphragm muscles) Therapy group B: Manual therapy by physiotherapists. No exercises for transverse abdominis, multius, diaphragm or pelvic oor prescribed Comparison group: Education information on anatomy, biomechanics, lifting, pathologies, education of tness, education and exercise

Cambron et al. [15]

LBP for >3 months

3, 6 and 12 months

Pain, disability

Celestini et al. [16]

Chronic LBP (duration not specied)

3, 6 and 12 months

Remission of pain

Fairbank et al. [17]

Chronic LBP > 12-month duration

6, 12 and 24 months

Pain, disability, quality of life

Friedrich et al. [18]

Chronic LBP of at least 4-month duration

3.5 weeks, 4, 12 and 60 months

Disability, pain, working ability

Golby et al. [19]

Chronic LBP >3 months

12 months

Pain frequency impairment, quality of life

486

2010 Blackwell Publishing Ltd

C. Smith and K. Grimmer-Somers

Treatment effect of LBP exercise

Table 1. Continued PEDro score 7 Length of follow-up 6 months Outcome measures Pain intensity, pain treatment

Author Jousset et al. [20]

Inclusion criteria LBP > 4-month duration

Intervention Therapy group: Functional restorations warm up, strengthening exercises, aerobic activity, occupational therapy, endurance training Control group: active individual therapy consisting of supervision by physiotherapist directing active exercises involving exibility, range of motion, pain coping strategies, strengthening exercises and functional training Therapy group: Multidisciplinary rehabilitation consisted of physical training (general tness training, muscle strengthening exercises performed for all muscle groups of the trunk and lower limbs, exercises to correct mobility of the spine, functional exercises to improve postural control) workplace interventions, back school, relaxation, and stress management techniques Control group: Individual physiotherapy included physical exercises and passive treatment administered over 10 hours. Included massage, spine traction, manual mobilization of spine, ultrasound, light exercise (muscle stretching, spine mobilization, deep trunk muscles exercise), swimming and walking recommended Therapy group: Local stabilizing exercises to the back and abdominal muscle exercises. Progressive increase in holding time and number of contractions. Control group: General exercise only. Exercise undertaken to activate the extensor (paraspinalis) and exor (abdominals) Both groups undertook exercises twice a week for 8 weeks, with a class duration 4560 minutes Therapy group: One to one treatment involving 30 minutes of manual therapy (mobilizations to the spine), and spine stabilization exercises Comparison group: 10 station exercise class involving aerobic exercise, spinal stabilization exercises and manual therapy Therapy group: Physiotherapy group: (30 minutes) individual physiotherapy focussed on improving functional capacity using strengthening, co-ordination and aerobic exercise and with instruction on ergonomic principles and home exercises Comparison group: Muscle reconditioning using devices or training machines. Progressive, iso-inertial loading to the trunk in the three cardinal planes Comparison group: Aerobics (1 hour, low impact) Therapy: Manipulative and stabilizing exercises 4 times a week over 4 weeks. Exercises focussed on correcting lumbo-pelvic rhythm Control group: Physician consultation plus a 25-page educational booklet on anatomy, physiology, basic exercises and how to cope with pain. Instruction on posture and 34 exercises to increase spinal mobility, muscle stretch, trunk muscle stability

Kaapa et al. [21]

LBP during previous 12 months

6, 12 and 24 months

Pain intensity, general wellbeing, working capacity

Koumantakis et al. [22]

Recurring LBP, repeat episodes of not <6-month duration

2 and 5 months

Pain perception, disability, cognitive status

Lewis et al. [23]

LBP for >3-month duration

6 and 12 months

Function, disability, pain

Mannion et al. [24]

Chronic LBP of >3-month duration

3, 6 and 12 months

Pain intensity, duration and frequency, disability, psychological disturbance

Niemisto et al. [25]

Chronic LBP

5 and 12 months

Pain intensity, disability

2010 Blackwell Publishing Ltd

487

Treatment effect of LBP exercise

C. Smith and K. Grimmer-Somers

Table 1. Continued PEDro score 7 Length of follow-up 3 and 12 months Outcome measures Pain, functional disability

Author Rasmussen-Barr et al. [26]

Inclusion criteria Chronic LBP

Intervention Therapy: Stabilizing training (controlled exercises stabilizing muscles). Six-week programme, once a week for 45 minutes. Instructions given on how to activate and control their deep abdominal and lumbar multidus muscles. Gradual loading was applied. 15 minutes training programme was undertaken at home, with a diary to monitor compliance Control group: Manual treatment. Six-week programme, once a week for 45 minutes. Manual techniques used including muscle stretching, segmental traction, soft tissue mobilization Therapy group: Whole body vibration exercise. During the exercise unit subjects perform slow movements of the hips and waist, with bending in the sagittal and frontal planes and rotation in the horizontal plane Comparison group: Isodynamic lumbar extension exercise, including resistance training of the abdominal and thigh muscles Saline plus exercise, saline plus normal activity, prolotherapy plus exercise, prolotherapy plus normal activity Exercise consisted of two sagittal loading exercises to be performed in standing-alternating exion and the extension of the hips to midrange with the spine held straight, and exion of the lumbar spine with the hips stationary. Ten repetitions performed four times daily

Rittweger et al. [27]

Chronic LBP for >6 months

6 months

Pain, disability

Yelland et al. [28]

10

Non-specic LBP over last 6 months

4, 6 and 12 months

Pain intensity, disability

LBP, low back pain.

indicated that two patients needed to be treated with exercise to avoid one recurrence of back pain.

Discussion
This review took a novel approach by synthesizing interval and dichotomous follow-up data from randomized controlled studies, to report the long-term effectiveness of physiotherapy exercise programmes. Therapists, patients and referrers are potentially just as interested in the long-term effectiveness of exercise treatment for LBP, as in short-term effects, particularly in terms of preventing LBP reoccurrences. Our ndings suggest that despite their heterogeneity, physiotherapy exercise programmes are effective for many CLBP sufferers in reducing pain scores, and preventing LBP reoccurrences (see Table 4). The most consistent evidence of effectiveness of exercise programmes was at 6-month follow-up. Our ndings of the signicance of follow-up pain score differences from physiotherapy exercise programmes are supported by a Cochrane review [10]. We have not identied any secondary evidence source which supports our summary of the long-term effectiveness of exercise therapy in preventing LBP reoccurrences. We found that few studies reported dichotomous follow-up data with the same attention afforded to the scaled data. Thus the
488

consistently favourable effects of physiotherapy exercise programmes to prevent recurrences of CLBP up to 6 months after treatment ceases were under-reported. While between-group differences in pain scores at 6 months were non-signicant in 50% of studies with relevant time-point data, differences in events of reoccurrence favoured the exercise group in 75% relevant studies. These differences may be related to the constructs of measures of pain chosen as study outcomes, and the appropriateness of their application to summary statistics. Dichotomous study outcomes rely on patients recall of memorable events such as LBP reoccurrence. The use of pain scales, however, relies on patients to apply a single numeric scale to assess his/her own experiences to the complexity of symptoms related to LBP. Using a pain scale for before-and-after research outcomes potentially introduces issues such as the validity of reducing the pain experience to one numeric scale, the appropriateness of applying equal interval statistics to what is essentially an individual subjective experience (albeit scored 010), recall bias in assigning repeated scores, the potential for regression to the mean, and the validity of the assumption of random errors in differences between study participants in their repeated use of the scale [33]. These issues may be less relevant when reporting LBP occurrences, which are individual to the sufferer. The ndings provided by the dichotomous data may give clinicians, patients and funders important insights into the effectiveness

2010 Blackwell Publishing Ltd

C. Smith and K. Grimmer-Somers

Treatment effect of LBP exercise

Table 2 Trials evaluating the effect of exercise using pain scales Change in outcomes between groups Favours treatment No difference No difference Favours motivation and exercise Favours treatment No difference No difference No difference

Author Cambron et al. [15] Cairns et al. [14] Fairbank et al. [17] Friedrich et al. [18]

Outcome measures Pain (VAS) Pain (short form Magill questionnaire VAS) Pain (Oswestry low back pain disability) Pain intensity (rating scale)

Treatment effect 6 months MD -2.09, 95%CI -2.47 to -1.71 12 months MD -0.51, 95%CI -0.82 to -0.21 12 months MD -0.9 (-1.2 to 0.5) 24 months MD 4.1 95%CI -1.67 to 10.0 After intervention: MD -1.87, 95%CI -2.42 to -1.32 12 months: MD -3.43, 95%CI -0.97 to -2.67 5 years: MD -6.07, 95%CI -7.35 to -4.79 6 months MD -0.04, 95%CI -0.41 to 0.33 12 months MD 0.08, 95%CI -0.30 to 0.46 24 months MD -0.46, 95%CI -0.87 to -0.05 2 months MD 0.8, 95%CI -2.15 to 3.75 5 months MD -2.5, 95%CI -5.86 to 0.86 Exercise versus individual treatment Lumbar exion 6 months MD 1.22, 95%CI 0.03 to 2.41 12 months MD 1.00 95%CI -0.16 to 2.16 Lumbar extension 6 months MD 1.5, 95%CI 0.17 to 2.83 12 months MD 1.1, 95%CI -0.28 to 2.48 Left side exion 6 months MD 1.2, 95%CI 0.01 to 2.39 12 months MD 0.9, 95%CI -0.29 to 2.09 Right side exion 6 months MD 1.6, 95%CI 0.34 to 2.86 12 months MD 1.0, 95%CI -0.06 to 2.26 Left SLR 6 months MD 1.6, 95%CI 0.44 to 2.76 12 months MD 0.6, 95%CI -0.66 to 1.86 Right SLR 6 months MD 1.9, 95%CI 0.56 to 3.24 12 months MD 0.7 95%CI -0.63 to 2.03 Exercise versus physiotherapy 6 months MD -0.09, 95%CI -0.51 to 0.33 12 months MD -0.05, 95%CI -0.47 to 0.37 Exercise versus devices 6 months MD 0.09, 95%CI -0.34 to 0.52 12 months MD 0.14, 95%CI -0.30 to 0.54 Whole body vibration exercise versus : isodynamic lumbar extension exercise 3 months MD -0.20, 95%CI -1.20 to 0.80

Kaapa et al. [21]

Low back pain intensity

Koumantakis et al. [22] Lewis et al. [23]

Short Form McGill Pain Questionnaire Pain VAS Pain (Qubec back pain VAS)

Favours treatment No difference Favours treatment No difference Favours treatment No difference Favours treatment No difference Favours treatment No difference Favours treatment Mannion et al. [24] Pain (VAS) No difference

Rittweger et al. [27]

Pain (VAS), pain disability index

No difference

VAS, visual analogue scale; MD, mean difference; CI, condence interval.

of the exercise programme, which are not provided by differences in scaled data. Comparing Tables 2 and 3 suggests that the clinical utility of the data in Table 3 is more readily interpretable than the SMD and effect sizes reported in Table 2, when assisting clinicians to implement research ndings into their practice. In particular the number needed to treat provides a powerful estimate of the effect of an exercise programme for individual patients. Knowing that therapists need to treat three patients to prevent one reoccurrence of LBP is persuasive information for patients, health care providers and funders. This information also provides a standard platform for communication between stakeholders.

Many of the trials included in our review demonstrated similar methodological weaknesses, which similarly reduced their critical appraisal scores. The heterogeneity of the exercise interventions makes it difcult to identify the most effective components of exercise interventions. However, given the similarity of effect sizes across studies during the follow-up periods (using either the scaled or dichotomous data), it could be argued that the nature of the exercise programme is perhaps not as important as the message it provides, that is, to keep active. As with many chronic health complaints, the symptoms of CLBP are complex and perhaps not readily estimated by one
489

2010 Blackwell Publishing Ltd

Treatment effect of LBP exercise

C. Smith and K. Grimmer-Somers

Table 3 Trials evaluating the effect of exercise: measures of risk and effect sizes Follow-up time period (months) 6 Control event rate (placebo) (%) 30.4 55 15 35.7 Experimental event rate (%) 66.6 13 52 39 Relative risk reduction (%) 119 45.5 246 9.2 Number needed to treat 2.7 2.3 2.7 33

Author Rasmussen-Barr et al. [26] Celestini et al. [16] Jousset et al. [20] Golby et al. [19]

Relative risk 2.19, 95%CI 1.11 to 4.32 3.81, 95%CI 1.26 to 11.53 2.04, 95%CI 1.28 to 3.26 1.16, 95%CI 0.78 to 1.73 Combined effect 1.74, 95%CI 1.33 to 2.27 0.87, 95%CI 0.41 to 1.85 1.09, 95%CI 0.62 to 1.91 3.2, 95%CI 0.96 to 10.66 1.31, 95% 0.45 to 3.81 1.2, 95%CI 0.30 to 4.80 Combined effect 1.18, 95%CI 0.81 to 1.72 10.61, 95%CI 4.27 to 26.57

Absolute risk reduction (%) 36.2 42 37.0 3.3

Golby et al. [19] Niemisto et al. [25] Rasmussen-Barr et al. [26] Yelland et al. [28] Celestini et al. [16]

12

46.4 58 30.4 39 25

43.7 56.8 58.3 41 25

5.8 2.0 91.7 5.1 0

2.7 1.2 27.9 2.0 0

37 83 3.5 50 0

Golby et al. [19] CI, condence interval.

24

20

73

265

53%

1.8

Table 4 Summary of long-term effectiveness of measures of outcome 6 months Author Cairns et al. [14] Cambron et al. [15] Celestini et al. [16] Fairbank et al. [17] Friedrich et al. [18] Golby et al. [19] Jousset et al. [20] Kaapa et al. [21] KoumantaKis et al. [22] Lewis et al. [23] Mannion et al. [24] Niemisto et al. [25] Rasmussen-Barr et al. [26] Rittweger et al. [27] Yelland et al. [28] Continuous Dichotomous 612 months Continuous Dichotomous >12 months Continuous Dichotomous

outcome measure. CLBP outcomes were variably expressed in the included studies as pain scores, episodes of pain, impaired function, days off work, return to work, reoccurrences and pain experiences. This review highlights the need for the use of standard denitions of a good outcome in short- and long-term studies, as well as the need to express change in multiple outcome measures in order to comprehensively express patients response to treatment.

standard exercise programmes, as well as report appropriately interpreted interval and dichotomous measures of longer-term outcome, so that the true nature of the long-term effectiveness of physiotherapy exercise interventions can be explored.

References
1. Sackett, D. L., Rosenberg, W. M. C., Muir Gray, J. A., Haynes, R. B. & Richardson, W. S. (1996) Editorial: evidence based medicine: what it is and what it isnt. British Medical Journal, 312, 7172. 2. Altman, D. G., Schulz, K. F., Moher, D., Egger, M., Davidoff, F., Elbourne, D., Gotzsche, P. C. & Lang, T.; CONSORT GROUP (Consolidated Standards of Reporting Trials) (2001) The revised CONSORT statement for reporting randomized trials: explanation and elaboration. Annals of Internal Medicine, 134 (8), 663694. 3. Shakespeare, T. P., Gebski, V. J., Veness, M. J. & Simes, J. (2001) Improving interpretation of clinical studies by use of condence

Conclusion
Our review highlights that when measures of clinical utility are calculated, physiotherapy exercise programmes are consistently effective in reducing reoccurrences of LBP at least up until 6 months post intervention. To improve the clinical utility of research ndings, future primary trials should aim to apply
490

2010 Blackwell Publishing Ltd

C. Smith and K. Grimmer-Somers

Treatment effect of LBP exercise

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

levels, clinical signicance curves, and risk-benet contours. Lancet, 357 (9265), 13491353. Cook, R. J. & Sackett, D. L. (1995) The number needed to treat: a clinically useful measure of treatment effect. British Medical Journal, 310, 452454. Pengel, L. H. M., Herbert, R. D., Maher, C. G. & Refshauge, K. M. (2003) Acute low back pain: systematic review of its prognosis. British Medical Journal, 327, 323. Wasiak, R., Kim, J. & Pransky, G. (2006) Work disability and costs caused by recurrence of low back pain: longer and more costly than in rst episodes. Spine, 31 (2), 219225. Katz, J. N. (2006) Lumbar disc disorders and low-back pain: socioeconomic factors and consequences. The Journal of Bone and Joint Surgery, 88A (Suppl. 2) 2124. Loney, P. L. & Stratford, P. W. (1999) The prevalence of low back pain in adults: a methodological review of the literature. Physical Therapy, 79, 384396. Bekkering, G. E., Engers, A. J., Wensing, M., Hendriks, H. J. M., van Tulder, M. W., Oostendorp, R. A. B. & Bouter, L. M. (2003) Development of an implementation strategy for physiotherapy guidelines on low back pain. Australian Journal of Physiotherapy, 49 (3), 208214. Hayden, J. A., van Tulder, M. W., Malmivaara, A. & Koes, B. W. (2005) Exercise therapy for treatment of non-specic low back pain. Exercise therapy for treatment of non-specic low back pain. Cochrane Database Systematic Review, 20 (3), CD000335. Sherrington, C., Herbert, R. D., Maher, C. G. & Moseley, A. M. (2000) PEDro. A database of randomized trials and systematic reviews in physiotherapy. Manual Therapy, 5 (4), 223226. Verhagen, A. P., de Vet, H. C., de Bie, R. A., Kessels, A. G., Boers, M., Bouter, L. M. & Knipschild, P. G. (1998) The Delphi list: a criteria list for quality assessment of randomised controlled trials for conducting systematic reviews developed by Delphi consensus. Journal of Clinical Epidemiology, 51 (12), 12351241. The Cochrane Collaboration (2008) Review Manager (RevMan) [Computer program]. Version 5. Copenhagen: The Nordic Cochrane Centre. Available at: http://www.cc-ims.net/RevMan (last accessed July 2008). Cairns, M. C., Foster, N. E. & Wright, C. (2006) Randomized controlled trial of specic spinal stabilization exercises and conventional physiotherapy for recurrent low back pain. Spine, 31 (19), 1E67081. Cambron, J. A., Gudavalli, M. R., Hedeker, D., McGregor, M., Jedlicka, J., Keenum, M., Ghanayem, A. J., Patwardhan, A. G. & Furner, S. E. (2006) One-year follow-up of a randomized clinical trial comparing exion distraction with an exercise program for chronic low-back pain. Journal of Alternative and Complementary Medicine, 12 (7), 659668. Celestini, M., Marchese, A., Serenelli, A. & Graziani, G. (2005) A randomised controlled trial on the efcacy of physical exercise in patients braced for instability of the lumbar spine. Europa Medicophysica, 41, 223231. Fairbank, J., Frost, H., Wilson-MacDonald, J., Yu, L. M., Barker, K. & Collins, R.; Spine Stabilisation Trial Group (2005) Randomised controlled trial to compare surgical stabilisation of the lumbar spine with an intensive rehabilitation programme for patients with chronic low back pain: the MRC spine stabilisation trial. British Medical Journal, 330 (7502), 12331239. Friedrich, M., Gittler, G., Arendasy, M. & Friedrich, K. M. (2005)

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

Long term effect of a combined exercise and motivational program on the level of disability and patients with chronic low back pain. Spine, 30, 9951000. Golby, L. J., Moore, A. P., Doust, J. & Trew, M. E. (2006) A randomized controlled trial investigating the efciency of musculoskeletal physiotherapy on chronic low back disorder. Spine, 31, 10831093. Jousset, N., Fanello, S., Bontoux, L., Dubus, V., Billabert, C., Vielle, B., Roquelaure, Y., Penneau-Fontbonne, D. & Richard, I. (2004) Effects of functional restoration versus 3 hours per week physical therapy: a randomized controlled study. Spine, 29 (5), 487493. Kaapa, E. H., Frantsi, K., Sarna, S. & Malmivaara, A. (2006) Multidisciplinary group rehabilitation versus individual physiotherapy for chronic non specic low back pain. Spine, 31, 371376. Koumantakis, G. A., Watson, P. J. & Oldham, J. A. (2005) Trunk muscle stabilization training plus general exercise versus general exercise only: randomized controlled trial of patients with recurrent low back pain. Physical Therapy, 85 (3), 209225. Lewis, J. S., Hewitt, J. S., Billington, L., Cole, S., Byng, J. & Karayiannis, S. (2005) A randomized clinical trial comparing two physiotherapy interventions for chronic low back pain. Spine, 30 (7), 711721. Mannion, A. F., Muntener, M., Taimela, S. & Dvorak, J. (2001) Comparison of three active therapies for chronic low back pain: results of a randomized clinical trial with one-year follow-up. Rheumatology (Oxford), 40 (7), 772778. Niemisto, L., Lahtinen-Suopanki, T., Rissanen, P., Lindgren, K. A., Sarna, S. & Hurri, H. (2003) A randomized trial of combined manipulation, stabilizing exercises, and physician consultation compared to physician consultation alone for chronic low back pain. Spine, 28 (19), 21852191. Rasmussen-Barr, E., Nilsson-Wikmar, L. & Arviddson, I. (2003) Stabilizing training compared with manual treatment in sub-acute and chronic low-back pain. Manual Therapy, 8, 233241. Rittweger, J., Just, K., Kautzsch, K., Reeg, P. & Felsenberg, D. (2002) Treatment of chronic lower back pain with lumbar extension and whole-body vibration exercise: a randomized controlled trial. Spine, 27 (17), 18291834. Yelland, M. J., Glasziuo, P. P., Bogduk, N., Schluter, P. J., McKernon, M. & Loeser, J. D. (2004) Prolotherapy injections, saline injections, and exercises for chronic low-back pain: a randomized trial. Spine, 29 (1), 91. Moseley, L. (2002) Combined physiotherapy and education is efcacious for chronic low back pain. Australian Journal of Physiotherapy, 48, 297302. Petersen, T., Kryger, P., Ekdahl, C., Olsen, S. & Jacobsen, S. (2002) The effect of McKenzie therapy as compared with that of intensive strengthening training for the treatment of patients with subacute or chronic low back pain: a randomized controlled trial. Spine, 27 (16), 17021709. Aure, O. F., Nilsen, J. H. & Vasseljen, O. (2003) Manual therapy and exercise therapy in patients with chronic low back pain: a randomized, controlled trial with 1-year follow-up. Spine, 28, 525532 Frost, H., Lamb, S. E. & Doll, H. A. (2004) Randomised controlled trial of physiotherapy compared with advice for low back pain. British Medical Journal, 329 (7468), 708713. Sackett, D. L. (1979) Bias in analytic research. Journal of Chronic Diseases, 32 (12), 5163.

2010 Blackwell Publishing Ltd

491

Вам также может понравиться