|Cyclobenzaprine and back pain: a meta-analysis
|Browning R, Jackson J L, O'Malley P G
To perform a systematic review of the effectiveness of cyclobenzaprine in the treatment of back pain.
MEDLINE (from January 1966 to December 1999), PsycLIT (from November 1887 to December 1999), CINAHL (from 1982 to 1999), EMBASE (from January 1974 to December 1999), AIDSLINE, HealthSTAR, Cancerlit and MICROMEDEX were searched. The searches were conducted using the textwords and keywords (all languages) 'cyclobenzaprine', and the MeSH terms 'dibenzocycloheptenes' and 'propylamines' combined with the following textwords: 'back pain', 'backache', 'low back pain', 'lumbago', 'acute muscle spasm' or 'muscle spasm'. Other databases searched were the Cochrane Controlled Trials Register for randomised trials, the Cochrane Database of Systematic Reviews, and Federal Research in Progress for unpublished literature. The reference lists of the reviewed articles were also examined and the drug manufacturer (Merck, Sharpe and Dohme) was contacted.
Study designs of evaluations included in the review
Randomised placebo-controlled trials were included.
Specific interventions included in the review
Cyclobenzaprine (10 to 60 mg dose) compared with placebo. The duration of the study was 7 to 18 days.
Participants included in the review
Adults with acute and chronic neck and/or back pain were included. Studies evaluating cyclobenzaprine in the treatment of muscle spasticity originating from the central nervous system were excluded. It was reported that one study included patients with an X-ray file-confirmed diagnosis of osteoarthritis of the cervical or lumbar spine. Other included studies did not report the participants' characteristics.
Outcomes assessed in the review
The outcomes included global improvement, five specific domains of back pain (local pain, muscle spasm, range of motion, tenderness to palpation, and activities of daily living) and adverse effects.
How were decisions on the relevance of primary studies made?
The authors do not state how the papers were selected for the review, or how many of the reviewers performed the selection.
Assessment of study quality
The validity of the studies was assessed using the method of Jadad et al. (see Other Publications of Related Interest no.1). The criteria were: description of randomisation; adequacy of blinding; description of the withdrawals and drop-outs; the appropriateness of the statistical analysis; description of the inclusion and exclusion criteria; and the method for assessing adverse treatment effects. The study quality was assessed independently in duplicate. Any disagreements were resolved by consensus.
The authors do not state how the data were extracted for the review, or how many of the reviewers performed the data extraction.
The abstracted data included the following: setting; country of origin; treatment characteristics (dose, duration and follow-up); demographics; the number of participants enrolled; losses to follow-up; adverse effects; and outcomes. For continuous data, the standardised mean difference was calculated for each study. For dichotomous data, the odds ratio (OR) was calculated for each study.
Methods of synthesis
How were the studies combined?
Summary ORs, risk differences and standardised mean differences were calculated using random-effects models. Publication bias was assessed using the method of Egger et al. (see Other Publications of Related Interest no.2). The sensitivity of the results to the existence of unpublished studies was assessed using a 'file drawer' test (see Other Publications of Related Interest no.3).
How were differences between studies investigated?
Heterogeneity was assessed visually using Galbraith plots, and chi-squared statistics were calculated. Several sources of heterogeneity were explored using meta-regression. These included the year of publication, type of syndrome, (acute versus chronic back pain), study quality scores, and the duration and dose of therapy.
Results of the review
Fourteen randomised controlled trials (n=3,024) were included.
The studies were of moderate quality with a mean quality score of 4.3. Kappa for inter-rater agreement was 0.79.
Global improvement (10 studies) favoured cyclobenzaprine: the OR was 4.7 (95% confidence interval, CI: 2.7, 8.1). The pooled risk difference was 0.37 (95% CI: 0.24, 0.50), which corresponds to 2.7 (95% CI: 2.0, 4.2) individuals needing treatment for one to experience symptom improvement. There was evidence of heterogeneity (chi-squared 25.90, p=0.002) but no evidence of publication bias (Egger p=0.37). The magnitude of the improvement was modest, with an effect size of 0.38 to 0.58 in the 5 outcomes assessed, i.e. local pain, muscle spasm, tenderness to palpation, range of motion, and activities of daily living. There was statistical evidence for publication bias among 4 of the continuous outcomes in the first few days. The file drawer test found that between 201 and 786 studies finding no benefit in the 5 specific back pain measures would be needed to negate the results.
Patients treated with cyclobenzaprine were significantly more likely to experience adverse effects: 53% of the patients treated with cyclobenzaprine experienced at least one adverse effect, compared with 28% receiving placebo. The most common adverse effects were drowsiness, dry mouth and dizziness. The sensitivity analysis found that the following had no effect on any of the findings: study quality; year of publication; the duration of treatment; the type of back pain; and whether the drug manufacturer had sponsored the study. Dropping individual studies from the meta-analysis also had no effect.
Cyclobenzaprine was more effective than placebo in the management of back pain. However, the effect is modest and comes at the price of greater adverse effects. The effect was greatest in the first 4 days of treatment, suggesting that shorter courses may be better.
This was a reasonably well-conducted systematic review. The research question was clearly stated, although the types of participants were unclear and the patient demographics were not reported. This information would have been helpful as patient variables such as age, gender, type of back pain and co-morbid conditions, may affect the outcomes. The literature search was comprehensive and efforts were made to identify unpublished research. However, the restriction to English language publications could result in some studies being missed.
Some details of the review process, such as the number of reviewers selecting the studies, were not reported. A validity assessment was carried out, reported and used in the investigation of heterogeneity. Study details were not fully reported in terms of the outcomes. It seemed appropriate to pool the data, although unexplained heterogeneity in the global improvement outcome means that this result should be treated with caution. For the continuous outcomes, the rationale and process for data extraction at three different time points was unclear. The adverse events comparisons were rudimentary.
The authors' conclusion, that the effect of cyclobenzaprine compared with placebo is modest and comes at the price of greater adverse effects, does follow from the results presented. The conclusion that shorter courses of cyclobenzaprine may be better does not seem to be based on strong evidence.
Implications of the review for practice and research
Practice: The authors suggest that shorter courses of treatment may be better. However, this may not be supported by the evidence presented.
Research: The authors state that studies comparing the relative value of acetaminophen, non-steroidal anti-inflammatory drugs and cyclobenzaprine, both individually and in combination, in the treatment of back pain are needed.
Browning R, Jackson J L, O'Malley P G. Cyclobenzaprine and back pain: a meta-analysis. Archives of Internal Medicine 2001; 161(13): 1613-1620
Other publications of related interest
1. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17:1-12. 2. Egger M, Smith GD, Schneider M, Minder C. BMJ 1997;315:209-16. 3. Rosenthal R. The 'file drawer problem' and tolerance for null results. Psychol Bull 1979;86:638-41.
Subject indexing assigned by NLM
Amitriptyline /adverse effects /analogs & Back Pain /drug therapy; Humans; Muscle Relaxants, Central /adverse effects /therapeutic use; Randomized Controlled Trials as Topic; derivatives /therapeutic use
Date bibliographic record published
Date abstract record published
This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.