|
Glucosamine and chondroitin for treatment of osteoarthritis: a systematic quality assessment and meta-analysis |
McAlindon T E, LaValley M P, Gulin J P, Felson D T. |
|
|
Authors' objectives To evaluate the benefit of glucosamine and chondroitin preparations for osteoarthritis (OA) symptoms using meta-analysis combined with systematic quality assessment of clinical trials of these preparations in knee and/or hip OA.
Searching MEDLINE (1966 to June 1999) and the Cochrane Controlled Trials Register were searched using the following as both MeSH terms and textwords: 'osteoarthritis', 'osteoarthrosis', 'degenerative arthritis', 'glucosamine', 'chondroitin' and 'glycosaminoglycans'. Additional studies were located through searching the citation lists of review articles and published manuscripts; and through handsearching the following supplement issues for abstracts: Arthritis and Rheumatism, the British Journal of Rheumatology, and Osteoarthritis and Cartilage published between 1978 and 1998. Unpublished data were sought through contacting content experts, study authors and the manufacturers of glucosamine or chondroitin. There were no language restrictions.
Study selection Study designs of evaluations included in the reviewDouble-blind randomised controlled trials (RCTs).
Specific interventions included in the reviewOral or parenteral glucosamine sulfate, glucosamine hydrochloride, or chondroitin sulfate against placebo (dose regimens not stated).
Participants included in the reviewIndividuals with hip or knee OA.
Outcomes assessed in the reviewThe hierarchy of outcome measures used in the analysis was based on those currently recommended for OA clinical trials (see Other Publications of Related Interest no.1): global pain score for index joint (visual analog or Likert scale), pain on walking for index joint (visual analog or Likert scale), Western Ontario and McMaster Universities (WOMAC) Osteoarthritis Index Pain Subscale (visual analog or Likert scale), Lequesne index, pain in index joint during activities other than walking (visual analog or Likert scale).
How were decisions on the relevance of primary studies made?Two rheumatologists independently reviewed all of the studies.
Assessment of study quality Fourteen criteria as developed/tested in studies by Chalmers et al and Rochon et al (see Other Publications of Related Interest no.2 and no.3) were used. The criteria included: control appearance, allocation concealment, patient blinding, observer blinding to treatment, observer blinding to results, prior estimate of numbers, compliance testing, inclusion of pre treatment variables in analysis, presentation of statistical end points, statistical evaluation of type II error, presentation of confidence limits around between group differences in statistical end points, quality of statistical analyses, withdrawals, and side effects discussion. In addition it was noted if the trial included an intention to treat analysis and if an industrial sponsor was used (i.e. drug manufacturer). Two rheumatologists independently assessed the quality of the studies according to the fourteen criteria. Disagreements were resolved through discussion and the mean of the postadjudication scores of the two reviewers was used in the analyses. Potential scores ranged from 0 to 68 for negative and from 0 to 65 for positive studies. These scores were expressed as a percentage of the maximum possible score for each trial.
Data extraction Two rheumatologists independently extracted data from the studies using a standardised form. Where necessary means and measures of dispersion were approximated from figures in the manuscripts. Authors were contacted for further information where required. Tables in the report include information about the number of participants, type of study, mode of drug administration, joint studied, study sponsor, quality score, allocation concealment, intention to treat analysis, details of outcome measures and effect size. Effect sizes were calculated using intention to treat data where possible and based on the difference in mean outcome value between the treatment and placebo arms at the end of the trial, divided by the standard deviation of the outcome value in the control group. To reduce bias, all effect sizes were multiplied by a correction factor that depended on sample size (see Other Publications of Related Interest no.4).
Methods of synthesis How were the studies combined?Effect sizes were pooled where appropriate using a random-effects model. A scale for effect sizes was used with 0.8 reflecting a large effect, 0.5 a moderate effect and 0.2 a small effect (see Other Publications of Related Interest no. 5). Funnel plots and regression analyses according to the methods of Egger et al. (see Other Publications of Related Interest no.6) were used to assess the possibility of publication bias.
How were differences between studies investigated?A statistical test for heterogeneity was performed and sensitivity analyses carried out to assess the effect of: the choice of primary outcome (i.e. study authors own primary outcome versus the highest level of outcome measure from the reviewers hierarchy of outcomes), differences in trial quality, 1-month results versus final results, excluding trials for which data imputations were made, and using only pain outcomes or algofunctional outcomes.
Results of the review Fifteen double-blinded RCTs involving a total of 1710 participants (911 glucosamine and 799 chondroitin) were included in the review.
Effect sizes (n=15 trials):
Moderate treatment effect sizes were observed for glucosamine (0.44; 95% CI: 0.24, 0.64) and large effect sizes for chondroitin (0.96; 95% CI: 0.63, 1.3). However the heterogeneity test for chondroitin was significant (P<0.001). This was mainly due to the inclusion of one outlying trial which when removed from the analysis diminished both the heterogeneity and the effect size (0.78; 95% CI: 0.60, 0.95; P=0.5).
Sensitivity analyses:
Using outcome measures from the pre-defined hierarchy (n=15 trials) (glucosamine 0.49; 95% CI: 0.24, 0.74; chondroitin 0.88; 95% CI: 0.67, 1.1).
Using 1-month outcomes (n=9 trials) (glucosamine 0.26; 95% CI: 0.10, 0.42; chondroitin 0.40; 95% CI: 0.17, 0.62).
Adjusting for trial quality (n=?) (glucosamine 0.46; chondroitin 1.0) Excluding four studies with imputations (n=11) (glucosamine 0.35; chondroitin 0.87).
Pain outcomes only (n=8) (glucosamine 0.51; 95% CI: 0.05, 0.96; chondroitin 0.86; 95% CI: 0.64, 1.09).
Trials reporting Lequesne index only (n=5)(glucosamine 0.41; 95% CI: 0.14, 0.69; chondroitin 0.63; 95% CI: 0.32, 0.94).
Low (i.e. below median 0.7) quality scores vs high quality scores for glucosamine studies (low quality 0.7; 95% CI: 0.4, 1.0); high quality 0.3; 95% CI: 0.1, 0.5).
Low (i.e. below median 0.7) quality scores vs high quality scores for chondroitin studies (low quality 1.7; 95% CI: 0.7, 2.7); high quality 0.8; 95% CI: 0.6, 1.0).
Small trials vs large trials for glucosamine (small trials 0.5; 95% CI: 0.1, 0.9; large trials 0.4; 95% CI: 0.1, 0.7). Small trials vs large trials for chondroitin (small trials 1.7; 95% CI: 0.5, 2.8; large trials 0.8; 95% CI: 0.6, 1.0).
Quality of studies:
The quality scores ranged from 12.3% to 55.4% with a mean (standard deviation SD) of 35.5% (12%). The level of agreement between the two reviewers assessing the studies was good according to intraclass correlation coefficients of 0.75 prior to and 0.92 after adjudication (P<0.01). Only one study provided sufficient information to determine that allocation concealment had been adequate and only two studies reported an intention to treat analysis. Seven studies did not present drop out rates and the remainder reported a mean (SD) rate of 1.2% (4.2%) per month. None of the studies reported independent funding from governmental or non-for-profit organisations, with six studies receiving direct financial support from a manufacturer and seven further studies including an investigator from a company as an author.
Publication bias:
The funnel plots showed significant asymmetry, reflecting a relative absence of trials with both small numbers and small or null treatment effects. In addition the regression analyses also showed strong evidence of publication bias (glucosamine, intercept estimate, 1.3, P=0.01; chondroitin, 3.8, P=0.002).
Authors' conclusions Trials of glucosamine and chondroitin preparations for OA symptoms demonstrate moderate to large effects, but quality issues and likely publication bias suggest that these effects are exaggerated. Nevertheless, some degree of efficacy appears probable for these preparations.
CRD commentary This is a very clearly presented review with explicit methods and inclusion criteria. The literature search was reasonable and made attempts to locate both published and unpublished data. In addition there were no restrictions on language. However, funnel plots and regression analyses did suggest the presence of publication bias despite these attempts to locate unpublished work.
At each stage of the review process (i.e. selection of studies, quality assessment of studies and extraction of data) two independent reviewers were used. Only double-blind RCTs were included and these were further assessed on an individual basis to determine their validity using a comprehensive fourteen point scale. Details of the included studies are clearly reported in tables. The studies were then appropriately pooled and their heterogeneity assessed statistically. To further investigate the robustness of the studies a series of sensitivity analyses were performed taking into account a wide range of possible variables. The authors also discuss in some details the possible limitations of their review. Overall, this would appear to be a thorough and valid analysis and the authors' findings/implications are supported by the data presented.
Implications of the review for practice and research Practice: The authors do not state any implications for practice.
Research: The authors state 'we recommend further high quality, independent studies to determine the actual efficacy and utility of these preparations'.
Funding National Institutes of Health, grant number AR20613.
Bibliographic details McAlindon T E, LaValley M P, Gulin J P, Felson D T.. Glucosamine and chondroitin for treatment of osteoarthritis: a systematic quality assessment and meta-analysis. JAMA 2000; 283(11): 1469-1475 Other publications of related interest 1. Altman R, Brandt KD, Hochberg MC, Moskowitz R. Design and conduct of clinical trials in patients with osteoarthritis: recommendations from a task force of the Osteoarthritis Research Society. Osteoarthritis Cartilage 1996;4:217-43. 2. Chalmers TC, Smith H, Blackburn B, Silverman B, Schroeder B, Reitman D, et al. A method for assessing the quality of a randomized control trial. Control Clin Trials 1981;2:31-49. 3. Rochon PA, Gurwitx JH, Cheung CM, Hayes JA, Chalmers TC. Evaluating the quality of articles published in journal supplements compared with the quality of those published in the parent journal. JAMA 1994;272:108-13. 4. Hedges LV, Olkin I. Statistical methods for meta-analysis. San Diego (CA): Academic Press; 1985. 5. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale (NJ): Lawrence Erlbaum Assoc.; 1988. 6. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629-34.
These additional published commentaries may also be of interest. Reginster J-Y. Review: glucosamine and chondroitin improve outcomes in osteoarthritis, but the magnitude of effect is unclear. Evid Based Med 2000;5:146. Kee CC. Review: both glucosamine and chondroitin are effective for osteoarthritis, but the magnitude of effect is unclear. Evid Based Nurs 2000;3:124. Denham AC, Newton WP. Are glucosamine and chondroitin effective in treating osteoarthritis? J Fam Pract 2000;49:571-2. Long L. Probable benefits of glucosamine and chondroitin preparations for patients with hip and/or knee osteoarthritis. FACT 2000;5:211-2.
Indexing Status Subject indexing assigned by NLM MeSH Chondroitin /therapeutic use; Clinical Trials as Topic; Data Interpretation, Statistical; Dietary Supplements; Glucosamine /therapeutic use; Humans; Osteoarthritis /drug therapy AccessionNumber 12000008200 Date bibliographic record published 31/03/2001 Date abstract record published 31/03/2001 Record Status 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. |
|
|
|