Glucosamine and chondroitin for treatment of osteoarthritis: asystematic quality assessment and meta-analysis.
McAlindon TE, LaValley MP, Gulin JP, Felson DT
JAMA 2000 Mar 15;283(11):1469-75
PDF Version for Download‘, ‘PDF Version for Download
CONTEXT: Glucosamine and chondroitin preparations arewidely touted in the laypress as remedies for osteoarthritis (OA),but uncertainty about their efficacyexists among the medicalcommunity. OBJECTIVE: To evaluate benefit ofglucosamine andchondroitin preparations for OA symptoms using metaanalysiscombinedwith systematic quality assessment of clinicaltrials of thesepreparations in knee and/or hip OA. DATASOURCES: We searched for human clinicaltrials in MEDLINE(1966 to June 1999) and the Cochrane Controlled Trials Registerusing the terms osteoarthritis, osteoarthrosis, degenerativearthritis,glucosamine, chondroitin, and glycosaminoglycans. Wealso manually searched review articles, manuscripts, andsupplements from rheumatology and OA journals and soughtunpublished data by contacting content experts, study authors,and manufacturers of glucosamine or chondroitin. STUDYSELECTION: Studies were included if they were published orunpublished double-blind, randomized,placebo-controlled trialsof 4 or more weeks” duration that tested glucosamineorchondroitin for knee or hip OA and reported extractable data onthe effect of treatment on symptoms. Fifteen of 37 studies wereincluded in the analysis. DATAEXTRACTION: Reviewersperformed data extraction and scored each trial using a qualityassessment instrument. We computed an effect size from theintergroupdifference in mean outcome values at trial end, dividedby the SD of the outcome value in the placebo group (0.2, smalleffect; 0.5, moderate; 0.8, large), and applied a correction factorto reduce bias. We tested for trial heterogeneity and publicationsizes using a random effects model. DATA SYNTHESIS: Qualityscores ranged from12.3% to 55.4% of the maximum, with amean (SD) of 35.5% (12%). Only 1 study described adequateallocation concealment and 2 reported an intent-to-treat analysis.Most were supported or performed by a manufacturer. Funnelplots showed significant asymmetry (P< or =.01) compatible withpublication bias.Tests for heterogeneity were nonsignificant afterremoving 1 outlier trial. The aggregated effect sizes were 0.44(95% confidence interval [CI], 0.24-0.64) forglucosamine and0.78 (95% CI, 0.60-0.95) for chondroitin, but they werediminished when only high-quality or large trials wereconsidered. The effect sizes were relatively consistent for painand functional outcomes. CONCLUSIONS:Trials of glucosamineand chondroitin preparations for OA symptoms demonstratemoderate to large effects, but quality issues and likely publicationbias suggest that these effects are exaggerated. Nevertheless,some degree of efficacy appears probable for thesepreparations.’, ‘Publication Types:ReviewReview, academic
Comment in: ACP J Club. 2000 Sep-Oct;133(2):58 JAMA. 2000
Mar 15;283(11):1483-4
PMID: 10732937
PDF Version for Download