Glucosamine and Chondroitin for Treatment of Osteoarthritis: A Systematic Quality Assessment and Meta-analysis.

McAlindon ET, LaValley MP, Gulin JP, Felson DT (The Arthritis Center, Boston University School of Medicine, Boston, Mass), JAMA. 2000;283:1469-1475.

Despite the existence of numerous therapeutic approaches directed at osteoarthritis (OA), few of these approaches are truly beneficial to patients with OA. Pharmacologic agents, such as nonsteroidal anti-inflammatory drugs, have been the drugs of choice for years; however, their adverse side effects are a major drawback. According to the authors, over the past decade, glucosamine and chondroitin preparations have gained increasing popularity among patients with OA, primarily due to efficacy claims made by the nonmedical community. Glucosamine and chondroitin are biochemical components of proteoglycan molecules in the articular cartilage matrix, and their role in the integrity of the cartilage has been established.

Pharmacologically, glucosamine and chondroitin are absorbed through the gastrointestinal tract and can moderately enhance the rate of cartilage proteoglycan synthesis. In theory, if this process can be substantially enhanced, then these agents can be potentially effective candidates in the management of patients with OA. According to the authors, the medical community is skeptical about the efficacy of these drugs because the published information largely comes from nonmedical sources, which do not meet the strict standard of scientific, controlled research.

To evaluate the therapeutic benefits objectively, the authors conducted a quality assessment and meta-analysis of the evidence from a number of well-designed clinical trials that used glucosamine and chondroitin preparations in the management of symptoms in patients with OA.

This study used electronic and manual methods of searching established sources of clinical medical science literature, such as MEDLINE and the Cochrane Controlled Trials Register, to analyze the clinical trials conducted over the past 33 years (1966-1999). Contacts were made with the authors of many publications to access their newly published or even their unpublished data or to obtain updates on the old data. Studies were included in this meta-analysis if they had clearly stated that they: (1) were controlled trials (double-blind and randomized) of glucosamine and chondroitin drugs on patients with OA, (2) lasted for at least 4 weeks, and (3) had used at least one established outcome measure. The authors compiled data from 15 clinical trials that fulfilled all of the inclusion criteria. These data were subjected to several levels of statistical analyses, which extensively examined the efficacy of the trials, the quality of the data and their assessment tools, the outcome, and the potential bias in each study.

Most of the trials were found to have potential bias and were either supported by commercial grants or were actually conducted by pharmaceutical companies. The indexes of efficacy (the aggregated effect sizes) were fairly consistent for pain and functional outcome for either drug. On a proposed scale where 0.2 means small therapeutic efficacy and 0.8 denotes large therapeutic efficacy, glucosamine appeared to be generally less efficacious than chondroitin (0.44 and 0.78, respectively). The quality scores for the trials varied significantly and, at best, ranged from 12.3% to 55.4% of the maximum.

The authors concluded that clinical trials of drugs containing glucosamine and chondroitin appear to be safe and have some degree of therapeutic effect and clinical application in the management of the symptoms of patients with OA. They believed that the findings of this meta-analysis suggest that the actual efficacy of these drugs may be substantially lower than what has been claimed by the drug manufacturers. Furthermore, it might take longer than 1 month for either glucosamine or chondroitin preparations to induce their full therapeutic potentials. The authors recommended further high-quality, independent studies to determine the actual efficacy and true indication of these drugs.

Kamran Tavakol, PT, PhD Howard University Washington, DC