The GAIT study was designed to include 1,588 patients to provide the study with statistical power to detect 1 or more clinically meaningful differences based on an assumed placebo response rate of 35%. When this placebo response rate nearly doubled, the number of participants needed to obtain a similar statistical power increased substantially. With far too few patients given its placebo response rate, the data were barely able to prove its control (celecoxib) in the primary outcome measure (P = .04) and was unable to do so in the moderate/severe pain subgroup. Furthermore, the choice of the product tested (GH) has been called into question given the fact that GS has been more rigorously studied in the literature. The GAIT authors also chose less sophisticated methods for dealing with missing data, using the last observation–carried forward method rather than the multiple imputation method.
Alekseeva et al.33 examined 90 women aged between 40 and 75 years with Kellgren-Lawrence stage II or III knee OA who had pain after 40 minutes of walking and regularly took NSAIDs for pain relief. The patients were randomly selected either to receive 500 mg of the glucosamine and CS supplements with optional diclofenac (50 mg) or to receive a placebo and optional diclofenac for a total of 3 months. The results were measured by the WOMAC, daily need for NSAIDs, and evaluation of efficacy by the patient and the physician after 1 and 3 months of treatment and again 3 months after the oral supplementation had been stopped. The true WOMAC score decreased after 3 months of therapy and 3 months after the supplementation had been stopped (P < .03). At the end of the 3 months of therapy, the study group exhibited decreases in pain scores (P = .008) and increases in subjective functional ability. The patients taking the glucosamine and CS supplementation required less diclofenac. After 1 month of therapy, 4.5% stopped taking diclofenac and nearly 40% stopped taking it by the end of the study. Although limited by its size and the small subgroup that was studied (older women), this study showed that combined medications offer significant safety and effective pain relief in the short term with long-lasting effects.
Messier et al.,34 in a double-blind 12-month GH/CS study with 80 patients, incorporated 6 months of exercise after 6 months of a non-exercising treatment. The primary end-point was the WOMAC and functional measures such as the 6-minute walk. At 12 months, there was no difference between groups for the 6-minute walk, knee strength, mobility, and function over the placebo treatment.
Meta-Analyses of Glucosamine and CS Studies
Several important meta-analyses have been published in recent years about the efficacy of glucosamine and CS therapy. By performing exhaustive searches in the literature and applying systematic quality assessment of these studies, these meta-analyses provided pooled information from the many pre-existing small studies.
McAlindon et al.35 examined 15 double-blind, randomized, placebo-controlled trials of 4 weeks' duration or longer for their impact on the symptoms of hip and/or knee OA. They included studies of glucosamine and CS with various routes of administration, including oral, intramuscular, intravenous, and intra-articular. Very few of the examined studies described adequate allocation concealment or use of an intention-to-treat analysis. They also found evidence of significant publication bias, likely because of manufacturer's sponsorship of trials and the financial interests of the authors. When only the larger high-quality studies were evaluated, the effects of glucosamine and CS persisted, although they were noticeably diminished. This study also suggested that the full therapeutic benefit of these supplements likely did not occur in the first 4 weeks and that longer studies would be of significant value.
Richy et al.36 examined both structural and symptomatic efficacy of CS and glucosamine. By examining structural changes via radiographic progression of joint space narrowing, this analysis was the first to evaluate the disease-modifying effects of these supplements. Evaluating the results of 15 studies that included data from 1,775 patients, the authors showed a statistically significant improvement in symptom scores with both glucosamine and CS therapy. They also were able to show a significant effect of glucosamine on the progression of joint space narrowing over a 3-year period, suggesting a disease-modifying effect of the compound (no such studies existed for CS). Importantly, the tolerance for these supplements was again shown to be equal to that of placebo.
Bjoral et al.37 reviewed 63 RCTs using opioids, NSAIDs, glucosamine, CS, and acetaminophen (Tylenol; McNeil Consumer Healthcare) for knee OA including some 14,060 total patients. Acetaminophen, GS, and CS had maximum efficacies at 1 to 4 weeks with mild pain improvements. Overall clinical effects from these knee pharmacologic arthritic interventions were found to be small and limited to the first 2 to 3 weeks after the start of treatment.
Distler and Angueloouch38 reviewed clinical evidence for glucosamine and CS studies analyzing RCTs. Their results were inconclusive regarding the continuous use of these nutraceuticals because of weak research design
|