ROTTERDAM, The Netherlands—Glucosamine accounts for nearly $1 billion per year in consumer sales, but the first randomized, controlled study of glucosamine sulfate in patients with relatively early hip osteoarthritis (OA) concluded that the supplement is no better than placebo for pain, function, or for preventing structural progression. The study, known as the Glucosamine in Osteoarthritis: Longterm Effectiveness (GOAL) trial, was reported by Rianne M. Rozendaal, MSc, and colleagues in the Annals of Internal Medicine.¹

"[T]his trial followed the guidelines for clinical trials in osteoarthritis, used blinded outcome assessment, had a high rate of completion, and was performed without pharmaceutical company support. In addition, the patients are representative of those using over-the-counter glucosamine. Given the lack of clinically important effects on pain, function, and stiffness over 24 months, our results suggest that glucosamine sulfate is not an effective therapy for patients with hip osteoarthritis," wrote Rozendaal, at the Erasmus Medical Center in Rotterdam, The Netherlands, whose institution funded the study.

No difference between glucosamine and placebo after 2 years

The investigators conducted a 2-year, blinded, randomized, placebo-controlled trial in 222 patients who met American College of Rheumatology (ACR) criteria for hip OA. Patients took either placebo or 1500 mg of oral glucosamine sulfate once daily.

Primary outcome measures after 24 months were the Western Ontario and McMaster Universities (WOMAC) pain and function subscales, and joint space narrowing. Secondary outcome measures after 3, 12, and 24 months were the WOMAC pain, function, and stiffness subscales. There were no significant differences in either pain or function scores over 24 months. The mean difference in pain scores between glucosamine and placebo groups was -1.54, and the mean difference in function scores was -2.01.

Progression of joint damage was similarly unaffected. The investigators wrote, "The outcomes for the 4 sites at which we measured joint space narrowing were inconsistent in whether they favored glucosamine sulfate or placebo group; however, none of the 95% CIs for differences were consistent with the minimal clinically important change of 0.25 mm in joint space narrowing."

Translating research into practice

An accompanying editorial² by Johannes W. J. Bijlsma, MD, PhD, and Floris P. J. G. Lafeber, PhD, of University Medical Center in Utrecht, The Netherlands, suggests that if there was any effect, it was too small to be detected in the study. Drs. Bijlsma and Lafeber pointed out that the rate of cartilage loss was slow in the placebo group—only 0.16 mm after 2 years. "For the glucosamine group to exceed the placebo group by 0.25 mm [which is the accepted criterion for clinical response in OA trials], the thickness of the cartilage would have had to increase, an unrealistic expectation," they said.

The problem reflects a "Catch 22" problem in OA clinical trial design. Rozendaal et al enrolled patients with relatively early OA (excluding patients who had a Kellgren and Lawrence score of 4) because such patients had benefited most from glucosamine in earlier clinical trials, which were mostly done in knee OA. In retrospect, enrolling patients with more severe OA, which progresses more quickly, might have been better for detecting possible beneficial effects.

"Although Rozendaal and colleagues' study shows that patients with early hip OA do not benefit from glucosamine, it still does not give a definitive answer on the possible effect of glucosamine over a more prolonged period in patients with more severe disease," Drs. Bijlsma and Lafeber concluded.

References