Should you Recommend Glucosamine for Osteoarthritis?

June 3, 2007
Daniel M. Keller, PhD

Internal Medicine World Report, January 2006, Volume 0, Issue 0

SAN DIEGO?"What's a nice molecule like this doing in a joint like yours?" That's what rheumatologists are still trying to figure out for their patients who ask about using oral glucosamine to reduce the pain associated with osteoarthritis (OA).

After at least 15 randomized, double-blind, placebo-controlled studies, as well as many other less rigorous ones, the use of glucosamine supplements for OA pain appears to have a benefit,

but the extent of the benefit is still hard to estimate. Results from new studies presented at the annual meeting of the American College of Rheumatology (ACR) produced conflicting results, and a meta-analysis of 13 previous trials offered some explanations for the differing study conclusions.

GAIT: Ambivalent Results

The first trial presented at the meeting was the NIH-sponsored multicenter Glucosamine/Chondroitin Arthritis Inter?vention Trial (GAIT), which showed no benefit for the primary outcome measure of a 20% reduction in Western Ontario and McMaster Uni?versities (WOMAC) Osteoarthritis Index pain score at 24 weeks compared with baseline pain score. The trial included patients (mean age, 58.6 years) who have had knee pain for at least 6 months and had x-ray evidence of OA.

The patients were randomized in a double-blind fashion to 1 of 4 treatment groups: glucosamine HCl, 500 mg, or sodium chondroitin sulfate, 400 mg, 3 times daily; a combination of the 2 agents; celecoxib (Celebrex), 200 mg daily; or placebo. All patients were allowed up to 4000 mg of acetaminophen as rescue analgesia, except within 24 hours of study visits.

Among the 1258 patients who completed the trial, only celecoxib was associated with a sta?tis-??tically superior re?sponse rate compared with placebo?70.1% versus 60.1%; P = .008. In contrast, use of glucosamine and/or chondroitin sulfate demonstrated no benefit compared with placebo.

However, among the 22% of patients with moderate-to-severe knee OA, only the combination of the 2 study compounds (glucosamine and chondroitin sulfate) was statistically superior to placebo (79.2% vs 54.3%; P = .002), whereas celecoxib was not.

Adverse events were generally mild and were similar among all groups. Of note, the investigators did not see any increase in cardiac events with celecoxib in this fairly short-term trial.

Lead investigator Daniel Clegg, MD, chief of rheumatology, University of Utah School of Medicine, Salt Lake City, said of the overall findings, "In conclusion?no efficacy was seen with glucosamine, chondroitin sulfate, or the combination." But he added "there are patients who respond, and we need to see what's different about those patients and really design a trial looking at those patients and see why they respond." One problem in interpreting the results is that patients in the placebo group had responses that were "much higher than we ?expected," he said.

Dr Clegg said he does not think that doctors yet have a basis to answer patients who ask if they should take glucosamine/chondroitin sulfate supplements. As more expected peer-reviewed results become available in the near future, physicians should have a better basis to advise their patients, he said.

GUIDE Trial More Positive

The second study presented was the European Glucosamine Unum In Die Efficacy (GUIDE) trial, which produced more positive results, showing that 1500-mg once-daily glucosamine sulfate was at least as effective as acetaminophen, 1000 mg 3 times daily, in controlling the pain of knee OA. Patients (n = 318) were randomized equally in a double-blind manner to glucosamine sulfate, acetaminophen, or placebo for 6 months. Ibu?profen was allowed as needed.

In the glucosamine sulfate group, 75.4% of patients had a response versus 40.4% in the placebo group, with 65.7% responding in the acetaminophen group on the Osteo?Arthritis Research Society In?ternational -A and -B scales of pain, function, and patient glo?bal assessment (all results were statistically significant vs placebo). Gluco?s?amine sulfate was also statistically superior to placebo on the Leq?ues?ne Index of pain and function and on the WOMAC score. Safety re?sults were similar across groups, except that 31 of the 108 acetaminophen users had slight elevations in liver function test results.

Lead investigator Gabriel Herrero-Beaumont, MD, of the Jim?nez D?az Fou?n?dation in Madrid, Spain, noted that acetaminophen is the preferred first-line oral symptomatic treatment in European and American OA practice guidelines and thus was important to include in the trial. "But glucosamine sulfate oral doses once a day?should be the preferred symptomatic medication for this type of patient" over the long-term, namely, 2 to 3 months, he said, reserving acetaminophen for acute pain relief.

In light of the elevated liver function test results in a significant proportion of the study patients taking acetaminophen, he questioned the drug's safety in patients who may be taking it for several years.

Analyzing the Data

David Felson, MD, MPH, professor of medicine and public health, Boston University School of Medicine, and colleagues presented a meta-analysis of 13 studies that showed overall that glucosamine is effective for OA pain but that the level of effect varied among the studies. The largest effects occurred when industry (Rotta Pharmaceuticals) had some involvement in the trials (ie, funding, drug supply, or data handling) compared with nonindustry studies, making it difficult to estimate effect level with confidence, according to the investigators.

Dr Felson called both GAIT and GUIDE "null studies." He asked Dr Herrero-Beaumont if he had a chance to independently analyze the GUIDE data, without the study sponsor and manufacturer of the supplements, Rottapharm (Monza, Italy) "looking over his shoulder." Dr Herrero-Beaumont said no.

In keeping with the name of the GUIDE trial, Dr Herrero-Beaumont emphasized the efficacy of once-daily dosing. He cited pharmacokinetic data supporting the idea that once-daily dosing produces higher plasma levels compared with divided doses 3 times daily.

Dr Clegg suggested that differences in glucosamine preparations (eg, hydro?chloride vs sulfate) may account for the divergent findings that emerged from the GAIT and GUIDE studies. Glucos?amine sulfate is slightly different from the hydrochloride version used in the United States and is a prescription agent in Europe.

Another study by Stefano Persiani, MD, of Tottapharm, Monza, Italy, and colleagues showed large increases in plasma and synovial fluid glucosamine concentrations among 5 patients with OA who took oral crystalline glucos?amine sulfate, 1500 mg once daily, for 14 days.

Mean baseline levels before ad?m?i?n?istration were 39.5 ng/mL in plasma and 15.1 ng/mL in synovial fluid. Three hours after the last administration, the levels (after subtracting baseline values) were 1419 and 1291 ng/mL, respectively. Dr Persiani noted that the postadministration synovial fluid level of 7.2 ?M was in the range of the 10-?M level that was shown effective in a cultured human chondrocyte model.

Dr Clegg said his study used pharmaceutical-grade glucosamine HCl and chondroitin sulfate, since it was conducted under an investigational new drug application for therapeutic use of the compounds rather than as dietary supplements. The investigators had hoped to find a suitable commercial preparation to use in the study. "We couldn't find one that we felt would predictably allow us to ensure the potency [and] purity?.So that's why we manufactured one," he explained.

Dr Herrero-Beaumont commented that "glucos?amine is important [to study] because many people?take [it]." There?fore, studies are important to physicians, because so many patients ask for advice about taking glucosamine. And the results of the long-awaited pain outcome measure of the US GAIT trial are not definitive, as x-ray data on disease progression are still to come. But Dr Herrero-Beaumont appear?ed convinced by the GUIDE re?sults that glucosamine can be beneficial in relieving the symptoms of OA.

So What Should You Advise Your Patients?

Since the US government does not require commercially marketed preparations to be manufactured to the same standards as the ones in the trial, Dr Clegg said the GAIT findings may not be generalizable to products obtained on the open market. Furthermore, such dietary supplements do not have to meet pharmaceutical guidelines for the claims they can make.

The National Center for Com?plementary and Alternative Medicine (NCCAM) and the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health, cosponsors of the GAIT study, are reserving judgment on those results. In a news release, NCCAM said that although it is committed to "supporting rigorous research of complementary and alternative medicine approaches and therapies," it is "refraining from ?com?ment on the study until the full results are published in the peer-reviewed literature."

And thus, for now at least it appears that physicians are left without clear answ?ers as to the extent of the benefits of these products. But as Dr Clegg suggests, using glucosamine helps some patients, and given the apparently benign side-effect profile of these compounds, many physicians at the meeting had no objection to their patients trying them at this point, before definitive conclusions are available.