The Glucosamine File: One Landmark Null, One Drug Dead Heat, and a Continent Split in Two

Consider the strange inventory of the glucosamine-and-chondroitin evidence base. Item: one amino sugar and one glycosaminoglycan, both structural building blocks of articular cartilage, classified together as symptomatic slow-acting drugs for osteoarthritis. Item: one landmark trial, published in the New England Journal of Medicine in 2006, that found the pair no better than placebo overall. Item: one subgroup inside that same trial — participants with moderate-to-severe knee pain — in which the combination possibly helped. Item: one European trial, MOVES, published in Annals of the Rheumatic Diseases in 2015, that found the combination non-inferior to the prescription anti-inflammatory celecoxib. Item: one Cochrane systematic review of 43 chondroitin trials. Item: one 2018 meta-analysis of randomized controlled trials. Item: one pilot trial of a third ingredient, MSM. Item: two medical guideline bodies advising against the supplements, one European society recommending them — provided the formulations are pharmaceutical grade.

What kind of evidence pile produces both a New England Journal null and a dead heat with a prescription drug? That is the precise question, and the most defensible answer is that these compounds are neither miracle nor junk: the effect, where it exists, is modest, concentrated in certain patients and certain formulations, and contested by serious reviewers on both sides.

What the big trials actually showed

The GAIT trial is the anchor. In painful knee osteoarthritis, glucosamine and chondroitin sulfate — alone and in combination — did not beat placebo across the full study population. The qualification that keeps the debate alive is the subgroup with moderate-to-severe pain, where the combination showed a signal worth noticing. A subgroup finding is hypothesis-generating, not proof, but it is the reason the story did not end in 2006.

MOVES then complicated the picture from the other direction. The combination of chondroitin sulfate and glucosamine proved non-inferior to celecoxib for painful knee osteoarthritis — meaning the supplement pair performed about as well as an established drug in that trial's terms. Non-inferiority is not superiority, and one trial is not a literature, but matching a prescription analgesic is not nothing.

Meta-analyses that refuse to agree

The Cochrane review of 43 chondroitin trials and the 2018 meta-analysis of randomized trials represent the synthesis layer, and the syntheses conflict. Some analyses find chondroitin and glucosamine sulfate show structure-modifying effects on joint-space narrowing — the slowing of cartilage loss visible on imaging. Other high-quality reviews judge the pair ineffective, or of unclear clinical importance. This is why the guidelines split: the Osteoarthritis Research Society International and the American College of Rheumatology advise against the supplements, while Europe's ESCEO recommends them in pharmaceutical-grade form. All three bodies read substantially the same trials.

The third ingredient, and the mortality question

MSM — methylsulfonylmethane, an organosulfur compound with anti-inflammatory and antioxidant properties — has a thinner file: a 2006 pilot clinical trial in knee osteoarthritis, plus a small 12-week randomized trial in which the glucosamine–MSM combination reduced pain and swelling more than either agent alone. Encouraging, and small. Separately, large cohort studies have linked regular glucosamine use to lower all-cause and cardiovascular mortality — the finding that fuels longevity interest — but at least one well-adjusted NHANES analysis found no such association, and every one of these mortality links is observational. People who take joint supplements differ from people who don't in ways no statistical adjustment fully captures.

What this means for a reader

If you have knee osteoarthritis, the evidence supports a modest expectation at most: these supplements are slow-acting, generally well tolerated, possibly helpful for moderate-to-severe pain, and endorsed by exactly one side of a transatlantic guideline divide. They are not a substitute for the fundamentals of joint care, and the longevity-mortality headline is association, not cause.

The inventory closes where it opened — with two cartilage building blocks, one null, one dead heat, 43 trials, and a map of Europe drawn in a different color from North America. The compounds have earned the argument. They have not yet earned a verdict.

Educational, not medical advice.