S-Adenosylmethionine for osteoarthritis of the knee or hip

Rutjes AW, Nüesch E, Reichenbach S, Jüni P · Cochrane Database of Systematic Reviews

Four placebo-controlled trials of SAMe, and not every outcome draws on all four. Pain pooled two trials and 533 participants at a standardised mean difference of -0.17 (95% CI -0.35 to 0.01), which the review converts to 0.4 cm on a 10 cm visual analogue scale, with no between-trial heterogeneity; the larger of the two, on 96 percent of the weight, came in at -0.17 (-0.35 to 0.02) on its own. Function pooled three trials and 542 participants at SMD 0.02 (-0.68 to 0.71), heterogeneity 54 percent. Adverse events pooled four trials and 623 participants at a relative risk of 1.27 (0.94 to 1.71), 79 events on SAMe against 59 on placebo, and withdrawals because of one, four trials and 656 participants, at 0.94 (0.48 to 1.86). No trial described serious adverse events at all, which the reviewers call concerning. Doses differed across the four: 200 mg three times daily in one, 400 mg three times daily in two, and the largest trial is recorded at 200 mg six times daily in its own included-study table while the review's narrative says 400 mg six times daily. Cochrane judged the methodological quality and the quality of reporting poor, called the review inconclusive, and advised against routine use.

Population

Patients with knee or hip osteoarthritis across 4 placebo-controlled trials

Intervention

S-adenosylmethionine (SAMe) at any dose or formulation, in four placebo-controlled trials

Comparator

Placebo

Limitations

Four small trials with poor reporting, and the pain interval touches zero — the review calls the evidence inconclusive rather than negative. None of the four was analysed by intention to treat, three of the four gave insufficient information on how allocation was generated or concealed, and the reviewers attribute the function estimate largely to a nine-patient trial whose unfavourable result they put down to chance. With four trials they could not examine dose, route, methodological quality or source of funding at all, and the review contradicts itself on the largest trial's dose. Two of the four had their drug supplied or their grant paid by a manufacturer, which the review marks as high risk on its funding item. The largest trial's population was 272 knees, 115 hips, 69 spines and 33 hands, so it is not confined to the joints in this review's title, and its naproxen arm was excluded from every analysis here. It is also the only placebo-controlled synthesis SAMe has in this disease: most SAMe trials use an NSAID comparator and could not be pooled here. For the TMG entry it tests SAMe rather than betaine, so it speaks to the methylation pathway rather than to the supplement.