Efficacy and safety of dietary polyphenols in rheumatoid arthritis: A systematic review and meta-analysis of 47 randomized controlled trials.

The short version

Whether the plant compounds often discussed for rheumatoid arthritis — turmeric, green tea, pomegranate, garlic — actually do anything is the question this meta-analysis set out to answer. This review gathered every randomized trial it could find on the question: 47 trials, nearly 3,900 people, fifteen different plant polyphenols, from Iran, China, Korea, Egypt, Japan, Brazil, India and the United States. The clearest picture came from two of them. The first is total glucosides of paeony, an extract of Chinese peony root that is a licensed prescription medicine in China. Twenty-three trials tested it — always added to standard RA drugs like methotrexate, never instead of them. Disease activity scores came down, and so did CRP, ESR and rheumatoid factor, the blood markers doctors track. The most interesting result: people taking it reported *fewer* side effects than people on the standard drugs alone. That fits how it is used clinically — it appears to soften the toll the conventional drugs take on the liver and stomach. The second is curcumin, the yellow compound in turmeric. Five trials, and the same direction: lower disease activity, lower inflammatory markers, no extra side effects. Green tea polyphenols looked promising across two trials. Saffron did not show a clear effect. Everything else — pomegranate, garlic, resveratrol, sesame, cinnamon, ginger, kudzu, cranberry, olive oil, quercetin — rests on one or two small trials apiece, which is encouraging but not yet evidence that bears weight. The honest caveat, which the authors state plainly themselves: the trials disagreed with each other a great deal, most were small, and when the researchers sliced the data by dose and duration some of the headline benefits faded. Nearly all of these were add-on studies. Nobody tested a polyphenol against methotrexate, so the evidence says nothing about replacing a prescribed medication. What it does support is turmeric and peony extract as the best-evidenced candidates, studied as additions to existing treatment rather than replacements for it.

Key findings

• Total glucosides of paeony, from Paeonia lactiflora root, dominates the evidence base with 23 of the 47 trials: DAS28 -0.92, CRP SMD -1.32, ESR -6.44 mm/h, rheumatoid factor SMD -2.01, all statistically significant.
• TGP arms recorded FEWER adverse events than control arms (RR 0.55, 95% CI 0.44-0.69, 15 trials) — it appears to reduce, not add to, the side-effect burden of conventional DMARDs.
• Curcumin lowered DAS28 by 1.10 points across 4 trials, with parallel reductions in CRP, ESR and rheumatoid factor and no significant increase in adverse events.
• Eleven of the fifteen polyphenols — pomegranate, garlic, resveratrol, sesamin, cinnamon, ginger, puerarin, hesperidin, cranberry, olive, quercetin — rest on only one or two trials each and were reported narratively rather than pooled.
• Every included trial tested polyphenols as an ADD-ON to conventional therapy (methotrexate, leflunomide, hydroxychloroquine); none tested them as a replacement, and the authors explicitly warn against extrapolating to RA patients in general.

Common questions