Non-pharmacological interventions in the treatment of rheumatoid arthritis: A systematic review and meta-analysis

To investigate the role of non-pharmaceutical therapies on disease activity in rheumatoid arthritis through systematic review and meta-analysis.

A review of Pubmed, EMBASE, Web of Science, and the Cochrane Library was performed from inception until March 26, 2019. Only randomized controlled trials which assessed oral, non-pharmacological interventions (e.g. diets, vitamins, oils, herbal remedies, fatty acids, supplements, etc.) in adult patients with rheumatoid arthritis, that presented clinically-relevant outcomes (defined as pain, fatigue, disability, joint counts, and/or disease indices) were included in our meta-analysis. Data were analyzed as mean differences between active and placebo and forest plots were performed. Heterogeneity was evaluated using I-squared statistics while funnel plots and Cochrane's risk of bias assessment evaluated bias.

8170 articles were identified in the search and 51 were RCTs were included. The mean difference in DAS28 was significantly improved in experimental group treated with diet (-0.46 [-0.91, -0.02], p = 0.04), zinc sulfate, copper sulphate, selenium, potassium, lipoic acid, turmeric, pomegranate extract, chamomile, and cranberry extract supplements (-0.77 [-1.17, -0.38], p < 0.001), A, B6, C, D, E, and K vitamins (-0.52 [-0.74, -0.29], p < 0.001), and fatty acids (-0.19 [-0.36, -0.01], p = 0.03). Other clinical metrics such as SJC, TJC, HAQ, SDAI, ACR20, and self-reported pain were decreased in the treatment groups. There was significant reporting bias in the studies.

Some non-pharmacological therapies may modestly improve some clinical outcomes in patients with rheumatoid arthritis. Many identified studies lacked full reporting. Further clinical trials that are well-designed, adequately powered, and sufficiently report ACR improvement criteria or EULAR response criteria outcomes are needed to confirm the efficacy of these therapies.