Abstract
OBJECTIVE: To systematically synthesize Mendelian randomization (MR) evidence exploring causal associations between various exposures and rheumatoid arthritis (RA). METHODS: Systematic searches were conducted to identify eligible two-sample MR studies evaluating causal links between exposures and RA. Data extraction encompassed exposure types, genetic instruments, and analytical methods. The quality of evidence was evaluated based on STROBE-MR guidelines and evidence strength grades. Meta-analyses were performed using random-effects model with restricted maximum likelihood estimation and Hartung-Knapp-Sidik-Jonkman confidence intervals to summarize causal estimates. RESULTS: A total of 117 studies were included, comprising 496 MR associations across 248 unique exposures. Of these, 200 exposure-RA associations across 65 unique exposures were eligible for meta-analysis. The primary analysis identified smoking behavior (OR = 1.39, 95% CI 1.30-1.49) and hypothyroidism (OR = 1.55, 1.34-1.81) as significant risk factors with high-quality evidence. While television viewing (OR = 2.27, 1.77-2.90) and cholangitis (OR = 1.14, 1.06-1.23) showed nominal positive associations. Conversely, cognitive function (OR = 0.74, 0.68-0.81) and Interleukins-1 receptor antagonist (OR = 0.82, 95% CI 0.73-0.92) were identified as a nominally protective factor. In subtype analyses, basal metabolic rate emerged as a potential shared risk factor for both seropositive and seronegative RA. Overall, 7.66% and 26.01% of all associations were classified as providing Robust (I) and Probable (II) evidence, respectively. CONCLUSION: This comprehensive review clarifies the causal landscape of RA by distinguishing between significant causal drivers and potential associations. The findings highlight smoking cessation and thyroid function management as critical, evidence-based targets for RA prevention. However, the high proportion of low-quality evidence underscores the need for future high-quality MR studies to validate weaker signals. SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/view/CRD42024573056, identifier CRD42024573056.