Coffee consumption and the risk of rheumatoid arthritis and systemic lupus erythematosus: a Mendelian randomization study

Abstract

We aimed to analyze the causal association between coffee consumption and the risk of rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE). We performed a two-sample Mendelian randomization (MR) analysis using the inverse-variance weighted (IVW), MR-Egger regression, and weighted median methods. We used publicly available summary statistics datasets of coffee consumption genome-wide association studies (GWASs) as an exposure variable and RA and SLE GWASs as outcomes. Four single-nucleotide polymorphisms (SNPs) from GWASs of coffee consumption were selected as instrumental variables (IVs) to improve inference: NCARD (rs16868941), POR (rs17685), CYP1A1 (rs2470893), and LAMB4 (rs382140). The IVW method showed a causal association between coffee consumption and RA (beta = 0.770, SE = 0.279, p = 0.006). MR-Egger regression revealed that directional pleiotropy was unlikely to be biasing the result (intercept = - 0.145, p = 0.451). While the MR-Egger analysis showed no causal association between coffee consumption and RA (beta = 2.744, SE = 1.712, p = 0.355), the weighted median approach demonstrated a causal association between coffee consumption and RA (beta = 0.751, SE = 0.348, p = 0.031). However, the associations based on the weighted median analyses after the Bonferroni correction were not significant (adjusted p values = 0.091). The IVW, MR-Egger analysis, and weighted median methods showed no causal association between coffee consumption and SLE risk (beta = 0.594, SE = 0.437, p = 0.209; beta = 3.100, SE = 3.632, p = 0.550; beta = 0.733, SE = 0.567, p = 0.196). MR analysis results do not support causal associations between coffee consumption and the development of RA and SLE.