Abstract
Previous observational studies have suggested an association between antibiotic use and rheumatoid arthritis (RA), though the causal relationship remains unclear. This study aimed to investigate the causal link between antibiotic use and RA in a European population using Mendelian randomization (MR). We utilized pooled genome-wide association study (GWAS) data on 12 antibiotics and RA from European populations, extracted from the GWAS Catalog. Both univariate MR and multivariate MR were employed to examine the causal relationship. Three analysis methods were applied: inverse variance weighting, MR-Egger, and weighted median, with inverse variance weighting as the primary method. Sensitivity analyses were conducted using Cochran Q statistics, MR-PRESSO, the MR-Egger intercept, and the leave-one-out test. Univariate MR revealed that tetracycline use was positively associated with RA (odds ratio = 1.013, 95% confidence interval = 1.001-1.024, P = .028), while none of the other 11 antibiotics exhibited a causal relationship with RA. However, further multivariate MR analysis found no causal association between tetracycline use and RA. Our results do not support a direct causal relationship between RA and antibiotic use, which may help alleviate some concerns among clinicians. Further MR studies are needed to validate these findings as additional datasets from other cohorts and GWASs with more detailed information become available.