A causal association between schizophrenia and bipolar disorder on rheumatoid arthritis: A two-sample Mendelian randomization study

BackgroundSchizophrenia (SCZ) and bipolar disorder (BD) are both associated with several autoimmune disorders including rheumatoid arthritis(RA). However, a causal association of SCZ and BD on RA is controversial and elusive. In the present study, we aimed to investigate the causal association of SCZ and BD with RA by using the Mendelian randomization (MR) approach. MethodsA two-sample MR (2SMR) study including the inverse-variance weighted(IVW), weighted median, simple mode, weighted mode and MR-Egger methods were performed. We used summary-level genome-wide association study(GWAS) data in which BD and SCZ are the exposure and RA the outcome. We used data from the Psychiatric Genomics Consortium(PGC) for BD(n= 41,917) and SCZ(n= 33,426) and RA GWAS dataset(n= 2,843) from the European ancestry for RA. ResultsWe found 48 and 52 independent single nucleotide polymorphisms (SNPs, r2 <0.001)) that were significant for respectively BD and SCZ (p <5x10-8). Subsequently, these SNPs were utilized as instrumental variables(IVs) in 2SMR analysis to explore the causality of BD and SCZ on RA. The two out of five MR methods showed a statistically significant inverse causal association between BD and RA: weighted median method(odds ratio (OR), 0.869, [95% CI, 0.764-0.989]; P= 0.034) and inverse-variance weighted(IVW) method (OR, 0.810, [95% CI, 0.689-0.953]; P= 0.011). However, we did not find any significant association of SCZ with RA (OR, 1.008, [95% CI, 0.931-1.092]; P= 0.829, using the IVW method). ConclusionsThese results provide support for an inverse causal association between BD and RA. Further investigation is needed to explain the underlying protective mechanisms in the development of RA. Key messagesO_LIMendelian randomization can offer strong insight into the cause-effect relationships in rheumatology. C_LIO_LIBipolar disorder had a protective effect on rheumatoid arthritis. C_LIO_LIThere is no inverse causal association between schizophrenia and rheumatoid arthritis contrary to the findings from observational studies. C_LI