The causal effects of genetically determined human blood metabolites on the risk of atrial fibrillation.

Abstract

Blood metabolites have been found related to atrial fibrillation (AF), but the causal role is still unclear. Mendel randomization (MR) can give information about the causality between blood metabolites and AF. Two-sample MR analysis was used to evaluate the causality between 486 blood metabolites and AF. Firstly, the genome-wide association study (GWAS) data for AF (from Nielsen et al.) was analyzed and some metabolites were identified. Then another GWAS data for AF (from Roselli et al.) was repeatedly analyzed to verify the results. Inverse variance weighted method was mainly used to determine the causality, and MR-egger, Weighted Median, and MR-PRESSO models were used as supplements of MR. Cochran's Q test was used to assess heterogeneity. And MR-Egger intercept and MR-PRESSO global test were performed to measure pleiotropy. The study used Bonferroni's corrected P value (P < 1.03 × 10-4) as the significance threshold. After MR analysis and replication analysis, we found two overlapped metabolites. Among which tryptophan betaine was the most significant causal metabolite in both AF GWAS data (from Nielsen et al.) (odds ratio (OR) = 0.83, 95% confidence interval (CI) = 0.76-0.90, P = 9.37 × 10-6) and AF GWAS data (from Roselli et al.) (OR = 0.82, 95% CI = 0.76-0.88, P = 2.00 × 10-7), while uridine was nominally significant metabolites in both AF GWAS data (from Nielsen et al.) (OR = 0.58, 95% CI = 0.40-0.84, P = 0.004) and AF GWAS data (from Roselli et al.) (OR = 0.56, 95% CI = 0.35-0.88, P = 0.01). And the results of sensitivity analysis showed that none of them had obvious heterogeneity or pleiotropy. The study identified several blood metabolites that were causally related to AF, which may provide new perspectives on the pathogenesis of AF.