No causal association between plasma homocysteine levels and atrial fibrillation: A Mendelian randomization study

Abstract

Background and aimsAlthough several studies have shown an association between plasma homocysteine (Hcy) levels and atrial fibrillation (AF), the causality remains unclear. We undertook a Mendelian randomization (MR) study to investigate the causal association between Hcy and AF. Methods and resultsSingle-nucleotide polymorphisms (SNPs) which genome-wide significantly associated with plasma Hcy levels were obtained from a genome-wide meta-analysis (N = 44 147). MR analyses including the random-effect inverse variance-weighted (IVW) meta-analysis, weighted median analysis, and MR-Egger regression were used to estimate the associations between the selected SNPs and AF based on a meta-analysis of genome-wide association study for AF (N = 588 190). The MR analyses revealed no causal role of genetically elevated plasma Hcy levels with AF risk (random-effect IVW, odds ratio per 1 SD increase in Hcy levels = 0.972, 95% confidence interval = 0.919 to 1.027, P = 0.308). The results were consistent with the weighted median method, MR-Egger and the analysis after excluding the pleiotropic SNPs. No heterogeneity and directional pleiotropy were observed in sensitivity analyses. ConclusionThe findings suggested that plasma Hcy levels were not causally associated with AF.