Association between atrial fibrillation/flutter and left ventricular failure: A bidirectional Mendelian randomization study

Background ：atrial fibrillation(AF) and heart failure(HF) share common pathophysiological mechanisms, thus frequently coexisting and mutually influencing disease progression.The combination of these conditions is associated with heightened mortality rates and unfavorable prognosis. Significant progress has been made in the study of AF-HF, but it remains unclear which approach provides the best long-term efficacy. In this study, our objective is to employ Mendelian randomization studies in order to investigate the causal relationship between atrial fibrillation/atrial flutter(AFL) and left ventricular failure(LVF), explore potential therapeutic targets for clinical application, and optimize the management and clinical outcomes of patients with AF/AFL and HF. Methods: The data of AF/AFL from the IEU OpenGWAS project.These data derive from a European population consisting of 463,010 participants drawn from the UK Biobank. Among them, 5,669 individuals had AF/AFL, and a total of 9,851,867 SNPs were considered.To ensure a matching number of SNPs between LVF and minimize population overlap effects, we implemented the most recent and largest genome-wide association study meta-analysis from the IEU OpenGWAS project. A total of 2046 cases and 460,964 controls were investigated,within a total of 9,851,867 SNPs. We adopted inverse variance weighted (IVW) as the main way to estimate the Mendelian randomization analysis. Results ：The preliminary results of IVW revealed postive causal effect of AF/AFL on LVF [OR =1.053, 95% CI: 1.023-1.084, P = 0.0006] Cochran's IVW Q test results show no significant heterogeneity among these IVs. The results of the MR-Egger regression intercept analysis indicate no significant horizontal pleiotropy. MR-PRESSO global test results revealed no horizontal pleiotropy. Additionally, the p-values of the MR PRESSO global test for AF/AFL on LVF were all greater than 0.05.The funnel plot presents a symmetricl shap,suggesting significant heterogeneity,indicating that there is no systematic bias between the study effect and its accuracy. The leave-one-out plot is shown that each IVs does not have a serious bias effect on the overall MR results.The results of the steiger test confirmed no causal effect of LVF on AF/AFL. Conclusions： This MR study presents novel genetic evidence supporting a causal association between AF/AFL and LVF, thus contributing to the advancement of our understanding in this field.This study underscores the importance of managing HF-AF patients by incorporating AF/AFL treatment alongside conventional anti-HF therapy and ventricular rate control, aiming to enhance LV function and achieve a more favorable prognosis.In addition, this study found that LVF did not have a significant impact on AF/AFL.