Table2.DOCX

Abstract

Background: Heart failure (HF) is a potential cause of ischemic stroke (IS), and previous studies have reported an association between HF and IS. This study aimed to analyze the causal link between HF and IS using bidirectional and multivariable Mendelian randomization (MR) studies. Methods: Genetic variants significantly associated with HF and IS were selected in the MR analysis from two large genome-wide association studies. Bidirectional and multivariable MR analyses were performed to evaluate the effect of HF on IS or the effect of IS on HF. Results: Two-sample MR analysis showed causal effects of HF on IS of all causes (odds ratio [OR] = 1.555, 95% confidence interval [CI]: 1.343-1.799, p = 3.35 × 10-9) and large artery atherosclerosis stroke (LAS) (OR = 1.678, 95% CI: 1.044-2.696, p = 3.03 × 10-5), while there was a suggestive effect of HF on cardioembolic stroke (CES) (OR = 3.355, 95% CI: 1.031-10.919, p = 0.044). Genetically predicted HF was not associated with small artery occlusion stroke. Bidirectional MR analysis showed causal effects of IS of all causes (OR = 1.211, 95% CI: 1.040-1.410, p = 0.014) and CES (OR = 1.277, 95% CI: 1.213-1.344, p = 6.73 × 10-21) on HF, while there were no causal effects of LAS on HF. Conclusions: This MR analysis provided evidence of the causal links between genetically predicted HF and IS. Subgroup analysis highlighted the causal or suggestive relationship between genetically predicted HF and LAS or CES. The potential causal links need further investigation with genetic information about other ancestries or etiologies of HF.