Stroke and Myocardial Infarction: A Bidirectional Mendelian Randomization Study

Wenxian Sun,1 Luyang Zhang,1 Weishi Liu,1 Mengke Tian,1 Xin Wang,1 Jing Liang,1 Yuying Wang,1 Lan Ding,1 Lulu Pei,1 Jie Lu,2 Yuming Xu,1 Bo Song1 1Department of Neurology, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan Province, People’s Republic of China; 2Department of Epidemiology and Health Statistics College of Public Health, Zhengzhou University, Zhengzhou, Henan Province, People’s Republic of ChinaCorrespondence: Bo SongDepartment of Neurology, The First Affiliated Hospital of Zhengzhou University, 1st Jianshe Eastern Road, Zhengzhou, 450052, Henan Province, People’s Republic of ChinaTel +86-13603983297Fax +86-371-66975140Email fccsongb@zzu.edu.cnBackground: Stroke and myocardial infarction (MI) are associated with each other, as demonstrated in observational studies. However, it is unclear whether this relationship is causal, and the purpose of this study was to explore the bidirectional causality between stroke and MI.Methods: Causality between stroke and MI was assessed using two-sample Mendelian randomization (MR). All genetic instruments related to stroke (40,585 cases; 406,111 controls) and MI (43,676 cases; 128,199 controls) were derived from large published genome-wide association study. The MR analysis was calculated with inverse-variance weighting, MR-Egger, weighted mode, weighted median, and simple mode methods, and sensitivity analyses are used to detect the heterogeneity or pleiotropy.Results: Genetically predicted large-artery stroke (LAS) was causally related to higher odds of MI (odds ratio [OR] = 1.13, 95% confidence interval [CI]: 1.06– 1.20, p = 1.0× 10− 4), and the causal effect of LAS on MI was significantly weakened (OR = 1.09, 95% CI: 1.02– 1.17, p = 0.017) after excluding the multipotent single-nucleotide polymorphisms (SNPs). MI phenotypes were genetically correlated with all ischemic strokes (OR = 1.15, 95% CI: 1.03– 1.28, p = 0.013) and LAS (OR = 1.39, 95% CI: 1.14– 1.71, p = 0.001); but a causal effect of MI on all ischemic strokes (OR = 1.00, 95% CI: 0.95– 1.28, p = 0.219) and LAS (OR = 1.26, 95% CI: 0.93– 1.69, p = 0.130) was not observed after excluding the multipotent SNPs.Conclusion: This MR analysis provides evidence to support the causal effect of LAS subtype on MI, and some factors act as confiding factors whereas others may act as mediators.Keywords: causal inference, single-nucleotide polymorphism, genome-wide association study, epidemiologic methods, large-artery stroke