Bidirectional 2-sample Mendelian randomization study of the causal relationship between hypothyroidism and cardiovascular diseases.

A bidirectional 2-sample Mendelian randomization (MR) analysis was used to investigate the causal relationship between hypothyroidism and multiple types of cardiovascular diseases (CVDs). Single nucleotide polymorphisms strongly associated with hypothyroidism and 4 CVDs (myocardial infarction, coronary artery disease, heart failure [HF], hypertension) were extracted from genome-wide association study summary data as instrumental variables, excluding linkage disequilibrium. First, a forward MR analysis assessed causal effects of hypothyroidism on cardiovascular outcomes using inverse-variance weighted, weighted median, MR-Egger regression, simple mode, and weighted mode methods. Heterogeneity was evaluated via inverse-variance weighted and MR-Egger Q-tests, pleiotropy via Egger intercept test, and sensitivity via leave-one-out analysis. Reverse MR tested causality from CVDs to hypothyroidism. A bidirectional MR framework ensured robustness, with comprehensive statistical validations (heterogeneity, pleiotropy, sensitivity) to minimize confounding bias and confirm reliability of causal inferences. Genetically predicted hypothyroidism causally increased risks of myocardial infarction (odds ratio [OR] = 3.003, P = .002), coronary heart disease (OR = 2.780, P = .001), HF (OR = 1.892, P = .009), and hypertension (OR = 1.102, P = .024). Reverse MR revealed HF reciprocally elevated hypothyroidism risk (OR = 1.006, P = .040). Analyses confirmed absence of pleiotropy (all P > .05) and robustness through leave-one-out sensitivity analysis. Causal relationships remained consistent across sensitivity methods, supporting thyroid dysfunction as cardiovascular risk modifier in bidirectional pathways. This study affirms that hypothyroidism heightens the susceptibility to multiple types of CVDs, and establishes a causal link between HF and hypothyroidism.