パーソナリティ障害と心血管代謝疾患：メンデルランダム化研究

Observational studies have suggested the associations between personality disorders (PDs) and cardiometabolic diseases (CMDs), though these may be biased by unmeasured confounding, including lifestyle factors or reverse causality. Mendelian randomization (MR) provides a novel approach to address these limitations by using genetic variants as instrumental variables, thereby minimizing environmental confounding and enabling causal inference. This study leverages MR to systematically evaluate causal relationships across 9 clinically relevant PDs and 8 major CMDs, offering genetic evidence to clarify previously observed epidemiological associations. Data for 9 kinds of PDs (including anankastic personality disorder, anxious personality disorder, dependent personality disorder, dissocial personality disorder, emotionally unstable personality disorder, histrionic personality disorder, paranoid personality disorder, organic personality disorder, and schizoid personality disorder) were obtained from the FinnGen consortium R10 release and data for 8 kinds of CMDs (including myocardial infarction, atrial fibrillation, type 2 diabetes, hypertension, stroke, coronary atherosclerosis, ischemic stroke and peripheral artery disease) were obtained from UK biobank. Instrumental variables were selected using single-nucleotide polymorphisms meeting a genome-wide significance threshold (P < 5 × 10⁻⁶) and clumped (R2 < 0.001) to ensure independence. Primary causal estimates were generated via inverse-variance weighted regression, with sensitivity analyses MR-Egger, Cochran Q intercept analysis, and leave-one-out analysis to assess robustness. Genetically predicted organic personality disorder had a positive association with coronary atherosclerosis (OR = 1.001, 95% confidence interval, 1.000-1.002, P = .029). No other significant differences (P > .05) were found. Results were stable across sensitive analyses. Our findings contrast with prior observational studies, suggesting that reported associations may reflect residual confounding from environmental or behavioral risk factors, rather than biological causation. Future research should investigate behavioral or environmental mediators underlying these epidemiological correlations.