Hypertension as a Common Cause of Ulcerative Colitis, Intracerebral Hemorrhage, and Major Depressive Disorder: A Causal Inference Study Integrating Cross-Sectional Analysis, Bidirectional Mendelian Randomization, and Drug-Target Genetic Proxies

Abstract

Patients with intracerebral hemorrhage (ICH) are often observed to have comorbid ulcerative colitis (UC), but whether this reflects a direct causal relationship or confounding by a common cause remains unclear. We aimed to evaluate the causal network among UC, ICH, hypertension, and major depressive disorder (MDD), testing two competing explanations: a shared common cause and a medication-related effect. We conducted a descriptive, hypothesis-generating cross-sectional analysis using NHANES 2009–2010 data to assess phenotypic associations. Bidirectional two-sample Mendelian randomization (MR) served as the primary causal inference tool using GWAS summary data from multiple consortia, with inverse variance weighting (IVW) as the primary estimator and meta-analysis across databases. Drug-target MR was further conducted using genetic proxies for antihypertensive drug targets, with FDR correction applied. In the descriptive NHANES analysis, hypertension and UC were phenotypically associated in the unadjusted model but not after covariate adjustment, limited by small UC event numbers. Bidirectional MR showed significant causal signals from genetically predicted hypertension to UC, ICH, and MDD, although ICH estimates were imprecise due to potential weak instrument bias. MDD showed a statistically significant but clinically limited positive signal for hypertension. No causal association was detected between UC and ICH. Drug-target MR suggested nominally significant, hypothesis-generating signals: CACNB2 proxies were associated with lower UC risk; ADRB1 and CACNA1D with lower ICH risk; and SLC12A2 and CACNA1D with higher MDD risk. These signals lacked colocalization validation and did not uniformly survive FDR correction. Genetic evidence is consistent with hypertension as a shared causal risk factor for UC, ICH, and MDD, rather than a direct causal pathway between UC and ICH. The clinical co-occurrence of UC and ICH is thus more likely confounded by hypertension. Although genetic proxies for antihypertensive drug targets did not increase UC or ICH risk, they were associated with increased MDD risk. Mental health monitoring should be integrated into hypertension management. Drug-target findings remain exploratory and require validation through colocalization analysis and clinical trials.