Causality between Alzheimer disease and delirium: A two-sample Mendelian randomization study and gene colocalization analyses

To evaluate the possible causality between Alzheimer disease (AD) and delirium, a two-sample Mendelian randomization (MR) method and gene colocalization strategy were employed. Genome-wide association study (GWAS) data accessible to the public were utilized. The preliminary findings revealed a causal link between AD (21,982 cases and 41,944 controls) and delirium (comprising 1269 cases and 209,487 controls). By performing MR estimates on an expanded sample, which included AD (90,338 cases and 1036,225 controls) as well as delirium (1269 cases and 209,487 controls), the subsequent analysis confirmed the results. The approaches utilized comprised Inverse-variance weighted (IVW), MR-Egger, weighted median, and weighted mode. To evaluate heterogeneity, Cochran Q test was employed, while the MR-Egger intercept and MR-PRESSO tests assessed pleiotropy. Sensitivity was evaluated via leave-one-out analysis. Colocalization results revealed shared genetic loci. The analysis using MR indicated a notable causal influence of AD on delirium occurrence (initial: OR = 1.319, 95% CI: 1.175–1.481, P < .001; replication: OR = 1.334, 95% CI: 1.204–1.479, P < .001). Meta-analysis confirmed this effect (OR = 1.320, 95% CI: 1.210–1.440, P < .01). No heterogeneity or pleiotropy was detected, and results remain robust in sensitivity analyses. Colocalization analysis showed strong shared genetic signals (PPH4 > 0.8), identifying NECTIN2 and TOMM40 as potential mediators. No evidence of reverse causality was identified linking delirium to AD. AD poses a risk for delirium, probably mediated by shared genetic variants NECTIN2 and TOMM40 that impact acetylcholine pathways. These findings highlight potential preventive targets for delirium in Alzheimer patients.