Genetic contributors to postoperative delirium and their implications for dementia outcomes

BACKGROUND: Postoperative delirium (POD) is a perioperative neurocognitive disorder that substantially impairs patient recovery. Unfortunately, its genetic risk profile and relationship with subsequent dementia remain unclear. We aimed to elucidate genetic contributors to POD identified via Hospital Episode Statistics codes and to examine its association with subsequent dementia. METHODS: We included 230,179 non-cardiac and 21,254 cardiac surgery subjects from the UK Biobank, defining POD using ICD-10 delirium codes recorded within the first 7 postoperative days. Genome-wide association studies (GWAS) were performed in the non-cardiac and cardiac cohorts and their prespecified subgroups, followed by functional annotation, gene prioritization and drug-target analyses. Associations between POD and subsequent dementia were estimated using Cox models. RESULTS: In the non-cardiac cohort, we identified one genome-wide significant locus at the APOE region, with rs429358 as the lead variant (P=5.00×10-28). Integrative gene prioritization analyses highlighted multiple genes within this locus. Exploratory drug-target analyses suggested potential subgroup-specific drug-target enrichment. In the cardiac cohort, no genome-wide significant signals were detected. POD was associated with all-cause dementia after both non-cardiac (HR=6.45, 95% CI 5.45-7.63) and cardiac (HR=2.95, 95% CI 1.71-5.08) surgeries. CONCLUSIONS: This study demonstrates APOE as a genetic risk locus for ICD-coded POD in the non-cardiac surgery setting and confirms an association between POD and subsequent dementia.