APOEε4, in-hospital delirium and long-term cognitive impairment: A longitudinal memory clinic study

Delirium is common among hospitalized elderly. Previous short-term studies reported inconsistent associations between APOEε4 allele, in-hospital delirium, and post-delirium cognitive impairment. We examined the association of APOEε4 allele with in-hospital delirium and long-term cognitive outcomes following delirium. The electronic medical records were linked to the Korean National Health Insurance Service database of all citizens from January 2002 to July 2019. The study population consisted of 1057 memory clinic visitors with APOE genotype, longitudinal neuropsychological tests, and hospitalization records. Incident in-hospital delirium was defined as the initiation of antipsychotics during hospitalization after excluding prevalent users. Incidence analysis was conducted using Cox proportional hazards models, while longitudinal outcomes were analyzed using multivariable mixed models with an interrupted time series design. At baseline, APOEε4 carriers (N = 298, 28.2%) performed poorly on cognitive tests compared to non-carriers (CDR-SB mean±SD: 3.3 ± 3.5 vs 2.8 ± 2.9, P = 0.016; MMSE 22.3 ± 5.8 vs 23.2 ± 5.2, P = 0.029). The carriers developed more in-hospital delirium than noncarriers after covariate adjustments (HR 1.96, 95%CI 1.30–2.96, P = 0.002). The APOEε4 allele also had a more detrimental impact on four out of the five cognitive and functional measurements after the delirium (beta estimates of post-delirium change by APOEε4 for CDR-SB = 3.20, P < 0.001; CDR = 0.60, P < 0.001; KIADL = 0.99, P < 0.001; SIADL = 14.07, P < 0.001). These findings remained robust even after adjusting for covariates. APOEε4 carriers demonstrated robust associations with in-hospital delirium and exhibited more post-delirium cognitive and functional impairment compared to non-carriers. Individuals with APOEε4 allele may need more attention to prevent in-hospital delirium and post-delirium cognitive and functional deterioration.