Apolipoprotein E phenotypes, dementia and mortality in a prospective population sample

OBJECTIVE: To study the relationships between apoE phenotypes, dementia, and mortality. SETTING: A population-based study in Helsinki, Finland (the Helsinki Ageing Study). DESIGN: A prospective birth cohort study with 5-year follow-up. PARTICIPANTS: A total of 550 subjects of three birth cohorts of 75 (n = 182), 80 (n = 185), and 85 (n = 183) years of age. MEASUREMENTS: ApoE phenotype was determined from baseline blood samples. The cognitive function of the subjects was tested at baseline and at a 5-year follow-up using the Mini-Mental State Examination (MMSE) and the Clinical Dementia Rating (CDR). Diagnosis and type of dementia were determined by a neurologist. The cohorts were followed for 5 years, and causes of death were determined. Cox proportional hazards model was used for survival analyses. Analyses were performed comparing the apoE e4 allele and others. RESULTS: At baseline, the apoE e4 allele was found in 148 of 550 subjects (27%), in 24% of nondemented persons, in 51% of patients with probable or uncertain Alzheimer's disease (AD), and in 34% patients with vascular dementia. The CDR score was worse among subjects with an e4 allele compared with others at baseline (P < .001) and after a 5-year follow-up (P = .007). The crude mortality rates of subjects with and without an e4 allele were 48% (n = 71) and 37% (n = 148), respectively. After controlling for age and gender, the hazard ratio of an e4 allele was 1.61 (95% CI, 1.21-2.14) for all-cause mortality, deaths caused by dementia 2.20 (95% CI, 1.03-4.72), and presence of AD 3.24 (95% CI, 1.67-6.25). CONCLUSIONS: In a population aged 75 to 85 years, the presence of an apoE e4 allele is associated with impaired cognitive function, clinical dementia, AD, and excess 5-year mortality resulting from dementia and all causes.