September 1997

Apolipoprotein E Phenotype and Cognitive Decline in a Prospective Study of Elderly Community Women

Author Affiliations

From the Departments of Psychiatry (Dr Yaffe), Medicine (Dr Browner), and Epidemiology and Biostatistics (Drs Yaffe and Browner), and Mt Zion Institute on Aging (Dr Sands), University of California, San Francisco; San Francisco Veterans Affairs Medical Center (Drs Yaffe and Browner); and Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pa (Dr Cauley).

Arch Neurol. 1997;54(9):1110-1114. doi:10.1001/archneur.1997.00550210044011

Objective:  To determine whether apolipoprotein E (Apo E) phenotype is associated with cognitive decline in community-dwelling nondemented elderly

Design:  Prospective cohort study.

Setting:  A university-affiliated clinic near Pittsburgh, Pa.

Patients:  A total of 1750 nondemented community-dwelling women, aged 65 years and older, who were enrolled in the Study of Osteoporotic Fractures.

Main Outcome Measures:  The women completed a baseline interview and performed 3 cognitive tests: the modified Mini-Mental State Examination, Trails B, and Digit Symbol. Serum samples were obtained for Apo E typing. Baseline cognitive scores and repeated scores approximately 6 years after study enrollment were compared in women with and without Apo E ε4. Cognitive decline, defined as the worst 10th percentile change scores, was assessed for each test and by phenotype group.

Results:  After adjustment for age, education, presence of severe tremor, and depression, baseline scores did not differ by Apo E ε4 status except for lower scores on Trails B in the homozygous ε4 group (mean score, 159.7 compared with 127.7 for the heterozygous ε4 group and 125.4 for the no ε4 group; P=.01). However, repeated test performance on follow-up examination was worse on all tests in those women with 1 or more ε4. Reduction on the modified Mini-Mental State Examination was 0% for no ε4 allele, 1.9% for 1 ε4 allele, and 3.7% for 2 ε4 alleles (P<.001); reduction on Digit Symbol was 6.2% for no ε4 allele, 9.0% for 1 ε4 allele, and 17.5% for 2 ε4 alleles (P=.04); and reduction on Trails B was 5.9% for no ε4 allele, 25.0% for 1 ε4 allele, and 10.9% for 2 ε4 alleles (P=.002). Women with at least 1 ε4 had an odds ratio of 1.6 (95% confidence interval, 1.1-2.3) of having cognitive decline during the study period.

Conclusion:  Apolipoprotein E ε4 is associated with cognitive decline in community-dwelling nondemented women.