To explore long-term predictors of avoiding β-amyloid (Aβ) deposition and maintaining unimpaired cognition as outcomes in the oldest old.
In a longitudinal observational cohort study, 100 former participants of the Ginkgo Evaluation of Memory Study (GEMS; 2000–2008) completed biannual Pittsburgh compound B-PET imaging and annual clinical-cognitive evaluations beginning in 2010. Most recent Aβ status and cognitive status were selected for each participant. Longitudinal outcomes included change in serial Aβ and cognitive tests. Baseline predictors from GEMS included neuropsychological tests, daily functioning, APOE genotype, lifestyle variables, occupational measures, health history, sleep, subjective memory, physical and cognitive activities, depressive symptoms, and physical performance and health indices, among others.
Mean age at the last cognitive evaluation was 92.0 (range 86–100) years. Mean follow-up time from baseline to last measured Aβ status was 12.3 (SD 1.9) years and to last cognitive evaluation was 14.1 (SD 1.9) years. The APOE*2 allele predicted last Aβ status (n = 34 Aβ negative vs n = 66 Aβ positive). Baseline cognition predicted cognitive status (n = 30 unimpaired vs n = 70 impaired). Predictors of cognitive status among Aβ-positive participants only (n = 14 normal cognition vs n = 52 impaired) were baseline cognitive test scores and smoking history. Baseline pulse pressure predicted longitudinal Aβ increase; paid work engagement and life satisfaction predicted less cognitive decline.
The APOE*2 allele and lower pulse pressure predict resistance to Aβ deposition in advanced aging. Cognitive test scores 14 years prior, likely reflecting premorbid abilities, predict cognitive status and maintenance of unimpaired cognition in the presence of Aβ. Several lifestyle factors appear protective.