Abstract
OBJECTIVES: To examine the association between vascular pathology and rate of cognitive decline in older adults independent of Alzheimer's disease (AD) pathology. DESIGN: Prospective cohort study. SETTING: Community sample. PARTICIPANTS: Individuals from the Einstein Aging Study autopsy series (N = 62). MEASUREMENTS: The Blessed Information-Memory-Concentration (BIMC) test was used to assess global cognitive status. AD pathology was quantified according to Braak stage (<3 vs ≥ 3). Vascular pathology was quantified using a previously reported macrovascular lesion (MVL) score. The association between vascular pathology and antemortem rates of cognitive decline adjusted for level of AD pathology was assessed using linear mixed-effects models. RESULTS: Mean age was 81.8 at enrollment and 89.0 at death. Participants with more than two MVLs had faster cognitive decline than those with no MVLs (difference in annual rate of change in BIMC 0.74 points/yr, P = .03). Braak stage was also associated with cognitive decline (difference 0.57 points/yr, P = .03). The difference in rate of cognitive decline between those with more than two MVLs and those free of vascular lesions persisted after adjustment for AD pathology (difference in rate of change in BIMC 0.68 points/yr, P = .04). The effect of vascular pathology on cognitive decline was not significantly different according to AD pathology. CONCLUSION: Vascular brain pathology is associated with rate of cognitive decline after adjusting for level of AD pathology.