Examination of the clinicopathologic continuum of Alzheimer disease in the autopsy cohort of the National Alzheimer Coordinating Center

Abstract

To test the hypothesis that Alzheimer disease (AD) is a clinical and pathologic continuum between normal aging and end-stage dementia, we selected a convenience sample of subjects from the National Alzheimer Coordinating Center 2005 to 2012 autopsy cohort (n = 2,083) with the last clinical evaluation within 2 years before autopsy and no other primary neuropathologic diagnosis. Demographic and neuropathologic characteristics were correlated with the Clinical Dementia Rating-Sum of Boxes in the 835 subjects meeting these criteria. Both neuritic plaques and neurofibrillary tangles independently predicted Clinical Dementia Rating-Sum of Boxes. Severe small-vessel disease, severe amyloid angiopathy, and hippocampal sclerosis were also independently associated with the degree of cognitive impairment. By contrast, education was a strong independent protective factor against cognitive deficits. The cause of mild to moderate dementia remained uncertain in 14% of the patients. Inverse probability weighting suggests the generalizability of these results to nonautopsied cohorts. These data indicate that plaques and tangles independently contribute to cognitive impairment, that concurrent vascular disease strongly correlates with cognitive dysfunction even in a sample selected to represent the AD pathologic continuum, and that education further modifies clinical expression. Thus, multiple concomitant etiologies of brain damage and premorbid characteristics contribute to the uncertainty of AD clinicopathologic correlations based only on tangles and plaques.

FIGURE 1

Flowchart showing the selection procedure of study subjects. Of the 2,083 subjects from the 2005 to 2012 National Alzheimer Coordinating Center (NACC) autopsy cohort, 835 met all inclusion criteria and none of the exclusion criteria. Other primary neuropathologic diagnoses refer to conditions different from Alzheimer disease and included frontotemporal lobar degeneration, dementia with Lewy bodies, hippocampal sclerosis, vascular dementia, prion disease, Parkinson disease, Huntington disease, hypoxia, ischemia, necrosis, hemorrhage, and other non-neurodegenerative diagnosis. NP Dx, neuropathologic diagnosis.

FIGURE 2

Summary of the main results of this study (full model). Forest plot represents odds ratio ([OR] symbols) and 95% confidence interval ([CI] bars) of the association between each predictive variable and the outcome variable Clinical Dementia Rating scale–Sum of Boxes (CDR-SOB). Nonstatistically significant associations are depicted in black, statistically significant negative (protective) associations in green, and statistically significant positive (risk) associations in red. Two ORs (95% CI) are shown for female sex, one for the comparisons between the adjacent CDR-SOB categories 0, 0.5 to 3, 3.5 to 6, 6.5 to 12, and 12.5 to 17 (ρ) and the other for the comparison between the adjacent CDR-SOB categories 12.5 to 17 and 18 (ψ). Two ORs (95% CI) are shown for Braak NFT stages III/IV and V/VI, one for the adjacent CDR-SOB categories 0 and 0.5 to 3 (γ) and the other for the adjacent CDR-SOB categories 3.5 to 6, 6.5 to 12, 12.5 to 17, and 18 (ϕ). CAA, cerebral amyloid angiopathy; HScl, hippocampal sclerosis; NFT, neurofibrillary tangles; NPs, neuritic plaques; SVD, small-vessel disease (arteriosclerosis).