Cholesterol and neuropathologic markers of AD - A population-based autopsystudy

Objective: To examine the association of plasma cholesterol (total and high -density [HDL] and low-density lipoprotein) levels with neuritic plaques (N P) and neurofibrillary tangles (NFT) in a population-based autopsy series o f 218 Japanese American men followed as a part of the Honolulu-Asia Aging S tudy. Methods: Cholesterol levels were measured in late life (average age a t death 84.6 years) in all subjects (n = 218) and in midlife (20 years befo re late life) in a subsample (n = 89); for the analyses, levels were catego rized into quintiles, with the lowest quintile serving as the reference. Ti ssue from four areas of neocortex and two areas of hippocampus was prepared with Bielschowsky silver-stained sections and evaluated by one of three ne uropathologists who were blinded to clinical information. Diffuse and neuri tic plaques and NFT were counted in field areas standardized to 1 mm(2). Fi elds were selected from areas with the highest numbers of lesions, and the field with the highest count was taken to represent the brain area. Results : After adjusting for age at death, education, APOE allele, dementia, neuro pathologic infarction, and blood pressure, a strong linear association was found for increasing late-life HDL cholesterol (HDL-C) levels and an increa sing number of neocortical NP (5th versus 1st quintile: count ratio [95% CI ] 2.30 [1.05 to 5.06]) and hippocampal (2.63 [1.25 to 5.50]) and neocortica l (4.20 [1.73 to 10.16]) NFT. Trends were similar for the midlife HDL-C lev els. Conclusions: The constituents of HDL-C may play a role in the formatio n of AD pathology, and these processes are reflected in peripheral measures .