Relation of neuropathology to cognition in persons without cognitive impairment

Abstract

Objective: A study was undertaken to examine the relation of Alzheimer disease (AD) pathology, cerebral infarcts, and Lewy body (LB) pathology to cognition in persons without cognitive impairment.

Methods: Persons without dementia from 2 cohort studies of aging, the Religious Orders Study and the Memory and Aging Project, agreed to annual clinical evaluation and brain donation. The studies had 19 neuropsychological performance tests in common that assessed 5 cognitive domains. For 296 persons without cognitive impairment who died and underwent postmortem assessment, we quantified AD pathology as a global pathology score, and as amyloid load, paired helical filament tau-positive (PHFtau) tangle density, cerebral infarcts, and LB pathology. Linear regression was used to examine the relation of neuropathology to cognitive abilities, controlling for demographics.

Results: Nearly all persons had AD pathology with >¾ exhibiting amyloid; 22% had macroscopic and 24% had microscopic infarctions, and 13% had LB pathology. The global measure of AD pathology was related to global cognition (p = 0.008), whereas infarcts and Lewy bodies were not. Amyloid load was related to global cognition (p < 0.05), with only a trend for tangles (p = 0.08). In analyses of cognitive domains, AD pathology (p = 0.006), PHFtau tangles (p = 0.03), and macroscopic infarctions (p = 0.02) were related to episodic memory, with a trend for amyloid load (p = 0.06); AD pathology (p = 0.02) and amyloid load (p = 0.03) were related to working memory. Findings for global cognition and episodic memory were stronger in additional analyses with neocortical amyloid and mesial temporal tangles.

Interpretation: AD pathology and macroscopic infarctions are common in older persons without cognitive impairment and are related to episodic and working memory.

Figure 1

Relation of global pathology to global cognition (A, left) and to episodic memory (B, right) from regression models controlling for age, sex, education, macroscopic and microscopic infarctions, and Lewy bodies.

Figure 2

Box plots showing the relation of macroscopic infarctions to episodic memory.

Figure 3

Relation of neocortical amyloid to global cognition (A, top left) and to episodic memory (B, top right), and of mesial temporal lobe PHFtau tangles to global cognition (C, bottom left) and to episodic memory (D, bottom right), from regression models controlling for age, sex, education, macroscopic and microscopic infarctions, and Lewy bodies.