Cognitive reserve proxies, Alzheimer pathologies, and cognition

Abstract

This study aimed to explore the moderating effects of the frequently used cognitive reserve (CR) proxies [i.e., education, premorbid intelligence quotient (pIQ), occupational complexity (OC), and lifetime cognitive activity (LCA)] on the relationships between various in vivo Alzheimer's disease (AD) pathologies and cognition. In total, 351 [268 cognitively unimpaired (CU), 83 cognitive impaired (CI)] older adults underwent multi-modal brain imaging to measure AD pathologies and cognitive assessments, and information on CR proxies was obtained. For overall participants, only education moderated the relationship between Aβ deposition and cognition. Education, pIQ, and LCA, but not OC, showed moderating effect on the relationship between AD-signature cerebral hypometabolism and cognition. In contrast, only OC had a moderating effect on the relationship between cortical atrophy of the AD-signature regions and cognition. Such moderation effects of the CR proxies were similarly observed in CI individuals, but most of them were not in CU individuals. The findings suggest that the proposed CR proxies have different moderating effects on the relationships between specific AD pathologies and cognition.

Keywords: Alzheimer pathology; Cognitive activity; Cognitive reserve; Education; Intelligence; Occupation.

Fig. 1.

Visual representations showing the relationships between CR proxies, AD pathologies, and cognition. (A) Moderation of education for the relationship between Aβ deposition and CERAD TS. Moderation of (B) education, (C) pIQ, and (D) LCA for the relationship between AD-CM and CERAD-TS. (E) Moderation of OC for the relationship between AD-CT and CERAD-TS. Each of education, pIQ, and LCA was stratified into tertiles: Educational level was stratified into high (≥ 13 years), middle (10~12 years), and low (≤ 9) levels; pIQ was divided into high (> 120), middle (115~120), and low (≤ 114) levels; and LCA was stratified into high (> 2.6), middle (2.0~2.6), and low (≤ 1.9) levels. OC was stratified into high (3~4), middle (2), low (0~1) levels. Age, sex, and APOE ε4 carrier status are controlled as covariates. *p < 0.05; **p < 0.05; ***p < 0.001. Key: CERAD, Consortium to Establish a Registry for Alzheimer’s Disease; CERAD TS, Total Score of CERAD neuropsychological assessment battery; Aβ, amyloid-beta; AD-CM, AD-signature region glucose metabolism; AD-CT, AD-signature region cortical thickness; pIQ, premorbid Intelligence Quotient; LCA, lifetime cognitive activity; OC, occupational complexity; APOE ε4, apolipoprotein E ε4.