Protective effects of sleep duration and physical activity on cognitive performance are influenced by β‐amyloid and brain volume but not tau burden among cognitively unimpaired older adults

Abstract

AbstractBackgroundSleep and physical activity have gained traction as modifiable risk factors for Alzheimer’s disease (AD). Sleep duration is linked to amyloid‐β (Aβ) clearance while physical activity is associated with brain volume maintenance. We investigate associations between sleep duration and physical activity and cognition by testing if sleep duration—cognition and physical activity—cognition are explained by Aβ deposition and brain volume, respectively. Additionally, we explore the mediating role of tau burden in these relationships.MethodThis cross‐sectional study obtained data from participants in the Anti‐Amyloid Treatment in Asymptomatic Alzheimer’s Disease (A4) study, a randomized clinical trial. Cognitively unimpaired participants underwent Aβ and tau PET and brain MRI prior to randomization. Cognitive performance was assessed using the Preclinical Alzheimer Cognitive Composite (PACC). Self‐reported nightly sleep duration and weekly physical activity were the primary predictors. Regional Aβ and tau pathologies and volumes were the proposed variables influencing sleep duration—cognition or physical activity—cognition relationships.ResultAβ PET data were obtained from 4322 participants (1208 with MRI, 59% female, 29% amyloid positive). Sleep duration was associated with an Aβ composite burden (β = ‐0.005,CI:(‐0.01,‐0.001)) and Aβ burden in the anterior cingulate (ACC) (β = ‐0.012,CI:(‐0.017,‐0.006)) and medial orbitofrontal cortices (MOC) (β = ‐0.009,CI:(‐0.014,‐0.005)) but not with brain structure. Composite (β = ‐1.54,CI:(‐1.93,‐1.15)), ACC (β = ‐1.22,CI:(‐1.54,‐0.90)) and MOC (β = ‐1.44,CI:(‐1.86,‐1.02)) Aβ burden were also associated with PACC. Sleep duration—PACC association was explained by Aβ burden in path analyses (Figure 1). Physical activity was not associated with Aβ but was associated with hippocampal (β = 10.57,CI: (1.06,20.08)), parahippocampal (β = 9.3,CI:(1.69,16.91)), entorhinal (β = 14.68,CI:(1.75,27.61), and fusiform gyral (β = 38.38,CI:(5.57, 71.18)) volumes, which were also positively associated with PACC (p<0.02 for hippocampus, entorhinal cortex, and fusiform gyrus). Physical activity—cognition relationship was explained by regional volumes (Figure 2). Tau was not related to sleep duration or physical activity, and it did not moderate relationships between cognition and sleep duration or physical activity.ConclusionSleep duration and physical activity are associated with cognition through independent paths of brain Aβ and brain volume, respectively, but not tau burden. Dementia risk reduction approaches that emphasize the adequate sleep duration and a physically active lifestyle may benefit those with risk for AD.