Premium
Physical activity is associated with increased resting‐state fcMRI in networks predictive of cognitive decline in clinically unimpaired older adults
Author(s) -
Pruzin Jeremy J.
Publication year - 2021
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.049869
Subject(s) - cognitive decline , cognition , context (archaeology) , effects of sleep deprivation on cognitive performance , psychology , dementia , prospective cohort study , gerontology , medicine , confounding , resting state fmri , neuroscience , disease , biology , paleontology
Background While physical activity (PA) may promote resilience to cognitive decline and AD dementia, the mechanisms underlying this effect are poorly understood. Our group previously demonstrated lower levels of functional connectivity (rs‐fcMRI) in the default, salience, and control networks were associated with greater prospective cognitive decline, both alone and synergistically with β‐amyloid (Buckley and Schultz et al., 2017). In this context, we assessed whether higher PA may be associated with greater connectivity, and whether PA effects on connectivity may mediate the relationship of PA to prospective cognitive decline. Method A waist mounted pedometer was used to record PA (steps/day during one week) in 167 cognitively unimpaired adults aged 63‐90 (mean 74.0; SD 6.1) participating in the Harvard Aging Brain Study. Participants underwent baseline rs‐fcMRI and β‐amyloid imaging with Pittsburgh Compound B PET. We analyzed rs‐fcMRI in the 3 networks previously associated with cognitive decline (default, salience, and control) as well as in 4 motor or sensory networks in which connectivity strength is not associated with cognitive change. Cognition was measured annually with the Preclinical Alzheimer Cognitive Composite (PACC). We used linear regression to examine associations between PA and connectivity in these networks, controlling for age, sex, and APOE e4 status. We conducted exploratory analyses using linear mixed effects models to examine if associations between PA and connectivity strength may partially explain PA associations with cognitive decline in the context of elevated β‐amyloid. Result Higher baseline PA was significantly associated with increased connectivity in the default, salience, and control networks (all p<0.017) after adjusting for covariates and controlling for multiple comparisons. PA was not associated with connectivity in any of the other networks examined. Greater connectivity in these networks was a weak partial mediator of PA's association with longitudinal PACC decline in participants with elevated β‐amyloid (figure 1). Conclusion Greater PA is associated with increased connectivity in three networks (default, salience, and control) in which greater connectivity is associated with less cognitive decline. One potential mechanism by which PA may promote resilience to cognitive decline is through increased functional connectivity in a core set of AD‐relevant cognitive networks.