Premium
Dynamic functional connectivity patterns associated with Alzheimer’s disease risk and protective factors
Author(s) -
Dautricourt Sophie
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.051782
Subject(s) - default mode network , dynamic functional connectivity , cognition , resting state fmri , psychology , effects of sleep deprivation on cognitive performance , human connectome project , disease , cognitive decline , alzheimer's disease , cohort , neuroscience , medicine , audiology , functional connectivity , dementia
Background Regions of high functional connectivity (hubs) are particularly sensitive to Alzheimer’s disease (AD). Dynamic functional connectivity (DFC) recently emerged as a powerful technique to reflect changes in functional connectivity over a short time. In this study, we used resting‐state DFC to investigate whether transient connectivity configurations are associated with AD risk and protective factors. Method We collected resting state fMRI data in 127 healthy older adults from the Age‐Well cohort. We estimated DFC using a sliding window approach to identify the transient connection between the default mode network (DMN), the salience network (SN) and the executive control network (CEN) and identify different states using k‐mean clustering. Stepwise regression analyses were used to assess the relationship between the time spent in each state, the number of transitions between states and AD risk factors and biomarkers, including cardiovascular factors, depression, lifestyle (Lifetime of Experiences Questionnaire (LEQ), Cognitive Activity Questionnaire (CAQ)), cognition (Preclinical Alzheimer’s Cognitive Composite 5 (PACC5)), neuroimaging biomarkers (amyloid burden, glucose metabolism and hippocampal volume) and APOE4 status. Result We identified four states (Figure 1). Longer time spent in a configuration characterized by strong connections between all networks (State 1), was associated with higher scores in the CAQ (β=0.27, P =0.001) and PACC5 (β=0.25, P =0.011). More time spent in State 2, where the DMN is disconnected from the other networks, was associated with higher scores in the LEQ (β=0.21, P =0.015) and absence of hypertension or cholesterol. The time spent in a configuration where all networks are weakly connected (State 3), was associated with lower scores in the CAQ (β=‐0.23, P =0.001) and LEQ (β=‐0.22, P =0.007). The time spent in State 4, where the SN is disconnected from the other networks, was associated with lower hippocampal volume (β=‐0.19, P =0.040). Finally, higher number of transitions between states was associated with higher amyloid burden (β=0.21, P =0.025). Conclusion Distinct DFC patterns are differentially associated with AD risk. We were able to identify two DFC states associated with increased AD risk and two states associated with AD protective factors. These findings shed new lights on the dynamic neural mechanisms underlying reserve and vulnerability in healthy older adults.