Divergent functional connectivity patterns of brain network in patients with or without post-stroke dementia : A resting-state fMRI study

Abstract

Recently, accumulating studies have reported brain functional abnormalities in patients with post-stroke dementia (PSD). Nonetheless, our understanding about neural mechanisms of PSD in the perspective of functional connectivity alternations is still rudimentary. Here, we applied graph theory to investigate brain network from resting-state fMRI data of 14 PSD patients, 15 post-stroke non-dementia (PSND) patients, and 14 healthy subjects with normal cognitive function (NC). Five global metrics [clustering coefficient, characteristic path length, global efficiency (E <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">global</inf> ), local efficiency (E <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">local</inf> ), and small-world properties (σ)] and one regional metric [nodal efficiency (E <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">nodal</inf> )] were compared to explore the alterations of brain network connectivity patterns among the three groups. In terms of global network metrics, we found that all three groups maintained economical σ. However, significantly disrupted functional connectivity of brain network were observed in PSD patients compared to NC, represented by significantly declined σ, E <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">global</inf> , and E <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">local</inf> . In addition, significantly declined E <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">nodal</inf> at specific regions of inferior frontal gyrus and temporal lobe further revealed the altered connectivity patterns of brain network in PSD patients. Our findings may provide salient neuroimaging biomarkers for screening groups at high risk of PSD and searching for new treatment targets of neuromodulation in the future.