Changes in brain functional connectivity and cognition related to white matter lesion burden in hypertensive patients from SPRINT

• Published in: Neuroradiology Vol. 63; no. 6; pp. 913 - 924
• Main Authors: Shah, Chintan, Srinivasan, Dhivya, Erus, Guray, Schmitt, James E., Agarwal, Adhish, Cho, Monique E., Lerner, Alan J., Haley, William E., Kurella Tamura, Manjula, Davatzikos, Christos, Bryan, Robert N., Fan, Yong, Nasrallah, Ilya M.
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.06.2021; Springer Nature B.V
• Subjects: Blood Pressure; Brain; Brain - diagnostic imaging; Cognition; Cognition & reasoning; Cognitive ability; Cortex (parietal); Cross-Sectional Studies; Functional magnetic resonance imaging; Functional Neuroradiology; Humans; Hypertension; Hypertension - diagnostic imaging; Imaging; Independent component analysis; Lesions; Magnetic Resonance Imaging; Mediation; Medicine; Medicine & Public Health; Networks; Neural networks; Neurology; Neuropsychological Tests; Neuroradiology; Neurosciences; Neurosurgery; Radiology; Regression analysis; Regression models; Risk analysis; Risk factors; Substantia alba; White Matter - diagnostic imaging; Hypertension; Functional connectivity; White matter lesions; Cognitive impairment
• ISSN: 0028-3940; 1432-1920; 1432-1920
• DOI: 10.1007/s00234-020-02614-6

Purpose Hypertension is a risk factor for cognitive impairment; however, the mechanisms leading to cognitive changes remain unclear. In this cross-sectional study, we evaluate the impact of white matter lesion (WML) burden on brain functional connectivity (FC) and cognition in a large cohort of hypertensive patients from the Systolic Blood Pressure Intervention Trial (SPRINT) at baseline. Methods Functional networks were identified from baseline resting state functional MRI scans of 660 SPRINT participants using independent component analysis. WML volumes were calculated from structural MRI. Correlation analyses were carried out between mean FC of each functional network and global WML as well as WML within atlas-defined white matter regions. For networks of interest, voxel-wise-adjusted correlation analyses between FC and regional WML volume were performed. Multiple variable linear regression models were built for cognitive test performance as a function of network FC, followed by mediation analysis. Results Mean FC of the default mode network (DMN) was negatively correlated with global WML volume, and regional WML volume within the precuneus. Voxel-wise correlation analyses revealed that regional WML was negatively correlated with FC of the DMN’s left lateral temporal region. FC in this region of the DMN was positively correlated to performance on the Montreal Cognitive Assessment and demonstrated significant mediation effects. Additional networks also demonstrated global and regional WML correlations; however, they did not demonstrate an association with cognition. Conclusion In hypertensive patients, greater WML volume is associated with lower FC of the DMN, which in turn is related to poorer cognitive test performance. Trial registration NCT01206062