Cardiovascular Disease Risk Factors, Tract-Based Structural Connectomics, and Cognition in Older Adults

2018-11-27T00:00:00Z (GMT) by Elizabeth A Boots
Cardiovascular disease risk factors (CVD-RFs) are associated with decreased gray and white matter integrity and cognitive impairment in older adults. Less is known regarding the interplay between CVD-RFs, structural integrity between gray and white matter regions, and cognition. Using data from non-demented/non-depressed older adults, we examined whether CVD-RFs were associated with measures of tract-based structural connectivity; if alterations in connectivity mediated the association between CVD-RFs and cognition; and whether educational quality was protective against connectivity alterations. Ninety-six participants (age=67.9 years; 53.1% female; 46.9% Black) underwent CVD-RF assessment, MRI, and cognitive evaluation. Framingham 10-year stroke risk (FSRP-10) quantified CVD-RFs. Graph theory analysis integrated T1-derived gray matter regions of interest (ROIs) and DTI-derived white matter tractography into connectivity matrices, which were analyzed for local efficiency and centrality. PCA of cognitive variables resulted in three rotated factor scores: memory (CVLT-II Trial 1-5 Total, Delayed Free Recall, Recognition Discriminability); executive function (FAS, Trail Making Test (TMT) B-A, Letter-Number Sequencing, Matrix Reasoning); attention/information processing (AIP; TMT-A, TMT-M, Digit Symbol). Linear regressions adjusting for word reading and intracranial volume revealed associations between FSRP-10 and centrality in six ROIs (p-values<.05), efficiency in seven ROIs (p-values<.05), and AIP (p=.02). Analyses revealed mediation of right hippocampal centrality on FSRP-10 and AIP (p=.04) and left caudal middle frontal gyrus efficiency on FSRP-10 and AIP (p=.01). No protective effects of educational quality were noted. In sum, stroke risk plays deleterious roles in connectivity that negatively impact cognition, suggesting the importance of multi-modal neuroimaging biomarkers in understanding brain-behavior pathways in preclinical aging.