Association between cardiovascular disease risk, regional brain age gap, and cognition in healthy adults.
Sriya Pallapothu, Roger D Newman-Norlund, Nicholas Riccardi, Raghav Pallapothu, Pranesh Rajesh Kannan, Leonardo Bonilha, Julius Fridriksson, Chris Rorden
Abstract
Open AccessBackground: Cardiovascular disease (CVD) and its associated risk factors accelerate neurodegeneration and cognitive decline. This study examined relationships between CVD risk, cognition, and Brain Age Gap (BAG)-the difference between MRI-predicted brain age and chronological age. While prior research has linked CVD risk factors to global (i.e., "whole-brain") BAG, we extend these findings by examining region-specific associations, offering more spatially precise insights into brain aging across the cortex. Methods: Cross-sectional data from 187 participants in the University of South Carolina's Aging Brain Cohort (ABC) were analyzed. T1-weighted MRI scans were processed with volBrain, an automated brain volumetrics pipeline, to calculate global and regional BAG. CVD risk was assessed using the QRISK3 calculator, which provides a 10-year CVD risk percentage and Heart Age value. The Heart Age Gap (HAG) was calculated as Heart Age minus chronological age. Cognitive function was assessed using the Montreal Cognitive Assessment (MoCA). Six data-driven brain aging factors were identified, and participant-level BAG scores for each factor were analyzed. Spearman correlations examined associations between CVD risk metrics, regional BAG factors, and cognition, controlling for age and sex. Results: 10-year CVD risk and HAG were significantly correlated with global BAG (p < 0.001), even after adjusting for covariates. The BAGs of Factors 3-6 showed significant positive correlations with 10-year CVD risk and HAG, indicating region-specific vulnerability. Total MoCA was negatively associated with the BAGs of Factors 4-6. In addition, the Language Index was negatively correlated with the BAGs of Factors 1, 4, and 5, while the Executive Index was negatively associated with Factor 5's BAG. No CVD risk-cognition associations remained significant after adjusting for age. Conclusion: CVD risk is associated with global and regional brain aging, with specific cortical regions demonstrating greater vulnerability to CVD risk burden than others. These findings highlight the added value of regional BAG analyses, which reveal heterogeneity in aging patterns not captured by global estimates alone and may clarify vascular contributions to brain aging.