The Role of Cardiorespiratory Fitness in White Matter and Cognition in Healthy Aging

Abstract

While certain cognitive domains remain relatively stable into old age in the absence of pathological processes, other domains are selectively vulnerable to healthy aging, including processing speed, executive functions, and memory. These age-related cognitive differences may be in part due to alterations in white matter responsible for propagating neural signals and efficiently connecting brain regions. Cardiorespiratory fitness (CRF) is an important indicator of vascular health that may mitigate age effects on cognition by protecting against age-related white matter degradation. In this study, I sought to evaluate the differential impact of CRF on tract-specific white matter integrity (WMI) and global free water (FW) as well as subsequent associations with cognitive aging in a cohort of healthy community-dwelling older adults. First, I identified regional network covariance patterns of CRF-related WMI across four diffusion metrics, axial diffusivity (AD), radial diffusivity (RD), mean diffusivity (MD), and fractional anisotropy (FA). The resulting network patterns were characterized by enhanced integrity related to higher CRF in white matter bundles connecting anterior brain regions to one another, anterior to posterior regions, and subcortical regions. I additionally investigated associations of these patterns with important demographic and vascular health characteristics, including age, sex, macrostructural white matter hyperintensity (WMH) lesion load, and vascular risk level. Age and WMH volume were significantly negatively associated with expression of all four CRF-related WMI network patterns while male sex was significantly associated with higher expression of the CRF-AD and CRF-FA patterns. Vascular risk level also demonstrated a significant relationship with the CRF-RD, CRF-MD, and CRF-FA network patterns, such that increased vascular risk burden was associated with reduced expression of these patterns. Second, I assessed the potential mediating effects of these CRF-related WMI network patterns in a pathway representing vascular impacts on cognitive aging. These results demonstrated significant partial mediating effects of the CRF-RD and CRF-MD patterns on age-related processing speed, but not executive function or memory. Specifically, I found that older age was sequentially associated with greater WMH lesion load, which was further associated with reduced expression of the CRF-RD and CRF-MD patterns. Reduced expression of these patterns, in turn, predicted worse processing speed performance. Third, I evaluated the impact of CRF on the relationship between cognition and a global metric of neural degradation, FW in white matter. Higher levels of FW were only predictive of worse processing speed and executive function at low (i.e., below average) levels of CRF, but not for memory. These results suggest that CRF may protect against brain-based cognitive differences in healthy aging at average and elevated levels of fitness. Further, differences in self-reported physical activity levels did not impact these effects. Altogether, these findings add to our understanding of how vascular health may impact structural brain differences in white matter that, in part, underly the effects of cognitive aging. They additionally highlight the importance of investigating both regional and global impacts of CRF on brain markers of cognitive aging.