Brain perivascular space imaging across the human lifespan

Enlarged perivascular spaces (PVS) are considered a biomarker for vascular pathology and has been observed in several neurological disorders and in normal aging; however, the age-related changes in PVS structure across the normative lifespan are not well understood. We sought to characterize the relationship between PVS anatomical characteristics and age in a large cohort (~1400) between 8 and 90 years of age from the Human Connectome Project (HCP) Lifespan datasets. PVS from the white matter regions and the basal ganglia (BG) were automatically segmented and quantified from vesselness maps derived from multimodal structural MRI data. PVS burden increased from childhood to advancing age in both the white matter and BG and was reflected by increasing PVS volume fraction (VF), count and mean cross-sectional diameter and decreasing mean solidity, with elevated PVS burden in males compared to females. Within white matter regions, PVS burden in childhood significantly predicted the rate of PVS expansion across the lifespan ( t (66)=-12.15, p <.001, R2 =.69), where regions with low PVS VF in childhood undergo rapid growth (e.g., temporal regions) and regions with high PVS VF in childhood undergo slowed growth (e.g., limbic regions). PVS anatomical alterations across the lifespan was not significantly associated with working memory performance. Together, our findings of spatially-varying patterns of PVS enlargement across the lifespan constitutes a feature of normal aging and may reflect alterations to waste clearance functionality. These results provide a normative reference for the spatial distribution of PVS enlargement patterns from which pathological alterations can be compared. ### Competing Interest Statement The perivascular space mapping technology is part of a pending patent owned by FS and JC, with no financial interest/conflict.