A Quantitative Assessment of MRI-Visible Perivascular-Space Burden Across the Cognitive Spectrum: Associations with APOE ε4 Status and Venous Sinus Volume
TL;DR
The findings of this cross-sectional observational MRI study support further study of venous sinus morphology as a potential contributor to MRI-visible PVS burden, while longitudinal and flow-sensitive studies are needed to clarify causality and glymphatic clearance mechanisms.
Abstract
Background/Objectives: Perivascular spaces (PVS) are MRI-visible compartments surrounding cerebral vessels and are increasingly interpreted as markers of cerebral small-vessel disease and glymphatic-system dysfunction. PVS burden is multifactorial, with possible contributions from vascular risk, atrophy, white matter disease, sleep, inflammation, diabetes, pathological protein deposition, genetic susceptibility, and venous drainage anatomy. This study evaluated whether apolipoprotein E (APOE) ε4 carrier status and venous sinus volumes are associated with quantitative PVS burden in patients across the cognitive spectrum. Methods: In this cross-sectional observational MRI study, 110 participants with normal cognition or varying cognitive impairment underwent 3T brain MRI and cognitive assessment. PVS lesion count and total PVS volume were quantified from T2-weighted (T2W) imaging using an automated deep learning segmentation approach. Venous sinus volumes were derived for the transverse, straight, and superior sagittal sinus segments and reported in mm3. APOE genotype was analyzed as ε4 carrier status. Non-parametric correlations, ordered cognitive-stage trend tests, APOE group comparisons, and adjusted partial Spearman analyses were performed. Results: Dorsal superior sagittal sinus (SSS-D) volume was associated with PVS count (ρ = 0.353, p < 0.001) and total PVS volume (ρ = 0.285, p = 0.009). These associations persisted after adjustment for age, sex, medial temporal atrophy (MTA), Fazekas score, and estimated intracranial volume (eTIV). APOE ε4 carrier status was not associated with PVS volume or count. PVS burden tended to decrease with greater cognitive-stage severity, while MTA increased across cognitive stages. Conclusions: In this cohort, SSS-D volume was associated with automated 3T T2W MRI-detectable PVS burden, whereas APOE ε4 carrier status was not a dominant predictor. These findings support further study of venous sinus morphology as a potential contributor to MRI-visible PVS burden, while longitudinal and flow-sensitive studies are needed to clarify causality and glymphatic clearance mechanisms.