Vascular and neurodegenerative imaging markers are associated with increased interstitial fluid diffusion in memory clinic patients

Abstract

Background Diffusion measurements in brain tissue may reveal various diffusion regimes reflecting different microstructural topologies, ranging from fast blood circulation, intermediate diffusion in wider interstitial spaces, to highly restricted inter‐ and intra‐cellular diffusion. Intravoxel incoherent motion MRI can detect an intermediate diffusion component, which is thought to be a proxy of brain tissue damage. This intermediate component might be relevant for studying amyloid clearance pathways, but currently remains incompletely understood. This study aims to increase our understanding of this intermediate component by studying associations with vascular (i.e. white matter hyperintensity (WMH) volume) and neurodegenerative (i.e. hippocampal atrophy) imaging markers in a sample of memory clinic patients, who display a wide range of cognitive abilities and variation in neurovascular and neurodegenerative pathology. Method Forty patients (Alzheimer’s disease dementia (AD, n=15), mild cognitive impairment (n=16), vascular cognitive impairment (VCI, n=9)) and 34 healthy controls underwent a 3T MRI protocol. The intermediate diffusion component was identified with spectral analysis using non‐negative least squares (Figure 1), and the contribution of the intermediate component to the signal in each voxel was determined by quantifying its volume fraction fint. The median of fint was determined for the following regions of interest (ROIs): cortical gray matter (GM), subcortical GM, WMH and normal appearing white matter (NAWM). A multivariate linear regression model incorporating age and gender was conducted to investigate group differences. Multiple linear regression was employed to investigate the association of fint per ROI with the vascular and neurodegenerative markers, additionally adjusting for total intracranial volume. Result AD and VCI patients showed increased f int in GM and NAWM relative to controls (Figure 2). The fint was found to be associated to WMH and/or hippocampal volume, dependent on the brain region (Table 1). Conclusion The intermediate component is associated with both vascular and degenerative markers, but this effect is region dependent. The group differences indicate a relation between fint and clinical phenotype. Brain regions where fint is affected by both vascular and degenerative markers show abnormally high fint in AD, which emphasizes the complexity to separate vascular and degenerative influences on AD pathology.