Associations between the retinal vasculature and brain white matter hyperintensities in mid-life persons at risk of developing Dementia

Abstract

Purpose : Brain white matter hyperintensities (WMH) which can be detected using MRI are lesions that are associated with cerebral small vessel disease (cSVD). Individuals with higher volumes of WMH are reported to be at a higher risk of developing dementia. This positions MRI as a promising tool for tracking cerebrovascular changes in individuals that transition from mild cognitive impairment (MCI) to dementia. Furthermore, studies have shown that cerebrovascular changes might be tracked by analysing the retinal vasculature. Optical Coherence Tomography Angiography (OCTA) is a non-invasive retinal imaging modality that captures the retinal vasculature quickly, non-invasively, and at a fraction of the cost of MRI. In this study, we investigated the relationship between retinal phenotypes and WMH burden globally in the brain and regionally in the deep white matter area.

Methods : This is a sub-study within PREVENT, of which 99 mid-life participants (mean age = 52) with and without a family history of dementia have had their WMH volumes (from MRI) and OCTA retinal phenotypes acquired and quantified cross-sectionally. Brain WMH were segmented by location into global and deep WMH, and volumes were normalised as a percentage of total intracranial volume to account for variable head sizes. 190 total eyes (including left and right) were imaged and binarized into a graph to model the vasculature to calculate retinal phenotypes. Retinal phenotypes that characterize Foveal Avascular Zone (FAZ) area, vessel skeleton density (VSD), and vessel resistance (VR) were quantified and included for this study. We fitted a generalized estimating equation model using covariates, sex, age, blood pressure, and body mass index, and each retinal phenotype as the independent variables and the WMH volume as the dependent variable.

Results : Increased global WMH volume was related to a decrease in FAZ area (p < 0.01, beta = -6.84) and VSD (p < 0.01, beta = -4.44), while deep WMH was related to a decrease in retinal VR (p < 0.01, beta = -3.68).

Conclusions : We add to a growing body of evidence relating changes to the retinal vascular with WMH. We uncover for the first time a relationship between WMH and a functional retinal vascular phenotype (VR). Future work should evaluate the predictive power of the retinal vasculature as a surrogate for MRI in tracking the transition from MCI to Dementia.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.