Radial peripapillary capillary network microvascular changes in Alzheimer's disease, mild cognitive impairment, and cognitively healthy controls

Background The retinal and optic nerve head microvasculature may represent a non‐invasive, inexpensive biomarker for Alzheimer’s disease (AD) and mild cognitive impairment (MCI). In this study, we characterize microvascular changes at the optic nerve head (radial peripapillary capillary network) using optical coherence tomography angiography (OCTA). Method Optic nerve head‐centered OCTA images were obtained from 66 eyes of 36 patients with AD, 119 eyes of 63 patients with MCI, and 513 eyes of 265 cognitively healthy controls in a cross‐sectional study (NCT03233646) using Zeiss Cirrus HD‐5000 OCTA. Exclusion criteria included diabetes, uncontrolled hypertension, glaucoma, non‐AD dementia, and other optic nerve pathology. Two metrics of vessel perfusion were calculated in a 2.5mm‐wide annulus centered on the optic nerve head using Zeiss AngioPlex (Version 11.0.0.29946) – capillary perfusion density (CPD) and capillary flux index (CFI). CPD represents the percentage of perfused capillaries in a unit scan area; CFI is a unitless ratio of perfused capillaries in a unit scan area weighted by intensity of flow signal. Retinal nerve fiber layer (RNFL) thickness was measured using a peripapillary optic disc cube scan. Groups were compared using multivariable generalized estimating equations, controlling for age and sex as covariates. Result Average CPD was significantly greater in AD vs controls (0.446 ± 0.015 vs 0.439 ± 0.017, p = 0.001) and in MCI vs controls (0.443 ± 0.020 vs 0.439 ± 0.017, p = 0.007) but not AD vs MCI (p = 0.69). CFI and average RNFL thickness did not significantly differ among groups (all p > 0.05). Among all groups, average CPD differed significantly (p = 0.001), as did CPD in the nasal and inferior peripapillary quadrants (p < 0.001 and p = 0.001, respectively). Conclusion Peripapillary CPD is increased in eyes of patients with AD or MCI compared to cognitively healthy controls, despite similar RNFL thickness between these groups. This finding represents a novel biomarker and contrasts with macular microvascular changes, which show decreased vessel and perfusion density in AD and MCI compared to controls. Integration of optic nerve head OCTA in models using retinal imaging to diagnose AD and MCI may improve specificity; further investigation is necessary.