Retinal microvascular density and inner thickness in Alzheimer's disease and mild cognitive impairment

Alzheimer's disease (AD) is a major healthcare challenge, with existing diagnostics being costly/infeasible. This study explores retinal biomarkers from optical coherence tomography (OCT) and OCT angiography (OCTA) as a cost-effective and non-invasive solution to differentiate AD, mild cognitive impairment (MCI), and healthy controls (HCs).

Participants from the CALLIOPE Research Program were classified as "Dem" (AD and early AD), "MCI," and "HCs" using neuropsychological tests and clinical diagnosis by a neurologist. OCT/OCTA examinations were conducted using the RTVue XR 100 Avanti SD-OCT system (VISIONIX), with retinal parameters extracted. Statistical analysis included normality and homogeneity of variance (HOV) tests to select ANOVA methods. Post-hoc analyses utilized Mann-Whitney U, Dunnett, or Tukey-HSD tests based on parameters' normality and HOV. Correlations with age were assessed via Pearson or Spearman tests. A generalized linear model (GLM) using Tweedie regression modeled the relationship between OCT/OCTA parameters and MMSE scores, correcting for age. Another ordinal logistic GLM (OL-GLM) modeled OCT/OCTA parameters against classes, adjusting for multiple confounders.

We analyzed 357 participants: 44 Dem, 139 MCI, and 174 HCs. Significant microvascular density (VD) reductions around the fovea were linked with MCI and Dem compared to HCs. Age-related analysis associated thickness parameters with HCs' old age. Our OL-GLM demonstrated significant thickness/volume reductions in Inner_Retina and Full_Retina layers. Foveal avascular zone (FAZ) area and perimeter were initially not correlated with cognitive decline; however, OL-GLM significantly associated FAZ perimeter enlargement with Dem and MCI groups. Significant average and inferior peripapillary RNFL thinning were linked to Dem and MCI groups.

This is the first study to examine VD changes in G grid sections among Dem, MCI, and HCs. We found a significant association between various VD parameters and cognitive decline. Most macular thickness/volume changes did not correlate with cognitive decline initially; however, our OL-GLM succeeded, highlighting the importance of the confounders' corrections. Our analysis excluded individual retinal layer parameters due to limitations; however, the literature suggests their value. Our study confirmed existing biomarkers' efficacy and uncovered novel retinal parameters for cognitive decline, requiring further validation.