Back to Authors

Dumitrascu, O. M.

Mayo Clinic

Mayo Clinic Hospital

Citations 1

Abstract 188: Retinal venular diameter as an indicator of Alzheimer's‐associated vasculopathy

Stroke: Vascular and Interventional Neurology | December 2, 2025 | Volume: 5 | Issue: S1 | Pages: No page available

10.1161/svi270000_188

Introduction The association between retinal vessel diameters and cerebrovascular disease is well known. Additionally, retinal vascular measures provide a direct and objective modality to study Alzheimer's disease (AD) ‐ associated vasculopathy. Herein, we aimed to study the association between automatically quantified retinal arteries and veins diameters and AD. Methods We have identified 1039 subjects from the UK Biobank that had good‐quality macula‐centered color fundus photographs (220 with AD, 202 with Parkinson's disease (PD), 516 normal controls (NC)), of which 86.9% were right and 71.2% left eyes. We quantified retinal vascular measures in Hubbard zones B and C (central retinal vein equivalent (CRVE), central retinal artery equivalent (CRAE) and arterio‐venous ratio (AVR)) using Automorph. We assessed inter‐eye correlation between the vascular features using Spearman correlation coefficient. We adjusted for confounding patient characteristics between groups where present (age, gender, hypertension, diabetes mellitus (DM)) and the presence of APOE4 carrier allele. Group‐level adjusted mean estimates and mean differences for AD vs NC and AD vs PD with 95% confidence intervals (CI) were constructed using regression modeling. Results Subjects with AD were significantly older than NC (mean (SD) age 64.9(4.4) vs 55.9(8.1) years, p<0.001), had similar gender distribution, similar incidence of hypertension (27.7% vs 22.7%, p=0.1428), significantly more DM (6.8% vs 3.5%, p=0.04) and more likely to be APOE4 carriers (64.4% vs 30.8%, p<0.001). When compared to PD subjects, AD were older (mean (SD) age 64.9 (4.4) vs 62.6 (5.3) years, p<0.001), less commonly male (46.4% vs 61.4%, p=0. 0007), had similar incidence of hypertension (27.7% vs 31.0%, p=0.4152) and DM (6.8% vs 6.3%, p=0. 8020) but were more likely to be APOE4 carriers (64.4% vs 29.6%, p<0.001). All CRAE, CRVE, and AVR measures showed significant correlations between right and left eyes, hence we report only right eye measures given greater sample size. Between AD and NC, age and DM adjusted retinal vascular measures showed greater CRVE in Hubbard zone B (mean difference (CI) 2.042 (0.61, 3.47), p=0.005) and zone C (mean difference (CI) 1.859 (0.41, 3.29), p =0.01). After additional adjustment for APOE4, CRVE in zone B only remained significantly greater in AD when compared to NC (mean difference (CI) 1.97 (0.15, 3.79), p=0.03). When we compared AD and PD subjects, age and gender adjusted retinal vascular measures also exhibited greater CRVE in AD in both zones B and C, that remained significant after additional adjustment for APOE carrier status (mean difference (CI) 2.531 (0.946, 4.117), p=0.0018). CRAE and AVR showed non‐significant differences between AD and NC, as well as AD and PD, in all zones (p>0.05 for all comparisons). Conclusions The central retinal vein equivalent was significantly greater in subjects with AD in the UK Biobank, after adjusting for demographic, vascular co‐morbidities and APOE4 carrier status. These results suggest that wider retinal venules are indicators of AD‐associated retinal vasculopathy and degeneration.