Ophthalmology Science · Nov 2026

Retinal vasculometry and diabetic retinopathy severity by age, sex and ethnicity: North East London Diabetic English Screening program

By Shakespeare RD, Welikala R, Fatjl J et al.

Objective To determine the association between Retinal Vasculometry (RV) and Diabetic Retinopathy (DR) severity by age, sex and ethnicity. Design Cross-sectional study. Participants Participants who attended the North East London Diabetic Eye Screening Programme (NEL DESP) between January 2021 and December 2022. Methods Adjusted means for RV were computed from 178,911 consecutive encounters from NEL DESP. Multilevel linear regression models adjusted for age, sex, ethnicity, diabetes type, diabetes duration, smoking status and Index of Multiple Deprivation (IMD) quintiles, allowing for repeated RV measures within person from multiple encounters for small retinal vessels (SRV) < 50 μm and large retinal vessels (LRV) ≥ 50 μm. Analyses were repeated, stratified by age, sex, and ethnic groups. Main outcome measures Arteriolar and venular width, area, tortuosity and Vessel Calibre Uniformity (VCU) in both SRV and LRV. Results For LRV, arteriolar width decreased linearly as DR severity increased. Conversely, venular width and tortuosity measures for both arterioles and venules increased linearly with DR severity. Both SRV and LRV areas and VCU appeared curvilinear with reductions being more prominent in advanced stages of DR. LRV arteriolar area in the youngest age group remained stable with increasing DR grade, and venular area in the oldest age group remained stable. SRV arteriolar tortuosity increased linearly while venular tortuosity showed an initial increase but remained stable across intermediate and advanced DR grades. Stratified analysis revealed differences by ethnicity and age group. Systematic differences by subgroups existed, but the direction of association was largely similar. Importantly, area and VCU measures declined with increasing age. Conclusion Strong associations between RV measures and DR grade demonstrate their potential as non-invasive prognostic predictors of DR progression.

doi.org/10.1016/j.xops.2026.101461Open accessBiomarkers

AgingResearch.News