Objective: To examine the relationship between retinal vascular caliber and incident retinopathy in children and adolescents with type 1 diabetes mellitus.
Design: Hospital-based case-control study with prospective outcomes.
Participants: Cases and controls were selected from a cohort of children and adolescents 12 to 20 years old with type 1 diabetes followed up at a tertiary diabetes clinic. Cases were patients who developed incident diabetic retinopathy (n = 166) after at least 1 year of follow-up (> or =2 clinic visits), and controls were patients who had not developed retinopathy (n = 165) after > or =2 years of follow-up (> or =3 clinic visits). Baseline retinal photographs of cases and controls were digitized, and retinal vascular calibers were measured using a computer-assisted program by a grader masked to case-control status. These measurements were combined into summary indices reflecting the average arteriolar and venular calibers.
Main outcome measure: Development of diabetic retinopathy.
Results: Incident retinopathy cases had retinal arteriolar calibers (mean +/- standard deviation [SD], 206.5+/-18.4 microm) significantly larger than those of controls (200.2+/-16.5 microm) (P = 0.004) but similar retinal venular calibers (329.1+/-14.7 microm in cases vs. 326.4+/-15.1 microm in controls, P = 0.312). After adjusting for age, gender, diabetes duration, glycated hemoglobin levels, blood pressure, body mass index, and pubertal stage, larger arteriolar caliber was predictive of risk of diabetic retinopathy (odds ratio, 1.44 per SD increase in arteriolar caliber; 95% confidence interval, 1.11-1.86).
Conclusion: Larger retinal arteriolar caliber predicts incident retinopathy in children and adolescents with type 1 diabetes, independent of conventional risk factors for retinopathy. Measurement of retinal vascular caliber may provide prognostic information regarding the subsequent risk of diabetic retinopathy.