OBJECTIVE To describe the relationship of retinal arteriolar and venular calibers to the long-term incidence of microvascular and macrovascular complications in people with type 2 diabetes. METHODS Population-based prospective study. METHODS One thousand three hundred seventy persons diagnosed to have diabetes at > or =30 years of age in south central Wisconsin participated in the baseline examination from 1980 to 1982, 987 in the 4-year follow-up, and 533 in the 10-year follow-up. METHODS Computer-assisted grading was used to determine the average caliber of retinal arterioles (central retinal arteriolar equivalent [CRAE]) and retinal venules (central retinal venular equivalent [CRVE]) at all examinations. METHODS Incidence and progression of diabetic retinopathy; incidence of proliferative diabetic retinopathy and macular edema; incidence of nephropathy, neuropathy, and lower extremity amputation; and ischemic heart disease, stroke, and overall mortality. RESULTS While adjusting for other factors, smaller CRAE was associated with the 14-year cumulative incidence of lower extremity amputation (odds ratio [OR], first vs. second to fourth quartiles, 2.20; 95% confidence interval [CI], 1.14-4.24; P = 0.02), 22-year all-cause mortality (hazard ratio [HR], 1.18; 95% CI, 1.02-1.38; P = 0.03), and 22-year stroke mortality (HR, 1.47; 95% CI, 1.04-2.07; P = 0.03) but not with the other end points. Larger CRVE was associated with the 14-year incidence of diabetic nephropathy (OR, fourth vs. first to third quartiles, 2.08; 95% CI, 1.47-2.94; P<0.001) and 22-year stroke mortality (HR, 1.71; 95% CI, 1.20-2.44; P = 0.003) but with none of the other end points. CONCLUSIONS Retinal vessel caliber is independently associated with risk of incident nephropathy, lower extremity amputation, and stroke mortality in persons with type 2 diabetes. Measurement of retinal vessel caliber from photographs may provide additional information for the prediction of these events.