The authors investigated retinal blood flow changes in patients with insulin-dependent diabetes mellitus (IDDM) and no diabetic retinopathy compared to age-matched subjects without diabetes. They also investigated whether blood glucose levels could modulate retinal blood flow in these patients with diabetes and whether this modulation would impact retinal blood flow data used in cross-sectional studies assessing changes in retinal blood flow. Retinal blood flow was measured using video fluorescein angiography, and blood glucose levels were manipulated using glucose clamp methodologies with continuous basal insulin replacement. Blood glucose levels were clamped at 100, 200, and 300 mg/dl. Retinal blood flow measurements were performed at each blood glucose level after subjects had been stabilized for an hour at each of the different blood glucose levels. Retinal blood flow was found to be significantly decreased (P< 0.01) in the group of patients with no diabetic retinopathy (19.4+/-4.6 arbitrary units [AU]) compared to retinal blood flow in subjects without diabetes (28.7+/-6.4 AU). During glucose clamp adjustment of blood glucose levels, it was found that as blood glucose levels were increased from euglycemia (100 mg/dl) to 200 mg/dl and to 300 mg/dl, retinal blood flow was significantly increased at the 200 mg/dl level (21.5+/-4.7 AU, P< 0.05) and at the 300 mg/dl level (25.9+/-8.8 AU, P< 0.01) compared to the 100 mg/dl level (16.3+/-3.8 AU). In addition, the retinal blood flow at the 100 and 200 mg/dl levels was significantly reduced (P< 0.01) compared to nondiabetic retinal blood flow (28.7+/-6.4 AU). Retinal blood flow was found to be decreased in patients with IDDM with no diabetic retinopathy, and acute elevations in blood glucose levels resulted in increased retinal blood flow in these patients. The acute modulation of retinal blood flow by blood glucose levels should be considered in cross-sectional studies investigating retinal blood flow changes in patients with diabetes. The results from this study indicate that if blood glucose levels are not accounted for in the analyses, larger populations would have to be studied to demonstrate statistically significant differences between groups with and without diabetes.