Both hyperinsulinemia and free oxygen radicals have been implicated in the pathogenesis of atherosclerosis, but the relationship between insulin levels or insulin action and the oxidant/antioxidant balance has not been explored. We measured the effect of physiologic hyperinsulinemia on plasma concentrations of vitamin E, a major free radical scavenger molecule. Isoglycemic clamps (at an insulin infusion rate of 6 pmol . min-1 . kg-1) were performed in four groups of subjects: (1) 12 non-insulin-dependent diabetic (NIDDM) patients, (2) eight patients with essential hypertension, (3) 11 nondiabetic obese individuals, and (4) 12 healthy subjects. In 10 healthy volunteers, a time-control experiment was performed by replacing the insulin infusion with normal saline. Vitamin E and plasma lipid levels were determined at baseline and after 2 hours of insulin/saline infusion. Insulin sensitivity was reduced in diabetic, obese, and hypertensive groups in comparison to healthy controls, but fasting plasma vitamin E concentrations were similar in all groups. A consistent decrement in plasma vitamin E concentrations (averaging 12% of baseline, P < .0001) was observed in all subjects receiving insulin regardless of the level of insulin sensitivity, whereas no significant changes in plasma vitamin E were seen in subjects receiving saline infusion (P < .001 v insulin infusion groups). The insulin-induced decrement persisted in all study groups when plasma vitamin E concentrations were corrected for total serum cholesterol levels (-8.9% +/- 1.2% v -0.4 +/- 2.3% of saline controls, P = .0004) or serum low-density lipoprotein (LDL(-10.0% +/- 1.2% v -0.4% +/- 2.2%, P = .0002). We conclude that insulin infusion acutely depletes vitamin E in circulating lipids regardless of insulin resistance. This effect may represent a physiologic means of transferring vitamin E into cell membranes; alternatively, it might reflect a pro-oxidant action of insulin in vivo.