We compared the effects of two ethanol doses on glucose kinetics and assessed the role of acetate as a mediator of ethanol-induced insulin resistance. Ten normal males were studied on four occasions, during which either a low (blood ethanol 4 +/- 1 mmol/l) or moderate (14 +/- 1 mmol/l) ethanol, acetate, or saline dose was administered. Both ethanol doses similarly inhibited (0.4-0.5 mg.kg-1.min-1, P less than 0.01) basal glucose production. The decrease in Ra was matched by a comparable decrease in glucose utilization (Rd), resulting in maintenance of normoglycemia. During hyperinsulinemia (approximately 70 microU/ml), glucose disposal was lower (1.2-1.7 mg.kg-1.min-1, P less than 0.01) in the moderate than the low-dose ethanol or saline studies. During acetate infusion, the blood acetate level was comparable with those in the ethanol studies. Acetate had no effect on glucose kinetics. In conclusion, 1) in overnight fasted subjects, ethanol does not cause hypoglycemia because its inhibitory effect on Ra is counterbalanced by equal inhibition of Rd;2) basal Ra and Rd are maximally inhibited already by small ethanol doses, whereas inhibition of insulin-stimulated glucose disposal requires a moderate ethanol dose; and 3) acetate is not the mediator of ethanol-induced insulin resistance.