After informed consent we studied effect of pharmacological suppression of insulin secretion on tissue sensitivity to insulin in eight subjects with moderate obesity (relative body weight 133 +/- 6 per cent). Glucose tolerance (100 g oGTT) and insulin sensitivity were studied on two separate days before and after treatment with 500 mg diazoxide/d for a period of 4 d. Insulin sensitivity was determined by means of a 1-h priming dose-constant insulin infusion technique (two 30-min periods of 8 and 16 mU/kg, insulin MC-Actrapid, initiated by a start injection of 1 and 2 mU/kg, respectively). The relative decrease of plasma glucose and nonesterified fatty acid concentration at comparable steady-state insulin levels has been taken as an estimate of body sensitivity to insulin. Diazoxide treatment resulted in a significant suppression of glucose-stimulated early (insulin area 0-30 min) (P less than 0.05), late (insulin area 30-120 min) (P less than 0.01) and total insulin response (insulin area 0-120 min) (P less than 0.01) as well as a mild deterioration of glucose tolerance. Insulin sensitivity increased by about 50 per cent in comparison to the pretreatment value. Similarly, there was a marked increase of the antilipolytic effect of insulin after short-term treatment with diazoxide (P less than 0.05). The results support the concept that decreased sensitivity to insulin in obesity may be secondary due to hyperinsulinaemia.