Hyperglycemia results from an imbalance between endocrine pancreatic function and hepatic and extrahepatic insulin sensitivity. We studied 57 well-matched Swedish men with normal glucose tolerance (NGT), impaired glucose tolerance (IGT), or mild diabetes. Oral glucose tolerance and insulin release were assessed during an oral glucose tolerance test (OGTT). Insulin sensitivity and glucose turnover were determined during a two-step euglycemic insulin clamp (infusion 0.25 and 1.0 mU. kg(-1). min(-1)). High-performance liquid chromatography-purified [6-(3)H]glucose was used as a tracer. During low-insulin infusion, the rate of endogenous glucose production (EGP) decreased more in subjects with NGT than in subjects with IGT or diabetes (delta rate of appearance [R(a)] 1.25 +/- 0.10 vs. 0.75 +/- 0.14 vs. 0.58 +/- 0.09 mg. kg(-1). min(-1), P < 0.001). The corresponding rates of glucose infusion during the high-dose insulin infusion (M values) were 8.3 +/- 0.6 vs. 5.4 +/- 0.9 vs. 4.7 +/- 0.4 mg. kg(-1). min(-1) (P < 0.001). A total of 56% of the variation in glucose area under the curve (AUC) during OGTT (glucose AUC) was mainly explained by delta R(a) (increase in multiple R(2) 0.42) but also by delta R(d) (rate of disappearance) (increase in multiple R(2) 0.05), and the early insulin response during OGTT contributed significantly (increase in multiple R(2) 0.07). When M value was included in the model, reflecting extrahepatic insulin sensitivity, it contributed to 20% of the variation in glucose AUC, and together with the incremental insulin response (increase in multiple R(2) 0.21), it explained 45% of the variation. In conclusion, insulin sensitivity of suppression of EGP plays the most important role in the determination of blood glucose response during OGTT.