An intravenous glucose-infusion of 0.3 g glucose per Kg body weight was administered over 1 min in nine healthy males with simultaneous blood sampling from the hepatic vein, femoral artery and a peripheral vein. Insulin secretion rates (ISR) were determined by the Eaton method and the ISEC method using C-peptide concentrations from arterial and peripheral venous blood. First phase (0–10 min), second phase (10–60 min), and total insulin secretion (0–60 min) were calculated as the incremental areas (iAUC) above baseline. The primary endpoint was first phase insulin response. The first phase insulin response in artery and venous blood did not differ with the Eaton method (p = 0.25), but was significantly greater with the ISEC method in arterial compared with venous blood (p < 0.05). The first phase insulin responses did not differ between methods in artery (p = 0.73) or venous blood (p = 0.73). The first phase responses of insulin and C-peptide were significant higher in the hepatic vein compared with those in the artery (p < 0.05) and peripheral vein (p < 0.05) but did not differ significantly between the artery compared with the peripheral vein for insulin (p = 0.09) or C-peptide (p = 0.26). Prehepatic insulin secretion rates did not differ between the Eaton and ISEC methods, but with the ISEC method the first phase insulin response was significantly greater in arterial compared with venous blood. The first phase insulin response differs when calculated from plasma insulin or C-peptide and depends on sample sites.