We investigated the effects of hyperinsulinemia and hyperglycemia on peripheral glucose uptake, hepatic glucose production, and splanchnic glucose uptake in man. Euglycemic and hyperglycemic clamp studies were carried out in 37 healthy subjects in combination with hepatic vein catheterization and [3H-3]glgcose infusion.
In the basal state, hepatic glucose production ([3H-3]glucose) exceeded net splanchnic glucose output (catheter) in every subject (2.3 ± 0.04 versus 1.7 ± 0.07 mg/min · kg, P < 0.001), indicating uptake of glucose by the splanchnic region at a rate of 0.6 ± 0.05 mg/ min · kg. In agreement with this estimate, [3H-3]glucose concentration was consistently lower in hepatic venous than in arterial blood, by 3.0 ± 0.2% (P < 0.001).
When plasma insulin levels were raised to 37 ± 2, 53 ± 2, 101 ± 2, 428 ± 37, and 1189 ± 14 μU/ml, with maintenance of euglycemia, total glucose uptake rose to 2.9 ± 0.4, 3.9 ± 1.0, 5.1 ± 0.4, 9.9 ±1.1, and 11.8 ± 1.3 mg/min · kg, respectively. The whole body glucose clearance rose significantly above baseline at each hyperinsulinemic plateau (P < 0.05 or less). Hepatic glucose production fell by 68% (P < 0.01) at the lowest hyperinsulinemic level, by 87% at insulin levels of 53 ± 2 μU/ml, and by over 95% with each higher insulin dose. Splanchnic glucose uptake was not significantly increased over basal values at any insulin concentration.
When plasma glucose levels were raised to 137 ± 3 and 224 ± 2 mg/dl peripheral plasma insulin levels rose to 20 ± 4 and 55 ± 5 μU/ml, respectively. Total glucose uptake was enhanced (2.5 ± 0.4 and 5.3 ± 1.0 mg/min · kg, P < 0.05 and P < 0.01, respectively). Suppression of hepatic glucose production was <90% at the lower hyperglycemic level, and virtually complete at the higher one. Splanchnic glucose uptake was not changed by mild hyperglycemia (0.5 ± 0.05 mg/min · kg), but rose significantly (1.3 ± 0.3 mg/ min · kg, P < 0.01) with further hyperglycemia. The latter effect resulted primarily from increased glucose delivery to the splanchnic area, since the splanchnic glucose extraction ratio (4.0 ± 0.3%) was not different from baseline (3.0 ± 0.3%).
When hyperglycemia (224 ± 1 mg/dl) was combined with a somatostatin infusion, thereby reducing plasma insulin from 15 ± 3 to 10 ± 1 μU/ml (P < 0.01), both total glucose uptake (2.8 ± 0.03 mg/min · kg) and clearance (1.3 ± 0.01 mg/min · kg) were significantly (P < 0.01) lower than in the hyperglycemic studies in which insulin secretion was not blocked. Hepatic glucose production, however, was effectively suppressed (by 74%, P < 0.001), whereas splanchnic glucose uptake was only slightly increased above baseline. Replacement of insulin (via an exogenous infusion at a rate of 0.3 mU/min · kg) restored total glucose uptake, splanchnic glucose uptake, and suppression of hepatic glucose production to the levels seen with hyperglycemia without somatostatin. When hyperglycemia (216 ± 2 mg/dl) was combined with somatostatin and glucagon replacement (no insulin), hepatic glucose production was still suppressed by 47 ± 1% to 1.18 ± 0.01 mg/kg · min (P < 0.001 versus hyperglycemia + SRIF without glucagon replacement). The results indicate that both hyperglycemia and hypoglucagonemia contribute to the decline in hepatic glucose production following somatostatin infusion.
In conclusion, hyperinsulinemia alone stimulates glucose uptake by peripheral but not splanchnic tissues. The dose-response characteristics of stimulation of peripheral glucose uptake and inhibition of hepatic glucose production by insulin are very different, the half-maxima being ∼120 and ∼50 μU/ml, respectively. Hyperglycemia enhances glucose uptake by both peripheral and splanchnic tissues, but this action requires an intact endogenous insulin response. In contrast, hyperglycemia can suppress endogenous glucose production even in the presence of low insulin levels.
