Impairment of Insulin-mediated Glucose Metabolism by Hyperosmolality in Man

The deterioration of glucose tolerance and insulin resistance observed in states of hypertonie dehydration are commonly ascribed to a concomitant increase in circulating insulin counterregulatory hormones. To examine the effect of hyperosmolality per se on carbohydrate metabolism, tissue sensitivity to insulin was assessed by means of the euglycemic insulin clamp technique and simultaneous 3H-3-glucose kinetic analysis. Eight healthy volunteers participated in three protocols: (1) In the hyperosmolal study, serum osmolality was raised from 280 ± 1 to 302 ± 1 mosm/kg by a primed continuous infusion of hypertonie mannitol. Following 2 h of hypertonicity, a euglycemic insulin clamp study was performed. The plasma insulin concentration was acutely raised and maintained at 147 ± 20 μU/ml, while plasma glucose was maintained at basal levels; (2) In the control insulin clamp study, isotonic sodium chloride was infused instead of mannitol. No significant change in serum osmolality was observed during the control study; (3) To examine the effect of mannitol per se on insulin-mediated glucose metabolism, isotonic mannitol was infused in a third study to raise the plasma mannitol level to a similar extent as observed during the infusion of hypertonie mannitol. The serum tonicity did not change during the isotonic mannitol infusion. Insulin-mediated glucose utilization (M) was 6.69 ± 0.51 mg/kg · min in the control study and decreased to 5.84 ± 0.40 mg/kg · min following the exposure to hyperosmolality (P < 0.05). When glucose uptake by the body was expressed per unit of plasma insulin (M/l ratio), the decrement in insulin-mediated glucose metabolism by hyperosmolality was 22% (5.56 ± 0.60 versus 4.30 ± 0.43 mg/kg · min per μM/ml × 100). This could not be accounted for by changes in the plasma levels of cortisol, growth hormone, or catecholamines. Basal hepatic glucose production (1.85 ± 0.08 mg/kg · min) as well as the degree of its suppression following hyperinsulinemia (95 ± 3%) were unchanged during hyperosmolality (1.84 ± 0.08 mg/kg · min and 94 ± 2%, respectively). The infusion of isotonic mannitol had no effect on insulin-mediated glucose metabolism or suppression of hepatic glucose production by insulin. We conclude that moderate elevations in serum osmolality, less than 10%, cause a significant decrease in insulin-mediated glucose metabolism. Thus, hyperosmolality observed following severe dehydration should be regarded as an additional factor contributing to the development of insulin resistance. This pathophysiologic concept may be of particular relevance to diabetics with severe derangements of metabolic control. Our findings underline the importance of proper rehydration in restoring to normal insulin's action on intermediary metabolism in patients who present with hyperosmolality.