Simulated postaggression metabolism in healthy subjects: metabolic changes and insulin resistance.

Postaggression metabolism (PAM) is difficult to study in critically ill patients. The objective of this study was to simulate PAM in healthy subjects to quantify insulin sensitivity under these conditions. Six healthy men (age, 24 +/- 1 years; body mass index, 22.0 +/- 0.7 kg/m2 [mean +/- SE]) received an intravenous (i.v.) infusion of insulin-counteracting hormones (epinephrine 100 ng/kg/min, glucagon 16 ng/kg/min, hydrocortisone 5 microg/kg/min, and growth hormone [GH]-releasing hormone 50 microg/h) for 4 hours in addition to glucose (270 mg/kg/h). Control experiments used glucose only. In additional experiments, insulin sensitivity was measured by a two-step hyperinsulinemic glucose clamp with and without concomitant hormone infusion (insulin infusion rate, 2.5 and 5.0 mU/kg/min for hormone infusion or 1.0 and 2.5 mU/kg/min for control experiments). Plasma stress hormones reached levels comparable to severe PAM (epinephrine, 1,085 +/- 89 pg/mL; glucagon, 1,100 +/- 114 pg/mL; cortisone, 1,004 +/- 32 ng/mL; and GH, 20.6 +/- 6.1 pg/mL) in the hormone infusion experiment. This resulted in hyperglycemia and hyperinsulinemia (steady-state blood glucose, 19.7 +/- 0.4 mmol/L; serum insulin, 352 +/-8 pmol/L) in comparison to the control experiments with glucose infusion only (maximal blood glucose 7.2 +/- 0.8 mmol/L; serum insulin, 110 +/- 16 pmol/L). The insulin sensitivity index (S(I)) was 88% +/- 6% lower during hormone infusion (0.6 +/- 0.4 mL/min/m2/microU/min) compared with the control experiments (4.5 +/- 1.3 mL/min/m2/microU/min). Infusion of insulin-counteracting hormones at high doses allows simulation of the changes in carbohydrate metabolism observed in PAM in healthy subjects. The observed profound decrease in insulin sensitivity explains the hyperglycemia observed in nondiabetic critically ill patients. With this experimental setup, standardized investigations of therapeutic interventions in PAM should be possible.