Abstract
Recent data suggest that critically-ill medical patients might not obtain as much benefit from insulin therapy as cardiothoracic patients. This may be because medical patients are more insulin resistant. We have investigated this in both groups. 10 medical patients (age 67.0±4.4 y; 7M:3F, BMI 24.0±1.7 kg/m2; studied within 36 hours of admission to the ICU) and 7 post-operative cardiothoracic patients (age 71.4±3.8 y; 5M:2F, BMI 26.4±1.1 kg/m2; studied within 4 hours of surgery) were recruited. Patients with diabetes mellitus, pancreatitis, liver failure and oral steroid use were excluded. Data are compared to a healthy control group (age 60.0±1.3 y, 4M:1F, BMI 25.0±0.9) undergoing 3-hour euglycaemic clamp. Medical patients received dextrose at 25 kcal/kg/day; postoperative patients at 6.5 kcal/kg/day. In both groups, variable-dose insulin was infused to maintain blood glucose at 7-9 mmol/L. Glucose rate of uptake (Rd) and production (Ra), and leucine Ra (representing proteolysis) were measured with a 3-hour primed infusion of [6,6-2H2] glucose (170 mg, 1.7 mg/min) and [1-13C]leucine (1 mg/kg, 1 mg/kg/hr). Steady state sampling was performed at 150 to 180 minutes. Endogenous glucose Ra was calculated by total glucose Ra minus dextrose infusion rate. Data are mean ± S.E.M. Insulin infusion rate was 3.7±0.8 U/hr (medical), 2.0±0.7 U/hr (surgical patients) and 2.2±0.2 U/hr in controls (NS). Despite the increased provision of dextrose and insulin, medical patients exhibited a higher rate of proteolysis than cardiothoracic patients. If the benefit of exogenous insulin is due to its anabolic action, these findings might explain why insulin therapy is less efficacious in medical than surgical patients.