BACKGROUND: Hyperglycemia and an increased ventilatory demand secondary to an increased CO2 production are frequent undesirable effects of total parenteral nutrition (TPN) in critically ill patients. This study was performed to assess whether sodium lactate as a metabolic substrate may affect these variables. METHODS: Five male patients with multiple trauma during the flow phase were studied during two consecutive 3-hour periods of isocaloric (1.1 x resting energy expenditure) TPN. Sixty-five percent of total calories was provided as carbohydrate, 15% as lipids, and 20% as amino acids during the first period (TPN-glucose), whereas 35% carbohydrate, 30% lactate, 20% lipids, and 15% amino acids (TPN-lactate) were substituted during the second period. Respiratory gas exchanges and net substrate oxidation were assessed by means of indirect calorimetry. Glucose kinetics was determined by primed-constant infusion of U-13C glucose. RESULTS: Compared with TPN-glucose, TPN-lactate decreased glycemia by 20%, insulinemia by 43%, net carbohydrate oxidation (assessed from indirect calorimetry) by 34%, and plasma glucose oxidation (assessed from 13CO2) by 54%. Respiratory oxygen exchange were increased by 3.7% due to a 20% thermic effect of lactate, but respiratory CO2 exchanges did not change. Pao2 decreased by 11.3 mm Hg, indicating that the increased O2 consumption was not matched by an appropriate increase in spontaneous ventilation. Arterial pH increased from 7.41 +/- 0.04 to 7.46 +/- 0.05. CONCLUSION: Sodium lactate as a metabolic substrate limits hyperglycemia but induces metabolic alkalosis and does not spare the ventilatory demand.