Objectives: To estimate the efficacy and metabolic effects of growth hormone substitution as intravenous pulses together with alanyl-glutamine supplementation and tight blood glucose control in prolonged critical illness. Design: Prospective double-blind, randomized trial with open-label control arm. Setting: Intensive care unit of tertiary level hospital. Patients: Thirty multiple trauma patients (median Injury Severity Score 34). Interventions: Patients were randomized, at day 4 after trauma, to receive intravenous alanyl-glutamine supplementation (0.3 g/kg.day(-1) from day 4 until day 17) and intravenous growth hormone (administered days 7-17, full dose 50 mu g/kg.day(-1) from day 10 onward) (group 1, n = 10) or alanyl-glutamine and placebo (group 2, n = 10). Group 3 (n 10) received isocaloric isonitrogenous nutrition (proteins 1.5 g/kg.day(-1)) without alanyl-glutamine. Measurements and Main Results: Cumulative nitrogen balance for the whole study period was -97 +/- 38 g of nitrogen for group 1, -193 +/- 50 g of nitrogen for group 2, and -198 +/- 77 g of nitrogen for group 3 (p <.001). This represents a daily saving of 300 g of lean body mass in group 1. Insulin-mediated glucose disposal, during euglycemic clamp, as a measure of insulin sensitivity, significantly worsened between days 4 and 17 in group 1 but improved in groups 2 and 3. Group 1 required significantly more insulin to control blood glucose, resulting in higher insulinemia (similar to 70 mIU in group 1 vs. similar to 25 mIU in groups 2 and 3). Despite this, growth hormone treatment caused an increase in plasma nonesterified fatty acid (similar to 0.5-0.6 mM in group 1 in comparison with similar to 0.2-0.3 mM in groups 2 and 3) but did not influence lipid oxidation. There were no differences in morbidity, mortality, or 6-month outcome among the groups. Conclusions: Treatment with frequent intravenous pulses of low-dose growth hormone together with alanyl-glutamine supplementation improves nitrogen economy in patients with prolonged critical illness after multiple trauma but worsens insulin sensitivity. Tight blood glucose control is possible but requires higher doses of insulin.