A double-blind, randomized controlled trial was performed to determine the effect of glutamine (GLN)-enriched intravenous feedings on the volume and distribution of body fluids in catabolic patients. Subjects with hematologic malignancies in remission underwent a standard treatment of high-dose chemotherapy and total body irradiation before bone marrow transplantation. After completion of this regimen, they were randomized to receive either standard parenteral nutrition (STD, n = 10) or an isocaloric, isonitrogenous nutrient solution enriched with crystalline L-glutamine (0.57 g/kg/day, GLN, n = 10). Extracellular water (ECW) and total body water (TBW), determined by bromide and heavy water dilution techniques, were measured before the conditioning treatment and after termination of the intravenous feedings that were administered for 27 +/- 1 days. In addition electrical resistance (R, in ohms, omega) and reactance (Xc, omega) of the body to a weak alternating current were measured at these time points. Both study groups were comparable for age, weight, height, sex, and diagnosis. Initial TBW was highly related to electrical resistance (r = -0.93, p less than 0.001). After conditioning therapy, bone marrow infusion, and intravenous feedings, a 20% expansion in ECW was observed in the STD group (ECW: 18.0 +/- 1.1 L vs. 14.9 +/- 1.0, p = 0.012), and this fluid retention was associated with a marked decrease in electrical resistance (R: 514 +/- 28 omega vs. 558 +/- 26, p less than 0.05). In contrast the extracellular fluid compartment in patients receiving GLN-supplementation did not change (ECW: 15.8 +/- 0.9 L vs. 15.4 +/- 0.8, p = 0.49), and the body's resistance was maintained (R: 552 +/- 27 omega vs. 565 +/- 23, p = 0.42). Expansion of ECW could not be related to differences in fluid or sodium intake, or to the use of diuretics or steroids. Patients receiving the STD solution, however, exhibited a greater number of positive microbial cultures (p less than 0.01) and a higher rate of clinical infection compared with the GLN patients (5/10 vs. 0/10, p less than 0.05); the fluid expansion in infected STD patients was greater compared with uninfected individuals (delta ECW: + 5.0 +/- 1.4 vs. 0.7 +/- 0.5, p = 0.007). In this model of catabolic stress, fluid retention and expansion of the extracellular fluid compartment commonly observed after standard total parenteral nutrition can be attenuated by administering glutamine-supplemented intravenous feedings, possibly by protecting the host from microbial invasion and associated infection.