Because several studies have shown a significant inverse correlation between depressed serum concentrations of albumin and hospital morbidity, a study with central total parenteral nutrition (TPN) with normal serum albumin (NSA) in hypoalbuminemic patients was conducted. Sixty-one patients who required central TPN were randomized into one of two groups: group 1 (n = 31) received TPN plus NSA (25 to 37.5 g/day) until their measured serum albumin was >3 g/dl, and group 2 (n = 30), who received TPN alone. AH patients were followed for hospital complications until discharge or death. The groups were well matched for age, sex, major diagnoses, initial serum albumin concentrations, hospital complications before TPN, and number of operative procedures. Both groups received comparable doses of energy (37.2 ± 8.9 vs. 33.0 ± 6.2 kcal/kg·day) and protein (1.6 ± 0.4 vs. 1.6 ± 0.3 g/kg·day). After initiation of TPN, there were significantly more hospital complications in group 2 (1 = 1.1 ± 1.4, n = 33; 2 = 2.6 ± 3.0, n = 80, p < .01). When complications in the patient groups were stratified, significantly more patients in group 2 developed pneumonia (18 vs. 9, p < .05) and septicemia (11 vs. 2, p < .05). Increasing serum albumin concentrations with NSA in hypoalbuminemic patients receiving central TPN appears to be associated with a reduction in hospital morbidity.