BACKGROUND Malnutrition is a common complication of end-stage liver disease. It is frequently not a priority of treatment before liver transplantation. The purpose of this study was to examine whether prospective preoperative nutritional assessment could predict resource utilization and outcome after liver transplantation. METHODS We retrospectively reviewed 109 sequential orthotopic liver transplants performed at our center between July 1996 and May 1999. Ten patients with fulminant hepatic failure were excluded from the study, leaving 99 patients. Nutritional status was determined at the time of transplantation using subjective global assessment. Wilcoxon rank sum test and rank analysis of variance were used to analyze the data. Results are reported as median (interquartile range). A P value <0.05 was considered significant. RESULTS Intraoperative transfusion requirements of packed red blood cells and cryoprecipitate was higher in the patients with severe malnutrition in comparison to the mild and moderate groups (severe vs. moderate, 5.5+/-5.5 vs. 3.0+/-6, P=0.026; vs. mild, 1.5+/-3, P<0.0001). The severe group required more fresh-frozen plasma intraoperatively than the mild group (mild vs. severe, 0+/-2 vs. 2+/-6, P=0.0007; vs. moderate, 1+/-4, P=0.071). Patients in the severe group had longer postoperative lengths of stay compared with patients in the moderate and mild groups (severe vs. moderate, 16+/-9 days vs. 10+/-5 days, P=0.0027; vs. mild, 9+/-8 days, P=0.0006). CONCLUSIONS Subjective global assessment is an excellent independent predictor of outcome in patients undergoing liver transplantation. Severely malnourished patients require more blood products during surgery and have prolonged postoperative length of stay in hospital. Our data suggest that if nutritional repletion is possible in patients with end-stage liver disease before transplantation, patient outcomes could be improved.