The postoperative courses of 115 liver transplant recipients were reviewed to monitor for outcomes of acute renal failure and mortality. An analysis of baseline (preoperative) variables with particular attention to baseline renal function was accomplished to establish predictive variables for a complicated postoperative course. Acute renal failure requiring dialysis occurred in 27 cases (23%) and was associated with a prolonged ICU stay, greater infectious complications, greater hospital charges and a high mortality rate (46 +/- 11% vs. 9 +/- 3%) as compared to patients who did not experience acute renal failure. Death occurred in 20 patients (17%) overall prior to discharge. In order to assess the contribution of renal function, the population was divided arbitrarily into two groups based on preoperative serum creatinine. Group 1 (n = 50) had a preoperative serum creatinine < 1.0 mg/dl (mean +/- SD = 2.2 +/- 0.2 mg/dl) and Group 2 (n = 65) had a preoperative serum creatinine < or = 1.0 mg/dl (0.7 +/- 0.1 mg/dl). The groups experienced similar operative courses. Group 1 patients experienced significantly longer ICU stays (18 +/- 3 vs. 10 +/- 2 days), higher rates of acute renal failure requiring dialysis (52 +/- 7 vs. 5 +/- 2%), higher hospital charges (231,454 +/- 17,088 vs. 178,755 +/- 14,744 $, US) and a greatly increased mortality rate (32 +/- 1 vs. 6 +/- 1%), as compared to Group 2 patients. A multifactorial regression analysis demonstrated that of all pretransplant factors analyzed, elevation in the serum creatinine was significantly associated and was the strongest predictor of both outcomes: acute renal failure requiring dialysis (ROC = 0.89) and death (ROC = 0.83). The presence or absence of hepatorenal syndrome did not influence the results of this analysis. This study demonstrates that cirrhotic patients with renal dysfunction, as indicated by an elevated serum creatinine, experience a poor surgical outcome following liver transplantation. These patients may require special attention in the perioperative period. Alternatively, these data may influence the selection of ideal candidates for liver transplantation, where scarce resources need to be applied appropriately.