Acute renal failure in critically ill patients is asociated with significant morbidity and mortality. With the development of continuous renal replacement therapy (CRRT), nephrologists are able to provide dialytic support for the critically ill patient with renal failure. The purpose of this study was to assess patient survival, renal function recovery, and the occurrence of CRRT-related complications in critically ill patients receiving CRRT. Additionally, various demographic, clinical, and laboratory factors were analyzed to determine whether they were associated with better patient outcomes. Methods. A retrospective study analyzing data on patients who received CRRT form January 1996 through June 1998 at the University of Rochester Medical Center was conducted. Results. Out of 103 patients, 41 (40%) survived and 62 (60%) died. The only significant finding among the group of patients who survived was the timing of CRRT initiation. Patients who survived were initiated on the 10th day (mean value) from admission compared to the 18th day (mean value) for those who died. Of the 37 patients without preexisting end-stage renal disease (ESRD) who survived, 32 (86.5%) had renal function recovery and 5 (13.5%) did not have renal function recovery. Heparin was used in 60% of patients, resulting in bleeding in 29% of patients and in heparin-induced thrombocytopenia in 10% of patients. Other CRRT complications occurred in 5 patients and were related to the catheter placement or to cardiac disturbances. There were no significant differences in complications in patients who survived versus those who died. Conclusions. In conclusion, earlier initiation of CRRT may improve the outcomes of these critically ill patients. Further, surviving patients are likely to recover renal function, and with the judicious use of heparin, CRRT appears to be a safe procedure.