BACKGROUND Acute renal failure (ARF) after cardiac surgery is a serious adverse event that is associated with high perioperative mortality and prolonged hospitalization. The aim of our study was to evaluate pre- and intraoperative risk factors for the development of ARF requiring hemofiltration after cardiac surgery. METHODS From February 2002 through February 2003, 913 patients underwent cardiac surgery at our institution. Seventy-three patients developed ARF (8.1%), 16 patients were excluded from the study because of chronic end-stage renal insufficiency. Patient characteristics and operative variables were analyzed. A multivariate logistic regression analysis was performed to determine risk factors for ARF. RESULTS Patients who developed ARF were older (P < .001; odds ratio [OR], 1.084; 95% confidence interval [CI], 1.0371.133) than patients who did not develop ARF. Furthermore, cardiopulmonary bypass duration (P = .007; OR, 1.013; 95% CI, 1.004-1.032) and emergent surgery (P = .011; OR, 6.667; CI, 1.538-28.571) were predictive for development of ARF. The strongest predictor for ARF was a preoperative creatinine level >or=2 mg/dL (P < .001; OR, 97.519; 95% CI, 22.363425.252). Most interestingly, even moderately elevated preoperative creatinine levels (1.3-1.99 mg/dL) independently predict ARF after cardiac surgery (P = .001; OR, 3.838; 95% CI, 1.793-8.217). CONCLUSIONS Our data indicate that emergent surgery as well as advanced age and long duration of cardiopulmonary bypass independently predict ARF after cardiac surgery. Most importantly, even slightly impaired preoperative creatinine levels predict the development of ARF requiring hemofiltration after cardiac surgery.