Background: Acute renal failure increases risk of death after cardiac surgery. However, the definition of acute renal failure is not standardized. The Acute Dialysis Quality Initiative Workgroup has outlined consensus criteria. The aim of the present study was to evaluate this determination of postoperative renal impairment in cardiac surgical patients, and its association with mortality.
Methods: The 813 consecutive patients undergoing cardiac surgery at Helsinki University Central Hospital were analyzed. According to the RIFLE classification (named by the severity of renal impairment: Risk, Injury, Failure, Loss, End-stage kidney disease) patients were divided into three levels based on either plasma creatine level or urine output. The discrimination with death within 90 days after surgery was determined.
Results: According to the RIFLE criteria, 19.3% of patients had renal impairment after cardiac surgery. Patients in the more severe category, RIFLE-F (failure), had a 90-day mortality rate of 32.5% compared with 8.0% for those in RIFLE-R (risk) and 21.4% for RIFLE-I (injury) patients. The RIFLE classification discriminated 90-day mortality quite well (area under curve 0.824) compared with the change of plasma creatinine and the change of estimated gromerular filtration rate (areas under curve 0.849 and 0.829, respectively). The results of the multivariate forward stepwise logistic regression analysis found that RIFLE classification was an independent risk factor for 90-day mortality (p < 0.0001), unlike change in glomerular filtration rate and change in plasma creatinine.
Conclusions: We propose that the RIFLE classification is a valuable method to evaluate acute renal failure after cardiac surgery. The severity of RIFLE classification may be associated with increased 90-day mortality rate.