Impact of minimal increases in serum creatinine on outcome in patients after cardiothoracic surgery: Do we have to revise current definitions of acute renal failure?*

Abstract

Traditional cutoff values of serum creatinine considered to define postoperative acute renal failure have been challenged recently. In a previous investigation we demonstrated that minimal changes in serum creatinine concentration were associated with a substantial decrease in survival after cardiac surgery. In this investigation, we assessed the impact of minimal absolute increases in serum creatinine in a second institution, and we analyzed whether relative changes, as in the RIFLE classification and, partially, in Acute Kidney Injury Network (AKIN) classification, confer a different prognostic potential. Prospective analysis. University hospital. All consecutive patients undergoing cardiac surgery in the University Hospital of Zurich (Center USZ) over a 46-month period. Patients were prospectively documented. We analyzed maximal changes in serum creatinine in the first 48 hrs postoperatively (DeltaCrea) regarding death within 30 days. Results were compared with those of the University Hospital Vienna (Center AKH). Moreover, the prognostic potential of DeltaCrea within 48 hrs vs. serum creatinine elements according to RIFLE and AKIN classifications was assessed. A total of 3,123 patients were evaluated from USZ. The majority of patients had decreased postoperative serum creatinine values (negative DeltaCrea) and the lowest mortality (1.8%). Minimal increases, [0, 0.5) mg x dL(-1), were associated with a more than doubled mortality in both centers (5%/6%). Mortality, according to RIFLE and AKIN classifications for both populations combined, was as follows: 7,023 (3.6%), 160 (29%), 43 (19%), and 15 (33%) for RIFLE Normal, Risk, Injury, and Failure; 6,644 (2.8), 463 (16.4), 3 (66.7), and 131 (1.8) for AKIN stage 0, 1, 2, and 3. Measuring repeat serum creatinine concentrations within 48 hrs and determining DeltaCrea were the most effective discrimination method to find patients at risk for adverse postoperative outcome after cardiac surgery, better than application of this sole criterion to the RIFLE (least discriminatory) or the AKIN classification.