Impact of preoperative serum creatinine on short- and long-term mortality after cardiac surgery: a cohort study.

Abstract

Preoperative renal insufficiency is an important predictor of mortality after cardiac surgery. This retrospective cohort study was designed to identify the optimal cut-off for baseline serum creatinine (bSCr) and estimated glomerular filtration rate (eGFR) to predict survival. Furthermore, we investigated the potential confounding effect of other perioperative risk indicators on short- and long-term survival. Data of 9490 cardiac surgical patients were prospectively collected between 1997 and 2008 (follow up to 2010) at the Medical University Vienna. We identified bSCr cut-off values and calculated uni- and multivariate hazard models for short- and long-term survival and compared the results with a validation set from Zurich. The estimated survival curves defined a distinct period of increased mortality until 150 days. Cut-off values of >115 µmol litre(-1) for bSCr and ≤50 ml min(-1) for eGFR were identified. Increased bSCr, associated with higher mortality [hazard ratio (HR) 2.61, 95% confidence interval (CI) 2.43-2.80, P<0.0001], was present in 19.5% of patients and remained predictive for short- (HR 1.59, 95% CI 1.38-1.83, P=0.0027) and long-term survival (HR 1.46, 95% CI 1.32-1.62, P<0.0001) in the multivariate hazard models. A cut-off of >120 µmol litre(-1) for bSCr was determined for the validation set. Decreased eGFR was present in 23.6% (HR 2.86, 95% CI 2.67-3.06, P<0.0001). In our patients, increased bSCr was an independent predictor of mortality, which may critically influence risk evaluation and perioperative treatment guidance.