Use of a new scoring system to predict the risk of chronic postoperative renal failure in cardiac surgery

Background. The aim of this study was to cross validate a new scoring system, based on preoperative risk factors, in predicting the risk of postoperative dialysis after cardiac surgery and recently proposed by Thakar et al. Methods. Between January 2003 and March 2006, 1642 consecutive patients underwent cardiac surgery. From our clinical database, using Thakar's method, we calculated the total score for each patient. Univariate analysis was applied to all the risk factors considered for the score to test their statistical power in predicting postoperative acute renal failure. The effectiveness of the scoring system was analyzed with univariate analysis and with the area under the ROC curve. Results. Postoperative dialysis was necessary in 22 patients. The factors that predicted acute renal failure, identified by univariate analysis, were the following: previous cardiac surgery, emergency surgery, coronary artery bypass graft associated with valve procedure, preoperative creatinine (1.2 to 2.1 mg/dl) and preoperative creatinine >= 2.1 mg/dl. The scoring system resulted statistically significant in predicting dialysis risk (p < 0.001), the area under the ROC curve was 0.82. The incidence of renal failure in the four scoring groups, identified by Thakar, was similar in the first two groups but higher in the last two groups. Conclusions. Thakar's scoring system was validated in our population and all the major risk factors considered for its calculation were statistically significant in predicting postoperative dialysis after cardiac surgery.