Coronary artery bypass grafting in non–dialysis-dependent mild-to-moderate renal dysfunction

Abstract Objectives: The effect of mild-to-moderate elevation of preoperative serum creatinine levels on morbidity and mortality from coronary artery bypass grafting has not been investigated in a large multivariable model incorporating preoperative and intraoperative variables. Our first objective was to ascertain the effect of a mild-to-moderate elevation in the preoperative serum creatinine level on the need for mechanical renal support; the duration of special care and total postoperative stay; the occurrence of infective, respiratory, and neurologic complications; and hospital mortality. Our second objective was to ascertain which patient variables contributed to an increase in the serum creatinine level in association with coronary artery bypass grafting. Methods: A total of 1427 patients who had no known pre-existing renal disease and who were undergoing first-time coronary artery bypass grafting with cardiopulmonary bypass were recruited for the study. Patients were divided, on the basis of preoperative serum creatinine level, into 3 groups as follows: creatinine level of less than 130 μmol · L –1 ; creatinine level of 130 to 149 μmol · L –1 ; and creatinine level of 150 μmol · L –1 or greater. A multivariable stepwise logistic regression analysis was used, and variables significant at the 5% level were included when developing the final multivariable models. Results: Multivariable analysis showed that elevation of the preoperative serum creatinine level to 130 μmol · L –1 or greater increased the likelihood of needing mechanical renal support postoperatively ( P P P –1 ( P =.045) and to 150 μmol · L –1 or greater ( P –1 ( P =.02), patients with preoperative serum creatinine levels of 150 μmol · L –1 or greater ( P =.001), hypertensive patients ( P =.007), patients with angina of New York Heart Association class III or greater ( P =.001), patients having a nonelective operation ( P =.002), and patients having a prolonged cardiopulmonary bypass time ( P =.008) had a significantly greater increase in the serum creatinine level as a result of coronary artery bypass grafting. Of particular note was the finding that the method of myocardial protection (cardioplegia or crossclamp fibrillation) did not significantly influence in-hospital mortality, need for mechanical renal support, or special care or total postoperative hospital stay. Conclusions: A mild elevation (130-149 μmol · L –1 ) in the preoperative serum creatinine level significantly increases the need for mechanical renal support, the duration of special care and total postoperative stay, and the in-hospital mortality. As the preoperative serum creatinine level increases further (≥150 μmol · L –1 ), this effect is more pronounced. No significant difference in outcome was observed between the use of cardioplegia or crossclamp fibrillation for myocardial protection. (J Thorac Cardiovasc Surg 2001;121:1083-9)