Effect of risk-adjusted, non-dialysis-dependent renal dysfunction on mortality and morbidity following coronary artery bypass surgery: a multi-centre study

Objective: As little is known about the impact of non-dialysis-dependent renal dysfunction on short- and mid-term outcomes following coronary surgery we have conducted a large multi-centre study comparing patients with no history of renal dysfunction to those with preoperative renal dysfunction. Methods: Data was prospectively collected on 19,625 consecutive patients undergoing isolated coronary surgery between 1997 and 2003 from four institutions. Sixty-seven patients had a history of dialysis support prior to coronary surgery, and were excluded from the main analysis of the study. The remaining 19,558 patients were divided into two groups based on preoperative serum creatinine level, patients with preoperative renal dysfunction with serum creatinine levels >200 μmol/L without dialysis support and control patients with preoperative serum creatinine levels <200 μmol/L. Case-mix was accounted for by developing a propensity score, which was the probability of belonging to the non-dialysis-dependent renal dysfunction group, and included in the multivariable analyses. Results: There were 19,172 patients with preoperative serum creatinine levels <200 μmol/L and 386 patients with serum creatinine levels >200 μmol/L without dialysis support. The propensity score included sex, body mass index, co-morbidity factors (respiratory disease, diabetes, cerebrovascular disease, hypertension, and hypercholesterolemia), ejection fraction, left main stem stenosis, emergency status, prior cardiac surgery, off-pump surgery, and the logistic EuroSCORE. After adjusting for the propensity score, patients with preoperative non-dialysis-dependent renal dysfunction had significantly higher in-hospital mortality (adjusted odds ratio 3.0, p < 0.001), stroke (adjusted odds ratio 2.0, p = 0.033), atrial arrhythmia (adjusted odds ratio 1.5, p = 0.003), prolonged ventilation (adjusted odds ratio 2.1, p < 0.001), and post-op stay > 6 days (adjusted odds ratio 2.6, p < 0.001). One thousand one hundred and eighty-three (6.1%) deaths occurred during 58,062 patient-years follow-up. After adjusting for the propensity score, the adjusted hazard ratio of mid-term mortality for non-dialysis-dependent renal dysfunction was 2.7 (p < 0.001). Conclusions: Patients undergoing coronary surgery with non-dialysis-dependent renal dysfunction have significantly increased perioperative morbidity and mortality. Mid-term survival is also significantly reduced at 5-years.