111 Development of a Novel Scorecard Predicting In-Hospital Mortality Following Redo Cardiac Surgery

Abstract

Redo cardiac surgery is known to be associated with a high rate of morbidity and mortality. Most of the literature, however, focuses on risk factors for adverse outcome following redo coronary surgery. The objective of this study was to identify independent predictors of in-hospital mortality following redo cardiac surgery of various types and to develop an easy to use predictive scorecard. All patients at a single institution who received cardiac surgery via sternotomy multiple times between 1995 and 2010 were included. Thoracotomy procedures, ventricular assist device implantations, and transplants were excluded from the study population. Clinical characteristics of patients who survived and patients who died after redo surgery were examined univariately. Variables with p-value <0.2 or those considered to be clinically relevant were retained for multivariate analysis. A logistic regression model was developed and variables were retained if the final p-value was <0.05. The final model was then used to create a scorecard. 1521 out of 17645 patients were included in the study. The unadjusted in-hospital mortality for patients undergoing redo surgery was 9.7% compared to 3.4% for those undergoing a first time procedure (p<0.001). The following variables were identified as independent predictors of in-hospital mortality following redo cardiac surgery of various types: a composite urgency variable that included active endocarditis, urgent or emergent redo surgery, myocardial infarction within 21 days of the redo procedure, or preoperative intra aortic balloon pump (OR=3.47); older age (70-79 OR=2.74, >=80 OR=3.32); >2 previous sternotomies (OR=2.69); current procedure other than isolated coronary or valve surgery (OR=2.64); preoperative renal failure (OR=1.89); and peripheral vascular disease (OR=1.55) - all p<0.05. These variables were used to generate a scorecard that stratifies patients undergoing redo cardiac surgery into 6 categories of in-hospital mortality risk ranging from <5% to >40% risk. Redo cardiac surgery is often associated with significantly higher in-hospital mortality than first time procedures. As far as we know, we have created the first preoperative scorecard for predicting in-hospital mortality specific to redo procedures of various types. The scorecard is easy to use and can assist clinicians in identifying the most appropriate candidates for redo surgery.