Predicting In-Hospital Mortality After Redo Cardiac Operations: Development of a Preoperative Scorecard

Background. The present study generated a risk model and an easy-to-use scorecard for the preoperative prediction of in-hospital mortality for patients undergoing redo cardiac operations. Methods. All patients who underwent redo cardiac operations in which the initial and subsequent procedures were performed through a median sternotomy were included. A logistic regression model was created to identify independent preoperative predictors of in-hospital mortality. The results were then used to create a scorecard predicting operative risk. Results. A total of 1,521 patients underwent redo procedures between 1995 and 2010 at a single institution. Coronary bypass procedures were the most common previous (58%) or planned operations (54%). The unadjusted in-hospital mortality for all redo cases was higher than for first-time procedures (9.7% vs. 3.4%; p < 0.001). Independent predictors of in-hospital mortality were a composite urgency variable (odds ratio [OR], 3.47), older age (70-79 years, OR, 2.74; >= 80 years, OR, 3.32), more than 2 previous sternotomies (OR, 2.69), current procedure other than isolated coronary or valve operation (OR, 2.64), preoperative renal failure (OR, 1.89), and peripheral vascular disease (PVD) (OR, 1.55); all p < 0.05. A scorecard was generated using these independent predictors, stratifying patients undergoing redo cardiac operations into 6 risk categories of in-hospital mortality ranging from < 5% risk to > 40%. Conclusions. Reoperation represents a significant proportion of modern cardiac surgical procedures and is often associated with significantly higher mortality than first-time operations. We created an easy-to-use scorecard to assist clinicians in estimating operative mortality to ensure optimal decision making in the care of patients facing redo cardiac operations.