Creation of a Scorecard to Predict In-Hospital Death in Patients Undergoing Operations for Acute Type A Aortic Dissection

Abstract

This study evaluated preoperative predictors of in-hospital death for the surgical treatment of patients with acute type A aortic dissection (Type A) and created an easy-to-use scorecard to predict in-hospital death. We reviewed retrospectively all consecutive patients who underwent operations for acute Type A between 1996 and 2011 at 2 tertiary care institutions. A logistic regression model was created to identify independent preoperative predictors of in-hospital death. The results were used to create a scorecard predicting operative risk. Emergency operations were performed in 534consecutive patients for acute Type A. Mean age was 61 ± 14 years and 36.3% were women. Critical preoperative state was present in 31% of patients and malperfusion of one or more end organs in 36%. Unadjusted in-hospital mortality was 18.7% and not significantly different between institutions. Independent predictors ofin-hospital death were age 50 to 70 years (odds ratio [OR], 3.8; p= 0.001), age older than 70 years (OR, 2.8; p=0.03), critical preoperative state (OR, 3.2; p < 0.001), visceral malperfusion (OR, 3.0; p= 0.003), and coronary artery disease (OR, 2.2; p= 0.006). Age younger than 50 years (OR, 0.3; p= 0.01) was protective for early survival. Using this information, we created an easily usable mortality risk score based on these variables. The patients were stratified into four risk categories predicting in-hospital death: less than 10%, 10% to 25%, 25% to 50%, and more than 50%. This represents one of the largest seriesof patients with Type A in which a risk model wascreated. Using our approach, we have shown that age, critical preoperative state, and malperfusion syndrome were strong independent risk factors for early death andcould be used for the preoperative risk assessment.