Predictive Factors of Mortality in Acute Aortic Dissection and Validity of the EuroSCORE Algorithm in a Small-Sized Cardiac Surgery Institution

INTRODUCTION Acute aortic dissection (AAD) is a devastating surgical emergency, with high operative mortality. Several scoring algorithms have been used to establish the expected mortality in these patients. Our objective was to define the predictive factors for mortality in our center and to validate the EuroSCORE and Penn classification system. METHODS Patients who underwent surgery for AAD from 2006 to 2016 were retrieved from the institution's database. Preoperative, operative and postoperative variables were collected. Observed and expected mortality was calculated by EuroSCORE. Logistic regression analysis and Cox regression analysis were performed to find predictors of operative mortality and survival, respectively. The receiver operating characteristic (ROC) curves were plotted for logistic EuroSCORE, and the area under the ROC curve (AUC) was calculated. RESULTS 87 patients (27.6% female) underwent surgery for AAD. The mean age was 58.6±9.7 years. Expected and observed operative mortality was 25.8±15.1% and 20.7%, respectively. Penn Aa, Ab and Abc shared similar observed/expected (O/E) mortality ratio. The only independent predictor of operative mortality (OR: 3.63; 95% CI: 1.19-11.09) and survival (HR: 2.6; 95% CI: 1.5-4.8) was female gender. EuroSCORE showed a very poor prediction capacity, with an AUC=0.566. CONCLUSION Female gender was the only independent predictor of operative mortality and survival in our institution. EuroSCORE is a poor scoring algorithm to predict mortality in AAD, but with consistent results for Penn Aa, Ab and Abc.