(1) Background: Our goal was to develop a risk prediction model for mortality in patients with moderate and severe aortic stenosis (AS). (2) Methods: All patients aged 40−95 years, with echocardiographic evidence of moderate and severe AS at a single institution, were studied over a median of 2.8 (1.5−4.8) years, between 2013−2018. Patient characteristics and mortality were compared using Chi-squares, t-tests, and Kaplan−Meier (KM) curves, as appropriate. The risk calculation for mortality was derived using the Cox proportional hazards model. A risk score was calculated for each parameter, and the total sum of scores predicted the individualized risks of 1-and 5-year mortality. (3) Results: A total of 1991 patients with severe and 2212 with moderate AS were included. Severe AS patients were older, had a lower ejection fraction %, were more likely to be Caucasian, and had lower rates of obesity and smoking, but had higher rates of cardiac comorbidities and AVR (49.3% vs. 2.8%, p < 0.0001). The unadjusted overall mortality was 41.7% vs. 41%, p = 0.6530, and was not different using KM curves (log rank, p = 0.0853). The models included only patients with complete follow-up (3966 in the 1-year, and 816 in the 5-year model) and included 13 variables related to patient characteristics, degree of AS, and AVR. The C-statistic was 0.75 and 0.72 for the 1-year and the 5-year models, respectively. (4) Conclusions: Patients with moderate and severe AS experience high morbidity and mortality. The usage of a risk prediction model may provide guidance for clinical decision making in complex patients.
Keywords: aortic; mortality; prediction; risk; stenosis.