A risk score for predicting mortality in patients with asymptomatic mild to moderate aortic stenosis

In English Background Prognostic information for asymptomatic patients with aortic stenosis (AS) from prospective studies is scarce and there is no risk score available to assess mortality. Objectives To develop an easily calculable score, from which clinicians could stratify patients into high and lower risk of mortality, using data from the Simvastatin and Ezetimibe in Aortic Stenosis (SEAS) study. Method A search for significant prognostic factors (p < 0.01) among SEAS patients was made by a combined judgemental and statistical elimination procedure to derive a set of three factors (age, gender and smoking) that were forced into the model, and four additional factors captured by the data: left-ventricular mass index, bilirubin, heart rate and natural logarithm of C reactive protein. Calibration was done by comparing observed with calculated number of deaths by tenths of calculated risk using coefficients from the simvastatin + ezetimibe group on placebo group patients. Results Discrimination was good with ROC area of 0.76 for all patients. Estimated probabilities of death were categorised into thirds. An optimised split point of estimated 5-year risk was about 15% (close to the upper 14% tertile split point), with risk 4 times as high in the upper compared to the two lower thirds. The SEAS score performed better than another established high risk score developed for other purposes. Conclusion A new seven factor model for risk stratification of patients with mild to moderate asymptomatic AS identified a high risk group for total mortality with good discrimination properties. Trial registration number ClinicalTrials.gov, NCT 00092677.