Abstract 14976: Doppler-Echocardiographic Predictors of Outcomes in a Large Series of Patients With Aortic Stenosis
BACKGROUND: Several studies have reported that peak aortic jet velocity and mean gradient (MG) are powerful predictors of outcomes in aortic stenosis (AS). However, these studies generally used aortic valve replacement (AVR) or the composite of AVR or death as primary end-point. Few studies have examined the relationship between these parameters and overall mortality. The aim of this study was thus to assess the value of the parameters of AS severity and LV function to predict AVR and mortality in AS population.
METHODS: 1065 patients (71±13 y.o., 59% men) with AS (53% severe) were included. ROC analysis was used to determine accuracy and optimal cut-point values (CPV) of parameters of AS severity (MG and aortic valve area [AVA]), LV systolic function (LVEF), and LV outflow (stroke volume index [SVi]) to predict mortality and AVR at 1- and 2-years follow-up.
RESULTS: During a mean follow-up of 5.4±3.7 yrs, 550 patients died and 584 underwent AVR. The predictors of AVR at 1- and 2-yrs were higher MG (HR=1.53 per ↑10 mmHg, p<0.001; 1-yr CPV: 31; 2-yrs CPV: 26 mmHg) and smaller AVA (HR=1.24 per ↓0.1 cm2, p<0.001; 1-yr CPV: 1.01; 2-yrs CPV: 1.01 cm2). The predictors of 1- and 2-yrs mortality were lower LVEF (HR= 1.14 per ↓5%, p<0.001; 1-yr and 2-yrs CPV: 56%), lower SVi (HR=1.22 per ↓5ml/m2, p<0.001; 1-yr CPV: 34.2; 2-yrs CPV: 36.5 ml/m2) and smaller AVA (HR=1.05 per ↓0.1 cm2, p=0.002; 1-yr CPV: 1.01; 2-yrs CPV: 1.05 cm2). MG was not associated with mortality. In multivariable analysis adjusted for age, gender, hypertension, coronary artery disease, diabetes, and renal failure, LVEF (HR=1.07 per ↓5%, p<0.001) and SVi (HR=1.07 per ↓5ml/m2, p=0.02) were independent predictors of mortality, whereas, after similar adjustment, only MG (HR=1.54, p<0.001) predicted occurrence of AVR.
CONCLUSION: MG is associated with increased risk of AVR but not of mortality. The most powerful predictors of mortality were low LVEF and low flow. Furthermore, these outcome data confirm the validity of the 35ml/m2 cut-point of SVi and they suggest that the cut-point of LVEF used to define LV systolic dysfunction in AS should be raised from 50 to 55%. Finally, the optimal CPV of AVA to predict outcomes was very similar to that proposed in the guidelines (1.0 cm2), whereas the CPV of MG was much lower (40 mmHg).
- © 2013 by American Heart Association, Inc.