Determinants and prognostic value of echocardiographic first-phase ejection fraction in aortic stenosis

Figures

• 

  Figure 1. Study flowchart study design demonstrating patient population and investigations at baseline and follow-up. AVR, aortic valve replacement; CMR, cardiovascular magnetic resonance; EF1, first-phase ejection fraction.

• 

  Figure 2. Distribution of EF1 and ejection fraction by aortic stenosis severity box plot demonstrates EF1 according to mean gradient (<20mm Hg: 30 (27–33)%, 20–39 mm Hg: 26 (20–30)%, ≥40 mm Hg: 15 (12–25)%). Density plots demonstrate the distribution of EF1 and ejection fraction among patients with aortic stenosis stratified by mean gradient. The dashed reference lines denote 25% and 50% for EF1 and ejection fraction, respectively. EF1, first-phase ejection fraction.

• 

  Figure 3. Association between EF1 and outcomes The Kaplan-Meier curve (left) demonstrates unadjusted survival free of AVR or death in the study cohort (n=149), stratified by high (≥25%) or low (<25%) EF1 (p value for log-rank test). The Cox regression curve (bottom right) demonstrates the predicted HR for AVR or death according to EF1, adjusted for age, mean gradient and indexed extracellular volume. The distribution of EF1 in the study cohort is presented in the rug plot below the regression curve. AVR, aortic valve replacement; EF1, first-phase ejection fraction.

• 

  Figure 4. Multivariable predictors of AVR or death forest plots of multivariable Cox proportional hazards models for the composite endpoint of AVR or death. HR scales are presented as logarithmic scales. NYHA was stratified according to class (reference: class I). (A) Model for the whole cohort. Aortic stenosis severity was stratified by mean gradient (<20 (reference), 20–39 and ≥40 mm Hg). iECV was stratified by tertiles (first tertile as reference). (B) Model for patients with a mean gradient of ≥40 mm Hg (n=44; 36 had AVR, 5 died before AVR). The model was restricted to these three variables to avoid overfitting. The mean gradient was analysed as a continuous variable in keeping with the lack of a clear threshold in clinical practice for these patients. AVR, aortic valve replacement; EF1, first-phase ejection fraction; iECV, indexed extracellular volume; LGE, late gadolinium enhancement; NYHA, New York Heart Association.