Risk stratification for mortality in heart failure with preserved ejection fraction (HFpEF) remains limited with currently used diagnostic scores. A real-world cohort study published in Heart developed and internally validated an echocardiography-based model to estimate long-term all-cause mortality in HFpEF. 

The analysis included 792 adults with HFpEF, defined by left ventricular ejection fraction (LVEF ≥50%), diagnosed between 2010 and 2016 using an echocardiography database linked to territory-wide electronic health records. Participants were randomly assigned to training (n=554) and validation (n=238) cohorts in a prespecified 70:30 ratio. The primary endpoint was all-cause mortality.

Model development used least absolute shrinkage and selection operator (LASSO) penalized Cox regression, followed by multivariable Cox refitting. The final nomogram included age, left ventricular posterior wall thickness at end-systole, mitral E velocity, E/e′ ratio, and pulmonary artery systolic pressure.

During follow-up exceeding five years in both cohorts, mortality occurred in 70.9% of the training cohort and 69.3% of the validation cohort. In the validation cohort, the model showed higher discrimination than Heart Failure Association Pre-test Assessment, Echocardiography and Natriuretic Peptide, Functional testing, Final aetiology (HFA-PEFF) and Heavy, Hypertensive, Atrial Fibrillation, Pulmonary Hypertension, Elder, Filling Pressure (H2FPEF). 

The 1-, 3-, and 5-year area under the curve values were 0.658/0.706/0.713 for the model versus 0.507/0.561/0.642 and 0.516/0.533/0.607 for HFA-PEFF and H2FPEF, respectively. Calibration was acceptable at 1 and 3 years but weaker at 5 years. Risk-score tertiles separated survival outcomes (log-rank p < 0.001).

The model showed higher discrimination than existing diagnostic scores in this cohort and may support mortality risk stratification. External validation is required before clinical implementation.