Heart failure & cardiomyopathies
Parsimonious echocardiography-based model for predicting long-term all-cause mortality in heart failure with preserved ejection fraction
Authors' conclusion
A five-variable echocardiography-based nomogram showed better discrimination for long-term mortality prediction than the repurposed diagnostic scores evaluated in this cohort. External validation is needed before clinical implementation.
Study context
- Population
- Adults with heart failure with preserved ejection fraction (HFpEF) (left ventricular ejection fraction ≥50%) diagnosed between 2010 and 2016adults with HFpEF · training cohort 554 · validation cohort 238
- Participants
- 792 adultsEvents: 393/554 (70.9%) training; 165/238 (69.3%) validation
- Predictors
- Age, left ventricular posterior wall thickness at end-systole, mitral E velocity, E/e′ ratio and pulmonary artery systolic pressure
- Predicted outcome
- all-cause mortality
Methodological detailsreal-world HFpEF echocardiography database linked to territory-wide electronic health records, randomly split in a prespecified 70:30 ratio, least absolute shrinkage and selection operator (LASSO)-penalised Cox regression refitted as a multivariable Cox model
- Real-world HFpEF echocardiography database linked to territory-wide electronic health records
- Randomly split in a prespecified 70:30 ratio
- LASSO-penalised Cox regression refitted as a multivariable Cox model
Model performance
As reported in the abstract
1/3/5-year
| Test or model | 1/3/5-year AUC |
|---|---|
| Nomogram | 0.658/0.706/0.713 |
| HFA-PEFF | 0.507/0.561/0.642 |
| H 2 FPEF | 0.516/0.533/0.607 |
1/3/5-year AUC
Nomogram0.658/0.706/0.713
HFA-PEFF0.507/0.561/0.642
H 2 FPEF0.516/0.533/0.607
Further findings
Calibration
- Calibration was acceptable at 1 and 3 years but weaker at 5 years
Survival by risk-score tertiles
- Risk-score tertiles separated survival in both cohorts
- log-rank p<0.001
Follow-up
Years
- Median follow-up, training cohort
- 5.17 (IQR 2.26-9.14)
- Median follow-up, validation cohort
- 5.75 (IQR 2.17-9.25)
Original publication
Parsimonious echocardiography-based model for predicting long-term all-cause mortality in heart failure with preserved ejection fraction. Heart. Published 24 March 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →