Personalized lifetime prediction of survival and treatment benefit in patients with heart failure with reduced ejection fraction: The LIFE-HF model

Pascal M. Burger, Gianluigi Savarese, Jasper Tromp, Carly Adamson, Pardeep S. Jhund, Lina Benson, Camilla Hage, Wan Ting Tay, Scott D. Solomon, Milton Packer, Xavier Rossello, John W. McEvoy, Dirk De Bacquer, Adam Timmis, Panos Vardas, Ian M. Graham, Emanuele Di Angelantonio, Frank L.J. Visseren, John J.V. McMurray, Carolyn S.P. LamLars H. Lund, Stefan Koudstaal, Jannick A.N. Dorresteijn, Arend Mosterd*

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Aims: Although trials have proven the group-level effectiveness of various therapies for heart failure with reduced ejection fraction (HFrEF), important differences in absolute effectiveness exist between individuals. We developed and validated the LIFEtime-perspective for Heart Failure (LIFE-HF) model for the prediction of individual (lifetime) risk and treatment benefit in patients with HFrEF. Methods and results: Cox proportional hazards functions with age as the time scale were developed in the PARADIGM-HF and ATMOSPHERE trials (n = 15 415). Outcomes were cardiovascular death, heart failure (HF) hospitalization or cardiovascular death, and non-cardiovascular mortality. Predictors were age, sex, New York Heart Association class, prior HF hospitalization, diabetes mellitus, extracardiac vascular disease, systolic blood pressure, left ventricular ejection fraction, N-terminal pro-B-type natriuretic peptide, and glomerular filtration rate. The functions were combined in life-tables to predict individual overall and HF hospitalization-free survival. External validation was performed in the SwedeHF registry, ASIAN-HF registry, and DAPA-HF trial (n = 51 286). Calibration of 2- to 10-year risk was adequate, and c-statistics were 0.65–0.74. An interactive tool was developed combining the model with hazard ratios from trials to allow estimation of an individual's (lifetime) risk and treatment benefit in clinical practice. Applying the tool to the development cohort, combined treatment with a mineralocorticoid receptor antagonist, sodium–glucose cotransporter 2 inhibitor, and angiotensin receptor–neprilysin inhibitor was estimated to afford a median of 2.5 (interquartile range [IQR] 1.7–3.7) and 3.7 (IQR 2.4–5.5) additional years of overall and HF hospitalization-free survival, respectively. Conclusion: The LIFE-HF model enables estimation of lifelong overall and HF hospitalization-free survival, and (lifetime) treatment benefit for individual patients with HFrEF. It could serve as a tool to improve the management of HFrEF by facilitating personalized medicine and shared decision-making.

Original languageEnglish
Pages (from-to)1962-1975
Number of pages14
JournalEuropean Journal of Heart Failure
Volume25
Issue number11
DOIs
Publication statusPublished - Nov 2023

Keywords

  • Heart failure with reduced ejection fraction
  • Individual treatment benefit
  • Lifetime risk
  • Personalized medicine
  • Risk prediction
  • Shared decision-making

Fingerprint

Dive into the research topics of 'Personalized lifetime prediction of survival and treatment benefit in patients with heart failure with reduced ejection fraction: The LIFE-HF model'. Together they form a unique fingerprint.

Cite this