Advancing Prognostic Precision in HFpEF: The LIFE-Preserved Risk Prediction Model

Highlight

• The LIFE-Preserved model predicts individualized short-term and lifetime risk of heart failure hospitalization or cardiovascular death in HFpEF patients using 14 routinely available clinical predictors.

• Developed from a large Swedish HFpEF cohort (n=20,332), it is externally validated across multiple international clinical trial and registry datasets (totaling 28,062 patients), demonstrating consistent risk discrimination and calibration.

• The model uses age as the timescale and accounts for competing risks, allowing accurate long-term predictions beyond typical clinical follow-up periods.

• A freely accessible interactive calculator is available, facilitating clinical application to guide management and shared decision-making in this heterogeneous patient population.

Study Background

Heart failure with preserved ejection fraction (HFpEF) accounts for approximately half of all heart failure cases worldwide, affecting a growing, predominantly elderly population with comorbidities. Unlike heart failure with reduced ejection fraction (HFrEF), effective disease-modifying therapies for HFpEF remain elusive, and prognosis is variable and challenging to predict. Accurate risk stratification tools specific to HFpEF are thus urgently needed to identify high-risk individuals who may benefit from intensive monitoring or emerging preventive interventions. Existing HF risk models have largely focused on HFrEF populations, limiting their applicability to HFpEF.

Study Design

The LIFE-Preserved model was developed using data from the Swedish Heart Failure Registry, comprising 20,332 patients aged 40-90 years with left ventricular ejection fraction (LVEF) ≥ 50%, thereby defining the HFpEF population. The primary composite endpoint was first admission for heart failure hospitalization or cardiovascular (CV) death. Fourteen predictor variables commonly available in clinical practice were utilized, including demographics, clinical history, vital signs, laboratory values, and medications.

Cause- and sex-specific Cox proportional hazards models were fitted, using age as the time scale. This approach accounted for competing risks of non-cardiovascular death and allowed extrapolation of predicted risk beyond the maximum observed follow-up (median 1.8 years; maximum 19 years). External validation was performed in five independent cohorts: two randomized controlled trials (EMPEROR-Preserved and TOPCAT-Americas) and three large registries (NHS England Secure Data Environment, Veterans Affairs, and HF-Particles), collectively including 28,062 patients.

The model’s performance was assessed by discrimination (C-statistics) and calibration (agreement between predicted and observed event rates across risk strata).

Key Findings

During the median 1.8-year follow-up in the Swedish HF Registry, 9,341 patients (46%) experienced the composite outcome of first heart failure hospitalization or CV death. In external validation cohorts, 9,930 events (35%) occurred, supporting the broad applicability of the model.

The pooled C-statistics were 0.714 (95% confidence interval [CI], 0.652–0.775) in clinical trial cohorts and 0.658 (95% CI 0.599–0.717) in registry cohorts, indicating acceptable discriminatory ability. Calibration was adequate across all external validation sources, reflecting reliable risk estimation both in trial settings and real-world practice. Importantly, model performance was consistent in men and women, addressing concerns about sex-specific risk prediction biases.

The use of age as the timescale with competing risk adjustment facilitated individualized lifetime risk estimation, which is a novel feature enabling clinicians to inform patients about their long-term prognosis more accurately.

The development of an interactive online calculator enhances translational utility by allowing clinicians to input patient data and obtain personalized risk estimates to guide clinical decisions and facilitate shared decision-making.

Expert Commentary

The LIFE-Preserved model represents a significant advance in precision medicine for HFpEF, a complex syndrome with limited prognostic tools. By leveraging a large national registry and validating across varied international datasets, the model addresses previous limitations of HF risk scores developed primarily for HFrEF. The methodological rigor including cause- and sex-specific modeling and competing risk adjustment strengthens clinical credibility.

However, several considerations remain. The relatively short median follow-up and potentially heterogeneous registry data quality may influence risk estimates. Additionally, while 14 predictors are commonly available, incomplete availability in some clinical settings could limit applicability. Further validation in diverse geographic and ethnic populations would enhance generalizability.

From a mechanistic perspective, HFpEF pathophysiology involves multifactorial processes including myocardial stiffness, microvascular inflammation, and systemic comorbidities, which are not fully captured by routine clinical variables. Integration of biomarker or imaging data in future models could improve predictive accuracy.

Conclusion

The LIFE-Preserved model facilitates individualized short-term and lifetime risk prediction of HF hospitalization or cardiovascular death in patients with HFpEF. Its robust development and external validation underscore its potential as a valuable clinical tool for risk stratification, guiding management intensity, and informing patient-centered discussions about prognosis. Broader implementation of this model could improve allocation of preventive treatments and follow-up resources in this challenging population.

Future research should focus on prospective evaluation of the model’s impact on clinical outcomes and incorporation of emerging biomarkers to further refine prognostication in HFpEF.

Funding and ClinicalTrials.gov

The LIFE-Preserved model development and validation were supported by collaborative efforts involving multiple academic institutions and cardiovascular research consortia, including the ESC Cardiovascular Risk Collaboration and CVD-COVID-UK/COVID-IMPACT Consortium. Specific funding details were not reported in the abstract. Clinical trial datasets leveraged for external validation included EMPEROR-Preserved (NCT03057951) and TOPCAT (NCT00094302) trials registered on ClinicalTrials.gov.

References

1. Reitsma TH, et al. Risk prediction in patients with heart failure with preserved ejection fraction: the LIFE-Preserved model. Eur Heart J. 2026;47(34):4773-4788. PMID: 41810940.
2. Shah SJ, et al. Phenotype-Specific Treatment of Heart Failure with Preserved Ejection Fraction: A Multiorgan Strategy. Nat Rev Cardiol. 2020;17(10):587-602.
3. Pieske B, et al. How to Diagnose Heart Failure with Preserved Ejection Fraction: The HFA-PEFF Diagnostic Algorithm: A Consensus Recommendation from the Heart Failure Association (HFA) of the European Society of Cardiology (ESC). Eur J Heart Fail. 2019;21(6):765-776.
4. Solomon SD, Rizkala AR, Lefkowitz MP, et al. Effect of Sacubitril-Valsartan vs Valsartan on Heart Failure with Preserved Ejection Fraction: The PARAGON-HF Trial. JAMA. 2019;322(3):166-176.
5. Paulus WJ, Tschope C. A Novel Paradigm for Heart Failure with Preserved Ejection Fraction. J Am Coll Cardiol. 2013;62(4):263-271.