Risk prediction in patients with heart failure with preserved ejection fraction: the LIFE-Preserved model

Area of Science:

• Cardiology
• Clinical Prediction Models
• Public Health

Background:

• Heart failure with preserved ejection fraction (HFpEF) is a growing concern with heterogeneous patient prognoses.
• Accurate risk stratification is crucial for identifying HFpEF patients who would benefit most from interventions.
• The LIFE-Preserved model aims to address this need for improved risk prediction.

Purpose of the Study:

• To develop and validate the LIFE-Preserved model for predicting individual short-term and lifetime risk of HF hospitalization or cardiovascular (CV) death in HFpEF patients.
• To provide a tool for identifying high-risk HFpEF individuals for optimized clinical management.

Main Methods:

• The LIFE-Preserved model was derived using data from 20,332 HFpEF patients in the Swedish HF Registry (SwedeHF).
• Cause- and sex-specific Cox models incorporated 14 routinely available predictors, using age as the timescale for extended predictions.
• External validation was conducted across two clinical trials and three registries, assessing discrimination and calibration.

Main Results:

• The model was validated in 28,062 HFpEF patients from external sources, demonstrating good performance.
• Pooled C-statistics were 0.714 in trials and 0.658 in registries, with adequate calibration across all validation datasets.
• Model performance was consistent between sexes, and an interactive calculator is available.

Conclusions:

• The LIFE-Preserved model effectively predicts both short-term and lifetime risks of HF hospitalization or CV death in HFpEF patients.
• This model can serve as a valuable clinical tool for identifying high-risk patients, guiding treatment decisions and shared decision-making.