A prediction model for sudden cardiac death in patients with heart failure and preserved ejection fraction

Aims: Sudden cardiac death (SCD) accounts for ∼ 25% of all deaths in heart failure with preserved ejection fraction (HFpEF). However, strategies to identify HFpEF patients at a higher risk of SCD have not been developed.

Methods and results: We studied 4128 patients with HFpEF enrolled in the Irbesartan in Patients with Heart Failure and Preserved Ejection Fraction (I-PRESERVE) trial. All SCDs were adjudicated by a clinical endpoint committee. Cumulative incidences of SCD were estimated counting other deaths as competing risks. Cox regression analysis was used to generate a risk model for SCD. During a mean follow-up of 4.1 years, 231 (5.6%) patients died suddenly and 650 (15.7%) died non-suddenly. A multivariable model in 3480 patients including age, gender, history of diabetes and myocardial infarction, LBBB on ECG, and the natural logarithm of NT-proBNP identified a subgroup of 837 (24%) patients with ≥10% cumulative incidence of SCD over 5 years, accounting for other deaths as competing risk (Harrell's C index 0.75). The 5-year cumulative incidences of SCD in the higher and lower risk groups were 11% and 4%, respectively. In the higher risk group, 32% of deaths were SCD compared with 26% in the entire I-PRESERVE cohort.

Conclusions: A multivariable prediction model identified patients with HFpEF who have a ≥10% risk of SCD over 5 years, similar to the risk of SCD in the Sudden Cardiac Death in Heart Failure (SCD-Heft) trial. This model may be useful for selecting patients with HFpEF for SCD prevention trials.