Background: The recent PREVENT-HF risk equations were developed to estimate 10-year risk of overall HF. While HF with preserved vs reduced ejection fraction (HFpEF; HFrEF) have common risk factors, distinct differences in risk conferred by older age, obesity, and coronary artery disease are known. In this context, we sought to evaluate the performance of PREVENT-HF equations relative to HFpEF vs HFrEF. Methods: We examined participants from 4 pooled community-based cohort studies: the Multi-Ethnic Study of Atherosclerosis (MESA), Framingham Heart Study (FHS) offspring exam 6, Cardiovascular Heart Study (CHS), and Prevention of Renal and Vascular End-stage Disease (PREVEND) study. We excluded those with ages 79 years, prevalent HF, prevalent CVD, and missing key clinical information. We compared the performance of the PREVENT-HF equation in predicting incident HFpEF vs HFrEF at 10 years using Uno’s C-statistics and area under receiver operating characteristic (AUROC). Results: Among 22,061 participants (age 59 ± 14 years, 54% women, BMI 27 ± 5 kg/m 2 ), 322 developed HFpEF and 464 HFrEF during a median follow-up of 10 years. The PREVENT-HF risk equation performed well for both HFpEF (c-statistic 0.81, 95% CI 0.80-0.83) and HFrEF (c-statistic 0.78 (95% CI 0.76-0.80), with similar improvement compared with age and sex alone (HFpEF: c-statistic 0.78, 95% CI 0.76-0.80; HFrEF: c-statistic 0.75, 95% CI 0.73-0.76, P<0.001 for delta for both). The AUROC remained stable across 1-10 years demonstrating consistently strong discrimination for both HFpEF and HFrEF. Time-dependent AUCs were generally higher for HFpEF compared with HFrEF, indicating slightly better discrimination for HFpEF across all time horizons ( Figure ). Conclusions: Our findings demonstrate overall similar performance of the PREVENT-HF risk model with respect to HFpEF and HFrEF. Future studies on subtype-specific risk models may improve upon current risk prediction tools for overall HF.
Parekh et al. (Tue,) studied this question.