Abstract 4141419: Predictors of Mortality and Hospitalization in Heart Failure with Preserved Ejection Fraction with Low NT-proBNP Levels

Introduction: Many patients with heart failure with preserved ejection fraction (HFpEF) have relatively low natriuretic peptide (NP) levels, often related to obesity. The predictors of mortality and hospitalization (CV outcomes) within HFpEF patients with low NT-proBNP levels remain unclear. In this study, we examined these predictors, comparing HFpEF with low NT-proBNP (HFpEF-LoBNP) and high NT-proBNP (HFpEF-HiBNP) groups. Methods: Patients from the Johns Hopkins HFpEF Clinic (July 2014 - May 2024) were categorized into HFpEF-LoBNP (<125 pg/mL) and HFpEF-HiBNP (≥125 pg/mL) groups. Diagnosis relied on clinical findings, standard echocardiography, and right heart catheterization when available. Group differences were analyzed using Chi-square or Mann-Whitney U tests. Cox models examined associations with CV outcomes, and Kaplan-Meier analysis assessed the combined factor impact on outcomes. Analyses compared within each BNP level category (HFpEF-LoBNP and HFpEF-HiBNP) those who experienced CV outcomes versus those who did not. Results: A total of 378 HFpEF patients were included, categorized as HFpEF-LoBNP (107 patients) and HFpEF-HiBNP (271 patients), with a median follow-up period of 36 months. Within the HFpEF-LoBNP group (mean age 59.5 years, 68% female), 26 (24.3%) experienced CV outcomes. HFpEF-LoBNP patients with CV outcomes had higher diabetes prevalence, HbA1c, RA pressure, PA systolic pressure, RV systolic pressure, PA diastolic pressure, PCWP, PVR, and lower albumin (Figure A) compared to HFpEF-LoBNP patients without CV outcomes. No significant differences were found in other characteristics. Within the HFpEF-HiBNP group (mean age 68.2 years, 59.4% female), 98 (36.2%) experienced CV outcomes. HFpEF-HiBNP patients with CV outcome were more likely to be on β-blocker therapy, had a higher E/e' ratio, NT-proBNP, cystatin C, uric acid, RV systolic pressure, PA systolic and diastolic pressures, PCWP, PVR, and lower albumin compared to those without CV outcomes (Figure B). There were no notable differences in other characteristics. Kaplan-Meier graphs showed poorer outcomes with an increasing number of factors in each subgroup (Figures C&D). Conclusion: NT-proBNP is linked to HF prognosis but often low in HFpEF. HFpEF-LoBNP patients still face high CV outcomes. Predictors of CV outcomes in HFpEF-LoBNP, similar to HFpEF-HiBNP, include higher pulmonary pressures. Alternative prognosis markers, like hemodynamic testing, are needed for HFpEF-LoBNP patients.