Introduction: Heart Failure preserved Ejection Fraction (HFpEF) is a major burden on the healthcare system and better understanding risk prediction in this population may help guide management. Our objective was to assess the complementary value of existing heart failure risk scores, Get with The Guidelines (GWTG), Intermountain Mortality Risk Score (IMRS) and the Meta-Analysis Global Group in Chronic Heart Failure (MAGGIC) Scores in an independent cohort of patients hospitalized with acute HFpEF.
Methods: Using Stanford Translational Research Integrated Database Environment and individual chart review, we identified 580 adult patients hospitalized with acute heart failure and left ventricular ejection fraction ≥ 50%. We excluded patients with advanced liver/kidney disease, valve replacement, transplantation, active malignancy, pulmonary arterial hypertension and hypertrophic cardiomyopathy. A sub-group of 341 patients had NT-proBNP levels available. Mortality status was determined using United States Social Security Death Index and chart review. Cox proportional hazard analysis was used to determine the heart failure risk scores predictive of all-cause mortality in HFpEF.
Results: The mean age was 76 ± 15 years with 334 (57.6%) females. Median length of hospital stay was 4 (IQR 2 - 9) days and follow up of 2.0 (IQR 0.2 - 4.6) years with a total of 139 (24%) deaths. On multivariate analysis, the GWTG, IMRS and MAGGIC scores were independent correlates of all-cause mortality with normalized hazard ratios of 1.48 (1.17 - 1.86), 1.41 (1.15 - 1.76) and 1.31 (1.04 - 1.70) respectively (Figure). In patients with NT-proBNP collected, NT-proBNP remained significantly associated with mortality with normalized hazard ratio of 1.52 (1.22 - 1.91) even after adjustment for the risk scores.
Conclusion: In patients hospitalized with acute HFpEF, established heart failure Risk scores together with NT-proBNP can play a complementary role in outcome prediction.