A-015 Prognostic utility of NT-proBNP for mortality and hospitalization after emergency department presentation for heart failure

Abstract

Heart failure (HF) is a leading cause of emergency department (ED) visits, yet risk stratification in the ED setting remains limited. NT-proBNP is routinely used for HF diagnosis, but its prognostic value at presentation is less defined. In this study, we prospectively assessed whether NT-proBNP levels collected in the ED independently predict 12-month all-cause mortality, cardiac-related mortality, or cardiac-related hospitalization in adults with new-onset or worsening HF. Subjects >18 years old presenting to the ED with a diagnosis of new onset or worsening HF were enrolled and followed for up to 12 months. The primary outcome was a composite outcome defined as the first occurrence of all-cause mortality, or heart-related hospitalization. Each component of the composite outcome was analyzed as a secondary outcome. The concentration of the Alere NT-proBNP assay for Alinity i at ED presentation was categorized into quartiles (Q1, Q2, Q3, and Q4) for prognostic evaluation. Kaplan-Meier curves with log-rank tests were constructed to assess time-to-event across quartiles. Cox proportional hazards regression estimated hazard ratios (HRs) for NT-proBNP using Q1 as the reference, with multivariable models adjusting for age, sex, diabetes mellitus, smoking status, BMI, eGFR category, NYHA class, and hypertension. Of the 861 subjects with HF included in the prognostic evaluation, 495 (57.5%) were male with a mean age of 62.0 years. Kaplan-Meier survival analysis demonstrated statistically significant stratification of the 12-month composite outcome (log-rank test P = 0.0004), with a clear gradient of increasing risk across quartiles of NT-proBNP. In multivariable Cox proportional hazards modeling (n = 816 subjects with complete covariate data), NT-proBNP remained a significant predictor of the composite outcome, with Q3 and Q4 showing elevated risk relative to Q1 (Q3: HR 1.44, 95% CI 1.08–1.93; Q4: HR 1.56, 95% CI 1.14–2.12). Secondary analyses of each individual outcome demonstrated consistent increases in risk across NT-proBNP quartiles, with the strongest associations observed for cardiac-related mortality. Adjusted HRs for cardiac-related mortality were 3.69 (95% CI 1.19–11.42), 3.84 (95% CI 1.22–12.04), and 8.24 (95% CI 2.66–25.50) for Q2, and Q4, respectively. Higher NT-proBNP quartiles at ED presentation were independently associated with increased 12-month risk for the composite of all-cause mortality, or cardiac-related hospitalization, with the steepest gradient for cardiac-related mortality (adjusted HR of 8.24 for Q4 vs Q1). These findings support the use of NT-proBNP levels in the ED for risk stratification, helping clinicians assess risk and guide patient management decisions in acute HF.