Admission NT-proBNP and in-hospital mortality in acute heart failure: a real-world emergency cohort study
D Andric, D Golubov, S Cemerlic Maksimovic, T Popov, A Gvozdenovic, S Andric, V Lecic, T Miljkovic, M Petrovic, A IlicAbstract
Background
Acute heart failure (AHF) remains a frequent cause of emergency admission and is associated with substantial in-hospital mortality. Readily available biomarkers and echocardiography may support early risk assessment in routine care
Purpose
To describe a real-world AHF emergency cohort and compare survivors vs non-survivors.
Methods
Observational study of consecutive emergency admissions with a primary diagnosis of AHF in 2025. Patients with acute myocardial infarction or severe aortic stenosis were excluded. Admission NT-proBNP and transthoracic echocardiography were collected (LVEF, E/e’, TAPSE, estimated RVSP) when available. Primary outcome was in-hospital death. Continuous variables are reported as median [IQR]; available-case comparisons used Mann–Whitney U.
Results
519 patients were included (mean age 67.8±13.1 years; 64.4% male). Mean LVEF was 38.3±14.7% (n=417), mean E/e’ 16.9±7.2 (n=306), mean TAPSE 1.58±0.46 cm (n=215), and mean RVSP 42.8±12.2 mmHg (n=390). In-hospital mortality was 10.8% (56/519). Non-survivors were older (73.5 [68.0–80.2] vs 69.0 [60.0–76.0] years; p=0.00028) and had shorter length of stay (4 [1–12] vs 7 [5–12] days; p=0.00057). Admission NT-proBNP was markedly higher in non-survivors (19,003.7 [8,932.8–35,000] pg/mL; n=52) than survivors (6,051.6 [2,679–14,434] pg/mL; n=409; p=3.45×10⁻⁸). Between-group differences in LVEF (p=0.77), E/e’ (p=0.58) and RVSP (p=0.29) were not pronounced; interpretation is limited by missing echocardiography in early fatal cases.
Conclusion(s)
In a large real-world AHF emergency cohort, in-hospital mortality was 10.8%. Older age and markedly higher admission NT-proBNP characterized patients with fatal in-hospital outcomes, supporting NT-proBNP as a key early risk marker in routine practice.