Journal Article Ahead of print
PubMed: 42888496 DOI: 10.1093/ejhf/xuag314
Diagnostic Accuracy of NT-proBNP thresholds in Dyspneic Patients in the Emergency Department according to patients' characteristics: A Systematic Review and Meta-Analysis.
Eur J Heart Fail
Authors: Guillaume Baudry, Luca Monzo, Vincent Faivre, Zohra Lamiral, Kevin Duarte, James L Januzzi, Antony Bayes-Genis, Nicolas Girerd
Affiliations
- Université de Lorraine, INSERM, Centre d'Investigation Clinique Plurithématique 1433, Inserm U1116, CHRU de Nancy and F-CRIN INI-CRCT, Nancy, France.
- REICATRA, Recherche et Enseignement en IC Avancée, Transplantation, Assistance , Vandœuvre-lès-Nancy, France.
- Heart and Vascular Institute, Mass General Brigham, Boston, Massachusetts, USA.
- Harvard Medical School, Boston, Massachusetts, USA.
- Baim Institute for Clinical Research, Brookline, Massachusetts, USA.
- Heart Institute, Hospital Universitari Germans Trias i Pujol, CIBERCV, Badalona, Spain.
Abstract
BACKGROUND
Concentrations of N-terminal pro-B type natriuretic peptide (NT-proBNP) is central to acute heart failure (AHF) diagnosis in dyspneic emergency departments (ED) patients, but optimal thresholds remain uncertain. This study aimed to evaluate the diagnostic accuracy of various NT-proBNP thresholds for AHF, across the general population and these key subgroups.
METHODS
A systematic review and meta-analysis were conducted, including studies published up to August 28, 2025, that reported the diagnostic performance of NT-proBNP thresholds for AHF in dyspneic ED patients.
RESULTS
The analysis included 31 studies comprising 26,948 patients. To exclude AHF, a fixed threshold of ≤300 ng/L yielded an area under the receiver operator characteristic curve (AUC) of 0.73 (0.69-0.77), with high sensitivity of 97.9% (96.9-98.5) and lower specificity of 46.2% (36.4-56.2); the negative predictive value of this threshold was 97.0% (96.5-97.4). Age-dependent thresholds (≥450 ng/L for patients aged <50 years, ≥900 ng/L for those aged 50-75 years, and ≥1,800 ng/L for those aged ≥75 years), evaluated in 14,410 patients, achieved an AUC of 0.84 (0.83-0.85) with a more favorable balance between sensitivity of 85.9% (85.0-86.7) and specificity of 82.0% (81.2-82.9). In patients with CKD (N=2,271), all fixed thresholds yielded AUC <0.80; age-dependent thresholds showed an AUC of 0.77 (0.75-0.78) with specificity of 62.6% (59.4-65.7). In patients with obesity (body mass index ≥30), only age-dependent thresholds were assessed, in 3,562 patients; AUC was 0.82 (0.81-0.84) and sensitivity was 81.2% (79.2-83.1), broadly comparable to the overall population.
CONCLUSION
A threshold ≤300 ng/L provides excellent rule-out performance for AHF, while age-dependent NT-proBNP thresholds offer superior diagnostic performance for ruling in AHF in the ED compared to fixed thresholds, and demonstrate consistent accuracy across subgroups. However, evidence in patients with CKD and obesity remains limited, both in terms of sample size and the range of thresholds evaluated.
MeSH
Emergencies, Heart Failure, Aged, Obesity, Emergency Service, Hospital