Background: The purpose of this study was to determine the diagnostic power of a newly available assay for amino-terminal pro-brain natriuretic peptide (NT-proBNP) to identify patients with acute heart failure. In addition, the influence of initial NT-proBNP measurements on economic consequences, diagnostic procedures and staff involvement was evaluated.
Methods and results: 401 patients presenting with acute dyspnea or peripheral edema in the emergency department were enrolled. NT-proBNP was measured after initial clinical evaluation. Clinical routine care and diagnostic assessment were blinded to NT-proBNP results. Two cardiologists independently validated the period of hospitalization, clinical examinations and medical therapies of each patient considering NT-proBNP results. The median NT-proBNP level among patients with acute congestive heart failure (CHF) (n=122) was 3497 pg/ml as compared to 320 pg/ml in patients without (n=279) (p<0.0001). An NT-proBNP cutoff level <300 pg/ml was optimal to rule out acute CHF (negative predictive value 96%; sensitivity 96%). NT-proBNP >or=300 pg/ml could strongly predict acute CHF when compared to patients' history or physical examination (odds ratio 9.5; p<0.0001) and diagnostic technical findings (odds ratio 14.7; p<0.05). In patients with NT-proBNP<300 pg/ml, 14% of the period of hospitalization could be saved, corresponding to savings of US $481 per patient. In addition, 9% of the number and time of staff involvement of clinical examinations and therapies could be saved, 10% of the costs of clinical examinations. Chest X-rays were saved in 34%, echocardiography in 9%.
Conclusions: Measurement of NT-proBNP leads to multiple saving amounts and optimizes diagnostic pathways and resource allocation.