Prognostic accuracy of natriuretic peptide levels at hospital admission versus discharge for all-cause mortality in patients presenting with acute dyspnea

Background The European Society of Cardiology recommends natriuretic peptide testing for confirmation or rule-out of acute heart failure (AHF) in patients presenting with acute dyspnea. However, the prognostic role of natriuretic peptides in patients presenting with acute dyspnea in general and the timing of biomarker measurement in particular is more controversial. We therefore aimed to directly compare the prognostic accuracy of natriuretic peptide levels obtained at hospital admission versus discharge for all-cause mortality in patients presenting with acute dyspnea. Methods This prospective diagnostic study enrolled unselected patients presenting with acute dyspnea to the emergency department. The final diagnosis causing dyspnea was centrally adjudicated by two independent cardiologists. Patients with available BNP and NT-proBNP plasma levels at hospital admission and discharge were included in this analysis. Results Among 1126 patients presenting with acute dyspnea, 657 (58%) had an adjudicated final diagnosis of AHF. All-cause mortality at 30, 180 and 360 days of follow-up was 5.4%, 15.3% and 22.4% for the overall cohort. Prognostic accuracy for 30-day all-cause mortality as quantified by the area under the receiver-operating-characteristics curve was 0.85 (95% CI 0.81-0.89) for NTproBNP plasma levels at discharge as compared to 0.79 (95% CI 0.74-0.83) for NT-proBNP plasma levels at admission (p<0.001, Figure 1A). With regards to BNP, prognostic accuracy for 30-day all-cause mortality as quantified by the area under the receiver-operating-characteristics curve was 0.82 (95% CI 0.76-0.87) for plasma levels at discharge as compared to 0.75 (95% CI 0.70-0.80) for plasma levels at admission (p<0.001, Figure 1B). When comparing plasma levels at discharge, there was a statistically significant prognostic superiority of NTproBNP as compared to BNP (p<0.032). The prognostic superiority of natriuretic peptide levels obtained at hospital discharge was consistent among predefined groups of 180 and 360days of follow-up as well as in patients with an adjudicated final diagnosis of AHF. Conclusion Natriuretic peptide levels obtained at hospital discharge have prognostic superiority in patients presenting with acute dyspnea and AHF.