BNP tests in the emergency department to diagnose congestive heart failure
Vandenbos, Fédéric.; Mejdoubi, N.Eddine.; Pinier, Y.; Epuron, J.; Grelier, M.
Presse Medicale 35(11 Pt 1): 1625-1631
2006
ISSN/ISBN: 0755-4982 PMID: 17086116 Document Number: 598140
B-type natriuretic peptide (BNP) testing is a useful tool for diagnosis of congestive heart failure (CHF). In our institution only physicians in the lung disease, cardiology, and intensive care departments were authorized to order BNP testing. Our study sought to determine the validity of this strategy. In this one-year prospective study, BNP testing was performed only in the lung disease and cardiology departments. The following details were recorded for each patient with a BNP assay: demographic information, clinical symptoms, physical examination, laboratory reports, chest radiography and ECG findings, and initial diagnosis (including in particular suspicion of CHF). The criterion for inclusion in this study was acute dyspnea on admission from the emergency department (ED), and the exclusion criterion was chest pain. We collected information about each patient's final diagnosis and outcome from the medical file. Overall, the study included 236 patients (100 admitted into the pneumology department and 136 into cardiology). Their mean age was 77.3+/-12.4 years and the sex ratio 1.29 (133 M/103 F). CHR was the final diagnosis for 170 patients (72%: 56% in pneumology and 84% in cardiology). Initial misdiagnosis led to erroneous orientation of 66 patients, 13 (5.5%) of whom were among the 39 (17%) patients who died. The diagnostic accuracy of the specialists' clinical judgment was 74%, compared with 87% and 91% for BNP cut-off levels of 100 and 250 pg/mL, respectively. Without BNP testing in the ED, more than a quarter of the patients with acute dyspnea were incorrectly oriented. Our study found BNP testing to be more effective than clinical judgment. In this series of patients, the cut-off value providing the best sensitivity and specificity was 250 pg/mL.