Diagnostic utility of N-terminal-proBNP in differentiating acute pulmonary embolism from heart failure in patients with acute dyspnea.

BACKGROUND: The plasma N-terminal pro-B-type natriuretic peptide (NT-proBNP) level is frequently elevated in dyspnoeic patients and increasingly used in emergency departments to assess the cause of acute dyspnea. In this study we prospectively tested NT-proBNP levels in patients with congestive heart failure (CHF) and/or acute pulmonary embolism (APE) and determined the utility of NT-proBNP for discriminating APE from CHF.

METHODS: A cohort of 177 dyspnoeic patients with a diagnosis of APE and/or CHF was prospectively studied between June 2010 and March 2013. NT-proBNP was measured by the electrochemiluminescence immunoassay (ECLIA). All patients were evaluated with transthoracic echocardiography (TTE). APE was diagnosed in the presence of thrombi signs in the pulmonary arteries with computed tomographic pulmonary angiography (CTPA) or a high-probability lung ventilation/perfusion scan. Risk stratification was based on the evaluation on admission according to the ESC guidelines from 2008. The diagnosis of CHF was based on the guidelines of the American College of Cardiology/American Heart Association and the European Society of Cardiology. Two physicians independently reviewed the records to determine the final diagnosis.

RESULTS: Fifty-nine patients met the criteria for dyspnea caused by APE, and 113 patients were diagnosed with CHF. Most of the APE patients (41, 69.5%) were intermediate-risk. The symptoms and signs, such as orthopnea, paroxysmal nocturnal dyspnea and rales in the lungs, were more common in patients with CHF than in patients with APE (P < 0.01). Median NT-proBNP was significantly lower in patients with APE compared to those in patients with CHF (2 855.9 pg/ml vs. 6 911.4 pg/ml, P < 0.01). We constructed the receiver operating characteristics (ROC) curve in predicting the diagnosis of APE. At a cut point = 1 582.750 pg/ml, NT-proBNP provided a specificity of 93% and a true positive rate (sensitivity) of 17% for the diagnosis. At a cut point = 3 390.000 pg/ml, NT-proBNP had a specificity of 83% and a sensitivity of 84% for the diagnosis of APE. At a cut point = 6 486.500 pg/ml, they were 54% and 93% respectively.

CONCLUSIONS: NT-proBNP can assist in excluding CHF patients from those admitted to the emergency department with acute dyspnea and identifying patients with a high probability of APE, which would reduce the missed diagnosis of APE. Larger studies are necessary to validate these findings.