The utility of aminoterminal pro-brain natriuretic peptide (NT-proBNP) testing in the emergency department to rule out acute congestive heart failure (CHF) and the optimal cutpoints for this use are not established. We conducted a prospective study of 600 patients who presented in the emergency department with dyspnea. The clinical diagnosis of acute CHF was determined by study physicians who were blinded to NT-proBNP results. The primary end point was a comparison of NT-proBNP results with the clinical assessment of the managing physician for identifying acute CHF. The median NT-proBNP level among 209 patients (35%) who had acute CHF was 4,054 versus 131 pg/ml among 390 patients (65%) who did not (p <0.001). NT-proBNP at cutpoints of >450 pg/ml for patients <50 years of age and >900 pg/ml for patients >or=50 years of age were highly sensitive and specific for the diagnosis of acute CHF (p <0.001). An NT-proBNP level <300 pg/ml was optimal for ruling out acute CHF, with a negative predictive value of 99%. Increased NT-proBNP was the strongest independent predictor of a final diagnosis of acute CHF (odds ratio 44, 95% confidence interval 21.0 to 91.0, p <0.0001). NT-proBNP testing alone was superior to clinical judgment alone for diagnosing acute CHF (p = 0.006); NT-proBNP plus clinical judgment was superior to NT-proBNP or clinical judgment alone. NT-proBNP measurement is a valuable addition to standard clinical assessment for the identification and exclusion of acute CHF in the emergency department setting.