Background: Both the brain natriuretic peptide (BNP) and the amino-terminal probrain natriuretic peptide (NT-proBNP) are well-established biomarkers for heart failure and cardiovascular prognosis. However, between two biomarkers, discrepancies are not uncommon and the significance of biological difference are not well recognized in clinical practice. Methods: We investigated the clinical characteristics of unselected consecutive 1,292 patients who complained chest discomfort or dyspnea and visited emergency department from Apr 2005 to Apr 2007. Biomarkers including BNP, NT-proBNP, and routine labs were obtained in all patients and analyzed immediately at emergency department. For measurement of BNP and NT-proBNP, triage meterpro by Biosite and Hitachi 7200 by Roche were used, respectively, and arithmetic or logarithm values of BNP and NT-proBNP were used for analysis. Results: BNP and NT-proBNP correlated well with each other (r=0.938, p<0.0001). Both BNP and NT-proBNP correlated marginally with the time interval from symptom onset (r=0.232, r=0.243, p<0.0001), and also the ratio of NT-proBNP/BNP did (r=0.066, p=0.024). Patients with myocardial damage, which was defined by elevation of troponin more than cutoff value (0.4 ng/ml), showed higher NT-proBNP, BNP (7860 +-11570 versus 3353 +- 7781 pg/ml, 837 +- 174 versus 361 +- 809 pg/ml), and higher NT-proBNP/BNP ratio (9.495 +- 7.34 versus 9.26 +- 18.47, analyzed using logarithm value) than non-elevated group (p<0.0001 for all). Conclusion: There was a close correlationship between BNP and NT-proBNP. Also there was a tendency of higher NT-proBNP compared to BNP in patients with late symptom onset or with myocardial damage. So both markers may not completely equivalent clinically, which should be reminded in clinical practice.