To determine the relation between plasma brain natriuretic peptide (BNP) and remodeling in terms of infarct-related artery (IRA) patency, 106 patients with a first anterior wall acute myocardial infarction with a patent IRA at 1 month were studied. The IRA reoccluded at 6 months in 17 patients (reoccluded IRA) and was patent in 89 patients (patent IRA). The 2 groups did not differ with respect to clinical characteristics, hemodynamic variables, and left ventricular function at 1 month, except for left ventricular end-diastolic and systolic volumes, which were significantly greater in the reoccluded IRA group. Plasma BNP concentration in the reoccluded IRA group (336 +/- 288 pg/ml) was significantly higher than that in the patent IRA group (116 +/- 106 pg/ml) at 1 month. BNP concentration decreased significantly at 6 months in the 2 groups (reoccluded IRA vs patent IRA 152 +/- 162 vs 44 +/- 58 pg/ml, p <0.05). The increase in left ventricular volume from 1 to 6 months was significantly correlated with plasma BNP concentration at 1 month in the patent IRA group (r = 0.314, p < 0.01) and the reoccluded group (r = 0.634, p < 0.01). Linear regression analysis showed that the correlation between the 2 parameters in the 2 groups was similar. Based on stepwise multivariate linear regression analysis, only plasma BNP concentration was significantly correlated with the increase in left ventricular volume from 1 to 6 months in the 2 groups. In conclusion, these results suggest that plasma BNP concentration predicts left ventricular dilation independently of IRA patency.