The correlation between Doppler deceleration time (DT) and brain natriuretic peptide (BNP) and their predictive value for detecting left ventricular (LV) remodeling in patients who are treated with primary percutaneous intervention for infarction and LV dysfunction are unknown. Fifty-six patients (64 +/- 12 years of age; 11 women) who had a first ST-segment elevation myocardial infarction and systolic dysfunction that was successfully treated with direct primary coronary intervention underwent 2-dimensional Doppler echocardiographic and plasma BNP evaluation at days 1 and 3 and 1 and 6 months after the index infarction. Repeat coronary angiograms were obtained at 1 and 6 months. Because of previous consistent evidence, 3 days after the index infarction was the time point of comparison between BNP and DT values. Echocardiographic LV remodeling was defined as an increase in end-diastolic volume index above baseline values of 2 x SD. Ventricular remodeling occurred in 20 patients (36%). Multivariate analyses that included BNP level, Doppler DT, echocardiographic measurements of systolic function, peak creatine kinase, and anterior infarct location showed Doppler DT to be the only predictor of LV remodeling (odds ratio 0.963, 95% confidence interval 0.936 to 0.990, p = 0.008). The optimal cutoff for DT in the prediction of 6-month LV remodeling was <136 ms (sensitivity 75%, specificity 97%, accuracy 81%, area under receiver-operating characteristic curve 0.90). Thus, in patients who have a first ST-segment elevation myocardial infarction and LV systolic dysfunction that is successfully treated with primary percutaneous coronary intervention, Doppler-derived DT 3 days after index infarction is more effective than BNP level in detecting patients who are at higher risk for 6-month LV remodeling.