OBJECTIVE The aim of this study is to investigate whether P wave dispersion (PWD), measured before, during and after fibrinolytic therapy (FT,) is able to predict successful reperfusion and infarct related artery (IRA) patency in patients with acute anterior MI who received FT. METHODS Sixty-eight patients who presented with acute anterior MI were enrolled in the study. An electrocardiogram was performed before and at 30, 60, 90 and 120 minutes after the start of FT. PWD was defined as the difference between maximum and minimum P wave duration on standard 12-lead surface electrocardiogram. A multivariate logistic regression model was used to assess whether PWD was predictor of IRA patency and ST-segment resolution (STR) on electrocardiogram. RESULTS PWD120 was significantly lower in patients with STR on electrocardiogram (38 patients) compared with those without STR (30 patients) (44.8±11.5 vs. 52.9±10.3 ms; p < 0.001). PWD120 was found to be significantly lower in patients with patent IRA (31 patients) compared to those with occluded IRA (37 patients) (42.3±9.7 vs. 53.5±10.6 ms; p < 0.001). Logistic regression analysis revealed that PWD120 significantly predicted STR and IRA patency. A ≥51.6 ms PWD120 can predict an occluded IRA with a 87% sensitivity, ≥51 ms PWD120 can predict no reperfusion with a 74% sensitivity. CONCLUSIONS PWD values, which were higher than 51 ms and 51.6 ms in patients who received fibrinolytic therapy, can serve as a marker of failed reperfusion and occluded IRA. PWD values, in combination with other reperfusion parameters, can contribute to the identification of rescue PCI candidates.