To evaluate the different contributions of infarct site and infarct extent in determining the in-hospital outcome and efficacy of thrombolytic therapy, 8,731 patients with a first Q-wave acute myocardial infarction (AMI) enrolled in the GISSI trial were studied. On the basis of the standard 12-lead electrocardiogram, the sample was classified in 2 ways: classic electrocardiographic site pattern (anterior, inferior, lateral and multiple location), and number of standard electrocardiographic leads with ischemic ST elevation (small, modest, large and extensive infarct in 2 to 9 leads). In-hospital mortality was evaluated according to infarct site, infarct extent and fibrinolytic treatment. The mortality rate was differently distributed in the various infarct sites. Streptokinase significantly reduced mortality only in anterior (13.8 vs 18.7%) and multiple site infarcts (8.1 vs 12.5%). According to the infarct extent observed, there is a progressive increase in the mortality rate--from 6.5% in small infarcts to 9.6% in modest, 14.3% in large and 21.7% in extensive. No significant benefit was obtained by streptokinase in small infarcts; in contrast, a significant decrease in mortality was achieved in modest (7.7 vs 11.4%), large (12.8 vs 16.6%) and extensive infarcts (19.5 vs 23.9%). Thus, the extent of myocardial injury seems to be more relevant than the site in determining in-hospital mortality and efficacy of thrombolytic therapy.