Myocardial expansion in acute myocardial infarction (AMI) is present in about 45% of the patients within the first 72 hours. This is associated with ventricular aneurysm formation, myocardial rupture, heart failure and early death. Experimental studies in animals with AMI have used late reperfusion to decrease the incidence of expansion with success. The present is a prospective, longitudinal, open and randomized study in 21 patients with anterior AMI, to evaluate if the late reperfusion (6 to 12 hours) can decrease the incidence of myocardial expansion graded quantitatively with bidimensional echocardiography. Two groups were made: group A (n = 12) who received thrombolysis with streptokinase 1.5 mill. IU plus oral aspirin 150 mg OD (n = 9). Both groups had the same characteristics of AMI and functional class of Killip and Kimball (I-II class). Intrahospital treatment was given freely in both groups. The expansion was evaluated with bidimensional echocardiography used Jugdutt's method. In group A, expansion was present in 25% of the cases, while in group B was 66.6% (p < 0.0005). The distortion area, distortion peak, septal thickness and large asynergic segment were more sensitive parameters to identify myocardial expansion. Our results are similar to some experimental studies. We conclude that late thrombolysis can be useful in decreasing the incidence of myocardial expansion. Bidimensional echocardiography is a useful, fast and safe method to identify myocardial expansion.