Direct (primary) coronary angioplasty for acute myocardial infarction has appeared in the literature since 1982, first as a rescue therapy in case of failed intracoronary thrombolysis, then as its complement, and finally as an alternative. In the late eighties, the upsurge of intravenous thrombolysis turned the focus off direct angioplasty until several randomized studies demonstrated the superiority of angioplasty in 1993. For an infarction primarily due to a thrombus, it is the treatment of choice. However, there is no accurate way to predict these lesions. Direct angioplasty yields results comparable with the best published results of thrombolysis. However, the patients included in the thrombolysis studies usually present less severe clinical pictures than those undergoing angioplasty. On the average, direct angioplasty is successful in 90%, reocclusion occurs in 11%, and emergency bypass surgery becomes necessary in 3%. Hospital mortality is 7%. The risk factors known from thrombolysis also apply to direct angioplasty, such as multivessel disease, old age, cardiogenic shock and duration of infarction. Failed angioplasty is also a harbinger of death. In randomized comparisons with thrombolysis, an initial 30-45 minutes delay from decision to treatment because of preparations for cardiac catheterization was noted which converts to a time gain until reperfusion of about 1 hour in favor of angioplasty. Hospital mortality is reduced with angioplasty. Cerebral bleedings were confined to thrombolysis. Ejection fraction at rest or on exercise was better after angioplasty in two studies, and there is a clearcut reduction in the need for further interventions, resulting in an economic benefit in favour of direct angioplasty. A randomized study demonstrated that it was deleterious to routinely complement direct angioplasty with intravenous streptokinase. Direct angioplasty establishes patency earlier than thrombolysis and has a better overall hospital and long term course, probably because it eliminates the lesion in most cases rather than just recanalizing the vessel. It should be used in all patients with acute myocardial infarction provided not more than 30 minutes are lost for its organization. Otherwise, thrombolysis is the preferred treatment. Direct angioplasty is particularly desirable in patients with large infarctions of short duration. Its role for infarctions presenting with cardiogenic shock is currently under evaluation.