Two strategies for treatment of acute myocardial infarction were compared in a randomised trial of 533 patients admitted within 4 h of the onset of symptoms. 264 patients were allocated to conventional treatment and 269 patients to a strategy aimed at rapid recanalisation of the occluded coronary artery. At first, intracoronary streptokinase (up to 250 000 U) was given immediately after angiography. In the last 117 patients intracoronary administration was preceded by intravenous streptokinase (500 000 U). No angiography was done in 35 patients allocated to thrombolytic therapy. Among the 234 patients who underwent angiography the infarct-related coronary artery was patent on admission or recanalised in 198 (85%). The median time between onset of symptoms and angiographic confirmation of a patent infarct-related artery was 200 min. Mortality was lower in patients randomised to thrombolysis than in controls at 28 days (16 v 31 patients) and at 8 months (23 v 42); 1-year survival was higher after thrombolysis (91%) than conventional treatment (84%). The clinical course in hospital was more favourable in patients allocated to thrombolysis, with a lower incidence of ventricular fibrillation (38 v 61), pericarditis (19 v 46), and cardiogenic shock (13 v 24), although they had a higher frequency of bleeding episodes (53 v 7). Non-fatal reinfarction was commoner after thrombolysis than after conventional therapy (36 V 16 patients). In both groups reinfarctions occurred predominantly in the same area as the initial infarct.