OBJECTIVE--To see whether early elective angiography with a view to coronary angioplasty or bypass grafting of a stenosed infarct related vessel would improve outcome in acute myocardial infarction treated by thrombolysis with anistreplase. DESIGN--Randomised study of two treatment strategies with analysis of results over 12 months. SETTING--21 district hospitals and regional cardiac centres in Britain and Ireland. SUBJECTS--800 of 993 patients presenting with clinical and electrocardiographic features of acute myocardial infarction up to three hours after the onset of major symptoms. TREATMENT STRATEGIES--Intravenous anistreplase 30 units followed by a standard regimen of heparin, warfarin, and timolol and (in patients so randomised) early angiography plus appropriate intervention. MAIN OUTCOME MEASURE--Death or reinfarction within 12 months. RESULTS--397 patients were randomised to receive early angiography plus appropriate intervention (coronary angioplasty in 169 cases, coronary grafting in 59) and 403 patients to receive conservative care (of these, 12 had angioplasty and seven bypass grafting during the initial admission). By 12 months mortality (5.8% (23 patients) in the intervention group v 5.0% (20) in the conservative care group; p = 0.6) and rates of reinfarction (15.1% (60 patients) v 12.9% (52); p = 0.4) were similar in the two groups. No significant differences in rates of angina or rest pain were found at 12 months. Left ventricular ejection fraction at three and 12 months was the same in both groups. Median hospital stay was longer in the intervention group (11 days v 10 days; p less than 0.0001). CONCLUSION--For most patients given thrombolytic treatment for acute myocardial infarction a strategy of angiography and intervention is appropriate only when required for clinical indications.