A multicentre, randomised, open trial with a 2 x 2 factorial design was conducted to compare the benefits and risks of two thrombolytic agents, streptokinase (SK, 1.5 MU infused intravenously over 30-60 min) and alteplase (tPA, 100 mg infused intravenously over 3 h) in patients with acute myocardial infarction admitted to coronary care units within 6 h from onset of symptoms. The patients were also randomised to receive heparin (12,500 U subcutaneously twice daily until discharge from hospital, starting 12 h after beginning the tPA or SK infusion) or usual therapy. All patients without specific contraindications were given atenolol (5-10 mg iv) and aspirin (300-325 mg a day). The end-point of the study was the combined estimate of death plus severe left ventricular damage. 12,490 patients were randomised to four treatment groups (SK alone, SK plus heparin, tPA alone, tPA plus heparin). No specific differences between the two thrombolytic agents were detected as regards the combined end-point (tPA 23.1%; SK 22.5%; relative risk 1.04, 95% Cl 0.95-1.13), nor after the addition of heparin to the aspirin treatment (hep 22.7%, no hep 22.9%; RR 0.99, 95% Cl 0.91-1.08). The outcome of patients allocated to the four treatment groups was similar with respect to baseline risk factors such as age, Killip class, hours from onset of symptoms, and site and type of infarct. The rates of major in-hospital cardiac complications (reinfarction, post-infarction angina) were also similar. The incidence of major bleeds was significantly higher in SK and heparin treated patients (respectively, tPA 0.5%, SK 1.0%, RR 0.57, 95% Cl 0.38-0.85; hep 1.0%, no hep 0.6%, RR 1.64, 95% Cl 1.09-2.45), whereas the overall incidence of stroke was similar in all groups. SK and tPA appear equally effective and safe for use in routine conditions of care, in all infarct patients who have no contraindications, with or without post-thrombolytic heparin treatment. The 8.8% hospital mortality of the study population (compared with approximately 13% in the control cohort of the GISSI-1 trial) indicates the beneficial impact of the proven acute treatments for AMI.