The objective was to compare the effect of thrombolytic therapy given either at home or in hospital on the recovery of left ventricular function after acute myocardial infarction. In a randomized double-blind trial, 311 patients with suspected acute myocardial infarction were given 30 units anistreplase intravenously either at home, or later, in hospital. The median time-saving made with domiciliary thrombolysis was 130 min. All patients were admitted to hospital where left ventricular stroke distance was measured daily using a simple bedside ultrasound technique, and expressed as a percentage of the age-predicted normal value. The last recorded inpatient stroke distance measurement was used to assess residual left ventricular function after recovery from myocardial infarction. The mean stroke distance in patients with confirmed myocardial infarction was 74% on the day of admission, rising to 83% on the last inpatient day; it did not change between discharge and 3 months after admission. For 180 patients assigned randomly to treatment within 2 h of the onset of symptoms, mean stroke distance was greater by 6.8% in those given active anistreplase at home rather than in hospital (95% confidence interval 1.0 to 12.7%, P = 0.02), but there was no significant difference in stroke distance following home or hospital thrombolysis in 111 patients assigned treatment after that time (difference -2.0%, 95% confidence interval -8.4 to 4.5%, P = 0.54). The efficacy of thrombolytic therapy is enhanced when administered within 2 h of the onset of symptoms.