In a randomized, double-blind study, rt-PA vs. placebo treatment in early suspected acute myocardial infarction (AMI) was evaluated in patients both in hospital and prehospitally. The inclusion criteria were as follows: (a) age less than 75 years; and (b) chest pain indicative of AMI, of no longer than 2.75 h duration before first examination. In the prehospital setting a mobile coronary-care unit, accompanied by a cardiologist, was sent out by the ambulance services to 350 patients, of whom 205 (59%) were classified as non-eligible when examined by the cardiologist. Of the 145 patients who fulfilled the inclusion criteria, 44 were excluded due to contraindications to thrombolytic treatment. Thus 101 patients were randomized to blinded treatment outside hospital. The median time interval between onset of pain and treatment was 75 min, 45 min less than for those subjects who were randomized in hospital. The prevalence of confirmed AMI was 42% in the prehospital group, compared to 66% in the hospital group. Bleeding and cardiac complications for prehospital treatment were few, and similar to those for hospital treatment. In conclusion, prehospital thrombolysis was feasible, and delay times prior to treatment were significantly reduced. However, the specificity and diagnostic accuracy were lower than those achieved with in-hospital therapy.