Cardiopulmonary resuscitation (CPR) is often considered a contraindication to thrombolytic therapy for acute myocardial infarction. Of 708 patients involved in the first 3 Thrombolysis and Angioplasty in Myocardial Infarction trials of lytic therapy for acute infarction, 59 patients required less than 10 minutes of CPR before receiving lytic therapy (CPR greater than 10 minutes was an exclusion of the trials) or required CPR within 6 hours of treatment. The patients receiving CPR were similar to the remainder of the group with respect to baseline demographics. The indication for CPR was usually ventricular fibrillation (73%) or ventricular tachycardia (24%). The median duration of CPR was 1 minute, with twenty-fifth and seventy-fifth percentiles of 1 and 5 minutes, respectively. The median number of cardioversions/defibrillations performed was 2 (twenty-fifth and seventy-fifth percentiles of 1 and 3 minutes, respectively). Patients receiving CPR were more likely to have anterior infarctions (66 vs 39%), the left anterior descending artery as the infarct-related artery (63 vs 38%) and lower ejection fractions on the initial ventriculogram (46 +/- 11 vs 52 +/- 12%) than those not receiving CPR. In-hospital mortality was 12 vs 6% with most deaths due to pump failure (57%) or arrhythmia (29%) in the CPR group and pump failure (38%) or reinfarction (25%) in the non-CPR group. At 7 day follow-up the CPR group had a significant increase in ejection fraction (+5 +/- 9%) compared with no change in non-CPR group. There were no bleeding complications directly attributed to CPR.(ABSTRACT TRUNCATED AT 250 WORDS)