Thrombolytic therapy for acute myocardial infarction (AMI) has reduced mortality at the expense of additional intracranial hemorrhages. To determine whether this trade-off has been optimized, a decision analysis was performed using pooled data to determine the further reductions in mortality required to justify increased intracranial hemorrhage rates from more potent thrombolytic and adjunctive antithrombotic regimens than intravenous streptokinase. Pooled data from large clinical trials suggest that streptokinase has a 0.07% nonfatal intracranial hemorrhage rate. Approximately 54% of these result in major/moderate disability and 46% in recovery or minor residual. The early mortality rate in all AMI patients treated with thrombolytic therapy is 9.8%; it is 6.8% in patients with inferior wall AMI and 17.9% in elderly patients. If a new thrombolytic regimen provides a 1% absolute reduction in early mortality compared with streptokinase therapy, approximately a ≥3.2% nonfatal intracranial hemorrhage rate is justified to obtain this survival benefit. For a 10% relative reduction in mortality risk, the maximal acceptable nonfatal intracranial hemorrhage rates are 2.2% for inferior wall AMI, 3.2% for all patients and 5.9% for elderly patients. Whereas intracranial hemorrhage is a catastrophic complication of thrombolytic therapy in the treatment of patients with AMI, thrombolytic regimens that result in significantly higher rates of intracranial hemorrhage than those observed with streptokinase may be preferable at surprisingly smaller additional reductions in mortality. In addition to evaluating new thrombolytic and antithrombotic regimens, this analysis, in conjunction with models that predict patient-specific intracranial hemorrhage risks and mortality benefits from thrombolytic therapy, can provide a framework for matching AMI patients with optimal thrombolytic regimens. © 1993.