BACKGROUND: Fibrinolytic therapy increases the risk of bleeding events. TNK-tPA (tenecteplase) is a variant of rt-PA with greater fibrin specificity and reduced plasma clearance that can be given as a single bolus. We compared the incidence and predictors of bleeding events after treatment with TNK-tPA and rt-PA. METHODS AND RESULTS: In the Assessment of the Safety and Efficacy of a New Thrombolytic (ASSENT)-2 trial, 16 949 patients with acute myocardial infarction were randomly assigned a single weight-adjusted bolus of TNK-tPA or a 90-min infusion of rt-PA. A total of 4.66% of patients in the TNK-tPA group experienced major non-cerebral bleeding, in comparison with 5.94% in the rt-PA group (P=0.0002). This lower rate was associated with a significant reduction in the need for blood transfusion (4.25% vs 5.49%, P=0.0003) and was consistent across subgroups. Independent risk factors for major bleeding were older age, female gender, lower body weight, enrolment in the U.S.A. and a diastolic blood pressure <70 mmHg. Females at high risk (age >75 years and body weight <67 kg) were less likely to have major bleeding when treated with TNK-tPA even after other risk factors were taken into account. A total of 0.93% of patients in the TNK-tPA and 0.94% of patients in the rt-PA group experienced an intracranial haemorrhage. Female patients >75 years of age who weighed <67 kg tended to have lower rates of intracranial haemorrhage when treated with TNK-tPA (3/264, 1.14% vs 8/265, 3.02%). CONCLUSIONS: The increased fibrin specificity and single bolus administration of TNK-tPA do not increase the risk of intracranial haemorrhage but are associated with less non-cerebral bleeding, especially amongst high-risk patients.