Objective: To evaluate the efficacy and safety of systemic thrombolysis administered to resuscitated patients after cardiac arrest (CA) due to an acute myocardial infarction (AMI), through a study of their mortality and haemorrhagic complications. Design: We studied a retrospective cohort of patients with ischaemic heart disease gathered from the database of the Spanish multi-centre project "Analysis of Delay in AMI" (ARIAM). Setting: Intensive care (ICU) and coronary care (CCU) units of 77 Spanish hospitals. Patients and participants: The study period was from 1 January 1995 to 1 January 2000, when 22,922 patients were included in the ARIAM database register; 13,704 were diagnosed with AMI and we studied 303 of these AMI patients admitted after resuscitation for CA. Measurements and results: Of the 303 patients studied, 228 were male (75.25%); the mean age was 64.57-12.48 years. Systemic thrombolysis was administered to 67 patients (group I) and the remaining 236 patients were managed without this treatment (group II). The ICU/CCU mortality rate of the series was 39.93% (121 patients); that of group I was 17.91% (12 patients) and that of group II 46.18% (109 patients) (P<0.00001). Group I required less mechanical ventilation (group I, 42.85% vs group II, 80.76%; P<0.00001) and fewer cardiopulmonary resuscitation attempts (33.34% vs 60.98%, P<0.0001). Group I also showed a lower incidence of cardiogenic shock (14.28% vs 39.01%, P<0.0001) and anoxic encephalopathy (8.62% vs 39.89% P=0.006). There were no fatal haemorrhagic complications in either group. Logistic regression analysis showed the administration of thrombolysis to be an independent variable that protected against mortality. Conclusions: The administration of thrombolysis to patients with AMI who require resuscitation may be efficacious in reducing mortality and is safe, with no increase in haemorrhagic complications.