Sensitivity analysis. Various sensi

Sensitivity analysis. Various sensitivity analysis were performed to test the robustness of the results. Analysis was performed: 1) combining acute myocardial infarction and postmyocardial infarction trail; 2) excluding trails that compared beta-blockers with active comparator; 3) using traditional meta-analysis with counts rather than patien-months; 4) restricting analysis to trail enrolling _> 400 patients; 5) excluding CIOpidogrel and metoprolol in myocardial infarction trail (COMMIT); and 6) based o the quality assessment of the trails. In addition, a meta-regression analysis was performed to evaluate the relationship of percentage of patients with reperfusion in each trail on the risk ratio of beta-blockers vs controls for mortality.