In somewhat less than a decade, prophylactic administration of antifibrinolytic drugs has achieved such widespread acceptance that in some institutions, my own included, antifibrinolytic infusions during cardiac operations requiring cardiopulmonary bypass have become a “standard of care.” This remarkable feat has occurred to a great extent in response to aprotinin. Pharmaceutical-sponsored efficacy trials of aprotinin have demonstrated significant reductions in blood loss and transfusion requirements. Unfortunately, as Casati and colleagues point out in their discussion, the costs associated with aprotinin therapy have led investigators to seek alternative, less costly antifibrinolytics for perioperative applications. In fact, the question has become not whether antifibrinolytics should be administered, but rather which antifibrinolytic offers the best combination of efficacy, cost, and safety. Numerous publications, including two recent meta-analyses, attest to the efficacy of both the synthetic antifibrinolytics and aprotinin to reduce bleeding associated with cardiac surgery. In most published trials, a single antifibrinolytic has been compared with a placebo. Interestingly, few direct comparisons have been made among the various antifibrinolytics. Consistent with results from a recent meta-analysis, Casati and colleagues report that aprotinin decreased bleeding to a greater extent than the synthetic antifibrinolytics; however, tranexamic acid produced similar hemostatic effects at significantly less cost. Somewhat surprising was the substantial difference in blood loss between the tranexamic acid and ϵ-aminocaproic acid (EACA) groups. Our group and others have generally reported 30%–35% reductions in blood loss associated with EACA therapy. To some extent, variations in dosing may account for the differences in reported efficacy. Unlike aprotinin, dosing for the synthetic antifibrinolytics has not been standardized and varies between institutions. Lack of a placebo control group among Casati’s patients further hinders assessment of the overall efficacy of antifibrinolytic therapy in these patients. Selection of a study population undergoing elective primary cardiac surgery may have limited the ability to detect differences in efficacy apparent with greater bleeding, as observed in reoperative and more complex operations. Despite the apparent similarity in efficacy between tranexamic acid and aprotinin, as reported by Casati and colleagues, the question of safety remains. A sample size of just over 200 patients is too small to detect subtle differences in morbidity and mortality between antifibrinolytic groups. Historically, lack of pharmaceutical funding has substantially limited the investigation of synthetic antifibrinolytic durgs, and in most cases, patient sample sizes have been inadequately powered to address safety issues. In contrast, considerable evidencehas accumulated to support the safety of aprotinin in the setting of adult cardiac operation. Casati and colleagues’ investigation addresses an important unanswered question for adult cardiac surgical patients. What is the most cost-effective pharmacologic approach to minimize bleeding and transfusion requirements in the perioperative period? The increasing popularity of minimally invasive and off-pump CABG operation appear certain to revitalize and further complicate this issue. Larger prospective placebo-controlled trials will be needed to further define the role of antifibrinolytic drugs in the setting of cardiac operations.