Tranexamic Acid Is Associated With Less Blood Transfusion in Off-Pump Coronary Artery Bypass Graft Surgery: A Systematic Review and Meta-analysis S.C. Adler Ma, BSc, William Brindle, MPhil, Gillian Burton, BSc, Stuart Gallacher, Fong Cheng Hong, Ilinda Manelius, Andrew Smith, MEng, Weiyang Ho, R. Peter Alston, MBChB, MD, FRCA, FFPMRCA, and Kausik Bhattacharya, FRCS Objective: Tranexamic acid reduces blood loss and trans- fusion in on-pump coronary artery bypass graft (CABG) sur- gery. Compared with on-pump, off-pump surgery is associ- ated with less blood loss and transfusion. Therefore, tranexamic acid may be less effective for off-pump surgery, and its safety proﬁle may be different in this setting. The aim of this study was to determine the efﬁcacy and safety of tranexamic acid for off-pump CABG surgery. Design: Systematic review and meta-analysis. Setting: University of Edinburgh. Interventions: The administration of tranexamic acid. Methods: A systematic review of randomized controlled trials administering tranexamic acid to patients undergoing off-pump CABG surgery. A meta-analysis of 24-hour blood loss, postoperative allogeneic transfusion, and thromboem- bolic events. Measurements and Main Results: Eight trials were identi- ﬁed. The lack of appropriate data limited the meta-analysis on blood loss. Tranexamic acid signiﬁcantly reduced the overall risk of allogeneic blood component transfusion (risk ratio  0.47; 95% conﬁdence intervals, 0.33-0.66; p < 0.0001) and packed red blood cell transfusions (risk ratio  0.51; 95% CI, 0.36-0.71; p  0.0001). No association was found between tranexamic acid and myocardial infarction, stroke, or pulmo- nary embolism. Population sizes of meta-analyses ranged from 466 to 544. Conclusions: Tranexamic acid reduces blood transfusion after off-pump surgery. Although no association with ad- verse events was found, the population sample size was too small to detect rare but clinically signiﬁcant adverse events. A well-designed randomized controlled trial with an appro- priate sample size is required to conﬁrm tranexamic acid effectiveness and safety in off-pump CABG surgery. © 2011 Elsevier Inc. All rights reserved. KEY WORDS: tranexamic acid, off-pump coronary artery by- pass, hemostasis, CABG, OPCAB T RANEXAMIC ACID (TA) inhibits ﬁbrinolysis by block- ing the binding site of plasminogen to ﬁbrin, increasing the clotting potential of blood, and subsequently reducing blood loss. 1 Since the European Medicines Agency suspended the marketing authorization for aprotinin and because aminocap- roic acid is not licensed in the United Kingdom, TA has become the agent of choice to reduce blood loss and transfusion associated with cardiac surgery in the United Kingdom. 2-5 Tranexamic acid reduces blood loss and transfusion associ- ated with coronary artery bypass graft (CABG) surgery per- formed on-pump (ie, using cardiopulmonary bypass). 6 Since the early 1990s, CABG surgery performed off-pump (OPCAB) (ie, without cardiopulmonary bypass) has become an estab- lished alternative surgical technique. 7,8 Although the long-term outcomes of the surgical techniques may be equivocal, OPCAB is associated with less blood loss and transfusion than when it is performed on-pump. 9 The efﬁcacy of TA in the setting of reduced blood loss and transfusion associated with OPCAB surgery is unclear. Additionally, it is possible that the safety proﬁle of TA may be different in OPCAB compared with on-pump CABG surgery. In particular, there is a greater level of activation of ﬁbrinogen and other acute-phase proteins associated with OPCAB compared with on-pump CABG that might result in a higher incidence of adverse thrombotic events. 10 Eight randomized controlled trials have examined the efﬁ- cacy and safety of TA in the off-pump setting. 11-18 However, the population sizes of these trials were small and statistically underpowered to detect clinically important adverse events. In 2006, Murphy et al 15 published a systematic review and meta- analysis of 4 randomized controlled trials and found that TA was associated with less red cell transfusion. A review of Murphy et al’s study by the Centre for Reviews and Dissemi- nation noted important limitations including ‘limited search, incomplete reporting of review methods, and lack of a quality assessment.” 19 In the same article, 15 Murphy et al published the ﬁndings of an additional randomized controlled trial (not in- cluded in their meta-analysis), and, subsequently, a further 3 trials have been published. Blood products are a scarce resource and expose the patient to the risks of blood-borne disease, graft-versus-host reactions, and acute hemolytic reactions, all of which increase patient mortality. 20-23 Therefore, reducing blood loss and transfusion is an important priority. The authors hypothesized that TA re- duces blood loss and transfusion associated with OPCAB sur- gery without increasing the risk of thromboembolic complica- tions. Accordingly, the aim of this study was to provide a current and robust systematic review of the literature and meta-analysis to examine the efﬁcacy and safety of TA in OPCAB surgery. METHODS Search Strategy The MEDLINE, Embase, Cochrane, and National Health Service electronic libraries were searched for randomized controlled trials in October 2009 using medical subject headings for the terms speciﬁed in Figure 1. In addition, a citation search was performed on all pertinent abstracts, articles, and reviews to ﬁnd other relevant trials. Two mem- bers of the group (WB and GB) then assessed the articles to identify those with exclusion criteria. Exclusion criteria were studies making no From the University of Edinburgh, College of Medicine and Veter- inary Medicine, Edinburgh, United Kingdom. Address reprint requests to S.C. Adler Ma, c/o University of Edin- burgh, Chancellor’s Building, 49 Little France Crescent, Edinburgh EH16 4SB, UK. E-mail: s.c.a.ma@sms.ed.ac.uk © 2011 Elsevier Inc. All rights reserved. 1053-0770/2501-0006$36.00/0 doi:10.1053/j.jvca.2010.08.012 26 Journal of Cardiothoracic and Vascular Anesthesia, Vol 25, No 1 (February), 2011: pp 26-35