CE: Madhur; ACO/200886; Total nos of Pages: 4; ACO 200886

SPECIAL COMMENTARY

CURRENT OPINION Should antifibrinolytics be given in all patients with trauma?

Marcel Levi

Purpose of review Hemorrhage is the second most important cause of death in patients with trauma, contributing to approximately 30% of trauma-related mortality. Pharmacological prohemostatic agents may be useful adjunctive treatment options in patients with severe loss. Recent findings was evaluated in a large international randomized controlled study in patients with trauma and severe blood loss. The was shown to reduce death due to bleeding, provided the treatment was given within 3 h after injury. Tranexamic acid treatment did not result in serious adverse events nor thrombotic complications. Summary In view of this efficacy and safety of this relatively cheap and simple drug, it may be recommended to put tranexamic acid in the first (maybe even prehospital) line of management of patients with severe traumatic hemorrhage. Keywords antifibrinolytics, hemorrhage, prohemostatic , tranexamic acid, trauma

INTRODUCTION physiological anticoagulation. This strategy may be Bleeding is a frequently occurring clinical problem. useful in the prevention and treatment of bleeding A substantial number of hospital admissions in in patients with defects but also in medical wards is related to bleeding and peri- patients with an a priori normal coagulation system, operative bleeding is one of the most frequent com- who experience severe (postoperative) bleeding or plications of surgery [1&&]. Bleeding is of particular are to undergo procedures known to be associated importance in trauma and is the second most with major blood loss [5]. important cause of death in trauma patients, The safety of prohemostatic therapy also deserves contributing to approximately 30% of trauma- some consideration. Interfering in the balance related mortality [2]. Trauma-related coagulopathy between coagulant and mechanisms proceeds via a myriad of mechanisms, including loss can indeed result in undesirable adverse effects. of factors and due to massive bleeding, The best illustration may be the higher risk of bleed- acidosis and hypothermia further compromising ing in patients receiving anticoagulant therapy. the coagulation system, dysregulation of mediatory Conversely, prohemostatic agents may, at least pathways, such as the activated protein C system theoretically, predispose for thrombotic compli- and , and systemic activation of throm- cations. The occurrence of such complications are bin generation [3&]. fortunately relatively rare. Obviously, the expected Management of bleeding consists of local benefit of the application of prohemostatic agents control measures to retain adequate circulation, and proper transfusion procedures [4]. In addition Department of Medicine, Academic Medical Center, University of to these strategies, prohemostatic treatment may, Amsterdam, Amsterdam, The Netherlands in some cases, support the treatment of (severe) Correspondence to Marcel Levi, MD, Academic Medical Center, bleeding. Pharmacological agents that are capable University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The of promoting hemostasis or fibrin formation, or Netherlands. Tel: +31 20 5662109; e-mail: mailto:[email protected] can block fibrinolytic activity, may interfere in Curr Opin Anesthesiol 2012, 25:000–000 the balance between activation of coagulation and DOI:10.1097/ACO.0b013e3283532b29

0952-7907 ß 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins www.co-anesthesiology.com Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. CE: Madhur; ACO/200886; Total nos of Pages: 4; ACO 200886

Special commentary

their fibrinolysis-inhibiting capacity. Indeed, tran- KEY POINTS examic acid (Cyklokapron, Pfizer, New York, NY, Bleeding is of particular importance in trauma and is USA) is at least 10 times more potent than e-amino- the second most important cause of death in trauma caproic acid (Amicar, Xanodyne Pharmaceuticals patients, contributing to approximately 30% of trauma- Inc., Newport, KY, USA). Tranexamic acid was shown related mortality. to be effective in reducing blood loss, transfusion requirement, and prevention of re-exploration due Prohemostatic strategies, including antifibrinolytic agents, may support the treatment of severe bleeding in to ongoing hemorrhage in a variety of clinical set- trauma patients. tings, including cardiac surgery, major orthopaedic surgery, and gynaecological conditions [11–13]. Randomized controlled trials showed a reduction of A recently conducted large randomized trial mortality in patients with major trauma and with, or at with tranexamic acid in 20 211 patients (admitted risk of, severe hemorrhage that received the antifibrinolytic agent tranexamic acid. to 274 hospitals in 40 different countries) with major trauma and with, or at risk of, severe hemor- Tranexamic acid is remarkably well tolerated in rhage showed a reduction of mortality in patients patients with trauma and is not associated with who received tranexamic acid compared with thrombotic complications or other adverse events. placebo [9&&]. Patients were treated within 8 h of It has been calculated that universal use of tranexamic injury with either tranexamic acid (1 g over 10 min acid may result in the aversion of death in more than followed by another gram over 8 h) (n ¼ 10.096), or 70 000 trauma patients each year worldwide. placebo (n ¼ 10.115). All-cause mortality at 4 weeks after admission was 14.5% in the tranexamic acid group as compared with 16% in the placebo group [relative risk (RR) 0.91, 95% confidence interval (CI) in distinct clinical situations should be balanced 0.85–0.97, P ¼ 0.0035]. The risk of death due to with the risk of thrombosis in that particular patient haemorrhage was reduced from 5.7% in controls population. to 4.9% in the tranexamic acid-treated patients Prohemostatic treatment can interfere with (RR 0.85%, 95% CI 0.76–0.96, P ¼ 0.0077). Interest- the coagulation system or promote primary hemo- ingly, there were no detectable differences in the rate stasis [1&&,6–8]. One of the best studied prohemo- of transfusion and the need for surgical (re)explora- static interventions is antifibrinolytic treatment. tion between the two groups. Tranexamic acid was Recently, the use of the antifibrinolytic agent remarkably well tolerated in this study. In the tran- tranexamic acid was shown to reduce mortality examic acid group, 168 patients (1.7%) had occlusive in trauma patients with excessive blood loss in a vascular events compared with 201 patients (2.0%) large international controlled multicenter trial [9&&]. in the placebo group. There were also no differences We will briefly report on the findings in this trial and in deaths due to vascular complications between the subsequently try to answer the question whether two groups. all patients with trauma should be treated with In an additional analysis, the authors found tranexamic acid. strong evidence that early administration of tran- examic acid was relatively more favourable in com- parison with administration at a later time after ANTIFIBRINOLYTIC TREATMENT IN trauma [14&]. Early (<1 h) treatment with tran- TRAUMA examic acid reduced the rate of death due to bleeding Agents that exert antifibrinolytic activity are to 5.3% compared with 7.7% in the placebo group and the group of lysine analogues [10]. (RR 0.68, 95% CI 0.57–0.82, P < 0.0001), whereas Lysine analogues, that is, e- and this RR decreased when the drug was given 1–3 h tranexamic acid are potent inhibitors of fibrinolysis after trauma to 0.79 (95% CI 0.64–0.97, P ¼ 0.03) [10]. The antifibrinolytic action of lysine analogues and was 1.44 (95% CI 1.12–1.84, P ¼ 0.004) when is based on the competitive binding of these tranexamic acid was given later than 3 h. There was agents to the lysine-binding sites of a fibrin clot, no evidence that the effect of tranexamic acid thereby competing with the binding of plasmino- on death due to bleeding was influenced by other gen. Impaired plasminogen binding to fibrin delays factors, including Glasgow Coma Score, type of the conversion of plasminogen to and injury, or SBP. Another analysis concerned the subsequent plasmin-mediated fibrinolysis, which subgroup of patients with . then proceeds at an inefficient and slow rate. In a nested case–control study, the authors studied Subtle molecular variations between different lysine 270 patients who also had traumatic brain injury in analogues may have important consequences for addition to their extracranial bleeding due to trauma

2 www.co-anesthesiology.com Volume 25 Number 00 Month 2012 Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. CE: Madhur; ACO/200886; Total nos of Pages: 4; ACO 200886

Antifibrinolytics in patients with trauma Levi

[15]. Mean intracranial hemorrhage growth at 24– to show time-related transfusion rates or (excessive) 48 h after admission was 5.9 ml (SD 26.8) in the transfusion-free survival rates, information which tranexamic acid group as compared with 8.1 ml is probably not available. Another (not unlikely) (SD 29.2) in the placebo group. There were new focal possibility is that transfusion protocols are not very cerebral ischemic lesions in 5% of patients treated well tailored to the bleeding status of the patient. with tranexamic acid and in 9% of patients in the control group. Death occurred in 11% of tranexamic acid-treated patients versus 18% in control patients CONCLUSION (adjusted odds ratio 0.47, 95% CI 0.21–1.04). Prohemostatic interventions appear to be effective in reducing perioperative blood loss and transfusion PUTTING THE CRASH-2 DATA IN requirements in specific situations and may be PERSPECTIVE helpful adjuncts in the management of severe spontaneous and postoperative bleeding. Early anti- The Clinical Randomization of an Antifibrinolytic in fibrinolytic treatment with tranexamic acid (1 g as a Significant Hemorrhage (CRASH)-2 study convinc- bolus in 10 min followed by another gram in 8 h ingly shows that administration of the antifibrino- intravenously) was clearly shown to be of benefit in lytic agent tranexamic acid results in less bleeding patients with severe trauma and major blood loss. and death due to bleeding in patients with severe Analysis of trial results indicate that this treatment trauma. Other clinical interventions have not should be given within 3 h after trauma. Remark- been proven to be effective in this setting, including ably, the administration of tranexamic acid was blood-saving strategies, resuscitation protocols well tolerated and did not result in major adverse using different plasma expanders, and hemodynamic events or thrombotic complications. This seems to intervensions, as well as strategies specifically & be another benefit of this intervention, in contrast focused at improvement of coagulation [16 ,17]. to other prohemostatic strategies that may carry the In a meta-analysis of four studies investigating anti- risk of thrombosis [21]. In addition, the treatment fibrinolytic agents in patients with acute traumatic with tranexamic acid is simple and relatively cheap, injury and (high risk of) bleeding, the authors found contributing to a beneficial cost-effectiveness, also that tranexamic acid reduced the risk of death by in resource-poor settings. 10–15%. Obviously, the results in this meta-analysis The next challenge is to translate these favour- were strongly driven by the CRASH-2 study results. able findings from clinical studies into real-life In an interesting cost-effectiveness study in three clinical practice [22]. There are many arguments types of countries participating in the CRASH-2 for placing tranexamic acid in the first line of study, the investigators were able to demonstrate defence in patients with severe traumatic injury that the incremental cost per life year gained by and blood loss. In some settings, this may mean the administration of tranexamic acid was US$ prehospital administration of the drug, which could 48, 66, and 64 in Tanzania, India, and the UK, be facilitated by the uncomplicated storage of the respectively [18]. Based on these calculations, it drug. If prehospital administration is not possible or may be concluded that tranexamic acid was highly required, it may be administered in the emergency cost-effective, regardless of the setting in high- room provided that the time window of 3 h after income or low-income countries. injury has not passed. Treatment should be initiated It may be argued that whereas the RR reduction on the basis of clinical findings and regardless of in traumatic death caused by tranexamic acid was laboratory tests, which are anyway not very helpful impressive, the absolute risk reduction was modest. in the early stage of the management of traumatic Although this may be true, a 1.5% reduction in bleeding. The inclusion criteria of the CRASH-2 death due to trauma translates to the saving of a study are sufficiently helpful for routine clinical very large number of people worldwide, as death due practice to guide its administration and it is not to trauma is very prevalent. It has been calculated likely that a certain amount of ‘overtreatment’ that universal use of tranexamic acid may result in would result in direct harm for the patient. Taken the aversion of death in more than 70 000 patients & together, there seems to be sufficient evidence to each year worldwide [19 ]. The lack of any effect put tranexamic acid in the first line of treatment of of tranexamic acid on transfusion rates remains patients who present with major blood loss after puzzling. However, one needs to realize that it trauma. may well be that surviving patients received more transfusion than patients who died in the early phase after trauma [20]. This ‘survivor bias’ could Acknowledgements only be properly analysed if the data would allow us None.

0952-7907 ß 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins www.co-anesthesiology.com 3 Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. CE: Madhur; ACO/200886; Total nos of Pages: 4; ACO 200886

Special commentary

10. Mannucci PM. Hemostatic drugs. N Engl J Med 1998; 339:245– Conflicts of interest 253. There are no conflicts of interest. 11. Levi M, Cromheecke ME, de Jonge E, et al. Pharmacological strategies to decrease excessive blood loss in cardiac surgery: a meta-analysis of clinically relevant endpoints. Lancet 1999; 354:1940–1947. 12. Henry DA, Carless PA, Moxey AJ, et al. Antifibrinolytic use for minimising perioperative allogeneic blood transfusion. Cochrane Database Syst Rev REFERENCES AND RECOMMENDED 2011:CD001886. READING 13. Fergusson DA, Hebert PC, Mazer CD, et al. A comparison of aprotinin and Papers of particular interest, published within the annual period of review, have lysine analogues in high-risk cardiac surgery. N Engl J Med 2008; 358:2319– been highlighted as: 2331. & of special interest 14. Roberts I, Shakur H, Afolabi A, et al. The importance of early treatment && of outstanding interest & with tranexamic acid in bleeding trauma patients: an exploratory analysis Additional references related to this topic can also be found in the Current of the CRASH-2 randomised controlled trial. Lancet 2011; 377:1096– World Literature section in this issue (pp. 000–000). 1101. Important substudy showing time window in which tranexamic acid is effective 1. Mannucci PM, Levi M. Prevention and treatment of major blood loss. N Engl J in trauma patients. && Med 2007; 356:2301–2311. 15. Effect of tranexamic acid in traumatic brain injury. a nested randomised, Review on efficacy and safety of prohemostatic interventions in various clinical placebo controlled trial (CRASH-2 Intracranial Bleeding Study). BMJ 2011; settings. 343:d3795. 2. Acosta JA, Yang JC, Winchell RJ, et al. Lethal injuries and time to death in 16. Curry N, Hopewell S, Doree C, et al. The acute management of trauma a level I trauma center. J Am Coll Surg 1998; 186:528–533. & hemorrhage: a systematic review of randomized controlled trials. Crit Care 3. Brohi K, Cohen MJ, Ganter MT, et al. Acute coagulopathy of trauma: 2011; 15:R92. & hypoperfusion induces systemic anticoagulation and hyperfibrinolysis. Overview article meta-analysing all studies on prohemostatic treatment in bleeding J Trauma 2008; 64:1211–1217. trauma patients. Interesting article on new pathophysiological insights into the role of coagulation in 17. Boffard KD, Riou B, Warren B, et al. Recombinant factor VIIa as adjunctive traumatic bleeding. therapy for bleeding control in severely injured trauma patients: two parallel 4. Levi M. Critical bleeding in surgery: conventional therapy and new prospects. randomized, placebo-controlled, double-blind clinical trials. J Trauma 2005; Minerva Anestesiol 2004; 70:267–271. 59:8–15. 5. Levi M, Vink R, de Jonge E. Prevention and treatment of bleeding by pro- 18. Guerriero C, Cairns J, Perel P, et al. Cost-effectiveness analysis of hemostatic treatment strategies. Wien Med Wochenschr2003; 153:421–425. administering tranexamic acid to bleeding trauma patients using evidence 6. Levi M, Eerenberg E, Kamphuisen PW. Periprocedural reversal and bridging from the CRASH-2 trial. PLoS One 2011; 6:e18987. of anticoagulant treatment. Neth J Med 2011; 69:268–273. 19. Sydenham E. Thousands of lives could be saved using tranexamic acid for 7. Levi MM, Eerenberg E, Lowenberg E, Kamphuisen PW. Bleeding in patients & patients with bleeding trauma. Inj Prev 2011; 17:211. using new or antiplatelet agents: risk factors and management. Article extrapolating the benefit of universal use of tranexamic acid in trauma Neth J Med 2010; 68:68–76. patients. 8. Lowenberg EC, Meijers JC, Levi M. -vessel wall interaction in health 20. Cap AP, Baer DG, Orman JA, et al. Tranexamic acid for trauma patients: and disease. Neth J Med 2010; 68:242–251. a critical review of the literature. J Trauma 2011; 71:S9–S14. 9. Shakur H, Roberts I, Bautista R, et al. Effects of tranexamic acid on death, 21. Levi M, Levy JH, Andersen HF, Truloff D. Safety of recombinant activated && vascular occlusive events, and blood transfusion in trauma patients with factor VII in randomized clinical trials. N Engl J Med 2010; 363:1791– significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. 1800. Lancet 2010; 376:23–32. 22. Ubbink DT, Vermeulen H, Knops AM, et al. Implementation of evidence- Very important large multinational randomized controlled trial on efficacy and safety based practice: outside the box, throughout the hospital. Neth J Med 2011; of tranexamic acid in trauma patients demonstrating reduction in mortality. 69:87–94.

4 www.co-anesthesiology.com Volume 25 Number 00 Month 2012 Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.