Thanks to evidence from the CRASH-2 trial, this inexpensive drug could radically change the way we manage blunt and penetrating trauma.
Over my career in emergency medicine, I cannot recall too many discoveries that have truly transformed the way we treat our patients. Certainly, there are advances in medicine, new drugs, old drugs with new indications and many incremental improvements. However, real game changers are rare. In trauma you might say that our approach has evolved with the availability of bedside ultrasound for FAST exams and even permissive hypotension, which has upended ATLS dogma. But now, 60 years after its original discovery, I think there is a transformational drug that will change the way we manage blunt and penetrating trauma. Meet TXA, tranexamic acid.
For those of you that have been using TXA since the CRASH-2 trial was published in Lancet 2010, I will be reporting old news and you might want to skip over this and just pat yourself on the back. For those who do not order a gram IV of TXA every time you write for a type and screen on a trauma patient, give me the opportunity to present you with interesting data.
What is TXA?
TXA is an antifibrinolytic. What does that mean? Recall that vascular structural integrity is achieved by a homeostatic control system. Mechanisms of clotting, via factors in the intrinsic and extrinsic pathways, leading to the final steps of factor X activation, and conversion of prothrombin to thrombin, which in turn catalyzes fibrin formation (from its precursor, fibrinogen). This clot-forming cascade is opposed and balanced by the fibrinolytic system, in which the ubiquitous plasminogen is activated by tissue plasminogen activator (tPA) to form the active enzyme plasmin. This tPA-plasmin complex is capable of breaking down the fibrin clot network, creating fibrin split products. Plasminogen has an active binding site for fibrin, which is blocked by the aminoacid lysine. Two synthetic derivatives of lysine are available that mimic the action of lysine, however bind to plasminogen much more effectively: ε-aminocaproic acid (EACA), about 75 times as potent as lysine and tranexamic acid (TXA) another 6-10 times more potent than EACA . This means that TXA is really effective at blocking plasmin’s attempts to break fibrin down.
Tranexamic acid is available as oral and IV formulations. Onset of action after IV administration is within 5-15 minutes, half life is listed as from 2 to 15 hours. TXA crosses the blood-brain and placental barriers. It is 5% metabolized and 95% excreted in the urine.
There is a wide variety of dosing regimes used in published trials. Loading doses range from 2.5 to 100mg/kg, followed by an infusion at 0.25 to 4 mg/kg/hr for one to 12 hours. For purpose of simplicity the largest trial, CRASH-2 used a standardized adult dose of 1gm IV load over 10 minutes followed by another 1gm IV over 8 hours. Pediatric dosing has not been established, however “limited data suggests that dosing instructions for adults can be used for pediatric patients needing tranexamic acid therapy.”
Cost of TXA is about $3 for a 1gm IV vial.
TXA: A Brief History
- TXA was co-discovered in Japan by Ms. Utako Okamoto and Mr. Shosuke Okamoto in 1955 in their research on antiplasmins.
- Clinical trials began in the late 60’s with multiple studies showing evidence of improved blood loss in menorrhagia and during surgery.
- In 1986 IV TXA was approved by the FDA for use in the prevention of bleeding after dental extraction in hemophiliacs.
- In 2009 the oral formulation received approval for heavy menstrual bleeding.
Cases, Indications, Studies
Out of the many patients I have treated with TXA, I would like to present three cases to illustrate potential applications and provide some of the relevant literature.
Case 1: 21 year-old female, restrained passenger in single vehicle MVC, hit a tree. Her vital signs on the scene were reported hypotensive and tachycardic, improving with 500cc NS. She came to our ED in full spinal immobilization. On arrival BP 104/56, HR 112, RR 22. She denied LOC. Lung sounds were equal. Abdomen diffusely tender to palpation. No obvious pelvic instability or long bone fractures. FAST exam reveals large amount of intraperitoneal blood and fractured spleen. TXA given, 1gm IV over 10 minutes, then 1gm on an eight hour drip. Vital signs stabilized in the ED, patient taken for CT, which demonstrated a laceration of the left lobe of the liver and fractured spleen. Patient transfused three units PRBC and transferred to the closest trauma center with interventional radiology capabilities. On arrival at the trauma center, three hours post injury; the patient became hypotensive and was taken to the OR for open laparotomy. Post-op course with good recovery.
Indication for TXA and dosing is based on the results of the CRASH-2 trial. Published in the Lancet in 2010, the study enrolled over 20,000 patients from 274 hospitals in 40 countries. Patients were randomized to TXA (1gm over 10 minutes followed by 1gm over 8 hours) or placebo. Included were adult patients up to eight hours after trauma with an initial BP less than 80mmHg or HR greater than 110 or at risk for shock. At four weeks all cause mortality (primary outcome) was 14.5% in the TXA and 16% in the placebo group (absolute risk reduction 1.5%, RR 0.81, NNT 67, p=0.0035). No difference in secondary outcomes, in particular blood transfusions (about 50% both groups) or thrombo-embolic events (1.7% in TXA, 2.0% in placebo group). Subgroup analysis indicates increased benefit for the more severely injured and if given early.
The study was followed up with an exploratory analysis, also published in the Lancet in 2011, looking at risk of death due to bleeding, stratified by timing of TXA dose. If TXA was given within an hour of injury, death due to bleeding occurred in 5.3% in the treatment group, 7.7% in the placebo group (RR 0.68), at one to three hours 4.8% vs 6.1% (RR 0.79) and after the three-hour window 4.4% vs 3.1% respectively (RR 1.44). All results were statistically significant.
In the retrospective observational study by the US military (MATTERs) published in 2012, the benefit of TXA is independently illustrated as reduction of unadjusted mortality (17.4% TXA, 23.9% placebo).
Interestingly, severe trauma is now recognized to cause a detrimental clotting disorder called the “acute coagulopathy of trauma” which appears to be independent of hypothermia and clotting factor dilution from IV fluids. This acute coagulopathy is associated with a fourfold increase in mortality. In a 2011 study published in the Annals of Surgery, Kashuk et al. associates the disorder with “primary fibrinolysis” as determined by the drop in the viscoelastic strength of the forming clot (measured by thrombelastography – TEG). This is most prevalent in the first hour after severe traumatic injury. The early inhibition of clot stability and elasticity, which does not manifest itself as abnormal PTT, INR or platelet count, provides an attractive theoretical framework to explain the benefit of TXA in severe trauma.
Case 2: 62 year-old morbidly obese patient with post-tonsillectomy bleed. H/o cochlear implant. Operated at outside university medical center four days prior to presentation. Bleeding began suddenly 45 minutes before arrival. On arrival vital signs are stable, but patient is sitting up, leaning forward and spitting up blood and clots in emesis bag every few seconds. Volume approximately 1500cc. Unable to visualize retropharynx due to large amount of blood and clots. Airway maintained, however Mallampati class IV, short, thick neck. Called anaesthesia backup and ENT surgeon on call to discuss best course of action. The patient received TXA 1gm IV over 10 minutes followed by TXA infusion, 1gm over 8 hours. 20 minutes later the bleeding had basically stopped. ENT surgeon surprised about sudden change in severity, “you gave TX-what?” Debated whether to take patient to the OR for definitive stabilization, however, due to cochlear implant not able to use unipolar cautery (no bipolar available). Patient finally, after multiple discussions with the original surgeon at the university medical center, transferred uneventfully to the medical center 2 and a half hours away. Patient arrived there with bleeding completely controlled. Admitted to the ICU. Rebleed 10 hours later, patient taken to the OR for definitive cautery. Post-op course uneventful.
Indication for TXA was the post-op bleed with few options to directly control with the resources available at the local ED. TXA treatment supported by multiple studies on intra- and post-operative hemorrhage. Chan et al. published a meta-analysis of seven studies on the use of TXA in tonsillectomy in 2013 and concluded that TXA significantly reduced blood loss. Poor differentiation between peri- and postoperative bleeding. A review of the utility in surgery of tranexamic acid by Dunn et al. in 1999 concluded a significant (29-54%) reduction in postoperative blood loss. Studies specifically comparing use of TXA vs. usual care in delayed post-tonsillectomy bleeding presenting to the ED were not found.
Case 3: 68 year-old male with severe GI bleed. Patient with dementia sent from nursing home, acute onset lower GI bleed. Maroon stools mixed with large amount of dark blood. No emesis. BP 86/49, HR 103, RR 22. Patient resuscitated with IV fluids, PPI IV bolus and infusion. TXA given 1gm IV load over 10 minutes followed by 1gm IV infusion over 8 hours. 30 minutes after initiating TXA rectal bleeding completely controlled with stool frequency dropping from every 10 to 15 minutes to none. Patient admitted to the hospital, received a total of 4 units PRBC and 2 units FFP. Upper and lower endoscopy did not reveal the source of bleeding. On hospital day 2 patient developed a DVT in the right lower extremity and was treated with placement of an IVC filter. During his hospitalization the patient had no further GI bleeding events.
The massive lower GI bleed was felt to be potentially life threatening and an indication for TXA. While literature can be found on upper GI bleeding, no studies exist that evaluate TXA in the treatment of lower GI bleeds. A 2014 Cochrane review looks at 8 studies published between 1973 and 2011. Grouped total mortality is reported as 4.9% in the TXA group compared to 8.4% in patients receiving placebo (p 0.007) and concludes that “tranexamic acid appears to have a beneficial effect on mortality”. A large trial of TXA in the treatment of GI bleeding is currently under way (HALT-IT).
The case presented highlights the need to assess the risk for thrombo-embolic complications and vascular occlusion. All available evidence, however, suggests that there is no increase in DVT’s, PE’s, strokes or MI’s in patients receiving TXA compared to those receiving placebo. Monitored as one of the endpoints in the CRASH 2 trial, thrombo-embolic events are reported as 1.7% in the TXA group and 2% in the placebo group, death due to vascular occlusion as 0.3% and 0.5% respectively. The differences do not reach statistical significance. However, as the mechanism of action of TXA would suggest the possibility of an increased risk of vascular occlusive events, a history of prior DVT’s or PE’s or known hyper-coagulability is considered a relative contraindication for the use of TXA.
There are a large number of trials evaluating the benefit of TXA in indications such as ante- and postpartum hemorrhage, menorrhagia, surgery, upper GI bleeding and hemophilia, uniformly reporting a reduction in blood loss without adverse events. Currently a large RCT is being conducted by the London School of Hygiene and Tropical Medicine on postpartum hemorrhage (World Maternal Antifibrinolytic Trial – WOMAN) to scientifically establish the role of TXA in postpartum hemorrhage which is estimated to claim approximately 100,000 lives annually.
TXA enjoys a large database of studies demonstrating its efficacy and safety. The largest trial to date is the CRASH-2 trial, which enrolled over 20,000 patients. As the largest RCT evaluating the use of TXA, this high quality trial is critically relevant for decision making in trauma patients presenting to our emergency departments. Other studies corroborate the results and suggest that it may be valid to expand the indications from acute trauma to other patients presenting with hemodynamically relevant bleeding. Currently ongoing studies such as CRASH-3 (head trauma), WOMAN (post-partum hemorrhage) and HALT-IT (GI bleed) are focused on establishing a solid basis on the use of TXA in other areas of life-threatening hemorrhage.
TXA appears to have the greatest impact if used early in the trauma patient with signs of hemodynamic compromise (hypotension, tachycardia) or who appear at high risk for traumatic bleeding. Given the excellent safety profile it appears reasonable and logical to consider initiating treatment with TXA by EMS in the pre-hospital setting.
TXA is not a replacement for definitive care of the injured patient. Two of the three cases presented had recurrent bleeds. Therefore patients with a life-threatening hemorrhage need to be observed or definitive care needs to be emergently arranged, regardless of whether the acute bleeding is controlled with TXA. It has been well established that care of the trauma patient in a level 1 trauma center improves the chances of survival. Routine use of TXA in non-trauma center ED’s possibly has a role in providing a sufficient time window for severe trauma patients to be transferred to a center specialized and ideally equipped for their care. More studies are required to establish when transfer to a trauma center is in the best interest of the patient.
In cases of minor bleeds it may be safe to discharge the patient after the bleeding is controlled. In a study on anterior epistaxis by Zahed et al., time to hemostasis, time in ED and patient satisfaction were improved in the group receiving topical atomized TXA. Rebleeding within 24 hours was 4.1% in the TXA group vs. 11% in patients who had received routine anterior nasal packing.
When I was told about TXA the first time several years ago by one of our well-read EMTs, my immediate response as a Vioxx-tainted physician was: well, if there was a substance as good as you say with essentially no side effects that can be used on bleeding trauma patients, I would have heard about it. After all, I do my CME, I visit national and international conferences, I lecture, I discuss relevant news with colleagues. So this must be a fad, some newly advertised treatment – along the lines of activated factor VIIa for trauma.
Well, I was wrong. It has been 60 years since the discovery of TXA as an antifibrinolytic, there is abundant surgical literature, now also a randomized, placebo controlled study with over 20,000 patients demonstrating its utility. And at $6 cost for load and eight hour infusion, this is a bargain. So why is it not part of every trauma protocol? Why is it not what our EMTs grab at the same time they start and IV and give their first fluid bolus to the hypotensive or tachycardic trauma patient? Where is the catch?
I wrote this because I can’t find the catch. The more I read about TXA, the more I wonder what is keeping us from using this life-saving drug. 10% relative risk reduction all cause mortality, 15% relative mortality reduction from bleeding. But there is no core measure to comply with supporting its adoption into our trauma toolbox both pre-hospital and in the ED. Not enough benefit? Well, we have core measures for tPA in stroke, marginal benefit in functional outcome in some (not all) studies, yet no established mortality benefit. Early goal directed therapy in presumed sepsis? Well, maybe there is an improvement in mortality, though a recent meta-analysis argues that there is no reduction in ICU days or total hospital stay, nor survival benefit. How about our OHCA patients? Epinephrine in the resuscitation bay? ACLS tells us to use 1mg Epi every 3-5 minutes, however, studies show that there is no survival benefit to hospital discharge even though it has been shown to improve ROSC and survival to admission. In contrast there is solid evidence supporting the use of TXA. I conclude that it has not become part of our trauma routine simply because no company or institution has sufficient financial interest to promote the $3 a dose life-saving drug.
My recommendation: Pull up the original data on CRASH-2, read about TXA. If you like what you read, use it. Tell your friends. I am convinced that you will welcome TXA as an invaluable addition to your treatment options for some of your sickest patients. And I believe that we as a specialty will be successful in promoting this medication through our professional networks based on its merits.
- Roberts I, et al. “Proposal for the inclusion of tranexamic acid (anti-fibrinolytic – lysine analogue) in the WHO list of essential medicines.” 18th Expert Committee on the Selection and Use of Essential Medicines 2011. WHO. http://www.who.int/entity/selection_medicines/committees/expert/18/applications/TRANEXAMIC_ACID_10_2.pdf?ua=1
- Helber S. “Bringing women to the forefront of science and medicine.” Lancet Editorial 2012 Mar 10;379(9819):867
- CRASH-2 collaborators, Shakur H, Roberts I, et al. “Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomized, placebo-controlled trial.” Lancet 2010 Jul 3;376(9734):23-32
- CRASH-2 collaborators, Roberts I, Shakur H, et al. “The importance of early treatment with tranexamic acid in bleeding trauma patients: an exploratory analysis of the CRASH-2 randomised controlled trial.” Lancet 2011 Mar 26;377(9771):1096-101
- Morrison JJ, et al. “Military application of tranexamic acid in trauma emergency resuschitation (MATTERs) study.” Arch Surg 2012 Feb;147(2):113-9
- Brohi K, et al. “Acute coagulopathy of trauma: hypoperfusion induces systemic anticoagulation and hyperfibrinolysis.” J Trauma 2008 May;64(5):1211-7
- Frith D, et al. “The acute coagulopathy of trauma shock: clinical relevance.” Surgeon 2010 Jun;8(3):159-63
- Kashuk JL, et al. “Primary fibrinolysis in integral in the pathogenesis of the acute coagulopathy of trauma.” Ann Surg 2010 Sep;252(3):434-42
- Chan CC, et al. “Systematic review and meta-analysis of the use of tranexamic acid in tonsillectomy.” Eur Arch Otorhinolaryngol. 2013 Feb;270(2):735-748
- Dunn CJ, et al. “Tranexamic acid: a review of its use in surgery and other indications.” Drugs 1999 Jun;57(6):1005-32
- Bennett C, et al. “Tranexamic acid for upper gastrointestinal bleeding.” Cochrane Database Syst Rev. 2014 Nov 21;11
- MacKenzie EJ, et al. “A national evaluation of the effect of trauma-center care on mortality.” N Engl J Med 2006 Jan 26;354(4):366-78
- Zahed R, et al. “A new and rapid method for epistaxis treatment using injectable form of tranexamic acid topically: a randomized controlled trial. Am J Emerg Med 2013;31(9):1389-92
- Uhrig D, et al. “Recombinant tissue plasminogen activator does not impact mortality in acute ischemic stroke. A meta-analysis of randomized controlled trials.” Stroke 2015;46:ATMP119
- Zhang L, et al. “Early goal-directed therapy in the management of severe sepsis or septic shock in adults: a meta-analysis of randomized controlled trials.” BMC Med 2015 Apr 3;13:71
- Lin S, et al. “Adrenaline for out-of-hospital cardiac arrest rescuscitation: a systematic review and meta-analysis of randomized controlled trials.” Resuscitation 2014 Jun;85(6):732-40