
Crash 2 and Tranexamic Acid
One of the most exciting and underutilized therapies for trauma is tranexamic acid (txa). TXA inhibits the breakdown of clot–it is an anti-fibrinolytic. Is there evidence for using this in trauma patients?
First came the Crash 2 Trial (Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. Lancet 2010; 376: 23–32),
then the subgroup reanalysis (Lancet. 2011 Mar 26;377(9771):1096) showing the benefit of treatment as early as possible.
Recently, the MATTERS trial (Arch Surg. 2012; 147:113-119) was published demonstrating the benefits of TXA in military situations, particularly massive transfusion.
How about an incredible review from the J Trauma (2011; 71(1) Supplement S9)
Then there is this paper describing current military protocol rationale.
To discuss TXA in Trauma, I got to interview Dr. Tim Coats, one of the primary authors of Crash 2.
Here is a List of Resources from the Crash 2 Investigators
This is the official resource page from Crash 2
We also discussed the concept of the pragmatic trial…
Future Research in Emergency Medicine: Explanation or Pragmatism
(Emerg Med J 2011;28(12):1004)
Listen to the podcast excerpt on pragmatic trials (mp3–right click the link and choose save-as if you want to download)
In an amazing demonstration of synchronicity, Jeff Guy of the ICU Rounds Podcast put out a tranexamic acid episode on the same day.
Minh Le Cong provided this prehospital protocol for TXA use.
An incredible review article can be found at this citation (Journal of TRAUMA 2011;71(1) July Supplement)
Update…
This article published in J Trauma (74(6), May 2013, p 1587–1598) gives an excellent summary of the current evidence as of 5/2013
Meta-Analysis shows NO Assoc with Thrombotic Event Increase [Association of Intravenous Tranexamic Acid With Thromboembolic Events and Mortality A Systematic Review, Meta-analysis, and Meta-regression. JAMA Surgery 156(6):3210884, 2021]
Summary: What Do We Know?
- TXA is associated with a 1.5% reduction in 28-day all-cause mortality in adult trauma patients with signs of bleeding (SBP < 90 mm Hg, heart rate > 110 beats per minute, or both, within 8 hours of injury) in a large pragmatic prospective randomized placebo-controlled trial.
- What is critical is the modest effect on the overall population: All-cause mortality was “significantly” reduced from 16.0% to 14·5% (NNT, 67). The risk of death caused by bleeding overall was “significantly” reduced from 5.7% to 4·9% (NNT, 121).
- TXA signal for benefit was in the most severe shock group (admission SBP <= 75 mm Hg), 28-day all-cause mortality of 30.6% for the TXA group versus 35.1% for the placebo group (RR, 0.87; 99% CI 0.76–0.99).
- 1,063 deaths (35%) were caused by bleeding in the CRASH-2 Trial.
- TXA had greatest impact on reduction of death caused by bleeding in the severe shock group (SBP <= 75 mm Hg) (14.9% vs. 18.4%; RR, 0.81; 95% CI, 0.69–0.95).
- Early TXA (<=1 hour from injury) was associated with the greatest reduction (32% reduction) in deaths caused by bleeding (5.3% vs. 7.7%; RR, 0.68; 95% CI, 0.57–0.82; p < 0.0001).
- TXA given between 1 hour and 3 hours after injury also reduced the risk of death caused by bleeding (4.8% vs. 6.1%; RR, 0.79; 95% CI, 0.64–0.97; p = 0.03).
- TXA given after 3 hours after injury was associated with an increased risk of death caused by bleeding (4.4% vs. 3.1%; RR, 1.44; 95% CI, 1.12–1.84; p = 0.004).
- TXA had no impact on TBI outcomes, but the study was limited by small sample size.
- TXA treatment is not associated with an increased risk of vascular occlusive events.
What Is Still Unknown?
- Whether TXA has any impact on trauma outcomes when damage-control resuscitation or MT protocols are used;
- The mechanism by which TXA reduced mortality in trauma in the CRASH-2 Trial. Fibrinolysis assessment and coagulation testing were not part of the study design, and determination of time to cessation of hemorrhage was not required in the study;
- Whether fibrinolysis testing should be performed before consideration of TXA treatment;
- What is the optimal dose and timing of TXA in trauma;
- Whether other antifibrinolytic agents could be substituted for TXA use in trauma;
- Whether TXA is associated with higher seizure rates in trauma or TBI patients. Increased postoperative seizures have been reported in cardiac surgery with TXA doses that are 2-fold to 10-fold higher than those used in CRASH-2.75–80 These seizures have been associated with an increased incidence of neurologic complications (delirium and stroke), prolonged recovery, and higher mortality rates. A proposed mechanism for seizures is TXA-mediated inhibition of glycine receptors as a potential cause of neurotoxicity.81,82 A recent warning has been added to the FDA drug label: “Convulsions have been reported in association with tranexamic acid treatment.”83
A Rational Approach for TXA use in Trauma
- In adult trauma patients with severe hemorrhagic shock (SBP <= 75 mm Hg), with known predictors of fibrinolysis, or with known fibrinolysis by TEG (LY30 > 3%);
- Only administer TXA if less than 3 hours from time of injury;
- TXA administration: 1 g intravenously administered over 10 minutes, then 1 g intravenously administered over 8 hours.
Also, MATTERS II was published, here is the summary from the same J Trauma review
MATTERs II Study excerpted from the same article
The MATTERs II study expanded the sample size of the MATTERs I study to further evaluate TXA and trauma outcomes. A review of 1,332 patients (identified from prospectively collected UK and US trauma registries) who required one or more RBC unit transfusion were analyzed to examine the impact of cryoprecipitate (CRYO) in addition to TXA on survival in combat injured patients.
Despite greater ISSs and RBC transfusion requirements, mortality was lowest in patients who received TXA (18.2%) or TXA/CRYO (11.6%) compared with CRYO alone (21.4%) or no-TXA/CRYO (23.6%). Logistic regression analysis confirmed that TXA and CRYO were independently associated with a similarly reduced mortality (OR, 0.61; 95% CI, 0.42–0.89; p = 0.01 and OR, 0.61; 95% CI, 0.40–0.94; p = 0.02, respectively). The combined TXA and CRYO effect versus neither in a synergy model had an OR of 0.34 (95% CI 0.20–0.58; p < 0.001), reflecting nonsignificant interaction (p = 0.21).
Additional Resources and Studies
- Tranexamic acid in trauma: How should we use it?
- Tranexamic Acid for Trauma Patients: A Critical Review of the Literature
Update
- Progressive intracranial haemorrhage was present in 21 (18%) of 120 patients allocated to TXA and in 32 (27%) of 118 patients allocated to placebo. The difference was not statistically significant [RR?=?0.65 (95% CI 0.40 to 1.05)]. (Tranexamic Acid for Patients With Traumatic Brain Injury A Randomized, Double-Blinded, Placebo-Controlled Trial BMC Emerg Med. 2013;13)
- Annals of Surgery 2015;261(2):390 Tranexamic Acid Use in Severely Injured Civilian Patients and the Effects on Outcomes: A Prospective Cohort Study Cole, Elaine MSc*; Davenport, Ross PhD*; Willett, Keith FRCS†; Brohi, Karim FRCS, FRCA* Abstract Objective: To characterize the relationship between tranexamic acid (TXA) use and patient outcomes in a severely injured civilian cohort, and to determine any differential effect between patients who presented with and without shock. Background: TXA has demonstrated survival benefits in trauma patients in an international randomized control trial and the military setting. The uptake of TXA into civilian major hemorrhage protocols (MHPs) has been variable. The evidence gap in mature civilian trauma systems is limiting the widespread use of TXA and its potential benefits on survival. Methods: Prospective cohort study of severely injured adult patients (Injury severity score > 15) admitted to a civilian trauma system during the adoption phase of TXA into the hospital's MHP. Outcomes measured were mortality, multiple organ failure (MOF), venous thromboembolism, infection, stroke, ventilator-free days (VFD), and length of stay. Results: Patients receiving TXA (n = 160, 42%) were more severely injured, shocked, and coagulopathic on arrival. TXA was not independently associated with any change in outcome for either the overall or nonshocked cohorts. In multivariate analysis, TXA was independently associated with a reduction in MOF [odds ratio (OR) = 0.27, confidence interval (CI): 0.10–0.73, P = 0.01] and was protective for adjusted all-cause mortality (OR = 0.16 CI: 0.03–0.86, P = 0.03) in shocked patients. Conclusions: TXA as part of a major hemorrhage protocol within a mature civilian trauma system provides outcome benefits specifically for severely injured shocked patients.
- Effect of treatment delay on the effectiveness and safety of antifibrinolytics in acute severe haemorrhage: a meta-analysis of individual patient-level data from 40?138 bleeding patients
(DOI: http://dx.doi.org/10.1016/S0140-6736(17)32455-8)
Additional New Information
More on EMCrit
PulmCrit- Tranexamic acid for traumatic brain injury (CRASH3)(Opens in a new browser tab)
Additional Resources
You Need an EMCrit Membership to see this content. Login here if you already have one.
- EMCrit 394 – CV-EMCrit – Inotrope Basics Part 2 – Specific Scenarios - February 7, 2025
- EMCrit RACC-Lit – January 2025 - February 4, 2025
- EMCrit 393 – CV-EMCrit – Inotrope Basics Part 1 - January 25, 2025
thanks Scott. A few aeromedical retrieval services in Australia have introduced a prehospital protocol for TXA. I drafted the protocol for my service based on the CRASH inclusion criteria and methodology. It was very simple and pragmatic as you have alluded to in this podcast, to follow. maybe you dont see too many obstetric patients with post partum bleeding but I believe this is another patient group that would benefit from TXA. The current WOMAN trial uses the same protocol as CRASH 2 for management of PPH in addition to routine care. In my neck of the woods we have… Read more »
Minh & Casey-Have you folks decided to give it to your severe GI bleeds as well?
Minh- can you share your inclusion in the prehospital realm
check out this great talk on the prehospital use of TXA
http://www.aeromedaustralasia.org/img.ashx?f=f&p=perth_2011%2fGILLON+Stuart+_Read-Only_.pdf
I need to ask the boss about releasing the protocol to you..get back to you. Suffice to say its basically the CRASH 2 inclusion criteria for age, vital signs etc, with a time window of 3 hrs from injury and the same dosing and infusion.
to be honest, had not considered its use for GIT bleeding. a quick review of pubmed however does reveal some low level data to indicate a possible mortality benefit. I am unaware of anyone in OZ using it routinely in GIT bleeding management. The situation we often find ourselves in remote OZ, is not having blood products available in the bleeding patient. TXA is a cheap and easily administered agent that has some good evidence to support benefit in bleeding patients. the earlier you give it the better it seems. stop the bleeding, reduce tranfusion needs. I have been using… Read more »
How about in bleeding from ectopic pregnancy complications, in a situation where there are delay to get patient to OR for definitive care , in addition to PPH mentioned by Minh? Can the Crash 2 finding extrapolate to this group of young healthy reproductive aged women?
i think it is an extrapolation, but a good one. I think once you buy the therapy for trauma patients, it is an easy fit for any patient requiring massive transfusion.
If I am out in the middle of nowhere, I got a crashing patient and I want to give blood but do not have any. I’m giving the TXA!
Is there any downside to pre-hospital administration? Do we need to study it in the pre-hospital setting before adoption…or are in-hospital results so impressive that we should adopt it pre-hospital (pending studies demonstrating benefit in that environment)? It seems like this might have great utility for ground EMS in rural settings where transport times to Level I centers are prolonged. If this is such a magic bullet why isn’t this being more widely adopted by centers in the US?
I think that if we accept that this is a valuable treatment in the hospital, then it is a better idea prehospital. We need to create protocols that will predict who will require blood transfusion, but I think it makes sense to get it in as early as possible. I think the uptake delays were the fact that there was only one study. Now that MATTERs verified the benefits, I think most of us are now considering implementation.
I’d like to see Prehospital US adopted (eFAST exams) along with TXA. Seems like the next big thing in civilian EMS treatment of trauma.
As above…it’s cheap, readily mixed up and the NNT is reasonable(ish). For us docs out in the boondocks of rural Oz, it makes sense to give at the roadside/in small rural hospitals early in the transfer process. Many of us don;t have access to blood products other than two units of O neg, then onto walking blood bank…
Hi Scott I think the direction the data is going we will be using it for: GI bleeds, PPH, hyphema (ask Rob O) in addition to trauma in the future. For me GI and Obs bleeds are the big killers. I think we have some good surgical options in trauma that work. We know that massive transfusion is a way of staying afloat until the life boat arrives – it is not a fix. In my world these are the ones where a ” fix” is not always available So TXA is a good addition to the quiver – esp.… Read more »
Hi Scott
Just read all the comments. Predicting massive transfusion is tough, trauma scores like TASH or ABC are ok. But none are good for medical or obs etc
As we have discussed – pulling the trigger on massive transfusion can do harm if done inappropriately (plasma seems to be the bad guy). However, what is the downside of a bolus of TXA….? None I can find yet. So why not give it if there is a chance of big bleed, earlier = better according to Prof. Coats.
Is this reasonable ?
Casey
Nobody can argue with that earlier = better. The question is can we extrapolate TXA with the first unit of blood, which seems justified in trauma patients to medical/obstetric patients. The answer is only maybe. In the acute phase of trauma, the patients are hypo-coagulable, then hyper-coaguable in the subacute phase. We don’t know if is an identical situation in post-partum patients or medical patients. I think we can say it is the case by the time you are massively transfusing these patients, but not necessarily with the 1st couple of units of blood.
This is what the WOMAN trial is answering as pregnancy is a state of relative hypercoaguability.My hunch is that it will be fine as per CRASH 2 in terms of adverse effects of thrombosis. Having prescribed oral TXA for years and having no complications I am reassured that in the emergency setting, it should be fine. But we shall see.
Hi Scott
The WOMAN trial will be fascinating !
We know pregnant women are ” hyper fibrinolytic” and that giving fibrinogen/cryoppt early in PPH seems to help. So TXA would seem to be a logical winner as it impairs fibrinolysis.
But….
The number 1 preventable cause of maternal mortality last triennium was PE. ( sepsis / endometritis pipped it this most recent survey)
So here we have a group where we just might see a bump in TXA- related VTE disease if the trial is big enough to demonstrate this possible effect.
Fingers crossed it doesn’t
Casey
What about studies looking at other uses for this compound…massive GI bleeding, intracranial bleeding, etc?
Hi,
I discussed the use of TXA and UGIB with Prof Coats recently. He was unsure about the theoretical efficay in that setting given the disease in trauma tends to be a large bed of bleeding and injured tissue, as opposed to a point bleeding origin in UGIB. He wondered if the factors liberated by the injured tissue, and the resulting hyperthrombolysis may differ significantly from the GI bleeder……if i have misquoted him I apologise!
Now that is extremely interesting. We are discovering (thanks to folks like Dr. Coats and Karim Brohi) that in order to get the coagulopathy of trauma, you need tissue malperfusion and tissue injury. Just one is not enough. But there are many traumatic injuries that don’t have much tissue injury, e.g. the rent brachial artery from the arm through a window, the stab wound to the groin that hit the femoral vein, etc.) By this same logic, should we be withholding the TXA in these folks. I mailed this question to Dr. Coats as well.
Here is the response from Dr. Coats: Interesting question. One of the exclusion criteria in the CRASH2 study was the patient in who you thought that the bleeding had stopped – these patients could not benefit therefore it was unethical to randomise them. So a stabbed arm that stops with pressure is not an indication for TXA, even if there was a lot of blood loss before effective first aid. There was no heterogeneity by type of injury, blunt/penetrating – it seemed to work equally well for both. I don’t see TXA as a treatment for the coagulopathy of trauma… Read more »
Dr. Coats,
Thanks for your response to my question. As a ED medical director with a significant number of GI bleeds, I have used TXA on many occasions based on the imprecise extension of the logic that it can be likened to a single vessel traumatic injury.
Thanks,
George Hutchison
Does anyone have any data or experience in using TXA in folks where massive transfusions are not needed, such as low-grade liver or spleen injuries? Might help our own anxiety while watching crits and vitals….
George, Dr. Coats has stated that his interpretation of the Crash-2 data is specifically not to only use in massive transfusion but in all patients receiving blood for trauma.
The subgroup I was thinking about is those who we sometimes don’t transfuse — the low-grade spleens, livers and pelvic fractures that we put in the ICU and watch crits on.
These folks may occasionally need a unit or two sometimes in the first 24 hours, but not always.
They would seem to be ideal folks for TXA, but I was just wondering if there is either data or some anecdotal experience.
I think if you are not sure up front whether they are going to get blood, hold off. The reanalysis of Crash-2 showed the advantage is in patients who were transfused and those who got it early. The low-grade spleens and livers do v. well without any additional treatments in most cases.
South western ambulance(UK) have already started using TXA.we are adopting it in our new prehospital care guidelines JRCALC which are due for release soon.Our HEMS as well as London HEMS are using it as well.
that’s great Trig!
Some links that might be interesting following on from the discussion:
The WOMAN trial (TXA in Post Partum Haemorrhage) is at http://www.thewomantrial.lshtm.ac.uk/
If you would like to express an interest in participation in the HALT IT trial (TXA in GI bleeding) please fill out the form at http://ctu.lshtm.ac.uk/invite.php and you will be put on the circulation list for more information as the study develops (I am afraid that centers in the USA will have to provide their own indemnity – the rest of the world can be covered on a central indemnity).
Hey Scott. I am a flight nurse for a Canadian HEMS program. The organization I work for adopted the use of TXA a little over a year ago, but my base is in a northern community that seems a little reluctant on whether or not we are acting as “cowboys” out in the field by working under protocols set forth by physicians in our bigger cities. I personally gave TXA for the first time in a trauma case last week and when I go back into the ER to work, there has been non-stop buzz around what this drug is?… Read more »
tough situation when you are ahead of the hospitals you are transporting to. I understood the skepticism when it was just a single trial, but MATTERS should have allayed it, IMHO. My belief is that first gram is the most important and the drip is bonus. Dr. Coats basically said similar; so even if you just give it in the field, I think you are doing good by your pts.
Terrific podcast and discussion. We’re covering CRASH-2 and MATTERs in our Journal Club this week in an effort to champion its use in our Level 1 Trauma Center. Anticipating questions about the external validity of CRASH-2 when applied to Trauma Centers in industrialized countries with sophisticated resuscitation measures. Professor Coats alluded to a sub-group analysis by continent; is this published, and have there been other responses to this concern? Amazing study-congratulations to all the investigators!
I have not seen that publication yet, but the recent re-analysis by severity is fantastic. (BMJ 2012;345:e5839)
Good Afternoon, Gents: I am an EMS fellow in Newark, NJ and we are considering using TXA in our truck. We utilize a prehospital physicians for large scale incidents, entrapments, as well as routine EMS calls whenever available. We have access to 4 units PRBCs and 4 units of FFP, by request. we delegate this resource to crush pts entraped for >1 hour and evidence of shock/Bywater’s syndrome, or those with massive hemorrhage. TXA is obviously more attractive because its easily transportable. I am also a LCDR in the military and have used TXA in Iraq with great success. My… Read more »
verifying address
just saw the post from Aug on similar question. Sorry for the redundancy.
Scott,
Have you come across any literature to suggest we should avoid TXA in the following:
Active use of any of the following agents:
? Anticoagulants such as warfarin or enoxaparin
? Direct thrombin inhibitors such as dabigatran
? Factor Xa inhibitors such as rivaroxaban or apixaban
? Administration of anti-hemophilic agents such as prothrombin complex concentrates (PCC) and/ or recombinant human factor VIIa
Hi scott, this is venkatesh from india. i am medical representative. i seen the all comments but i have a question to ask you? The TXA injection is can use as Topical in emergency?( to reduce blood loss )
I have only heard of this for epistaxis and for bleeding after tooth extraction.
Are there any trials that evaluate the effect of just giving the initial 1 g bolus (within 3 hours from the injury)? For example, if a flight crew were to give 1 g of TXA, but the receiving hospital did not give the subsequent maintenance dose, would there be any mortality benefit?
I think there good evidence for TXA for sick patients regardless of concurrent resources e.g. MATTERS, Brohi’s London trial. I don’t agree with Coates’ recommendation that we give it to anybody with trauma. As you widen the pool of candidates, your NNT increases. Even though the unit lost is low for TXA, eventually the cost per beneficial outcome increases. Keeping widening it, the NNT possibly will eventually exceed NNH. His article on explanatory and pragmatic trials goes back to the age old debate of lumping and splitting diseases. But the statistical effects observed on an over-represented sub-group of people may… Read more »