Tranexamic Acid



Lets get this bit over and done with quicksmart.


TXA is a synthetic derivative of the amino acid lysine.It says so in the CRASH 2 paper. Now plasminogen has lysine binding sites on it that allow it to be activated to plasmin so it can get down to the business of busting up clots and saving stroke patients and making people bleed to death and such. TXA competes for these binding sites preventing activation of plasminogen. So this means it is an antifibrinolytic.

Because the coagulation cascade is an incredibly complex system the precise effect of this on the patient will depend on a whole heap of stuff but in essence it acts to stop clot breakdown.


Oral bioavailability is about 40%, it is distributed in all compartments and filtered at the glomerulus, elimination is via the kidney with a half life of about 4 hours (although effect half life is much longer at about 80 hours) and it is excreted in very small quantities in breast milk (Nillson, J Clin Pathol, 33, Suppl (Roy Coll Path), 14, 41-47).

CRASH 2. What's all the fuss?

What did they do?

20,000 odd patients who were within 8 hours of trauma and whose doctors were considering transfusing got 1g of TXA stat and 1g over 8 hours or placebo. Only a handful of patients needed to be excluded from analysis (either because the withdrew consent after randomisation or it turned out they never met the inclusion criteria in the first place).

Primary outcome was all cause mortality at 4 weeks

They also pre-defined some subgroups that they thought might be meaningful. This offers them more validity than the fishing expedition that is the post hoc subgroup analysis however they are still only subgroups and if anyone can explain to me the difference between these and secondary endpoints I will be very grateful. They divided into death due to bleeding, vascular occlusion (AMI, stroke, PE), multi-organ failure, head injury and other and they divided the patients on the basis of some baseline characteristics (time since injury < 1hr, 1-3hr, 3-8hr; sBP <76, 76-89, >89; GCS 3-8, 9-12 and 12-15; and penetrating vs blunt trauma).There were some actual secondary outcome measures too but these weren't too interesting.

What did they find?

The TXA group had a death rate fo 14.5% and the placebo group, 16%. That means a 9% relative risk reduction, a 1.5% absolute risk reduction and a number needed to treat of 67. Because the trial was huge the result is statistically significant.

The results were best in the people who were sicker (you would struggle to improve mortality in a group of people who were not going to die anyway) and in the people treated in the first hour. in his interview with Scott Weingatt on CRASH 2 investigator, Dr Tim Coats, opined that in the first hours there was the most to gain, that the impact of TXA on the clotting system may be different early vs. late and that early on after trauma people tend to be anticoagulated while later they tend to get more "sticky" (which would fit with clinical experience of smashed up trauma patients being oozy in resus and theatre but getting DVT/PE in the coming days).

What is more, they found that while there was a statistically significant decrease in bleeding deaths there was a trend toward a decrease in vascular occlusion deaths as well.

And the drug costs about $10-$100 per treatment depending on where you are so you can save a life for as little as $670.

What could possibly be wrong with this?

First of all I am bothered by a drug that is great for stopping bleeding but which also probably stops clotting, which has no side effects, and which has the same dose for every patient no matter how big or small or sick or well or complex or simple. To me that sounds like naturopathy and I can't help but be suspicious that the trial has failed to show the full picture. That is not a reason to discount the findings of the trial but I think that when something sounds too good to be true it often is so you had better tread carefully.

Second, I am not totally sure how to relate findings in trauma patients from the countries in which the trial was conducted to patients from Victoria, Australia where we have 5 helicopters covering a quarter of a million square kilometers with 3/4 of the population living in major cities and most of the rest clustered around some major towns, where early access to excellent first aid, rapid transport and decent trauma care is the norm and where that care is likely to involve early access to surgical interventions whenever required. It would be interesting to see which areas got the best results in the trial and whether the bulk of the benefit was not seen in places where people suffer delayed response from a disorganised ambulance service and then there is no organised trauma systerm. Below is a chart I have made demonstrating where the patients were recruited from (the data is sourced from the acknowledgements section of the original paper).



So Who is Excited About It?

The early adopters amongst us certainly are. Scott Weingart is all over it. He is even talking about giving aminocaproic acid (a similar substance whcih is more available in the US) to his SAH patients (although why you don't just get on a clip/coil the damn aneurysm I don't know). Minh Le Cong of RFDS Cairns is looking at pre-hospital use.

So Who Isn't?

Well The Alfred Emergency and Trauma Centre here in Victoria are not using it yet for this indication (although I have heard a rumour that it is going to become part of their massive transfusion protocol). Sceptics like Jerry Hoffman aren't too impressed.

What Am I Going to Do?

Well, I'm sceptical. I suspect there is likely to be little harm if we give the drug early, before hypercoagulability become a problem. It is cheap and generic so it is not a huge burden on the system. It is a huge study of really sick patients which is never likely to be repeated. If I am ever working in the third world I will be looking to see where the TXA is kept. If I get a nasty trauma case in Bendigo where immediate surgical attention is not going to be possible (say a bleeding pelvic fracture that our surgeons are reluctant to mess with) then I'll most likely give it before the patient gets loaded on the helicopter. If my organisation decides that it should be part of our massive transfusion regimen for trauma patients then I won't grumble. But to be honest, I think there are higher priorities in trauma care where I am than this.

I also think you should watch out for the fancy, expensive, 1g minijet version of TXA.

And you should really watch out for it on the ACEM exams in the coming year or so!