Delayed Post-operative Bleeding – An Update on Tranexamic Acid

Photo by Rich Paice.

by John Dillberger, DVM, PhD

Reprinted from the March/April 2019 issue of The Claymore.

Using tranexamic acid to treat Delayed Bleeding Syndrome in Deerhounds.

Like their Greyhound cousins, Deerhounds are prone to develop delayed post-operative bleeding. While this is a rare occurrence, it can be life-threatening if not promptly recognized and appropriately treated.Delayed post-operative bleeding occurs when the blood clotting system gets out of balance.  In a healthy dog, the clotting system is balanced between making and dissolving blood clots.  This balance allows a clot to form quickly when needed to seal a damaged blood vessel but insures that any clot which forms in the wrong place is quickly dissolved so that blood circulation is maintained.  When tissue damage occurs from an injury or during surgery, both parts of the clotting system – the half that forms clots in the right place and the half that dissolves them in other places – are stimulated.  Occasionally after surgery, the system gets out of balance and starts dissolving useful clots before blood vessel damage is repaired.  The result is bleeding that begins one to three days after surgery.

At least two drugs are available to prevent or treat delayed post-operative bleeding:  epsilon amino caproic acid, sold under the brand name Amicar®, and tranexamic acid, sold under the brand name Lysteda® in the USA and Cyklokapron® in the USA and Canada.  Both drugs are available in generic form in each country. 

In 2015, I wrote a column about the use of tranexamic acid for post-operative bleeding in Deerhounds.  Since then, new information has emerged about the side effects and proper dosage of tranexamic acid in dogs and humans.  In this month’s column, I update my recommendations for using tranexamic acid.

This article is fairly long.  I have minimized the medical jargon to make it reader-friendly.  However, the article is intended to inform not only Deerhound owners but also veterinarians with Deerhound patients, and so some medical terminology and explanations are necessary. 

If you are a bottom-line sort of person, simply skip to the end for my recommendations about using tranexamic acid in Deerhounds.  And whether or not you have the time or inclination to read the whole article, I suggest you copy it and give it to your regular veterinarian.  In fact, you might make several copies so you can grab one in case you have to take your dog to an emergency clinic. 

How Tranexamic Acid Works

Tranexamic acid acts like a brake on fibrinolysis by binding to plasminogen.  This binding blocks the ability of TPA to activate plasminogen.

Tranexamic acid does not affect the clot-forming system.  Specifically, it does not make clots more likely to form or stronger.

Recent History

In 2010, tranexamic acid gained a lot of attention in human medicine when results were published from a multi-national, clinical trial named the Clinical Randomization of an Antifibrinolytic in Significant Hemorrhage (CRASH- 2) (Goobie 2017).  This trial studied tranexamic acid in adult trauma patients with significant bleeding.  Results were dramatic.  Tranexamic acid reduced the risk of death from bleeding by about a sixth and the risk of death from any cause by about a tenth.  

The next year, the World Health Organization recommended tranexamic acid as an essential medicine for the treatment of acute bleeding in patients with injuries, patients undergoing surgery that utilized cardiopulmonary bypass, and patients who had postpartum bleeding.  In 2013, European guidelines recommended tranexamic acid be given before major surgery to reduce blood loss and the need for blood transfusion.  In 2015, the American Society of Anesthesiologists recommended tranexamic acid as one option for surgical patients with excessive bleeding. 

As you can imagine, all of these events led to more widespread use of tranexamic acid in human medicine.  They also prompted the US Special Operations Command to fund a clinical trial at Cornell and NC State Veterinary Hospitals to see if tranexamic acid could reduce the risk of death from bleeding in dogs that were injured, as it did in humans.  The hope is that the results might advance trauma care for military dogs.  To learn about the trial, I spoke with Principal Investigator Dr. Dan Fletcher.  He told me that it has been difficult to find injured dogs with enhanced fibrinolysis to enroll in the trial.  Apparently, injured dogs tend to keep their clotting system in balance better than people do.  The trial is ongoing.

Side Effects

Every drug has side effects.  Sometimes the side effects are a direct consequence of the drug’s desired activity.  For example, one theoretical side effect of tranexamic acid is that giving too much could damp down fibrinolysis, so that clots which form in the wrong places are not quickly dissolved.

Other side effects have nothing to do with the drug’s desired activity.  This is why drugs are tested extensively in animals (for veterinary drugs) and in animals and humans (for human drugs) before they are allowed on the market – to identify potential side effects.  But even with such testing, new side effects can come to light when a drug becomes widely used for indications other than that for which it was approved.  This is the case with tranexamic acid, where two new side effects have been identified:

With any drug, using the lowest effective dose and proper dose regimen will reduce the chance of side effects.  Unfortunately, for tranexamic acid, this can be difficult for two reasons.  First, there is little information in veterinary or human medicine about the lowest effective dose or best dose regimen.  Second, the lowest effective dose is one that just brings the blood clotting system back into balance, and this will be different for each patient and situation.

Let’s examine the potential side effects of tranexamic acid in detail.

Blood Clots in the Wrong Place

A blood clot that forms in the wrong place is dangerous.  If it forms in a small blood vessel, it can block blood flow.  If it forms in a large vessel, it can break free and be carried to a small blood vessel, blocking it.  When a blood vessel is blocked, the tissue supplied by the blood vessel dies from lack of oxygen. This is one way in which a heart attack occurs:  a clot blocks a blood vessel supplying the heart, and the affected part of the heart muscle dies. 

Theoretically, tranexamic acid could prolong the life of a clot that forms in the wrong place, thereby increasing the risk of a blocked blood vessel.  This is very unlikely to occur in a dog with delayed post-operative bleeding, where fibrinolysis is over-stimulated.  But what about when tranexamic acid is used in a dog with normal fibrinolysis; for example, when the drug is given to prevent, not treat, delayed post-operative bleeding?  Or in a situation where clot formation and fibrinolysis are both over-stimulated, such as disseminated intravascular coagulation? 

In human medicine, clinical trials and retrospective analyses have found no evidence for an increased risk of blood clots in the wrong place with the use of tranexamic acid.  However, many of the trials excluded patients at high risk for this.  A recent review (Goobie, 2017) suggests that tranexamic acid might be contraindicated in human patients at high risk of forming blood clots in the wrong place because of a pre-existing condition.

In the veterinary literature, I can find no reports of blood clots in the wrong place in dogs given tranexamic acid.  But keep in mind that absence of evidence is not evidence of absence, as the old saying goes, and the information in the literature is limited.  The evidence is basically this:

  1. A 2009 study in 32 Beagles done by Xanodyne Pharmaceuticals to support FDA approval of Lysteda® tablets for human use.  Beagles were given tranexamic acid in capsules twice a day for 39 weeks at dose levels of 100 mg/kg (8 dogs), 300 mg/kg (8 dogs), or 600 mg/kg (16 dogs) and then were necropsied.  No evidence of blood clots in the wrong place was found. 
  2. A 1991 study by Moser et al. in 6 mongrel dogs given tranexamic acid by mouth twice a day for 40 days at 110 mg/kg.  Necropsy after the last dose revealed no evidence of blood clots in the wrong place. 
  3. A 2013 review by Kelmer et al. of the medical records of 68 dogs with various bleeding disorders that were given tranexamic acid three or four times a day at an average dose of 9 mg/kg (range 6 to 17 mg/kg) every 6 to 8 hours.  Treatment lasted an average of two days but as long as 21 days.  There was no clinical evidence of blood clots in the wrong place, but the authors do not say if any of the dogs underwent a necropsy examination.


As tranexamic acid became more widely used in surgical patients, physicians noticed that more patients were having seizures, usually within the first hours after surgery.  Retrospective analyses confirmed that this was true.  In cardiac patients, for example, the seizure incidence had increased from 0.5–1.0% to 6.4–7.3% (Lecker et al., 2016).  Subsequent studies have confirmed that tranexamic acid can indeed increase the risk of seizure. 

In retrospect, the increase in seizures with widespread tranexamic acid use isn’t a complete surprise.  When FDA approved Lysteda® tablets for human use in 2009, reviewer Kimberly Hatfield noted that tranexamic acid had been reported in 2002 to cause hyperexcitability and seizures in rats at a dose more than 100 times lower than the dose tested in humans.  But she considered the risk of seizure in patients to be “of low concern” because there were no instances of seizure in patients in clinical trials.

To make a long (and elegant) story short, it turns out that tranexamic acid interferes with communication between brain cells (neurons) that use the amino acid glycine to transmit messages.  Where two such neurons connect, there is a small gap.  To send a message, one cell releases glycine into the gap, which drifts over and attaches to glycine receptors on the other neuron’s surface.  Tranexamic acid also binds to glycine receptors.  In doing so, it blocks the attachment of glycine, cutting communication.  As it happens, neurons that use glycine are important for damping down the spread of electrical impulses in the brain; consequently, any chemical that interferes with the glycine communication system increases the risk of a seizure.  Perhaps the best known such chemical is strychnine.

In cardiac surgery, the usual practice is to infuse tranexamic acid into the blood at a steady rate during surgery and then stop when surgery ends.  When infusion stops, the blood concentration of tranexamic acid rapidly declines.  So why do most seizures occur hours after infusion ends, rather than during infusion, when blood concentration is highest?  For two reasons:

  1. Two of the most commonly used general anesthetic agents (propofol and isoflurane) rapidly and completely reverse the effects of tranexamic acid on glycine receptors.  During surgery, these agents prevent seizures.
  2. When surgery ends, the patient stops receiving anesthesia and tranexamic acid.  The anesthetic agents are quickly cleared from the blood and brain, and the patients wakes up.  Tranexamic acid is also quickly cleared from the blood but not from the brain; instead, the brain concentration of tranexamic acid remains steady or even increases for several hours after infusion ends.  With the anesthetic agents gone, there is nothing to counteract the seizure-promoting effects of tranexamic acid.

Given that tranexamic acid can cause seizures in laboratory animals and humans, there is every reason to think it also could do so in a dog in certain circumstances.  Certainly, the mechanism by which tranexamic acid causes seizures in humans – interfering with communication between brain cells that use glycine – would be equally vulnerable in dogs. 

In fact, there is a recent report (Orito et al., 2017) of a seizure in a dog given tranexamic acid.  It happened during a clinical trial to investigate the use of tranexamic acid to induce vomiting in dogs suspected of having recently swallowed something poisonous.  The investigators found that an intravenous dose of tranexamic acid at 50 mg/kg caused vomiting in 129 of 137 (94.2%) dogs. 

One dog in the trial, an 11-year-old Bichon Frise that had swallowed an unknown amount of cocoa powder, had a convulsion after injection of tranexamic acid.  The role played by tranexamic acid is impossible to pin down because the main ingredient of cocoa powder (theobromine) can itself cause seizures, and stress alone can cause seizures in dogs with an anxious or nervous temperament.  Nevertheless, I agree with the authors that “further studies are warranted to investigate the proconvulsant effect of tranexamic acid in dogs.”

An unanswered question in humans and dogs is whether tranexamic acid might pose a greater risk to an individual with a predisposition to seizure; for example, in a patient with a history of seizure, epilepsy that is poorly controlled with medication, or a recent head injury.  The report by Orito et al. of a convulsion in a Bichon Frise dog raises this possibility.  In her 2017 review, Dr. Goobie suggests that tranexamic acid might be contraindicated in human patients with “known uncontrolled seizure disorder.”


Two studies have explored using tranexamic acid to induce vomiting/regurgitation in a dog suspected of swallowing poison.  It works:

  1. In 2014, Kakiuchi et al. tested tranexamic acid injected intravenously as a single dose at 20, 30, 40, or 50 mg/kg or as three doses, 5 minutes apart, at 10, 20, and then 30 mg/kg.  All doses caused vomiting.  The higher the dose level, the sooner vomiting occurred.  Vomiting ended within 4 minutes postdose.
  2. In 2017, Orito et al. tested tranexamic acid given as a single intravenous injection at 50  mg/kg and found it to be very effective at inducing vomiting. 

While vomiting is desirable in a dog that has swallowed poison, it is not desirable in a dog with delayed post-operative bleeding, where vomiting puts a strain on surgical incisions that are not yet healed.  Moreover, vomiting always carries with it the risk of accidental aspiration of stomach contents into the lungs and subsequent pneumonia.

In 2013, Kelmer et al. published a retrospective study of the use of tranexamic acid to control bleeding in dogs.  The average tranexamic acid dose level was 9 mg/kg (range 6 to 17 mg/kg) every 6 to 8 hours.  The only adverse effect of tranexamic acid was vomiting immediately after injection in two dogs that were dosed at 11 and 16.5 mg/kg.

In 2015, the same group reported the results of a prospective study to evaluate the “clot-protecting” activity of tranexamic acid in healthy dogs.  The first dog vomited immediately after an intravenous injection of the drug at 20 mg/kg.  The second dog vomited immediately after injection at 15 mg/kg.  For subsequent dogs, the dose level was reduced to 10 mg/kg and injected more slowly, and vomiting did not occur.  

Tranexamic Acid in Deerhounds

The right way to use tranexamic acid would be to:

  1. Give it only to a Deerhound at risk for, or experiencing, delayed post-operative bleeding or with other evidence of over-stimulated fibrinolysis
  2. Use the lowest dose that restores balance between clot formation and fibrinolysis, and
  3. Treat only as long as necessary. 

While this is easy to write, it can be very difficult to execute for a given dog.  I will discuss each of these points in turn.

Identifying Deerhounds at Risk

Although it is easy to tell if a Deerhound is experiencing delayed post-operative bleeding, there presently is no way to tell in advance if a Deerhound is likely to develop this problem.  But that may soon change. 

Dr. Michael Court at Washington State University has identified a genetic variant associated with delayed post-operative bleeding in Greyhounds.  The SDCA is collaborating with Dr. Court and helping fund his research. 

As reported in the last issue of The Claymore, Dr. Court has found this same genetic variant in Deerhounds that experienced delayed post-operative bleeding.  This gives us the hope of a future genetic test to identify Deerhounds at risk of this problem.  Until then, however, an owner whose Deerhound undergoes surgery must decide whether to give the dog tranexamic acid to reduce the chance of delayed post-operative bleeding and, if so, then at what dose level and for how long.

Tranexamic Acid Dose in Deerhounds

Figuring out the lowest dose of tranexamic acid that will restore balance to the blood clotting system is not easy.  The goal is to give enough tranexamic acid to slow fibrinolysis but not completely inhibit it.  Clinical practice in humans suggests that inhibiting fibrinolysis by about 80% is sufficient, which requires a tranexamic acid plasma concentration of about 10 micrograms/mL (Nilsson 1980). 

In a test tube, it takes 10 times more tranexamic acid to inhibit fibrinolysis in dog plasma (about 100 micrograms/mL) than in human plasma (Fletcher at al., 1994), presumably because the normal plasma concentration of TPA (the protein that stimulates fibrinolysis) is much greater in dogs than humans (Couto and Odunayo, 2018).  But this does not mean that the effective dose of tranexamic acid is 10 times greater in dogs than humans.  Why? 

For both dogs and humans, the baseline clot-forming and clot-dissolving activities are in balance.  In other words, the clot-forming activity also is much greater in dogs than humans.  In both species, it takes a similar shift toward fibrinolysis to cause problems and a similar dose of tranexamic acid to correct the shift and restore balance.

As an analogy, think of two balances like the ones depicted in statues of the Goddess of Justice.  One balance represents a human’s clotting system; the other, a dog’s clotting system.  For each balance, one side is the clot-forming system, and the other side is the clot-dissolving system.  The human balance has a 0ne-pound weight on each side, and the dog balance has a 10-pound weight on each side.  If you add the same amount of weight to the clot-dissolving side of each balance, you will throw each one out of whack.  To restore balance, you need to remove the same amount of weight from the heavy side of each balance – not 10 times more weight from dog balance than the human balance.

When given intravenously, the recommended dose of tranexamic acid for humans is 10 mg/kg three or four times a day.  This dose is effective in dogs, too.  A single intravenous dose at 10 mg/kg protects clots in dogs for at least 6 hours (Osekavage et al., 2018), and a single intravenous dose at 50 mg/kg protects clots for at least 3 hours (Kakiuchi et al., 2014).  At 50 mg/kg, clot protection is greatest at 20 minutes postdose, less at 3 hours postdose, and gone by 24 hours postdose.  In a 2016 review of tranexamic use in Greyhounds, the authors (Yap and Aertsens) recommend giving tranexamic acid intravenously at 7 to 10 mg/kg to prevent delayed post-operative bleeding. 

When given by mouth, the recommended dose of tranexamic acid for humans is 25 mg/kg three or four times a day.  In dogs, an oral dose of tranexamic acid at 15 mg/kg protects clots for at least 6 hours (Osekavage et al., 2018).  Maximum protection occurs later and lasts longer when tranexamic acid is given by mouth than when it is given intravenously, presumably because it takes some time for an oral dose to be absorbed into the body.  In their 2016 review, Yap and Aertsens recommend giving tranexamic acid by mouth at only 5 to 10 mg/kg to prevent delayed post-operative bleeding in Greyhounds. 

When tranexamic acid is given by mouth, dogs absorb it better than humans; specifically, dogs absorb the entire dose (Osekavage et al., 2018), but humans absorb only a third of the dose (Pilbrant et al., 1981).  Therefore, to correct a similar degree of over-stimulation in fibrinolysis, the effective oral dose of tranexamic acid would be less in a dog than a human.  But because it is impossible to gauge the degree to which fibrinolysis is over-stimulated in any given situation, this species difference in absorption can be ignored for practical purposes.


As a breed, Deerhounds are prone to develop delayed post-operative bleeding.  This occurs infrequently, but can be life-threatening when it does. 

One drug that can be used to prevent or treat this problem is tranexamic acid, which is sold under the brand name Lysteda® in the USA and Cyklokapron® in the USA and Canada and also is available in generic form in both countries.  It is available in injectable form and in tablet form.

Based on the information available at present, my recommendations for using tranexamic acid in Deerhounds to treat or prevent delayed post-operative bleeding are just below.  You may want to copy this page, or the entire article, and give it to your regular veterinarian. 

Recommendations for Tranexamic Acid in Deerhounds

To TREAT delayed post-operative bleeding, give tranexamic acid intravenously (IV) or by mouth at 10 mg/kg three or four times a day until bleeding has been controlled and for three days thereafter. 

To PREVENT delayed post-operative bleeding, give tranexamic acid IV or by mouth at 5 to 10 mg/kg two or three times a day for three to five days after surgery, with the first dose given within 12 hours after surgery ends. 

If only the injectable product is available, then this can be given either IV or by mouth. 

Inject tranexamic acid slowly to reduce the chance of vomiting.

If at all possible, tranexamic acid should be used only in a Deerhound that is under the care of a veterinarian.

Besides being useful for delayed post-operative bleeding, tranexamic acid also may be beneficial in other situations where the clotting system balance has (or is suspected to have) shifted too far in favor of fibrinolysis, such as disseminated intravascular coagulopathy with enhanced fibrinolysis secondary to traumatic injury, heat stroke, etc. (Gando et al., 2013; Asakura 2014).


Asakura H (2014).  Classifying types of disseminated intravascular coagulation:  Clinical and animal models.  Journal of Intensive Care 2: 20-26.

Couto J, Odunayo A (2018).  Epsilon aminocaproic acid: A new tool for treating bleeding dogs.  At

Fletcher DJ, Blackstock KJ, Epstein K, Brainard BM (1994).  Evaluation of tranexamic acid and ε‑aminocaproic acid concentrations required to inhibit fibrinolysis in plasma of dogs and humans.  Am J Vet Res 75(8): 731-738.

Gando S, Wada H, Thachil J (2013).  Differentiating disseminated intravascular coagulation (DIC) with the fibrinolytic phenotype from coagulopathy of trauma and acute coagulopathy of trauma-shock.  Journal of Thrombosis and Haemostasis 11: 826–835.

Goobie SM (2017).  Tranexamic acid:  Still far to go.  British Journal of Anaesthesia 118 (3): 293–295.

Kakiuchi H, Kawarai-Shimamura A, Fujii Y, Aoki T, Yoshiike M, Arai H, Nakamura A, Orito K (2014).  Efficacy and safety of tranexamic acid as an emetic in dogs.  Am J Vet Res 75(12): 1099-1103.

Kelmer E, Marer K, Bruchim Y, Klainbart S, Aroch I, Segev G (2013).  Retrospective evaluation of the safety and efficacy of tranexamic acid (HexakapronR) for the treatment of bleeding disorders in dogs.  Israel Journal of Veterinary Medicine 68 (2): 94-100.

Kelmer E, Segev G, Papshvilli V, Rahimi-Levene N, Bruchim Y, Aroch I, Klainbart S (2015).  Effects of intravenous administration of tranexamic acid on hematological, hemostatic, and thromboelastographic analytes in healthy adult dogs.  J Vet Emerg Crit Care (San Antonio) 25(4): 495‑501.

Lecker I, Wang D-S, Whissell PD, Avaramescu S, Mazer D, Orser BA (2016).  Tranexamic acid–associated seizures:  Causes and treatment.  Ann Neurol 79: 18–26.

Moser KM, Cantor JP, Olman M, Villespin I, Graif JL, Konopka R, Marsh JJ, Pedersen C (1991).  Chronic pulmonary thromboembolism in dogs treated with tranexamic acid.  Circulation 83: 1371-1379.

Nilsson IM (1980).  Clinical pharmacology of aminocaproic and tranexamic acids.  J Clin Pathol 33, Suppl (Roy Coll Path), 14: 41-47.

Orito K, Kawarai-Shimamura A, Ogawa A, Nakamura A (2017).  Safety and efficacy of intravenous administration for tranexamic acid-induced emesis in dogs with accidental ingestion of foreign substances.  J Vet Med Sci 79(12): 1978–1982.

Osekavage KE, Brainard BM, Lane SL, Almoslem M, Arnold RD, Koenig A (2018).  Pharmacokinetics of tranexamic acid in healthy dogs and assessment of its antifibrinolytic properties in canine blood.  Am J Vet Res 79(10): 1057-1063.

Pilbrant A, Schannong M, Vessman J (1981).  Pharmacokinetics and bioavailability of tranexamic acid.  Eur J Clin Pharmacol 20(1): 65-72.

Yap F, Aertsens A (2016).  Delayed post-operative bleeding in greyhounds.  At