REVIEW

Tranexamic Acid–Associated Seizures:

Causes and Treatment
1 1 2
Irene Lecker, PhD, Dian-Shi Wang, PhD, MD, Paul D. Whissell, PhD, Sinziana
3,4 1,3,4 1,3,5
Avramescu, PhD, MD, C. David Mazer, MD, and Beverley A. Orser, PhD, MD

Antifibrinolytic drugs are routinely used worldwide to reduce the bleeding that results from a wide range of hemor-rhagic
conditions. The most commonly used antifibrinolytic drug, tranexamic acid, is associated with an increased incidence of
postoperative seizures. The reported increase in the frequency of seizures is alarming, as these events are associated
with adverse neurological outcomes, longer hospital stays, and increased in-hospital mortality. How-ever, many
clinicians are unaware that tranexamic acid causes seizures. The goal of this review is to summarize the incidence, risk
factors, and clinical features of these seizures. This review also highlights several clinical and preclinical studies that
offer mechanistic insights into the potential causes of and treatments for tranexamic acid–associated seizures. This
review will aid the medical community by increasing awareness about tranexamic acid–associated seiz-ures and by
translating scientific findings into therapeutic interventions for patients.
ANN NEUROL 2016;79:18–26

A ntifibrinolytic drugs are used worldwide to decrease the

requirement for blood transfusions, reduce the risk of
eral case reports indicate that TXA-associated seizures also
occur in nonsurgical patients.
10,22,27
Seizures in post-
reoperation for bleeding, and lower mortality asso- operative cardiac surgery patients have been reported to
1–3 be associated with a 2-fold increase in hospital length of
ciated with hemorrhage following major trauma. The
26
most commonly used antifibrinolytic drugs include tra- stay and a 2.5-fold higher mortality rate. An increase
nexamic acid (TXA), E-aminocaproic acid (EACA), and in the incidence of delirium and stroke, and a reduced
1 35
aprotinin. TXA and EACA are synthetic derivatives of quality of life have also been reported.
the amino acid lysine that exert their hemostatic effects The goal of this review is to increase awareness about
4–6
by binding to plasminogen. This binding prevents the seizures associated with antifibrinolytic drugs and provide
conversion of plasminogen to plasmin and reduces the mechanistic-based prevention and treatment rec-
4–6
degradation of fibrin-containing blood clots. ommendations. The review focuses on TXA, the most
Aprotinin, conversely, is a serine protease inhibitor that commonly used and widely studied antifibrinolytic drug.
7
binds directly to plasmin and inhibits its function. First, the incidence, risk factors, and clinical features of
Antifibrinolytic agents are considered to be safe and TXA-associated seizures are summarized. Next, preclini-
affordable drugs with few serious adverse effects. 1 How- cal and clinical studies that offer insights into the under-
ever, observational clinical trials and case reports have lying causes of seizures are reviewed. In particular, a study
shown that TXA, and to a lesser extent EACA, but not that measured the concentration of TXA in the cerebral
aprotinin, are associated with seizures.8–34 Most TXA- spinal fluid (CSF) of patients undergoing major
associated seizures occur in patients who have undergone cardiovascular surgery is considered. The study then
cardiac procedures.16–18,20,21,24–26,30–32,34 However, sev- compared TXA concentrations in the CSF to TXA

View this article online at wileyonlinelibrary.com. DOI: 10.1002/ana.24558

Received Jul 13, 2015, and in revised form Nov 3, 2015. Accepted for publication Nov 10, 2015.

Address correspondence to Dr Orser, Department of Physiology, University of Toronto, Room 3318, Medical Sciences Building, 1 King’s College Circle,
Toronto, Ontario, Canada, M5S1A8. E-mail: [email protected]

From the Departments of 1Physiology, 2Psychology, and 3Anesthesia, University of Toronto; 4Department of Anesthesia, Keenan Research Centre for
Biomedical Science and Li Ka Shing Knowledge Institute, St Michael’s Hospital; and 5Department of Anesthesia, Sunnybrook Health Sciences Centre,
Toronto, Ontario, Canada

18 VC 2015 The Authors Annals of Neurology published by Wiley Periodicals, Inc. on behalf of American Neurological Association
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits
use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications
concentrations that modulate the activity of neurotrans-
mitter receptors in the brain in vitro. Based on these
findings, treatment strategies to mitigate TXA- Lecker et al: TXA-Associated Seizures
associated seizures in patients are proposed.
patients undergoing “open chamber surgery” (eg, aortic
Clinical Indications, Incidence, and Risk 16–18,20
Factors valve replacement). The risk is also increased in
patients with deep hypothermic circulatory arrest, long
TXA was originally approved by the US Food and Drug cardiopulmonary bypass time, or prolonged aortic cross-
Administration for the treatment of patients with hemo-
clamp time.16,18,20,26
philia undergoing dental surgery and for women suffer-
36,37 Several case reports indicate that seizures are not
ing from heavy menstrual bleeding. The clinical restricted to cardiac surgery patients. For example, a patient
indications of TXA have rapidly expanded and now include with chronic kidney failure who was treated with TXA
multiple “off-label” uses, including cardiac, gas- 10
experienced a convulsive seizure. Another patient who
trointestinal, and orthopedic surgery as well as treatment of
38–41 underwent a craniotomy for meningioma had tonic–clonic
postpartum hemorrhage. The World Health 22
convulsions after the administration of TXA. A third
Organization (WHO) recently included TXA in its “Model
42 patient who was admitted for hemoptysis had a focal
List of Essential Medicines.” The WHO rec-ommended seizure after TXA treatment, which pro-gressed to a
that TXA be used to reduce blood loss in patients 27
generalized seizure. None of these patients had a history
undergoing cardiopulmonary bypass procedures, in trauma
of seizure disorders and no abnormalities
patients with significant hemorrhage, and in patients with
42 were detected on subsequent electroencephalography
postpartum hemorrhage. 10,22,27
(EEG) or computed tomographic scans. Collec-
The broad introduction of TXA into surgical care has
tively, these case studies indicate that a wide range of
resulted in an increased reported incidence of seiz-ures,
particularly during the early postoperative period after
patients may be vulnerable to TXA-associated seizures.
21 Increasing global “off-label” use of TXA may further
cardiac surgery. Retrospective analyses have shown that
increase the incidence of TXA-associated seizures.
the incidence of seizures in postoperative cardiac patients
has increased from 0.5–1.0% to 6.4–7.3% with the use of Clinical Features and Diagnosis
8,17
higher doses of TXA. Additionally, several multicenter
A clear understanding of the clinical features of TXA-
retrospective studies confirm increased seiz-
associated seizures will aid in their diagnosis. Reports of
ures in postoperative patients who received TXA, with an patients who received an accidental intrathecal injection
16,18,20,26
incidence ranging from 0.9% to 2.5%. A single of TXA have offered rare insights into the clinical mani-
prospective trial found that seizures occurred in 3% (3 of 11,13,15,19,23,29
festation of TXA-associated seizures. These
100 patients) of post–cardiac surgical patients treated with
34
patients experienced severe back pain that radiated below
TXA. Although the incidence of TXA-associated seizures the waist, with burning pain in the lower limbs and glu-teal
after cardiac surgery varies between studies, treat- 11,13,15,19,23,29
region. Involuntary motor activity, such
ment with TXA was a strong independent predictor of as a “jerking” of the lower extremities (referred to as
16,20,26
seizure. myoclonic movements) and twitching of facial muscles,
Retrospective studies have identified several risk fac- 11–13,15,19,23
was also observed. These abnormal move-
tors for TXA-associated seizures. These include higher
ments rapidly progressed to generalized tonic–clonic seiz-
doses of TXA, such as those recommended in the BART 11–13,15,19,23
ures. Myoclonic movements may serve as a
study (Blood conservation using Antifibrinolytics in a warning sign of impending seizures.
16,20,26,43
Randomized Trial; 80–100mg/kg total dose).
In postoperative cardiac surgery patients, TXA-
Female gender, increased age, and poor overall health also associated seizures are typically generalized tonic–clonic
16,17,20,26
predispose patients to seizures. Seizures are 20,24,26
events, although focal and mixed seizures also occur.
observed more frequently in patients older than 70 years,
Approximately 20% of these seizure patients
and those with a high disease severity score, as measured 24
experience myoclonic activity. Seizures usually occur
by an APACHE II index (Acute Physiology and Chronic within the first 5 to 8 hours after surgery, during the
16,20
Health Evaluation II) > 20. Patients with renal dys- period of weaning from intravenous sedation in the
function or prior neurological and cardiovascular disor- intensive care unit.
17,20,24,26
Seizure events typically per-
16,17,20,26
ders are also at increased risk. Other important sist for a few minutes
17,20
and do not progress into status
risk factors for TXA-associated seizures include the type 34
epilepticus. About 30 to 60% of patients have recur-
and duration of surgery. Most seizures are reported in rent episodes during the first 24 to 48 hours after
surgery.16,17,20
The diagnosis of TXA-associated seizures may be
facilitated by EEG monitoring in the early postoperative
period. EEG monitoring could help distinguish between

January 2016 19
ANNALS of Neurology

shivering, myoclonic movements, and seizures, and

17,18,26
thereby prevent a misdiagnosis. EEG monitoring
may also detect subclinical seizures that are not apparent by
observing sedated patients. Continuous EEG monitor-ing
following cardiac surgery identified 1 patient with EEG
evidence of seizure who exhibited no convulsive
34
behaviour. Finally, EEG monitoring may be particularly
useful for the diagnosis of TXA-associated seizures in
patients cotreated with a neuromuscular blocker agent such
as rocuronium. These drugs inhibit motor activity and mask
the behavioral correlates of network hyperexcit-ability. In
the absence of EEG monitoring, the incidence of TXA-
associated seizures may be underestimated.

TXA Concentrations in the Central Nervous

System of Patients
The proconvulsant properties of TXA likely result from
direct effects on the central nervous system (CNS), as
application of TXA to the cortex or injection into the
cisterna magna in experimental animals causes general-ized
44–46
seizures. In an effort to identify the mechanism
underlying TXA-associated seizures, the concentration of
TXA in the CNS of patients was measured. One study took FIGURE 1: Tranexamic acid (TXA) concentrations measured in
advantage of a unique clinical scenario where CSF was the cerebral spinal fluid (CSF) and serum of patients cause
47
hyperexcitability in vitro. (A) The time course of TXA levels in
intermittently sampled during surgery. Specifically, an the CSF and serum of 1 patient who experienced a seizure is
indwelling catheter was inserted into the lumbar intra- shown on the left. The decline of TXA levels in the brain lags
behind that in the blood. The timeline at the bottom of each
thecal space of patients undergoing repair of thoracoab-
figure indicates key surgical events during cardiopulmonary
dominal aneurysms to allow the CSF to be intermittently bypass (CPB). The red arrow highlights the concentrations
drained. The purpose of this procedure is to prevent spi-nal when TXA administration was terminated. On the right are the
cord ischemia by decreasing the volume of CSF and summarized data of TXA concentra-tions in the CSF and serum
48 during key surgical events (n 54). TXA levels in the serum
reducing intrathecal pressure. (2mM) are 10-fold higher than those in the CSF (200 mM). (B)
Measurements of TXA levels in the CSF from these Clinically relevant con-centration of TXA (200 mM) causes
hyperexcitability by increasing the frequency of seizure-like
patients produced unexpected results. After infusion of the
events in neocortical slices. *P < 0.05.
drug was discontinued, the concentration of TXA in the
CSF failed to decline and in some cases continued to
increase, reaching peak concentration of about 200 mM Molecular Mechanism of TXA-Associated
47
(Fig 1A). In contrast, TXA levels in the serum peaked Seizures
following cardiopulmonary bypass, then rapidly declined Studies of animal models have offered insights into the
47
after the drug infusion was terminated. The peak serum molecular mechanisms underlying TXA-associated seiz-
concentration of TXA (2 mM) was about 10 times higher ures. Application of a clinically relevant concentration of
TXA (200 lM) to slices of neocortex markedly increased
than the concentration of TXA in the CSF (200 mM).
47
Notably, 1 patient with a high TXA concen-tration in the field responses to excitatory stimuli. TXA also increased
CSF experienced postoperative seizures. The time course the frequency of spontaneous epileptiform field potentials
47
of TXA concentrations in the CSF and serum from this or “seizurelike events” (see Fig 1B). Another study
patient is illustrated in the top half of Figure 1A. The showed that application of TXA (1mM) to mouse amyg-
average concentrations from the CSF and serum of 4 49
dala slices caused widespread neuronal depolarization.
patients are shown in the bottom half of Fig-ure 1A. These Collectively, these studies show that TXA directly
results suggest that seizures could arise due to persistently increases the excitability of neuronal networks. Increasing
high concentrations of TXA in the brain during the early evidence suggests that this hyperexcitability produced by
postoperative period. TXA results from reduced inhibitory neurotransmission

20 Volume 79, No. 1

 Lecker et al: TXA-Associated Seizures

FIGURE 2: Tranexamic acid (TXA) is a competitive antagonist of glycine (Gly) receptors. (A) Glycine and TXA are structural ana-
logues, suggesting that TXA competes with glycine at the agonist binding site of glycine receptors. (B) TXA (1mM) inhibits gly-cine (100
mM)-activated currents in cortical neurons. The concentration–response plots for glycine current recorded in the absence and presence
of TXA are shown. The results indicate that TXA is a competitive antagonist of glycine receptors.

47,49
or “disinhibition.” c-Aminobutyric acid type A (GABAA) inhibitory currents (IC50 5 1mM) was similar. Thus,
receptors and glycine receptors are major mediators of although inhibition of GABAA receptors may contribute
50,51
inhibition in the CNS. These transmitter-gated to TXA-associated seizures, higher affinity target recep-
anion channels, which are well-known targets for a vari- tors are likely to exist in the CNS.
ety of proconvulsant and anticonvulsant agents, are Our research group searched for novel receptors that
52–56
plau-sible targets for TXA. are sensitive to clinically relevant concentrations of TXA.
The effects of TXA on GABA A receptors were Because TXA is a structural analogue of glycine, we
57 hypothesized that TXA competitively inhibits glycine
examined first by Furtmuller and colleagues. They
showed that TXA is a competitive antagonist of GABA A receptors (Fig 2A), and this action contributes to seizures.
In support of this hypothesis, glycine receptor antagonists,
receptors and that it inhibits recombinant GABA A recep-
tors (a1b2c2) with a half-maximal inhibitory concentra-tion such as strychnine, cause myoclonic movements and
57
(IC50) of 7mM. Other investigators showed that TXA twitching, particularly in the lower limbs, as well as mus-cle

inhibits native GABAA receptors in cortical and spi-nal spasms and convulsions.55,56,63,64 Interestingly, the pat-
47 tern of twitching, myoclonus, and seizures observed in
cord neurons (IC50 5 1.5 and 1mM, respectively).
Collectively, these results demonstrated that TXA inhibits patients treated with TXA is similar to the pattern of the
11,22,27
GABAA receptors, but only at concentrations that are proconvulsant effects of strychnine. We found that
higher than the concentration detected in the CSF of TXA acts as a competitive antagonist of glycine receptors,
47
patients (200 lM). GABAA receptors generate 2 distinct with an IC50 of 1mM (see Fig 2B). Similar to GABAA
forms of inhibition, synaptic and tonic, which could exhibit receptors, glycine receptors generate both synaptic currents
58–61 61
different sensitivities to TXA. Synaptic cur-rents are and tonic inhibitory currents (Fig 3A). Thus, we com-
fast, transient events that are activated by near-saturating pared the potency of TXA for inhibition of spontaneous
62 miniature inhibitory postsynaptic currents and a tonic cur-
concentrations of agonist. In contrast, tonic currents are
62 47
generated by low, ambient concentrations of transmitter. rent in spinal cord neurons (see Fig 3C). Tonic glycine
Synaptic and tonic currents are mediated by different current was found to be 10-fold more sensitive to TXA
receptor subtypes that often exhibit different 47
(IC50 5 90 mM) than synaptic currents. Therefore, the
60
pharmacological properties. Surprisingly, the potency of potency of TXA is greatest for glycine receptors that gen-
TXA for synaptic currents (IC50 50.8mM) and tonic erate a tonic inhibitory current (see Fig 3B).

January 2016 21
ANNALS of Neurology

FIGURE 3: Tonic glycine current is highly sensitive to tranexamic acid (TXA) inhibition. (A) Inhibitory receptors are expressed in
synaptic and extrasynaptic regions of the neuron. These receptors are composed of different subunits and have distinct phar-
macological properties. Extrasynaptic receptors mediate a tonic inhibitory conductance. (B) Summary table of the half-maximal
inhibitory concentration (IC50) values for TXA inhibition of synaptic and tonic currents mediated by glycine and c-aminobutyric acid type
A (GABA) receptors. (C) TXA (1mM) inhibits synaptic and tonic glycine currents in a similar manner as the competitive glycine
antagonist, strychnine. Synaptic currents were studied by recording miniature inhibitory postsynaptic currents. Tonic currents were
evoked by applying a low concentration of glycine (10 mM), similar to the ambient concentration present in the extracellular fluid, to the
bath solution. SEM 5standard error of the mean.

47
Finally, others have studied the effects of TXA on the rent. In addition, field recordings from slices of mouse
49,57
activity of excitatory amino acid receptors. Binding cortex showed that isoflurane (250 mM) and propofol (1
assays and electrophysiological studies show that TXA mM) completely reversed the hyperexcitability produced
57 47
(5mM) does not directly inhibit the N-methyl-D-aspartate by TXA. Therefore, isoflurane and propofol, as well as
or a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid other anesthetics that increase glycine receptor function,
49 might be effective either for treating or for preventing
subtypes of glutamate receptors. Thus, the proconvulsant
properties of TXA are likely mediated by disinhibition of TXA-associated seizures (Fig 4).
tonic glycine current.
Prevention and Treatment of
Anesthetics Reverse TXA Inhibition of TXA-Associated Seizures
Glycine Receptors Currently, there are no recommended treatments for TXA-
Because the tonic current generated by glycine receptors is associated seizures. Given the low incidence and variable
highly sensitive to TXA inhibition, drugs that reverse this clinical manifestations of TXA-associated seiz-ures,
inhibitory effect may mitigate TXA-associated hyperexcit- randomized controlled clinical trials that compare the
ability. There are no commonly used selective glycine efficacy of various anticonvulsants treatments are likely to
receptor agonists that can be administered intravenously to be impractical. Nevertheless, results from animal studies
patients in the intensive care unit. However, several general have shown that TXA inhibition of tonic glycine current is
anesthetics, including the inhalational agents iso-flurane, rapidly and completely reversed by the general anesthetics
sevoflurane, and desflurane and the intravenous anesthetic 47
isoflurane or propofol. These results suggest that general
propofol, act as positive allosteric modulators of glycine anesthetics may be useful to consider for the first-line
65,66 treatment for TXA-associated seizures in patients. For
receptors. Whole-cell recordings of currents in spinal
cord neurons showed that clinically relevant con- example, TXA-associated seizures could be prevented by
centrations of isoflurane (150 and 250 mM) and propofol (3 simply prolonging the delivery of anes-thetics during the
mM) fully reversed TXA inhibition of tonic glycine cur- early postoperative period. Notably,

22 Volume 79, No. 1

 Lecker et al: TXA-Associated Seizures

Finally, reducing the dose of TXA during surgery

may be the simplest and most practical strategy to pre-vent
TXA-associated seizures. TXA is a competitive antagonist
of glycine and GABAA receptors. Thus, a lower dose of
TXA is less likely to cause seizures, as lower
concentrations would be “outcompeted” by endog-enous
neurotransmitters at the agonist-binding sites of glycine and
GABAA receptors. The notion that higher
FIGURE 4: The molecular mechanism underlying tranexamic
acid (TXA)-associated seizures and the reversal of TXA- doses of TXA increase the risk of seizures is supported
45 16,20,26
mediated inhibition by anesthetics. TXA binds to the glycine by animal and human studies. Specifically, in
receptors, resulting in a decrease in inhibitory current. This cardiac surgery patients, the use of higher doses of TXA
reduction in anion conduction increases excitability, which gives 16,20,26
rise to seizures. Anesthetics reverse the effect of TXA by drastically increased the incidence of seizures.
increasing glycine receptor function and thereby prevent or Lowered TXA dosing should also be considered for
reverse TXA-induced seizures. patients with renal dysfunction, as renal excretion is the
major route of TXA elimination. Case reports indicate that
patients experience seizures most often in the first few patients treated with TXA while undergoing dialysis
hours after admission to the intensive care unit. At this experience generalized seizures and myoclonic move-
47 10,70
time, TXA levels are either peaking or declining slowly. ments. Interestingly, TXA administered to cardiac
In contrast, anesthetic levels are declining rapidly in the surgery patients at doses recommended in the BART Trial
67
CNS, as drug delivery has been terminated. Thus, the resulted in higher than expected plasma concentra-
anesthetic is no longer available to provide anticonvulsant tions, which exceeded the recommended therapeutic lev-
effects. Consistent with this notion that anesthetics pro-tect 71,72
els. Consistent with these findings, lowering TXA
against TXA-induced seizures, many patients first 20
doses reduced the frequency of postoperative seizures.
develop seizures during emergence from propofol seda- Therefore, decreasing the dose of TXA is likely the sim-
17,24
tion. Also, case reports indicate that propofol is plest and most effective strategy to reduce the incidence
effective for treating seizures in patients who inadver- and/or severity of postoperative seizures. However, the
15,23
tently received an intrathecal injection of TXA. benefits of reducing TXA dose need to be balanced
Although treatment with a general anesthetic is likely to against the possibility of reducing the drug’s antifibrino-
be effective, these drugs should only be administered lytic effects.
under conditions that allow their safe use. This review
does not provide specific recommendations regarding Summary and Outstanding Questions
doses of anesthetics, as treatment of patients during the In summary, TXA-associated seizures occur most fre-
early postoperative period in the intensive care unit is quently during the early postoperative period after car-diac
highly complex. Treatment must be guided by the judg- surgery but also occur in patients undergoing noncardiac
ment and skill of the care providers. surgery and other medical treatments. To reduce the risk of
If the use of propofol or other anesthetics is deemed seizures, the lowest effective TXA dose should be
to be unsafe or if these drugs are unavailable, alternative considered and dosing should be adjusted for clinical
therapies can be considered. A second-line treatment for conditions such as renal dysfunction. A high index of
TXA-associated seizures includes com-pounds that suspicion is required to detect seizures, and EEG
increase GABAA receptor activity, which may compensate monitoring may be considered for patients who experience
for a reduction in glycinergic inhibition. Benzodiazepines myoclonic movements or twitching or show evidence of
(lorazepam, midazolam, diazepam, and clonazepam), focal seizures. Based on results from preclini-cal studies,
which do not modify glycine receptors but rather general anesthetics including propofol and isoflurane may
68,69
upregulate GABAA receptor function, have be considered as the first line for preven-tion and/or
been used to treat seizures following inadvertent intrathe- treatment. In high-risk patients, terminating the TXA
27–29 16,26
cal injection of TXA or after cardiac surgery. infusion early and/or prolonging the adminis-tration of
Lorazepam may be considered for the treatment of seiz- anesthetics may prevent seizures.
ures, rather than other benzodiazepines that have shorter Although progress has been made in our under-
duration of action. standing of the causes underlying TXA-associated seiz-
ures, many questions remain unanswered. First, it is
uncertain why cardiac surgery patients are more vulnera-
ble to TXA-associated seizures. One potential factor is

January 2016 23
ANNALS of Neurology
the high doses of TXA administered during cardiac sur-
73
gery. Also, cardiac surgery can cause intensive systemic
inflammation that increases the permeability of the blood– Author Contributions
74
brain barrier. A jeopardized blood–brain barrier could
C.D.M. proposed the concept for the review and con-
facilitate the entry of TXA into the CNS. Second, it is
tributed to the writing. I.L. wrote the review and created
important to understand the mechanism by which TXA
the figures. D.-S.W., P.D.W, S.A., and B.A.O. helped
gains entry into the CNS, as such knowledge could aid in
structure and edit the review. All authors read and
the development of neuroprotective strategies that reduce
TXA penetration. Third, it is of interest to know whether approved the final manuscript.
TXA dosing should be reduced or avoided in patients with
a previous history of a seizure disorder or those with
clinical conditions, such as traumatic brain injury, that Potential Conflicts of Interest
damage the blood–brain barrier and predis-pose to seizures. Nothing to report.
Also, antibiotics such as penicillins and cephalosporins
inhibit GABAA receptors and it is unknown whether these
drugs exacerbate the proconvul-sant properties of TXA.
References
Finally, future studies are needed to determine 1. Henry DA, Carless PA, Moxey AJ, et al. Anti-fibrinolytic use for
minimising perioperative allogeneic blood transfusion. Cochrane
whether antifibrinolytic drugs other than TXA also cause
Database Syst Rev 2007;(3):CD001886.
seizures. Interestingly, EACA is a structural analogue of
2. Dunn CJ, Goa KL. Tranexamic acid: a review of its use in surgery
the amino acid glycine and case reports show that EACA and other indications. Drugs 1999;57:1005–1032.
33
causes seizures. Our electrophysiological studies demon- 3. Roberts I, Shakur H, Afolabi A, et al. The importance of early
strated that EACA acts as a competitive antagonist of treatment with tranexamic acid in bleeding trauma patients: an
47 exploratory analysis of the CRASH-2 randomised controlled trial.
glycine receptors (IC50 5 12mM) in mouse neurons. The Lancet 2011;377:1096–1101.
potency of EACA for glycine receptors is 10-fold
4. Hoylaerts M, Lijnen HR, Collen D. Studies on the mechanism of the
lower than that of TXA; however, EACA is often admin- antifibrinolytic action of tranexamic acid. Biochim Biophys Acta
75,76 1981;673:75–85.
istered at 10-fold higher doses than TXA to patients.
Aprotinin is structurally distinct from TXA and EACA. 5. Iwamoto M. Plasminogen-plasmin system IX. Specific binding of
tranexamic acid to plasmin. Thromb Diath Haemorrh 1975;33: 573–
We observed that aprotinin does not inhibit glycine cur- 585.
47
rents, even at a very high concentration (10mM). 6. Thorsen S. Differences in the binding to fibrin of native plasmino-
Antifibrinolytic drugs remain an important and gen and plasminogen modified by proteolytic degradation. Influ-
ence of omega-aminocarboxylic acids. Biochim Biophys Acta
effective, low-cost intervention that reduces blood loss, 1975;393:55–65.
morbidity, and mortality. Understanding the cause of
7. McEvoy MD, Reeves ST, Reves JG, et al. Aprotinin in cardiac sur-
TXA-associated seizures, recognizing the early warning gery: a review of conventional and novel mechanisms of action.
signs of impending seizures, and using anesthetics may Anesth Analg 2007;105:949–962.

reduce the incidence and severity of seizures and lead to 8. Berman M, Cardone D, Sharples L, et al. Safety and efficacy of
better patient outcomes. aprotinin and tranexamic acid in pulmonary endarterectomy sur-
gery with hypothermia: review of 200 patients. Ann Thorac Surg
2010;90:1432–1436.

Acknowledgment 9. Bertholini DM, Butler CS. Severe hyponatraemia secondary to des-

mopressin therapy in von Willebrand’s disease. Anaesth Intensive
This work was supported by operating grants from the Care 2000;28:199–201.
Canadian Institute of Health Research (MOP-38028, 10. Bhat A, Bhowmik DM, Vibha D, et al. Tranexamic acid overdosage-
induced generalized seizure in renal failure. Saudi J Kidney Dis
MOP-79428), a Canada Research Chair (B.A.O.), a
Transpl 2014;25:130–132.
grant from the Canadian Anaesthesiologists’ Society
11. Butala BP, Shah VR, Bhosale GP, et al. Medication error: subar-
(B.A.O.), the CIHR Frederick Banting and Charles Best achnoid injection of tranexamic acid. Indian J Anaesth 2012;56:
Doctoral Award (I.L.), a CIHR Fellowship (S.A.), a Cli- 168–170.

nician Scientist Transition Award from the Department 12. de Leede-van der Maarl MG, Hilkens P, Bosch F. The epilepto-
genic effect of tranexamic acid. J Neurol 1999;246:843.
of Anesthesia, University of Toronto (S.A.), a Mentored
Research Award from the International Anesthesia 13. Garcha PS, Mohan CV, Sharma RM. Death after an inadvertent
intrathecal injection of tranexamic acid. Anesth Analg 2007;104:
Research Society (S.A.), and a Merit Award from the 241–242.
Department of Anesthesia, University of Toronto 14. Ichikawa J, Kodaka M, Nishiyama K, et al. A case of postoperative
(C.D.M.). convulsive seizure following tranexamic acid infusion during aortic
valve replacement. Masui 2013;62:186–189.

15. Kaabachi O, Eddhif M, Rais K, et al. Inadvertent intrathecal injec-

tion of tranexamic acid. Saudi J Anaesth 2011;5:90–92.

24 Volume 79, No. 1

16. Kalavrouziotis D, Voisine P, Mohammadi S, et al. High-dose tra-
nexamic acid is an independent predictor of early seizure after
cardiopulmonary bypass. Ann Thorac Surg 2012;93:148–154.
17. Keyl C, Uhl R, Beyersdorf F, et al. High-dose tranexamic acid is
Lecker et al: TXA-Associated Seizures
related to increased risk of generalized seizures after aortic valve
replacement. Eur J Cardiothorac Surg 2011;39:114–121.
37. Wellington K, Wagstaff AJ. Tranexamic acid: a review of its use in
18. Koster A, Borgermann J, Zittermann A, et al. Moderate dosage of the management of menorrhagia. Drugs 2003;63:1417–1433.
tranexamic acid during cardiac surgery with cardiopulmonary
bypass and convulsive seizures: incidence and clinical outcome. Br 38. Ducloy-Bouthors AS, Jude B, Duhamel A, et al. High-dose tra-
J Anaesth 2013;110:34–40. nexamic acid reduces blood loss in postpartum haemorrhage. Crit
Care 2011;15:R117.
19. Mahmoud K, Ammar A. Accidental intrathecal injection of tranexa-
mic acid. Case Rep Anesthesiol 2012;2012:646028. 39. Rahman Z, Hoque R, Ali A, et al. Blood conservation strategies for
reducing peri-operative blood loss in open heart surgery. Mymen-
20. Manji RA, Grocott HP, Leake J, et al. Seizures following cardiac singh Med J 2011;20:45–53.
surgery: the impact of tranexamic acid and other risk factors. Can J
Anaesth 2012;59:6–13. 40. Sabovic M, Lavre J, Vujkovac B. Tranexamic acid is beneficial as
adjunctive therapy in treating major upper gastrointestinal bleeding
21. Martin K, Wiesner G, Breuer T, et al. The risks of aprotinin and in dialysis patients. Nephrol Dial Transplant 2003;18:1388–1391.
tranexamic acid in cardiac surgery: a one-year follow-up of 1188
consecutive patients. Anesth Analg 2008;107:1783–1790. 41. Zufferey PJ, Miquet M, Quenet S, et al. Tranexamic acid in hip
fracture surgery: a randomized controlled trial. Br J Anaesth 2010;
22. Merriman B, Mayson K, Sawka A, et al. Postoperative seizure in a 104:23–30.
neurosurgical patient: should tranexamic acid be on the differen-
tial? Can J Anaesth 2013;60:506–507. 42. Roberts I, Kawahara T. Proposal for the inclusion of tranexamic acid
(anti-fibrinolytic-lysine analogue) in the WHO model list of essential
23. Mohseni K, Jafari A, Nobahar MR, et al. Polymyoclonus seizure medicines. Geneva, Switzerland: World Health Organization, 2010.
resulting from accidental injection of tranexamic acid in spinal
anesthesia. Anesth Analg 2009;108:1984–1986. 43. Fergusson DA, Hebert PC, Mazer CD, et al. A comparison of
aprotinin and lysine analogues in high-risk cardiac surgery. N Engl
24. Murkin JM, Falter F, Granton J, et al. High-dose tranexamic acid is J Med 2008;358:2319–2331.
associated with nonischemic clinical seizures in cardiac surgical
patients. Anesth Analg 2010;110:350–353. 44. Pellegrini A, Giaretta D, Chemello R, et al. Feline generalized epi-lepsy
induced by tranexamic acid (AMCA). Epilepsia 1982;23:35–45.
25. Sander M, Spies CD, Martiny V, et al. Mortality associated with admin-
istration of high-dose tranexamic acid and aprotinin in primary open- 45. Schlag MG, Hopf R, Zifko U, et al. Epileptic seizures following
heart procedures: a retrospective analysis. Crit Care 2010;14:R148.
cortical application of fibrin sealants containing tranexamic acid in
rats. Acta Neurochir (Wien) 2002;144:63–69.
26. Sharma V, Katznelson R, Jerath A, et al. The association between
tranexamic acid and convulsive seizures after cardiac surgery: a mul- 46. Yamaura A, Nakamura T, Makino H, et al. Cerebral complication of
tivariate analysis in 11 529 patients. Anaesthesia 2014;69:124–130.
antifibrinolytic therapy in the treatment of ruptured intracranial
aneurysm. Animal experiment and a review of literature. Eur Neu-
27. Wang CS, Yang CJ, Chen SC, et al. Generalized convulsion rol 1980;19:77–84.
resulted in hyperammonemia during treatment with tranexamic acid
for hemoptysis. Ir J Med Sci 2011;180:761–763. 47. Lecker I, Wang DS, Romaschin AD, et al. Tranexamic acid concen-
trations associated with human seizures inhibit glycine receptors. J
28. Wong JO, Yang SF, Tsai MH. Accidental injection of tranexamic Clin Invest 2012;122:4654–4666.
acid (Transamin) during spinal anesthesia. Ma Zui Xue Za Zhi
48. Estrera AL, Sheinbaum R, Miller CC, et al. Cerebrospinal fluid
1988;26:249–252.
drainage during thoracic aortic repair: safety and current manage-
29. Yeh HM, Lau HP, Lin PL, et al. Convulsions and refractory ventricu- ment. Ann Thorac Surg 2009;88:9–15.
lar fibrillation after intrathecal injection of a massive dose of tra-
49. Kratzer S, Irl H, Mattusch C, et al. Tranexamic acid impairs c-
nexamic acid. Anesthesiology 2003;98:270–272.
aminobutyric acid receptor type A-mediated synaptic transmission
30. Casati V, Romano A, Novelli E, et al. Tranexamic acid for trauma. in the murine amygdala: a potential mechanism for drug-induced
Lancet 2010;376:1049–1050. seizures? Anesthesiology 2014;120:639–649.

31. Jimenez JJ, Iribarren JL, Brouard M, et al. Safety and effectiveness 50. Olsen RW, Tobin AJ. Molecular biology of GABA A receptors.
of two treatment regimes with tranexamic acid to minimize inflam- FASEB J 1990;4:1469–1480.
matory response in elective cardiopulmonary bypass patients: a
51. Sigel E, Steinmann ME. Structure, function, and modulation of
randomized double-blind, dose-dependent, phase IV clinical trial. J
GABAA receptors. J Biol Chem 2012;287:40224–40231.
Cardiothorac Surg 2011;6:138.
52. Greenfield LJ Jr. Molecular mechanisms of antiseizure drug activ-ity
32. Makhija N, Sarupria A, Kumar Choudhary S, et al. Comparison of at GABAA receptors. Seizure 2013;22:589–600.
epsilon aminocaproic acid and tranexamic acid in thoracic aortic
surgery: clinical efficacy and safety. J Cardiothorac Vasc Anesth 53. Henschel O, Gipson KE, Bordey A. GABA A receptors, anesthetics
2013;27:1201–1207. and anticonvulsants in brain development. CNS Neurol Disord Drug
Targets 2008;7:211–224.
33. Rabinovici R, Heyman A, Kluger Y, et al. Convulsions induced by
aminocaproic acid infusion. DICP 1989;23:780–781. 54. Sillito AM. The effectiveness of bicuculline as an antagonist of
GABA and visually evoked inhibition in the cat’s striate cortex. J
34. Gofton TE, Chu MW, Norton L, et al. A prospective observational Physiol 1975;250:287–304.
study of seizures after cardiac surgery using continuous EEG mon-
itoring. Neurocrit Care 2014;21:220–227. 55. Karkar KM, Thio LL, Yamada KA. Effects of seven clinically impor-
tant antiepileptic drugs on inhibitory glycine receptor currents in
35. Hunter GR, Young GB. Seizures after cardiac surgery. J hippocampal neurons. Epilepsy Res 2004;58:27–35.
Cardiothorac Vasc Anesth 2011;25:299–305.
56. Lynch JW. Molecular structure and function of the glycine recep-tor
36. Sindet-Pedersen S, Stenbjerg S. Effect of local antifibrinolytic chloride channel. Physiol Rev 2004;84:1051–1095.
treatment with tranexamic acid in hemophiliacs undergoing oral
surgery. J Oral Maxillofac Surg 1986;44:703–707. 57. Furtmuller R, Schlag MG, Berger M, et al. Tranexamic acid, a
widely used antifibrinolytic agent, causes convulsions by a c-
aminobutyric acid A receptor antagonistic effect. J Pharmacol Exp
Ther 2002;301:168–173.

58. Chebib M, Johnston GA. GABA-activated ligand gated ion chan-

nels: medicinal chemistry and molecular biology. J Med Chem
2000;43:1427–1447.

January 2016 25
ANNALS of Neurology 59. Macdonald RL, Olsen RW. GABAA receptor channels. Annu Rev
Neurosci 1994;17:569–602.
60. Mody I. Distinguishing between GABA A receptors responsible for
tonic and phasic conductances. Neurochem Res 2001;26:907–913.

61. Lynch JW. Native glycine receptor subtypes and their physiologi-cal
roles. Neuropharmacology 2009;56:303–309. 68. Tan KR, Rudolph U, Luscher C. Hooked on benzodiazepines:
GABAA receptor subtypes and addiction. Trends Neurosci 2011;
62. Farrant M, Nusser Z. Variations on an inhibitory theme: phasic and 34:188–197.
tonic activation of GABA A receptors. Nat Rev Neurosci 2005;6:
215–229. 69. Curtis DR, Game CJ, Lodge D. Benzodiazepines and central gly-
cine receptors. Br J Pharmacol 1976;56:307–311.
63. Parker AJ, Lee JB, Redman J, et al. Strychnine poisoning: gone but
not forgotten. Emerg Med J 2011;28:84. 70. Hui AC, Wong TY, Chow KM, et al. Multifocal myoclonus secondary
to tranexamic acid. J Neurol Neurosurg Psychiatry 2003;74:547.
64. Young AB, Snyder SH. Strychnine binding associated with glycine
receptors of the central nervous system. Proc Natl Acad Sci U S A 71. Bojko B, Vuckovic D, Cudjoe E, et al. Determination of tranexamic
1973;70:2832–2836. acid concentration by solid phase microextraction and liquid
chromatography-tandem mass spectrometry: first step to in vivo
65. Downie DL, Hall AC, Lieb WR, et al. Effects of inhalational general analysis. J Chromatogr B Analyt Technol Biomed Life Sci 2011;
anaesthetics on native glycine receptors in rat medullary neurones 879:3781–3787.
and recombinant glycine receptors in Xenopus oocytes. Br J Phar-
macol 1996;118:493–502. 72. Sharma V, Fan J, Jerath A, et al. Pharmacokinetics of tranexamic
acid in patients undergoing cardiac surgery with use of cardiopul-
66. Hales TG, Lambert JJ. The actions of propofol on inhibitory amino monary bypass. Anaesthesia 2012;67:1242–1250.
acid receptors of bovine adrenomedullary chromaffin cells and
rodent central neurones. Br J Pharmacol 1991;104:619–628. 73. Fox MA. Tranexamic acid: how much is enough? Anesth Analg
2010;111:580–581.
67. Belelli D, Callachan H, Hill-Venning C, et al. Interaction of positive
allosteric modulators with human and Drosophila recombinant 74. Merino JG, Latour LL, Tso A, et al. Blood-brain barrier disruption
GABA receptors expressed in Xenopus laevis oocytes. Br J Phar- after cardiac surgery. Am J Neuroradiol 2013;34:518–523.
macol 1996;118:563–576.
75. Dubber AH, McNicol GP, Douglas AS. Amino methyl cyclohexane
carboxylic acid (AMCHA), a new synthetic fibrinolytic inhibitor. Br J
Haematol 1965;11:237–245.

76. Dubber AH, McNicol GP, Douglas AS, et al. Some properties of the
antifibrinolytically active isomer of amino-methylcyclohexane
carboxylic acid. Lancet 1964;2:1317–1319.

26 Volume 79, No. 1