Keywords
tranexamic acid - traumatic brain injury - intracerebral hemorrhage - chronic subdural
hematoma - subarachnoid hemorrhage - perioperative period
Introduction
Tranexamic acid (TxA) is a synthetic lysine analogue that competitively blocks the
lysine-binding residues on fibrin clot and prevents the conversion of plasminogen
to plasmin, thus inhibiting clot lysis.[1] The middle of the twentieth century was the era of controlling surgical bleed using
chemical hemostatic agents, with a keen interest in antifibrinolytics (AF). There
was overwhelming literature on AF use in various surgical specialties with the aim
of reducing perioperative bleed and transfusion requirements, particularly in cardiac
surgical patients. Aprotinin, epsilon-amino-caproic acid (EACA), and TxA were the
commonly studied AF agents. The spirits were soon dampened as reports of an association
between aprotinin and renal failure and increased mortality trickled in, and this
drug was banned in 2007.[2] The ubiquitous use of AF was viewed with doubts about the possible life-threatening
adverse effects. Similarly, there was a topsy-turvy use of AFs in the 1970s and 1980s
in neurosurgery with a barrage of trials on its use in conservative management of
cerebral aneurysms to prevent rebleeding.[3] Over the next decade, administration of TxA in aneurysmal subarachnoid hemorrhage
(SAH) was marred by controversy with reports of increased risk of delayed cerebral
ischemia.[4] The largest trauma trial of this century, CRASH-2 (Clinical Randomisation of an
Antibrinolytic in Significant Haemorrhage 2) and the nested CRASH-2 IBS (Intracranial
Bleeding Study) opened the floodgates and revived this old drug with renewed interest
in the field of trauma including head trauma.[5]
[6] There are over 400 publications of its use in neurosurgery, and the spectrum of
its use is ever expanding in conditions such as chronic subdural hematoma (CSDH),
spontaneous intracerebral hemorrhage (ICH), SAH, skull base and vascular tumors, craniosynostosis,
and corrective spine surgery. In this narrative review, Google Scholar, PubMed, and
EMBASE databases were searched to identify publications relevant to the current use
of TxA in varied neurosurgical and critical care settings. The authors begin this
review with a brief discussion on the physiology of fibrinolysis and pharmacology
of TxA, followed by its use in various neurosurgical conditions. In the end, this
review will highlight the complications associated with TxA administration and conclude
the review by discussing ongoing and future research on TxA with suggestions for its
use in neuroanesthesia and critical care practices.
Fibrinolytic System
Coagulation and fibrinolysis are the two pillars of the hemostatic system that work
in perfect harmony to maintain vascular integrity, clot localization, and vascular
patency.
Tissue injury stimulates the coagulation cascade with thrombin burst, platelet activation,
and adhesion and conversion of fibrinogen to fibrin that in turn polymerizes to form
a stable fibrin clot. This fibrin clot provides a hemostatic seal at the site of wall
breach.
Simultaneously, the fibrinolytic system gets activated to localize the clot. Plasminogen
and tissue plasminogen activator (tPA) complex bind to the lysine residues on the
fibrin clot and generate the active serine protease, plasmin. Plasmin lyses the fibrin
clot to soluble fibrin degradation products and D-dimers.[7] The fibrinolytic activity is kept in check by the circulating serine protease inhibitors
that include plasminogen activator inhibitors 1, 2, and 3, α2-antiplasmin, α2-macroglobulin, and thrombin-activated fibrinolysis inhibitor.
This fine-tuned hemostatic system goes haywire in the event of massive trauma, surgery,
and cardiopulmonary bypass resulting in excessive systemic fibrinolysis and coagulopathy.[8] This speculation forms the basis for AF use in the above-mentioned settings.
Tranexamic Acid Pharmacology
Tranexamic Acid Pharmacology
Tranexamic acid was independently described by Okomato et al[9] and Melander et al.[10] TxA is a trans-stereoisomer of a synthetic derivative of amino acid lysine (trans-4-aminomethyl-cyclohexane-carboxylic
acid: AMCHA).[11] Being 7 to 10 times more potent than its congener EACA, it was soon embraced into
clinical practice for blood conservation.[1] With a half-life of 80 to 100 minutes, one-third of the drug is eliminated unchanged
in the urine within 1 hour and over 90% in 24 hours. It crosses the blood–brain barrier
(BBB) with cerebrospinal fluid (CSF) concentration reaching up to one-tenth of the
plasma concentration.[1]
[2] Though a complete fibrinolytic shutdown occurs at a plasma concentration of 100
to 150 mg/L, it has been suggested that 80% fibrinolysis inhibition is sufficient
for a clinically significant effect and is achieved at a plasma concentration of 10
mg/L.[12]
[13] In clinical practice, wide range of dosages have been described for TxA use in various
clinical settings. However, it has been suggested that a total dose of 14 mg/kg or
1 g of TxA is adequate for blocking fibrinolysis and use of higher doses may be unnecessary.[14] A recent trial, based on in vivo pharmacokinetic/pharmacodynamic model for TxA,
suggested lower TxA concentrations (up to 20 mg/L) to inhibit tPA fibrinolysis cascade.
However, higher concentrations (up to 150 mg/L) are required to inhibit urokinase-type
plasminogen activator (uPA)-induced fibrinolysis and proinflammatory potential of
plasmin.[15]
Trauma
One-third of the mortality in trauma patients is due to hemorrhage. It is estimated
that coagulopathic trauma patients are eight times more likely to die in the first
24 hours of injury.[16]
Rationale
The earlier notion was that overzealous resuscitation and hypothermia-coagulopathy-acidosis
(HCA) triad was the primary driver of bleeding in trauma patients. However, current
research has shifted the focus toward an endogenous process of acute trauma-induced
coagulopathy (ATC) as the root cause of bleeding and mortality in trauma patients.[17] ATC commences within minutes of injury as a thrombin burst and is accompanied by
a massive endothelial release of tPA, which manifests as disseminated intravascular
coagulation (DIC) with a fibrinolytic phenotype. Over hours, this process translates
into a state of fibrinolytic shutdown as the plasminogen inhibitors take control of
the systemic fibrinolysis.[18] It is noted that up to 35% of the trauma patients are coagulopathic at admission
to the emergency department. In a thromboelastometry (ROTEM)–based analysis of 300
trauma patients, it was noted that more than one-half of the patients had elevated
plasmin–antiplasmin complex levels, suggesting fibrinolytic activation. These patients
experienced higher 28-day mortality (12% vs. 1%, p < 0.001).[19]
Evidence
This insight into the pathogenesis of trauma, with a window of hyperfibrinolysis,
was identified as a potential target for AF application. It paved the way for the
largest civilian trauma trial, CRASH-2, that enrolled over 20,000 adult trauma patients
within 8 hours of injury across 274 hospitals in 40 countries.[5] Trauma patients with significant hemorrhage or at risk of hemorrhage were randomly
allocated to receive either TxA 1 g loading dose, followed by another 1 g over the
next 8 hours or placebo. The all-cause mortality at 28 days was significantly lower
in the TxA group compared with the placebo (14.5% vs. 16%). The authors observed maximal
survival benefit in patients who received TxA within 3 hours of injury (mortality
relative risk [RR] 0.68 [≤ 1 hour of injury] and 0.79 [1–3 hours of injury]).[20] It is likely that trauma patients beyond the 3-hour window might have been more
acidotic, hypothermic, and in a prothrombotic or fibrinolytic shutdown phase of the
injury. The CRASH-2 trial, however, has been marred by serious criticism for various
reasons. These include randomization design based on the uncertainty principle, the
study population including patients from low- to medium-income group countries with
varying levels of trauma systems, an absolute reduction in mortality by 0.8% due to
bleeding between groups with no reduction in transfusion requirements, lack of data
on injury severity scores, laboratory tests for fibrinolysis, coagulation derangements
or inadequate perfusion (acidosis, lactate), and 100% follow-up in spite of the vast
spread of study population. MATTERs (Military Application of Tranexamic acid in Trauma
Emergency Resuscitation) study, a retrospective observational study conducted in Afghanistan,
enrolled combat trauma patients requiring blood transfusion.[21] In a total of 293 patients who received TxA, the unadjusted mortality was lower
in TxA group despite being more severely injured (TxA vs. non-TxA group; 17.4% vs.
23.9%, respectively; p = 0.03). This effect was more evident in the subgroup requiring massive blood transfusion
(14.4% vs. 28.1%, respectively; p = 0.004). TxA administration was associated with better survival (odds ratio [OR]
= 7.23; 95% confidence interval [CI]: 3.016–17.322) and less coagulopathy (p = 0.003). This study demonstrated a clear survival benefit with a number needed to
treat (NNT) of 1:7 compared with the larger CRASH-2 trial where the NNT was approximately
1:67. A Cochrane review explored the effect of AFs in acute traumatic injury and concluded
that AFs reduced the risk of death from any cause (RR = 0.90, 95% CI: 0.85–0.96).[22]
The Antifibrinolytics Collaboration trial did a meta-analysis on 40,138 patients to
evaluate the effect of TxA (pool of CRASH-2 and WOMAN trial subjects).[23] It found a survival benefit of 70% in patients who were treated early with TxA.
The survival benefit degraded by 10% for every 15-minute delay up to 3 hours after
which there was no mortality benefit with TxA administration.
TxA is included in the list of essential medicines by the World Health Organization
for the management of acute bleeding in patients with trauma, cardiopulmonary bypass,
and postpartum hemorrhage, and early administration of TxA is now a part of trauma
guidelines worldwide.[24]
Comments
In the light of the available evidence, it may be prudent to administer TxA 1 g early
(≤ 3 hours of injury) in major trauma to attain maximal benefit.
Traumatic Brain Injury
Rationale
Following traumatic brain injury (TBI), patients experience a hypercoagulable state
in the early period, followed by an increased fibrinolytic activity. Both hyper- and
hypocoagulable states can lead to secondary brain insults. The pathogenesis at the
heart of TBI-induced coagulopathy is a matter of extensive debate. Massive release
of tissue factor, DIC, hyperfibrinolysis, protein C-thrombomodulin activation, and
platelet dysfunction have all been implicated.[25] TBI-associated coagulopathy results in higher odds of mortality and unfavorable
outcome.[26] Another phenomenon that is unique to TBI is the progression of hemorrhagic lesion
by either expansion or development of new, noncontiguous hemorrhages, commonly occurring
in the initial 6 to 12 hours of injury and referred to as hemorrhagic progression
of contusion (HPC).[27] In a prospective study of 153 TBI patients, the frequency of early HPC (within 6
hours of primary insult) was 43.5% and carried a fivefold hazard ratio of unfavorable
outcome.[28] Understanding the pathogenesis of TBI-induced coagulopathy resulted in increased
use of procoagulants that include fresh frozen plasma (FFP), cryoprecipitate, platelet
concentrate, factor concentrates such as prothrombin complex concentrate (PCC), and
pharmaceutical agents such as recombinant factor VII and AFs.
Evidence
The potential role of TxA in TBI has been explored in two trials till date. IBS was
a randomized placebo-controlled trial nested within the CRASH-2 trial that enrolled
270 TBI patients.[6] The primary outcome studied was total hemorrhage growth on computed tomography (CT)
of the head obtained at admission and after 24 to 48 hours. The mean total hemorrhage
growth was lower in the TxA group as compared with the placebo group (5.9 mL [standard
deviation (SD) 26.8] vs. 8.1 mL [SD 29.2]) with the adjusted difference of −3.8 mL
(95% CI: −11.5 to 3.9). The other randomized placebo-controlled trial recruited 240
moderate to severe TBI patients (postresuscitation Glasgow coma scale [GCS] 4–12)
where a CT scan was obtained within 8 hours of injury.[29] All patients received 1 g bolus TxA, followed by 1 g over 8 hours. Progressive ICH
(new lesion or significant growth ≥ 25%) was noted in 18% patients who were treated
with TxA as compared with 27% in the placebo group (RR = 0.65; 95% CI: 0.4–1.05).
A Cochrane review pooled the above trials and concluded that in TBI patients, TxA
may reduce mortality, but the quality of evidence is low and there is substantial
uncertainty.[22] The clinical implication of these findings is limited as the reduction in mortality
may be related to the effect of TxA on associated extracranial injuries.
Ongoing Research
Currently three trials are underway to investigate the effect of TxA in isolated TBI
patients. CRASH-3 trial (NCT01402882) plans to recruit 13,000 isolated TBI patients
within 8 hours of injury, and the results are expected by mid-2019.[30] A nested randomized controlled trial (RCT) within the CRASH-3 trial (CRASH-3 IBMS)
is the mechanistic substudy to evaluate the effect of TxA on intracranial hemorrhage
and cerebral ischemia in isolated TBI.[31] The other trial is investigating the application of TxA in a prehospital setting
in TBI patients within 2 hours of injury (NCT02645552, NCT01990768).[32]
Comments
The evidence is robust for the use of TxA in TBI patients with associated exsanguinating
extracranial injuries where early TxA administration reduces the risk of death. Results
from CRASH-3 are expected to throw light on the management of isolated TBI patients.
One should keep in mind the possible thrombotic complications in TBI patients when
using TxA.
Nontraumatic Intracerebral Hemorrhage
Nontraumatic Intracerebral Hemorrhage
Rationale
A recognized modifiable risk factor in ICH is hematoma expansion (HE), defined as
an absolute increase in size by > 12.5 mL or proportional increase by ≥ 33%, and has
been the target of therapeutic interventions.[33] Approximately one-third of ICH patients experience HE within 6 hours of symptom
onset, and it strongly predicts poor outcome.[34] Strategies to limit HE have primarily focused on lowering blood pressure, whereas
a wide variety of hemostatic therapies including TxA have been tried with variable
results.
Evidence
Observational studies have shown less HE with TxA, with 2 g showing better results
than the conventional 1 g dose.[35] A Cochrane review concluded that TxA led to a nonsignificant reduction in HE (RR
= 0.76; 95% CI: 0.56–1.05) and a non-significant increase in death/unfavorable outcome
with a relative risk of 1.25 (95% CI: 0.57–2.75).[36] Recently published TICH-2 trial, the largest reported trial evaluating TxA in spontaneous
ICH, recruited 2,325 patients across 124 hospitals in 12 countries.[37] The treatment arm received TxA 1 g over 10 minutes, followed by 1 g over 8 hours
with a matching placebo. The functional status quantified by modified Rankin scale
(mRS) and mortality at day 90 did not differ significantly between the treatment and
placebo arms (adjusted odds ratio [aOR] for functional status: 0.88; 95% CI: 0.76–1.03;
p = 0.11, aOR for mortality: 0.88, 95% CI: 0.76–1.03, p = 0.11). Notably, adverse events were lower in the treatment arm, and fewer patients
in the treatment arm had HE on day 2 (25% vs. 29%, respectively; aOR = 0.80; 95% CI:
0.66–0.98; p = 0.03). Though this study failed to show improved clinical outcome with TxA in acute
ICH patients, it can be speculated that the small perceived benefit may be better
reciprocated in patients with larger baseline hematomas, younger population, spot
sign positive patients, and earlier administration of TXA (< 3-hour window).[38]
Ongoing Research
Seven RCTs are currently investigating the role of TxA in reducing ICH. These include
Stopping Hemorrhage With Tranexamic Acid for Hyperacute Presentation Including Mobile
Stroke Units (STOP-MSU NCT03385928), the Spot sign and Tranexamic acid On Preventing
ICH growth–Australasia Trial (STOP-AUST NCT01702636), The Norwegian Intracerebral
Hemorrhage Trial (NOR-ICH, EudraCT number 2012–005594–30), Tranexamic Acid for Acute
ICH Growth prEdicted by Spot Sign (TRAIGE, NCT02625948), Tranexamic acid for IntraCerebral
Hemorrhage secondary to Novel Oral AntiCoagulants (TICH-NOAC NCT02866838), Tranexamic
Acid for Spontaneous Acute Cerebral Hemorrhage (TRANSACT NCT 03044184), and Tranexamic
acid and biomarkers in Emergency management of Spontaneous IntraCerebral Hemorrhage
(EsICH). STOP-AUST trial was completed in July 2018, and results were unavailable
at the time of writing this review.[39]
Comments
Though the TICH-2 trial failed to show functional improvement with TxA following ICH,
the study showed reduced HE and no increase in adverse events. Until the ongoing research
results are published, it is reasonable to consider early TxA administration only
in moderate to large acute nontraumatic ICH.
Chronic Subdural Hematoma
Chronic Subdural Hematoma
Rationale
Another area that has invited research on AFs is the CSDH. The pathophysiology of
CSDH is complex with an intertwined cascade of inflammation, growth factors upregulation,
angiogenesis, recurrent micro bleeds, local coagulopathy, hyperfibrinolysis, and exudation.
At the site of CSDH, evidence shows that there is existence of local hyperfibrinolysis
with abundance of tPA from adjoining endothelium, low plasminogen levels, and increased
fibrinogen degradation products (FDPs) as fibrin monomers and D-dimers when compared
with serum.[40]
[41] In the light of this evidence, it is hypothesized that TxA might inhibit the fibrinolytic
activity and kinin–kallikrein inflammatory cascade-induced vascular permeability,
translating into gradual resorption of the subdural hematoma.
Evidence
In a nonrandomized study on 21 patients with CSDH managed conservatively, TxA was
administered orally at a dose of 750 mg/day orally.[42] The median hematoma volume at presentation was 55.6 mL (interquartile range [IQR]:
7.5–140.5 mL). At 8 weeks, the median volume for all 21 patients was 3.7 mL (IQR:
0–22.1 mL). Similar results have been reported in CSDH patients treated with subdural
evacuating port system (SEPS), followed by oral TxA postoperatively.[43] Moreover, few case series have reported TxA as an adjunct to surgery, particularly
to avoid recurrences and in patients who are at high risk for surgical intervention.[44]
Ongoing Research
Currently many prospective studies are ongoing in CSDH patients to study the effect
of TxA on CSDH resolution, recurrence rates, associated thromboembolic complications,
and its benefits on anticoagulated patients. These studies include Tranexamic Acid
in Chronic Subdural Hematomas (TRACS) trial (NCT02568124), Tranexamic acid in the
Treatment of Residual Chronic Subdural Hematoma (TRACE) trial (NCT03280212), and Tranexamic
acid to prevent Operation in Chronic subdural Hematoma (TORCH) trial (NTR 6758), a
Dutch study.[45]
Comments
With the available results that are primarily retrospective, small sample size and
with the exclusion of anticoagulated patients, the role of TxA in this setting remains
uncertain. Its effect on hematoma resolution/recurrence and adverse thromboembolic
events will be clarified with the publication of results of ongoing studies.
Spontaneous Subarachnoid Hemorrhage
Spontaneous Subarachnoid Hemorrhage
Rationale
Rerupture of an intracranial aneurysm significantly increases the odds of in-hospital
mortality by fivefold and reduces the chance of survival with functional dependence
by half.[46] Rerupture within the first 24 hours following aneurysmal SAH occurs in approximately
9 to 17% of patients, over two-thirds occurring within 3 to 6 hours of index bleed.[47] Rebleeding is one of the preventable causes of mortality. The use of AFs after SAH
was first proposed by Mullan and Dawley in 1968.[3] Fibrinolytic activation around the clot, formed at the time of initial hemorrhage,
was thought to be a culprit in the pathogenesis of rebleed, and AFs were found to
be a strong case to mitigate this sinister event. Evidence of reduced CSF plasminogen
and elevated FDP levels following 2 weeks of TxA treatment in ruptured aneurysm patients
further supported this speculation.[48] Late surgery in aneurysmal SAH being the norm in those times, AFs found a strong
foothold in the prevention of rebleed in these patients.
Evidence
A couple of trials reported a statistically significant reduction in the incidence
of rebleeding and a nonsignificant trend toward development of delayed cerebral ischemia.[49]
[50] The International Cooperative Study on the timing of aneurysm surgery trial analyzed
the outcome data of patients in the early surgery group between those who received
AFs versus controls and noted that the cumulative 2-week rebleed rate was lower in
the AF group (11.7%) compared with those who were not administered AFs (19.4%). However,
there was a marked increase in the risk of delayed cerebral ischemia (32% vs. 23%),
and no differences in the functional outcome were observed between the two groups.[4] Due to increased incidence of DCI and hydrocephalus, enthusiasm over use of TxA
in aneurysmal bleed showed a decadence.[51] The final nail in the coffin was driven by the Cochrane 2003 review that included
1,399 patients across nine RCTs and noted that AF treatment had no beneficial effect
(odds of a poor outcome; OR = 1.12; 95% CI: 0.88–1.43). Though there was reduced rebleeding
risk (OR = 0.55; 95% CI: 0.42–0.71) with no significant association with hydrocephalus,
the accrued benefit got offset by the increased risk of cerebral ischemia (OR = 1.39;
95% CI: 1.07–1.89).[52] This review soon faced considerable criticism for including decade-old trials when
cerebral ischemia preventive therapies were still not in vogue, AF treatment was used
for prolonged periods (2–6 weeks) and was initiated well beyond the 72-hour window.
Also, the review failed to include a Swedish trial in which TxA was given early to
505 patients and for a short period (< 72 hours) until aneurysm was secured.[53] The rate of rebleeding in the treated group was significantly lower when compared
with no treatment group (2.4% vs. 10.8%, p = 0.01). Notably, there was no increase in ischemic events or vasospasm noted in
the TxA group.
A subsequent Cochrane review in 2013 included 10 trials with 1,904 participants and
found no improvement in all-cause mortality or poor outcome with the use of AFs.[54] Though the reviewers noted a reduced risk of rebleed (RR = 0.65, 95% CI: 0.44–0.97),
they concluded that evidence does not support AFs use in SAH to prevent rebleeding
even with the application of cerebral ischemia preventive measures. This Cochrane
review also had several limitations such as inclusion of studies in which late surgery
was the norm, use of AF for prolonged duration, and lack of consistency on cerebral
ischemia diagnosis methodology between studies.
In the current era, early obliteration of aneurysm is recommended to aggressively
manage DCI and prevent rebleed. Hence, there is a resurgence of interest in the AF
use in the pre-intervention period of ruptured cerebral aneurysms.
A meta-analysis that included 17 studies with 2,872 participants addressed the issue
by reviewing these studies in three groups: studies with short-term AF and medical
prevention of vasospasm, studies with long-term AF and medical prevention of vasospasm,
and studies with long-term AF and no medical prevention of vasospasm.[55] The reviewers noted that patients receiving short-term AF therapy with cerebral
protection strategies had a better outcome with a pooled OR of 0.79 (95% CI: 0.57–1.08).
It may be plausible that short-term AF (< 72 hours) therapy with early aneurysm exclusion,
use of aggressive hypertensive therapy, and calcium channel blocker therapy may reduce
rebleeding and offset the incidence of delayed cerebral ischemia with improved clinical
outcomes. Moreover, the stigma of thrombotic events associated with AF use in spontaneous
intracranial bleeding has been cleared, albeit sparingly, by another meta-analysis
of 57 studies including 5,049 participants, of which 72% of patients had a spontaneous
SAH.[56] The incidence of limb ischemia, pulmonary embolism, and myocardial infarction was
noted to be around 1% with TxA use. The rate of DCI, however, was 9.7% (95% CI: 5.5–14.8)
in SAH patients.
Guidelines
Antifibrinolytic use is still a gray area in SAH patients. European Stroke Organization
guidelines advise against its use and the American Heart Association/American Stroke
Association guidelines advocate optional use of short-term (< 72 hours) AF therapy
with TxA or aminocaproic acid when there is unavoidable delay in aneurysm occlusion
and when there are no medical contraindications to AF use.[57]
[58] Neurocritical Care Society guidelines also suggest use of AF therapy for less than
72 hours.[59]
Ongoing Research
An ongoing Dutch trial, ULTRA, endeavors to address the issue of whether ultra early
and short-term TxA use added to the standard SAH management will improve functional
outcome at 6 months. The study will recruit 950 adult aneurysmal SAH patients randomized
to TxA treatment (1 g IV immediately followed by 1 g every 8 hours until aneurysm
obliteration or a maximum of 24 hours, whichever is earlier).[60]
Comments
The available evidence is still sparse and is of low quality. It is clear that rebleeding
has a high case fatality rate and morbidity. AFs reduce the rebleeding risk by 35
to 40%. There is no uniform international consensus on TxA use in SAH. Though more
conclusive evidence toward an improved clinical outcome is eagerly awaited, in the
absence of contraindications to the use of TxA, it is reasonable to consider early
and short-term therapy (< 72 hours) if unavoidable delay in securing the aneurysm
is anticipated.
Perioperative Period
The role of TxA in reducing surgical bleed and transfusion requirements has been proven
in a variety of perioperative settings. A systematic review of 129 randomized clinical
trials including 10,488 patients showed that TxA administration reduced the probability
of blood transfusion (pooled risk ratio RR = 0.62; 95% CI: 0.58–0.65; p < 0.001).[61] General recommendations for perioperative use include a intravenous bolus dose of
10 to 20 mg/kg prior to skin incision, followed by a maintenance infusion (1–10 mg/kg/h).[1]
Intracranial Tumors
Rationale
With the background knowledge of hemostatic dysfunction associated with primary brain
tumors, particularly meningiomas, which are known to induce local tPA leading to fibrinolysis,
it seems plausible to consider TxA perioperatively.[62] TxA administration may provide additional benefit of reducing the incidence of postoperative
hematoma, especially in surgical excision of arteriovenous malformations (AVMs), tumors,
TBI, and CSDH. Trials on AF use in intracranial tumors were published more than a
decade ago.[63] These, however, failed to generate interest, and publications are restricted to
few case reports.[64] One of the cited reasons for limited scope of prophylactic use of TxA is that not
all neurosurgical procedures bleed excessively; the other concern being TxA-associated
seizures.
Evidence
A prospective RCT in 100 adult patients undergoing excision of intracranial tumors
noted a statistically significant reduction in total blood loss in TxA group compared
with placebo (817 mL ± 423.3 mL vs. 1,084 mL ± 604.8 mL; p = 0.01).[65] Though the amount of allogeneic blood transfused was similar in the two groups,
hemoglobin decline was significant in the placebo group (p = 0.02). The fibrinogen levels postoperatively were significantly elevated in the
TxA group supporting the fibrinolysis inhibition theory for reduced blood loss with
AFs. Another prospective double-blind RCT randomized 60 patients undergoing excision
of large meningiomas (size > 4 cm) noted reduced blood loss in TxA group compared
with placebo (830 mL vs. 1,124 mL; p = 0.03).[66] The investigators additionally reported better hemostasis in the TxA group on a
5-grade surgical hemostasis scale (p = 0.007). There were no thromboembolic events noted with the use of TxA. Complex
skull base surgeries are another unique group with significant blood loss due to problematic
surgical access and extension of tumor pathology into neighboring sinuses. A retrospective
study of 245 patients with skull base tumors who received TXA was compared with matching
control (274 patients).[67] TxA-treated patients had a 6% reduction in the absolute risk of transfusion.
Application of TxA topically at surgical site, though prevalent in other surgical
scenarios, in intracranial surgery, has been limited because of its tendency to induce
seizures.
Ongoing Research
A phase 3 clinical trial on efficacy of TxA in brain tumor resection (NCT01655927)
intends to determine whether TxA is effective or not in the reduction of intraoperative
blood loss in brain tumor resections.
Procedures on Spine
Rationale
Spine surgery is unique considering that it entails stripping the paraspinal muscles
off the lamina leaving raw muscle edges and periosteum to ooze. The older adults with
thin periosteum and osteoporotic bone with large vascular channels tend to bleed more.
Additionally, adult spine surgery usually involves instrumentation at several segments
and the rate of revision surgery is high. Whereas laminectomy is a low transfusion
risk procedure, spinal fusion surgery entails a blood loss approximating 800 mL in
noninstrumental fusion versus over 1.5 L in instrumental fusions.[68] Overall, perioperative blood loss and patient outcome in spine surgery are related
to the duration of surgery, anterior–posterior sequential approach, number of vertebrae
fused, number of osteotomies, and protocol-based transfusion practice. Moreover, the
ensuing coagulopathy can lead to postoperative neuraxial hematoma with potential neurologic
compromise.[69]
Evidence
A study in spine patients included 147 adult patients undergoing elective posterior
thoracic/lumbar instrumented spinal fusions and noted one-third reduction in perioperative
blood loss, but its efficacy on allogeneic transfusion requirement and clinical outcome
failed to reach significance.[70] Subsequent trials showed contradictory findings for outcomes of blood loss and blood
transfusion.[71]
[72] A 2017 meta-analysis comprehensively explored the efficacy and safety of AFs in
spine surgery.[73] It included 11 studies and showed reduced total blood loss (standard mean difference
[SMD] −0.75; 95% CI: −0.93 to −0.57; p = 0.000). The intraoperative and postoperative blood losses were significantly reduced
with TxA. However, all studies had significant heterogeneity for the studied outcomes
regarding drug dosage regimes, surgical procedure, duration of surgery, methods of
blood loss estimation, transfusion trigger, and transfusion protocols. A recent multicenter,
prospective randomized double-blind trial compared TxA with placebo in posterior instrumentation
spine surgery (> 3 segmental fusions) for efficacy based on number of transfused units
and perioperative blood loss.[74] There was significant reduction in the need for transfusion (42% in TxA vs. 67%
in placebo group; p < 0.05) and intraoperative blood loss.
With the current evidence, the European Society of Anaesthesiology recommends TxA
administration to attenuate perioperative blood loss in orthopedic and spine surgery
if the risk-benefit ratio is favorable.[75] Caution is advocated in women, previous history of venous thromboembolism, and older
adults (age above 60 years).
Another aspect of TxA use that is gaining popularity is its topical application into
the surgical wound site. Intuitively, the lower plasma TxA levels and thromboembolic
complications are advantageous.[76] Though there is literature supporting this topical application in varied surgical
settings, the literature in spine procedures are limited.[77]
Currently, there is interest in combination therapy with topical TxA added to systemic
TxA.
Ongoing Research
There are currently two ongoing trials to delineate the optimum dose of TxA required
to achieve maximal benefit with minimal harm ClinicalTrials.gov NCT 02053363 and NCT02188576.[78]
Comments
TxA can be administered in major complex multilevel spine surgery, revision surgery,
prolonged procedures, and where substantial blood loss is expected (> 20% total blood
volume), provided no contraindications exist. The optimal dose for surgery remains
unclear; however, a loading dose of 1 g may be reasonable followed by maintenance
infusion. Topical TxA may be a reasonable choice to consider in the wake of proven
safety and efficacy profile in cardiac surgical settings.
Safety Concerns
The adverse events of TxA are seen with prolonged oral administration and include
headache, nausea, vomiting, diarrhea, dyspepsia, dysmenorrhea, dizziness, back pain,
numbness, phosphenes, and anemia.[79] TxA is contraindicated if there is a history of preexisting active thromboembolic
event/disorder, DIC, renal failure, and recent coronary or vascular stent.[80] The two common worrisome adverse events are thrombotic risks and seizures.
Thrombotic Complications
TxA-induced fibrinolytic inhibition theoretically entails a prothrombotic state and
a plausible concern for thrombotic complications such as deep vein thrombosis (DVT),
pulmonary embolism (PE), myocardial infarction (MI), renal cortical necrosis, and
stroke. Though the majority of these are reported as case reports and case series,
large trials have refuted this concern.[81] With the available evidence in trauma and perioperative scenario, the thrombotic
potential of TxA remains more of a theoretical concern.
The neurosurgical patients are unique in this aspect as the reported incidence of
DVT is between 18 and 50% and PE in 0 to 25% patients.[82] Additionally, there is risk of delayed cerebral ischemia seen with wide range of
neuropathologies such as SAH, and TBI.
In the meta-analyses of patients undergoing arthroplasties and prostate cancer patients,
who are inherently at a high risk of thromboembolic events, no increased incidence
of VTE was noted.[83] Even following a high-dose administration of TxA (50–100 mg/kg), 30-day thrombotic
events were similar to the placebo group.[84]
Seizures
Epileptogenic potential of cortical TxA has been documented in animal studies and
with accidental intrathecal TxA administration.[85] Typically, TxA-associated seizures are myoclonic or generalized tonic-clonic seizures
occurring within the first 5 to 8 hours of surgery as the patient is weaned off the
anesthetics.[86] A plausible mechanism is elucidated by the remarkable structural homology between
TxA, gamma-aminobutyric acid type A (GABAA), and glycine receptors. It is possible that TxA acts as a competitive antagonist
of GABA type A receptors and blocks the chloride channels responsible for neuronal
hyperpolarization.
This manifests as increased excitability of the neuronal networks resulting myoclonus.[87]
[88] Finally, owing to the high affinity of TxA for glycine receptors, it is suggested
that the proconvulsant effects are probably mediated by the disinhibition of the glycine
receptor–mediated tonic current in the neuronal circuitry.[89] The incidence rate of TxA-associated seizures was noted to be dose dependent with
incidence rate varying from 1.4% in low-dose group to 5.3% in the high-dose group.[90] High incidence rate in cardiac surgical patients might be due to the frequent coexistence
of renal dysfunction in these patients.[91]
[92] The relationship between TxA and seizure activity in noncardiac surgery populations
remains undefined. An odd case report of postoperative seizure in a neurosurgical
patient who received TxA intraoperatively was attributed to TxA.[93]
However, the occurrence of seizures has not been noted in the CRASH-2 trial.
Future Research
The research into optimal use of AFS is a steep uphill task and needs to define the
optimal dosage of TxA and possible mechanisms of mortality benefit with its use. Dose
regimens are varied, and there are no in vivo pharmacokinetic studies to support any
of the regimens. Low doses tend to be questionable in terms of reducing allogeneic
transfusion but may tend to cause less harm compared with high doses. Another related
issue is to find out the minimum therapeutic concentration to bring about clinical
inhibition of fibrinolysis and titrate the dosage regimens accordingly.
Second, there is some evidence suggesting increased mortality with inadvertent use
of TxA in patients with bleeding trauma. In a database of 2,540 trauma patients, different
fibrinolytic phenotypes were noted, fibrinolysis shutdown being most common (46%),
followed by physiologic fibrinolysis (36%) and hyperfibrinolysis (18%).[94] In patients who received TxA, mortality increased in normal fibrinolysis and shutdown
subtypes, and there was no mortality benefit in hyperfibrinolysis phenotypes.[95]
[96] An evidence-based approach to the use of TxA based on hyperfibrinolysis assessment
is the need of the hour.
A randomized placebo-controlled TAMPITI (Tranexamic Acid Mechanisms and Pharmacokinetics
in Traumatic Injury) trial (NCT02535949) intends to evaluate the effects of TxA on
the immune system (viz. activated monocytes, cytokines), pharmacokinetics, safety
and efficacy of two different doses of TxA in severely injured trauma patients.[97]
Third, evidence is robust denying a prothrombotic potential of TxA in trauma and other
perioperative settings. However, there are sparse data on specific patient population
who are at higher risk for thromboembolic events, which may be the case in neurocritical
care settings. This has been primarily because expert consensus and drug formulary
contraindicate its use in this subset of patients.
Fourth, in the absence of any direct measurements of TxA effect, surrogate measures
such as reduced allogeneic transfusion have shown wide variability in patients and
different surgical settings. It is likely that genetic variability may be responsible.[98]
Finally, in the absence of ready availability of direct tests of fibrinolysis and
poor reliability of standard laboratory tests in detecting fibrinolysis, this remains
a major gray area for ubiquitous administration of TxA. Though rotational ROTEM/thromboelastography
(TEG) is the most widely used clinical test to detect hyperfibrinolysis, the sensitivity
is poor in detecting mild to moderate degrees of fibrinolytic activation.[19]
An interesting speculation that is gaining momentum is the anti-inflammatory and antineoplastic
potential of TxA by inhibiting plasminogen and plasmin at high doses.[99]
Conclusion
Tranexamic acid as a fibrinolytic inhibitor appears promising in exsanguinating trauma,
postpartum hemorrhage, and cardiopulmonary bypass-dependent cardiac surgery with a
reasonable safety profile. However, the evidence for universal application in neuroanesthesia
and neurocritical care is less robust. As the results of various ongoing trials are
awaited, there are few, if any, level I recommendations for TxA use in neurosurgical
patients. It may have a case where the benefits justify the harm as in large vascular
intracranial tumors, multilevel spine corrective surgery, pre-intervention period
in aSAH, and TBI with significant extracranial exsanguinating trauma. Basic premise
that needs to be followed is “give early, stay low.” The picture is elusive until
the results of CRASH-3 trial throw some light on the future of TxA in neuroanesthesia
and neurocritical care.
