Keywords
traumatic brain injury - contusion - anticoagulants - antiplatelets
Introduction
A normal individual with normal hemostasis maintains a balance between thrombus formation
and destruction using a complex interaction between the smooth vascular endothelium,
the coagulation cascade, the platelet aggregation system, and the fibrinolysis mechanism.
However, in patients who are on either antiplatelet drugs (APDs) or anticoagulants
(ACDs), this normal homeostasis is altered. This is further altered with traumatic
brain injury (TBI) and thus, we need specific guidelines to address this subpopulation
to decide the length of observation, avoid unnecessary hospitalization, and relieve
the economic burden. To be on the safer side, physicians often err on the side of
caution and allow for a prolonged period of observation and obtain repeat head imaging.
This can lead to prohibitively higher caseloads and even higher costs with undue radiation
exposure for such cases.
There exists a very few randomized controlled trials (RCTs) for this clinical question
and hence the evidence is weak at best. Despite this, there remains an acute need
for clear and precise guidelines on how to manage such cases. A thorough risk–benefit
analysis for each patient is prudent before making clinical decisions. The need is
further compounded when we take into account the number of elderly who suffer from
TBI—up to 7% in one study.[1] There are a significant number of individuals in this age group who are on anticoagulants
due to cardiovascular diseases and to make matters worse, mortality rates in TBI are
higher in this age group. Thus, the intake of these drugs can increase the risk of
intracerebral bleed further in these cases and with the absence of specific guidelines,
it only makes matters worse. We suggest the following recommendations based on available
literature; however, they cannot be given any level of evidence as these are not based
on any RCTs due to paucity of such studies.
Mechanism of Action of Commonly used Antiplatelet Drugs and Anticoagulants
Historically, vitamin K antagonists (VKAs), such as warfarin, were the only anticoagulants
widely available for human use. They act by blocking the vitamin K–epoxide reductase,
thereby preventing the formation of the active form of the vitamin K–dependent clotting
factors. The VKAs have an initial prothrombotic effect, by initially blocking proteins
C and S, followed by a delayed antithrombotic effect, through the inhibition of coagulation
factors II, VII, IX, and X. Antithrombin III is a peptide that inhibits several of
the activated clotting factors. Unfractionated heparin (UFH) binds to and increases
the activity of antithrombin III by inducing a conformational change to factor Xa,
which ultimately leads to inhibition at Xa and IIa in a 1:1 ratio. Unfractionated
heparin also has some inhibition on factors IXa, XIa, and XIIa. Low molecular weight
heparins (LMWH), which also bind AT3, are smaller and have a higher proportional impact
on Xa versus IIa, in a 3:1 or 2:1 ratio. Apart from these major classes, there are
antiplatelet drugs like aspirin which is a cyclooxygenase (COX) inhibitor that irreversibly
inhibits COX1 and, in higher doses, COX2. P2Y12 receptors are adenosine diphosphate
(ADP) receptors expressed on the surface of thrombocytes; they can be blocked chemically
by clopidogrel, prasugrel, and ticagrelor, which are in use, and cangrelor, which
has recently been licensed. Other drugs such as glycoprotein IIb/IIIa inhibitors like
abciximab, which is a humanized monoclonal mouse antibody, tirofiban, and eptifibatide
are synthetic GpIIb/IIIa inhibitors. Cilostazol and dipyridamole are phosphodiesterase
inhibitors and protease-activated receptor-1 antagonists; agents such as vorapaxar
and atopaxar, inhibit platelet activation through alternative routes. The various
drugs and mechanism of action are detailed in [Table 1]. Few caveats may be important to understand like heparin-induced thrombocytopenia
risk is same with UFH and LMWH. Protamine sulfate can only help to partially reverse
the effect of heparin: at most 60% of the anticoagulation effect of LMWH. In such
a case, the best option is to use antifactor X.[2] Given the plethora of agents that any case of TBI can be on, it can be daunting
to decide on the right course of management. Given below is a clinical scenario—stage-wise
management on the options and best course of management.
Table 1
Commonly used anticoagulants and antiplatelets with their mechanism of action
Mechanism of action
|
Drug name
|
Class of drug
|
COX-1 inhibitor
|
Aspirin
|
Antiplatelet drugs
|
P2Y12 inhibitors
|
Clopidogrel
|
Prasugel
|
Ticlopinine
|
Ticagrelor
|
Cangrelor
|
Gp IIb/IIIa inhibitors
|
Abciximab
|
Tirofiban
|
Eptifibatide
|
PAR-1 inhibitors
|
Vorapaxar
|
PDE-3 inhibitors
|
Dipyridamole
|
Cilastazol
|
Vitamin K antagonist
|
Warfarin
|
Anticoagulants
|
Heparins
|
Unfractionated Heparin
|
Enoxaparin
|
Dalteparin
|
Tinzaparin
|
Factor Xa inhibitor
|
Fondaparinaux
|
Non–anti Vitamin K anticoagulants
|
Rivaroxiban
|
Apixaban
|
Fibrinolytics
|
Alteplase
|
Reteplase
|
Tenecteplase
|
Urokinase
|
Initial Management
Management of head injuries depends on the severity of head injuries. Once a patients
on anticoagulants presents with a head injury, we need to approach the management
with care and the steps we need to follow are subdivided and detailed as follows:
-
Initial management and investigations for head injury—based on whether they are:
-
Mild head injury (GCS 14–15) or moderate and severe head injury GCS 13 and below.
-
Investigations.
-
Necessity for in-patient admission.
-
Serial follow-up (both clinical and radiological).
-
Management
Initial Management of Head Injuries
Initial Management of Head Injuries
Internationally renowned guidelines such as Canadian CT head rule, New Orleans Criteria,
NICE clinical guideline, and NEXUS II are based on the factors like mechanism of injury,
patient age, comorbidities, and the neurological exam, but do not take anticoagulant
use into consideration. While mild head injuries may still have a predictable course,
the dilemma becomes deeper in case of moderate and severe injuries. However, for any
patient who is on anticoagulants or has liver dysfunction or other coagulation disorders
due to any medical or surgical reasons, a plain CT scan is justified in all cases.
This is because the exact indications have not been clearly elucidated and not doing
the same can risk missing a serious underlying pathology in cases with mild head injury.
Only VKAs have been shown to increase mortality in TBI cases and other anticoagulants
have had this effect only in very few studies.[3]
[4]
[5] Hence, it may be appropriate to perform a CT scan in all cases to prevent missing
an intracranial pathology.
Similarly, the course of a hematoma in such patients has not been demonstrated in
literature. The only factors that can be thought to influence a hematoma are findings
of the initial CT scan, the underlying risk factors, and the evolution of neurologic
state and these may be considered while performing a repeat CT scan. A safer methodology
would be to admit such a case for 24 hours despite the normal neurological examination
and unremarkable CT scan. The only caveat would be aspirin monotherapy and a normal
neurological and CT scan examination. The arbitrary limit of 24 hours is based on
the risk of delayed bleeding which has been reported to affect 0.2 to 6% of TBI patients
on VKAs or clopidogrel with normal findings upon repeat CT.[5] Thus, a safe methodology would be to perform a CT scan in those cases where a repeat
examination shows deterioration and/or cannot be performed due to intubation, sedation,
or dementia within 24 hours; and cases where such an examination does not show any
findings the cases can be discharged after an observation period of 24 hours. The
various scenarios are summarized in [Table 2].
Table 2
Steps and scenarios of initial clinical management of a case of traumatic brain injury
(TBI) on anticoagulants
Question
|
Recommendation
|
Reasoning
|
Abbreviations: GCS, Glasgow coma scale; VKA, vitamin K antagonist; CT, computed tomography;
ICH, intracerebral hematoma; CCT, cranial computed tomography.
|
Should a CT scan be performed in all patients with suspected or known TBI and potential
or known intake of oral anticoagulants? All mild TBIs too?
|
All patients with suspected or known TBI and potential or known intake of oral anticoagulants
require a CT scan irrespective of anamnesis or neurological examinations (e.g., Glasgow
coma scale [GCS])
|
|
-
Of the available anticoagulants, only VKAs are proven to increase TBI mortality, and
in several guidelines, these drugs are identified as an indication to perform a CT.
|
|
|
Should a follow-up CT scan be performed routinely? If yes, when?
|
|
|
|
|
Should a patient with a normal CT scan be admitted for monitoring of the neurological
state? If yes, for how long and what kind of monitoring should be used?
|
|
|
-
A follow-up CT is indicated only in case of neurological deterioration (changes in
GCS and pupil responses or FOUR score, as determined by specialists in neurology,
trauma surgery, neurosurgery, or intensive care medicine).
|
|
|
|
|
|
|
|
-
Patients whose antithrombotic therapy comprises ASA monotherapy only may be discharged
immediately under the following conditions: normal initial CCT scan, GCS 15, absence
of other risk factors, and guaranteed observation by nursing home staff or suitably
instructed close family/friends.
|
|
|
|
How should we proceed with patients with a normal CT scan who cannot be examined neurologically
(intubated, sedated, delirious, or noncooperative)?
|
All patients with TBI and potential or known intake of oral anticoagulants with a
normal CCT scan who cannot be examined neurologically (e.g., due to intubation, sedation,
or dementia) require a follow-up CCT within 6–24 h after trauma.
|
Since initial CT scan was negative a further CT scan in cases where neurological examination
is not possible would be beneficial to rule out any new injury/pathology.
|
Investigations
After the initial management of the patient, need arises for them to undergo investigations
related to the effect of anticoagulant drugs. Among the routine investigations that
were performed are bleeding time (BT) and clotting time (CT). BT is defined as the
time taken from the infliction of a wound to the arrest of bleeding (N = 2–5 min) and reflects platelet function and number. The clotting time (CT) has
been defined as the time taken by a sample of blood to coagulate in vitro under standard
conditions (N = 8–15 min). These tests, though useful, are abnormal only in severe disease and
do not correlate with intraoperative bleeding. These factors along with a lack of
standardization have resulted in them going out of common practice; but they are still
used in many centers as a useful bedside indicator. Prothrombin time (PT) which is
a test for the extrinsic pathway (N = 11–15 s) is used to test the activity of factors VII and X. The internationalized
normalized ratio (INR) is the ratio of PT that would be obtained if the international
reference thromboplastin has been used to test the patient. Activated prothrombin
time (aPTT) is the time required for plasma to clot when maximal surface contact activation
and optimal phospholipid and calcium concentration are provided (N = 35–45 s). Thrombin time (TT) is the time required for the formation of a stable
clot after the addition of thrombin to citrated plasma (N = 12–14 s), that is, for the fibrinogen to form fibrin strands in vitro. Apart from
these standard tests, there are mixing studies which are usually done in cases with
an abnormally prolonged PT or aPTT. Here, adding an equal volume of the patient’s
citrated plasma to normal pooled plasma can perform a mixing study of the specimen.
If the abnormal value normalizes after mixing, it indicates clotting factor deficiency
in the plasma. Persistent prolongation of PT or aPTT after mixing suggests the presence
of inhibitors in the plasma.[3]
When dealing with antiplatelets, the scenario might be different. These include platelet
function tests like platelet function analyzer (PFA), impedance aggregometry (multiplate),
and VerifyNow. These tests are capable of detecting and/or ruling out the presence
of a platelet inhibitor. The intensity of platelet inhibition can be assessed; this
allows for an estimation of the bleeding risk. This might be useful especially when
the patient’s medication is unknown. Platelet function tests are established methods
for detecting disorders of the primary hemostatic capacity (e.g., von Willebrand syndrome)
and monitoring antiplatelet drugs. Therapeutic ranges are established for the principal
drugs (PFA, seconds [Siemens Package insert 2012–10]; Multiplate, the area under the
curve [Roche Diagnostics Package insert 2016–12, V3.0 German]; and VerifyNow). However,
these are certain drawbacks too. Notably, there is no evidence regarding residual
inhibitory effects and the probability of accelerated bleeding when test results are
outside the therapeutic range. These tests may (NOT ALWAYS) help differentiate nonresponders
from noncompliant patients. Also, a poor correlation has been seen when these tests
are compared to each other.[4]
Newer anticoagulants require special investigations. For example, TT or dilute TT
(dTT) is used to rule out the presence of dabigatran anticoagulation. TT within the
reference range excludes (remaining) dabigatran-associated anticoagulation while dTT
(hemoclot) level <30 ng/mL excludes (remaining) dabigatran-associated anticoagulation.
TT measurement within the normal range excludes the presence of dabigatran anticoagulation
since even low dabigatran concentrations (30–40 ng/mL) cause significant prolongation
of the TT. It is not possible to use TT measurements for the quantitative determination
of dabigatran or assessment of the risk of bleeding from dabigatran-mediated thrombin
inhibition. Dabigatran level <30 ng/mL, likely to be observed >4 hours postdose, excludes
a relevant risk of bleeding. It should be noted that the detection limit of the available
dTT assays is 30 ng/mL. When using apixaban, edoxaban, or rivaroxaban measuring antiactivated
factor X (anti-Xa) activity—calibrated to LMWH or the specific “xaban” of interest—is
the method of choice. Calibrated to LMWH, an anti-Xa activity below the detection
limit of the respective laboratory excludes (remaining) xaban-associated anticoagulation,
whereas calibrated to the particular xaban, anti-Xa activity <30 ng/mL excludes (remaining)
xaban-associated anticoagulation.[2] The various investigations are detailed in [Table 3].
Table 3
Investigations for coagulation function according to drug being consumed
Drug
|
Test
|
Usefulness
|
Abbreviations: BT, bleeding time; CT, clotting time; PT, prothrombin time; aPTT, activated
prothrombin time; CCT, cranial computed tomography; INR, internationalized normalized
ratio; PFA, platelet function analyzer.
|
Any anticoagulant
|
BT, CT
|
Bedside test
|
Any anticoagulant or antiplatelet
|
PT, aPTT, INR, mixing studies
|
|
|
Antiplatelets
|
|
|
|
|
|
|
|
|
|
|
|
|
Dabigatran
|
Thrombin time and dilute thrombin time
|
Even low dabigatran concentrations (30–40 ng/ml) cause significant prolongation of
the TT
|
Apixaban edoxaban, rivaroxaban
|
Anti–activated factor X (anti-Xa)
|
Anti-Xa activity < 30 ng/mL excludes xaban-associated anticoagulation
|
Treatment
Any goal of surgical treatment entails three main steps: maintaining target INR, the
reversal of anticoagulation, and the use of adjuncts; the first and foremost of these
is to decide on a target INR ([Table 4]). Textbooks have often mentioned a value of <1.4 INR for performing procedures but
there is no substantiation of the same.[6] A simpler methodology would be to reverse the INR in cases where a pathological
finding is found on the CT scan of a TBI case. The reversal needs to be urgent as
hemorrhagic lesions often progress during the early hours after trauma, and hemorrhagic
progression of a contusion impairs clinical outcomes.[7]
[8]
[9]
[10] In such cases, a target INR less than 1.5 and platelet count more than 1.35 lakhs/dL
would be ideal, as any value other than these has been shown to be predictive of both
radiographic and clinical worsening. Here a special mention of platelet concentrate
transfusion needs to be made as to the findings as controversial. Downey et al[11] investigated the effect of platelet transfusion in a retrospective study of 328
TBI patients aged >50 years on Acetyl Salicylic Acid (ASA) or clopidogrel. Patients
who received platelet transfusion had a similar mortality rate to those who were not
treated with platelets (17.5% vs. 16.7%, respectively; p = 0.85). Similarly, Ducruet et al[12] studied 66 patients on antiplatelet therapy (ASA and/or clopidogrel) who suffered
a primary ICH. Here, hematoma expansion was similar in transfused versus nontransfused
patients. In another study, Briggs et al[13] studied the effect of platelet transfusion in TBI patients, 12 on ASA and 5 not
on ASA. ASA-induced component of platelet dysfunction but not the trauma-induced component
was ameliorated by platelet transfusion.
Table 4
Steps and scenarios to be followed in the treatment of case of TBI on anticoagulants
Scenario
|
Recommendation
|
Reason
|
Abbreviations: TBI, traumatic brain injury; INR, internationalized normalized ratio
VKA, vitamin K antagonist; CT, computed tomography; ICH, intracerebral hematoma; ASA,
acetyl salicylic acid; IV, intravenous fluid; PCC, prothrombin complex concentrate;
GCS, Glasgow coma scale ; rFVIIa, recombinant factor VIIa.
|
What is the target INR in patients receiving VKAs when the initial CT scan gives a
positive result?
|
INR < 1.5.
|
|
|
|
When should reversal of anticoagulants be done?
|
|
|
|
|
|
|
|
|
Should platelet concentrate be administered to reverse the effect of platelet inhibitors?
|
|
|
|
|
Should vitamin K be administered to reverse the effect of VKAs?
|
|
|
|
|
|
|
|
|
Should PCC (prothrombin complex concentrate) and/or plasma be used for reversal of
VKAs?
|
|
|
|
|
|
|
Should recombinant activated factor VII (rFVIIa) be used for the reversal of VKA anticoagulation?
|
The available evidence shows no benefit from using rFVIIa versus PCC for the reversal
of VKA in hemorrhagic TBI.
|
|
Two small retrospective studies:
|
-
First study: time to INR reversal was similar with both treatments (PCC, 784 min;
rFVIIa, 980 min), but INR rebound occurred more frequently in the rFVIIa group.[7]
|
|
|
Should desmopressin (DDAVP) be administered to reverse the effect of platelet inhibitors?
|
|
|
|
|
|
|
|
|
Should tranexamic acid (TXA) be administered to reverse the effect of platelet inhibitors?
|
|
|
|
|
|
|
Should tranexamic acid (TXA) be administered to all head injury cases?
|
|
|
|
|
|
-
Only 9,202 patients received TA within 3 h which did not meet the objective of 10,000
subjects.
-
After excluding patients of GCS 3 and bilaterally non-reactive pupils (NON-SALVEGABLE
CASES) the death rate was 12.5% in tranexamic acid group vs. 14% in the placebo group,
so there is no statistical difference in outcome in head injury-related deaths.
|
|
|
|
|
|
|
Platelet transfusion and coadministration of DDAVP has not been associated with a
decreased risk of hemorrhage progression or mortality. In a prospective study,[14] DDAVP was administered to 10 patients with ICH who had been receiving ASA, and platelet
function was improved; this effect was short-lived and not statistically significant.
Naidech et al[15] reported a study in patients with spontaneous intracerebral hematoma and reduced
platelet activity. In patients (n = 7) treated with DDAVP within 12 hours of ICH symptom onset, a modest reduction
in intracranial hematoma volume was observed (median 0.5 mL).
No studies have investigated TXA in TBI patients on platelet inhibitors. A subgroup
analysis of TBI patients (n = 270) recruited in the CRASH-2 study[9] where the effect of TXA on ICH in patients with TBI was analyzed. CT scans performed
before randomization and after 24 to 48 hours showed comparable mean total hemorrhage
growth with versus without TXA (TXA group 5.9 ± 26.8 mL; placebo group, 8.1 ± 29.2
mL). A RCT (double-blind, placebo-controlled trial) by Yutthakasemsunt et al[16] enrolled 238 patients with moderate to severe TBI (GCS 4–12) and no coagulopathy,
and found that no significant difference in ICH progression was observed between the
TXA group and placebo patients (risk ratio [RR] = 0.65). The risk of death from all
causes and the risk of an unfavorable outcome on the Glasgow outcome scale (GOS) were
similar between groups (RR = 0.69 and RR = 0.76, respectively). Chakroun-Walha et
al[17] conducted a prospective, randomized trial of TXA in 180 TBI patients and found that
mortality and 28-day GOS were similar in patients who received or did not receive
TXA.
In the recently published CRASH-3 study[18] which was conducted in 175 hospitals and 29 countries across the world,[1] the authors have reported the results of a randomized, placebo-controlled trial
of 12,737 adults with TBI (mean age 41.7 years [SD 19.0]; 80% men, 20% women) in which
cases treated within 3 hours of injury had a reduced risk of head injury-related death,
which was 18.5% in the tranexamic acid group versus 19.8% in the placebo group (855
vs. 892 events; RR 0.94 [95% CI 0.86–1.02]). TA given very early (<3 h) in cases of
acute severe bleeding (traumatic and postpartum hemorrhage) significantly increased
overall survival from bleeding and survival benefit decreased by 10% for every 15
minutes of treatment delay until 3 hours, after which there is no benefit. Even after
excluding patients of GCS 3 and bilaterally nonreactive pupils the death rate was
12.5% in tranexamic acid group versus 14% in the placebo group, so there is no statistical
difference in outcome in head injury-related deaths (RR 0.94, CI 0.86–1.02). In this
study, it was found that there was no significant reduction in death rates in severe
TBI patients and a significant reduction in death rates in mild and moderate head
injury patients with intracranial bleed. Effect of tranexamic acid on head injury-related
death stratified by time to treatment is suggestive of the fact that earlier the treatment
is given, better is the prognosis but severity of head injury is confounding factor
and, thus, the trial failed to show an effect on two major parameters, that is, death
and disability in head injury patients, but there was no increase in vascular occlusive
events and morbidities. Hence, giving tranexamic acid to isolated TBI patients may
not have a major change in their prognosis.
When to Restart Anticoagulation?
When to Restart Anticoagulation?
As with all chronic illness, the need for anticoagulation is life long and TBI forms
an event in the due course of the drug. Here comes the all-important question of optimal
timing and preferred agent for pharmacological prophylaxis in patients after hemorrhagic
TBI. The updated Brain Trauma Foundation guidelines[19] state that anti-thromboembolism medication can be reinitiated 24 hours after injury
in patients who have a clinically and radiographically stable TBI. It also recommends
LMWH as the agent of choice, at a dose suitable for patients with a high risk of thrombosis
(e.g., subcutaneous enoxaparin 4,000 IU once daily). The reasons for such a recommendation
are manifold. The reported incidence of venous thromboembolic events in isolated TBI
varies from 3 to 25% when thromboembolism prophylaxis is delayed or not administered.
Brain Trauma Foundation guideline recommends LMWH or unfractionated heparin (UFH)
in combination with mechanical prophylaxis but the time frame for this treatment is
not specified. In a systematic review of 14 studies it was shown that pharmacologic
thromboembolism prophylaxis, administered 24 to 72 hours after injury, is well tolerated
in patients with stable TBI, and 4 studies suggested that administering thromboembolism
prophylaxis within 24 hour of injury does not lead to progressive traumatic ICH.[19]
While prophylaxis is understood, should therapeutic anticoagulation be resumed after
hemorrhagic TBI? If yes, what should be the optimal timing? Here the evidence fails
us. There is insufficient evidence to support or discourage the resumption of therapeutic
antithrombotic treatment following TBI. Expertise from a multidisciplinary team with
experience of clinical practice should be sought to guide decision-making on a case-by-case
basis. A survey of practice patterns[20] in patients with central nervous system hemorrhage and a history of atrial fibrillation
and ischemic stroke showed that the most common times for restarting anticoagulation
after the index hemorrhage was 1 month (43.5%) followed by 1 week (33.7%), respectively.
Only 13.3% of respondents indicated they would prefer an earlier restart time (3 days)
and 8% indicated they would not restart anticoagulation. Interestingly, 47.7% of respondents
indicated that they face dilemmas at least once per week concerning anticoagulation
restart time and intensity, and 59.4% stated that they relied predominantly on intuition
or experience. Thus, the answer lies in performing more clinical trials in this rarely
discussed scenario.[21]
Best Practice Guidelines
Based on our review we can propose the guidelines given in [Table 5] in cases of TBI on anticoagulants.
Table 5
Best practice guidelines in a case of TBI on anticoagulants
Abbreviations: CT, computed tomography; GCS, Glasgow coma scale; INR, internationalized
normalized ratio; LMWH, low molecular weight heparins.
|
Diagnosis
|
CT scan
|
|
|
Neurological examination
|
|
|
Tests and targets
|
Vitamin K antagonists
|
|
Non–Vitamin K anticoagulants
|
|
|
Platelet inhibitors
|
Platelet function tests preferably multiplate assays: no target values
|
Reversal
|
Vitamin K antagonists
|
Plasma concentrates to target INR < 1.5 with Vitamin K 5–10 mg IV dose
|
Non–vitamin K anticoagulants
|
|
|
Platelet inhibitors
|
No candidate for reversal
|
Resumption of anticoagulation
|
Prophylaxis
|
LMWH after 24rs: 4,000 IU s.c once daily
|
Thrombolysis
|
Case-by-case as per multidisciplinary team
|
Conclusion
The guidelines given borrow heavily from previously conducted trials and consensus
guidelines[20] and the reason remains poor availability of well-conducted trials in this field.
We have attempted to provide a pragmatic and practical approach to such cases with
the hope that it will ensure minimum risks with the best possible patient outcomes.
The entire journey from patient presentation to follow-up has been covered in this
article and we hope this would be useful to all practicing in the field of neurotrauma.