Keywords
trauma - coagulopathy - trauma-induced coagulopathy - bleeding
Schlüsselwörter
Trauma - Koagulopathie - trauma-induzierte Koagulopathie - Blutung
Introduction
Trauma is one of the major global public health challenges. For the year 2013, it
was estimated that 973 million people suffered injuries that warranted some type of
medical service and 4.8 million people died. Injuries accounted for 10.1% of the global
disease burden in 2013, mainly affecting the younger age group.[1] Although advances in trauma and intensive care as well as the widespread implementation
of damage control principles resulted in a decline in death rates,[2] the death toll due to trauma is still high. According to a recent retrospective
analysis of a large cohort, the prevalence of trauma death has declined since 1983,
but the majority of deaths (56%) still occur within the first 24 hours after injury.[3]
Uncontrolled post-traumatic bleeding, which is a major concern in the early phase
of trauma resuscitation, is the leading cause of potentially preventable death.[4]
[5]
[6] There are two mechanisms of bleeding following an injury: anatomical bleeding as
a result of blood vessel damage and bleeding due to coagulopathy. Anatomical bleeding
can be frequently managed through damage control strategies, which resulted in a decline
in the mortality rate attributed to haemorrhage.[7] However, the coagulopathy following the injury may lead to a haemostatic dysfunction
that can significantly contribute to further bleeding, organ dysfunction and poor
clinical outcome.
During further course of trauma management, particularly in patients with severe injuries,
hypercoagulopathy can ensue, which is associated with thromboembolic events and multiple
organ dysfunction. Trauma induces local and systemic inflammation, similar to that
described in sepsis. Inflammation, in turn, leads to initiation and propagation of
the coagulation cascade, while coagulation also influences inflammation. Although
their incidence has declined as a result of improved trauma management, multiple organ
dysfunction and sepsis are frequent causes of late death following trauma.[8]
Depending on the criteria applied and the severity of injury, coagulopathy can be
detected in up to 56% of trauma patients in the early phase of trauma ([Table 1]).[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17] However, defining a coagulopathy based on laboratory abnormalities alone does not
necessarily translate into a clinically relevant coagulopathy.[18]
Table 1
Prevalence of trauma-induced coagulopathy (TIC)
|
Ref.
|
N
|
TIC criteria
|
Prevalence (%)
|
|
Brohi et al[9]
|
1,088
|
PT or aPTT or TT
|
24.4
|
|
MacLeod et al[10]
|
10,790
|
PT
|
28.0
|
|
Maegele et al[11]
|
8,724
|
PT or platelet count
|
34.2
|
|
Hess et al[12]
|
15,728
|
INR, aPTT, platelet count,
or fibrinogen concentration
|
5.5 up to 43.1, depending on injury severity
|
|
Floccard et al[13]
|
45
|
ISTH DIC score
|
56.0
|
|
Cohen et al[14]
|
1,198
|
INR
|
41.6
|
|
MacLeod et al[15]
|
701
|
PT
|
16.3
|
|
Hagemo et al[16]
|
808
|
INR
|
11.0
|
|
Fröhlich et al[17]
|
61,212
|
PT and/or platelet count
|
24.5
|
Abbreviations: aPTT, activated partial thromboplastin time; DIC, disseminated intravascular
coagulation; INR, international normalized ratio; ISTH, International Society on Thrombosis
and Haemostasis; PT, prothrombin time; TT, thrombin time.
Mechanisms of Trauma-Induced Coagulopathy
Haemostasis is a complex process at the centre of the body's defence and wound healing
system. Physiologic haemostasis is a balance between pro- and anti-coagulation as
well as fibrinolysis and anti-fibrinolysis. This process takes place on the surface
of endothelial cells and platelets.[19] The intact endothelium is anti-thrombotic, this property being maintained by several
mediators, including tissue factor pathway inhibitor, endothelial protein C receptors,
endothelial glycocalyx layer (EGL), thrombomodulin, nitric oxide, and tissue plasminogen
activator. The EGL is a matrix of macromolecules, which is not only anti-coagulant
but also important to maintain microvascular integrity. The most prominent among these
molecules are heparan sulfates, accounting for 50 to 90% of the total pool, the rest
being mainly chondroitin sulfates and hyaluronic acid.[20] The EGL is a sort of endothelial gatekeeper.[21]
An injury to a vessel wall exposes the subendothelial collagen that provides an adhesion
platform for circulating platelets as well as for the interplay between the cellular
and humoral components of the haemostatic system. This pro-coagulant activity is controlled
by counter-regulatory anti-coagulant cascades. The net effect of these two opposing
systems may be generation of pro-coagulation at the site of endothelial injury, while
preventing uncontrolled microvascular thrombosis and tissue hypoperfusion by means
of endogenous anti-coagulation and fibrinolysis. Thrombin plays a central role in
this process by activating both coagulation and anti-coagulation as well as contributing
to the crosstalk with the inflammatory response.[22]
The activated protein C (APC) system has been considered as a major player in the
development of trauma-induced coagulopathy (TIC). APC is a physiologic anti-coagulant
that irreversibly inactivates the pro-coagulant factors Va and VIIIa. It is also profibrinolytic
by inhibiting plasminogen activator inhibitor-1 (PAI-1). APC is also cytoprotective
through anti-inflammatory and anti-apoptotic mechanisms. Results from a mouse model
indicated the role of APC in the endogenous coagulopathy following trauma.[23] In a single-centre study in 203 patients with major trauma (mean injury severity
score [ISS]: 25.2), a marked increase in APC was observed in those with an ISS >15
and a base deficit >6. The same study showed a correlation between protein C depletion
in the first 12 hours and increased risk of ventilator-associated pneumonia.[24] A subsequent publication using data from the Prospective, Observational, Multicenter,
Major Trauma Transfusion (PROMMTT) study supported the assumption that APC may have
an important role in the development of TIC. In the PROMMTT study, which included
1,245 severely injured patients (mean ISS: 26.2), acute traumatic coagulopathy identified
on arrival to the emergency department was associated with a depletion of the pro-coagulant
factors I, II, V, VII, VIII, IX and X on the one hand and activation of the protein
C system on the other.[14] A recent study by Davenport et al also showed that elevated APC concentration was
associated with excess mortality and morbidity following TIC. In the second part of
the same study using a murine model, mortality rate following trauma haemorrhage in
transgenic mice with 1,000-fold reduced capacity to activate protein C was almost
half that of wild-type mice, suggesting a central role of the protein C pathway in
early TIC.[25] However, the data on APC in trauma are inconsistent. A recent study comparing 25
trauma patients with hyperfibrinolysis identified using thromboelastography with 14
healthy controls showed a significant early increase in tissue plasminogen activator
but not PAI-1 following a severe injury.[26] This finding thus challenges the apparent role of APC-mediated hyperfibrinolysis.
A recent systematic literature review concluded that there may not be a direct cause–effect
relationship between APC and increased fibrinolysis.[27] This contradiction is not surprising, taking into consideration the complexity of
the haemostatic system and the crosstalk with the myriad of mediators released following
a trauma.[14]
Trauma activates the neurohumoral system that results in a catecholamine surge. This
increase in catecholamine levels leads to an endothelial damage and glycocalyx degradation,
generally known as endotheliopathy. A prospective study using blood samples from adult
trauma patients directly admitted to a trauma centre showed that adrenaline level
was increased in non-survivors, and this was independently associated with an increase
in syndecan-1, which is a marker of glycocalyx degradation. The increase in adrenalin
also correlated with biomarkers of endothelial damage and hyperfibrinolysis.[28] A recent study confirmed that high adrenaline levels and glycocalyx damage are associated
with hypocoagulopathy and hyperfibrinolysis.[29] A high level of syndecan-1 in trauma patients was also associated with increased
inflammation and endothelial damage.[30] Shedding of EGL components may also contribute to autoheparinization, which was
observed in approximately 5% of trauma patients, and this was associated with a high
ISS.[31] Endotheliopathy may also contribute to a capillary leak following trauma.[32] A secondary analysis of the data from 512 patients included in the Pragmatic Randomized
Optimal Platelet and Plasma Ratios (PROPPR) trial showed that, after adjusting for
ISS and transfusion requirements, elevated serum syndecan-1 levels were independently
associated with the development of sepsis during hospitalization.[33] In a recent prospective observational study in 424 trauma patients, Johansson et
al showed that both adrenaline and syndecan-1 were independent predictors of <24 hours,
7-day and 28-day mortality.[34] An animal study using male Sprague-Dawley rats showed that sympathetic denervation
was anti-inflammatory, anti-fibrinolytic and endothelial protective in rats with acute
traumatic coagulopathy, confirming the role of catecholamines in TIC.[35]
Hypofibrinogenemia as a consequence of TIC is frequently observed in the early phase
of trauma. In a multicentre observational study from the United States,[36] including 1,133 trauma patients who arrived in a hospital within 180 minutes post-injury,
a fibrinogen concentration <2 g/L measured by the Clauss method was identified in
19.2% of the patients. In another observational study on major trauma patients from
Australia, 12.9% of the patients had fibrinogen levels <1.9 g/L during the initial
resuscitation.[37] Low fibrinogen level was in both studies associated with a poor outcome. The age-associated
increase in fibrinogen level should be kept in mind when interpreting results.[38]
Platelets contain a plethora of proteins involved in coagulation and fibrinolysis.
It is not yet clear how these contradictory platelet secretions are exactly affecting
TIC. Data on platelet function testing in trauma patients are scant, because platelet
sample handling and availability of specific assays are complicated. One older small
prospective observational study using thromboelastography (TEG)-based platelet functional
analysis in whole blood samples collected from trauma patients at risk of TIC immediately
after injury and before blood or substantial fluid administration showed a significantly
impaired platelet aggregation in response to adenosine diphosphate (ADP) and arachidonic
acid, whereby the ADP inhibition was more pronounced.[39] Another study on 101 trauma patients using multiple-electrode impedance aggregometry
also showed platelet dysfunction in response to ADP, arachidonic acid, collagen and
thrombin receptor activating peptide (TRAP), with response to the first three agonists
markedly reduced within the first 24 hours.[40] In that study, platelet dysfunction was observed in 45.5% of trauma patients on
admission and 91.1% at some time during their intensive care unit (ICU) stay. In another
small more recent study on 40 trauma patients, a significantly decreased ADP- and
TRAP-mediated platelet aggregation was observed, suggesting that thrombin receptor
pathway plays an important role in trauma-induced platelet dysfunction.[41] Platelet dysfunction seems to be ubiquitous even in minor trauma. In a recent study,
platelet dysfunction identified with TEG-platelet mapping was reported on 459 patients
with minor trauma (median ISS: 5).[42] However, the mechanisms and the implication of this finding are not clear. Anaemia,
be it due to haemorrhage or dilutional, can also affect platelet adhesion. In vitro
experiments have shown the dependence of platelet adhesion on haematocrit.[43] Considering the available evidence, endotheliopathy and anaemia may be the triggers
for platelet dysfunction in trauma.
While hyperfibrinolysis has been discussed as a major event in early TIC, a trauma-induced
fibrinolytic shutdown is also demonstrated.[44] A study by Moore et al showed that there is a U-shaped distribution of mortality
related to fibrinolysis in response to a major trauma, warning that inadvertent exogenous
inhibition of fibrinolysis may have an adverse effect on survival.[45] Fibrinolysis shutdown may even be an adaptive physiologic response to trauma.[46]
Shock is an independent risk factor for trauma coagulopathy.[14]
[47]
[48] The true incidence of shock in trauma is not clear, since systolic blood pressure
was frequently used in several studies as a determinant of hypoperfusion. Traumatic
brain injury further hampers the use of blood pressure measurements as a sign of hypoperfusion.[49]
Hypothermia, acidosis and haemodilution are also associated with the development of
TIC. Hypothermia can ensue following a trauma as a result of heat loss, reduced heat
production and administration of fluids. Clinically significant reduction in platelet
function and coagulation factor activity ensues at a core body temperature below 33°C.[50]
[51]
[52]
Metabolic acidosis, mostly associated with shock, leads to a reduction in the activity
of coagulation factors.[50]
[53] In a swine model, acidosis, but not hypothermia, resulted in a decrease in plasma
fibrinogen concentration by 18%.[54] Acidosis accelerated fibrinogen consumption, while it did not affect fibrinogen
production.[55] The administration of bicarbonate to correct acidosis is not associated with a reversal
of the coagulopathy.[56] Finally, pre-clinical fluid administration also contributes to the development of
a dilutional coagulopathy,[11]
[14]
[57] with the incidence of coagulopathy rising with increasing amount of fluid administered.[11]
In summary, the pathogenesis of early coagulopathy is multifactorial, with the balance
between pro- and anti-coagulation influenced by injury severity, hypoperfusion, acidosis,
hypothermia, underlying disease conditions, medication history and genetic predispositions
as well as emergency treatment ([Fig. 1]).
Fig. 1 Mechanisms of early trauma-induced coagulopathy (TIC). Patient factors (medication,
comorbidities, genetic polymorphism) as well as environmental factors (heat loss)
or iatrogenic injury (e.g. haemodilution) modify TIC.
Late Coagulopathy
The haemostatic reaction following trauma returns to baseline during recovery in patients
without complications, while patients with severe injury may suffer from the sequelae
of massive coagulopathy. Recovery of the coagulopathy after severe trauma may be delayed
in such patients.[58] There is a massive reorganization of the human genome following trauma, which leads
to a multitude of changes in innate immunity and inflammation.[59]
[60] Trauma also results in an uncontrolled local and systemic release of damage-associated
molecular patterns (DAMPs) that trigger an inflammatory response.[61]
Similar to the haemostatic response, it remains difficult to distinguish between adaptive
and maladaptive systemic inflammatory responses to injury. From clinical viewpoint,
the feasible means to identify maladaptive systemic inflammation is at present the
identification of organ dysfunction, in analogy to the current definition of sepsis.[62] Almost 30% of severely injured patients develop a multiple organ dysfunction syndrome
(MODS) and this is associated with a high risk for nosocomial infections.[60]
[63] Similar to TIC, inflammation and infection result in a chain of coagulopathies that
may culminate into a disseminated intravascular coagulopathy (DIC),[64] which is associated with a significantly high mortality rate.[65] The late hypercoagulopathy following trauma also contributes to an increased risk
of venous thromboembolism ([Fig. 2]).[66]
[67]
[68]
[69]
[70]
Fig. 2 Consequences of late coagulopathy in trauma. Damage to the endothelium plays a crucial
role in inflammation and infection.
Diagnosis of Trauma-Induced Coagulopathy
The diagnosis of TIC is still based on laboratory abnormalities that may not necessarily
correspond to a distinct clinical phenotype. Despite significant advances in coagulation
research, there is no adequately valid test to predict and identify a clinically relevant
acquired coagulopathy. Published reports on TIC have been mostly based on the evidence
of abnormal laboratory findings of prothrombin time, activated partial thromboplastin
time, plasma fibrinogen concentration, platelet count, either alone or in combination.
However, an important caveat in this regard is comparing data from patients with those
of the healthy population without due consideration of the physiological adaptive
changes of the haemostatic system in trauma.
Viscoelastic point-of-care tests are increasingly used to diagnosis and manage TIC-associated
bleeding. However, published data are mostly retrospective and/or observational.[71]
[72]
[73]
[74] Despite these drawbacks, viscoelastic point-of-care tests can help trauma caregivers
to establish algorithms for rapid diagnosis of relevant bleeding diathesis and appropriate
use of haemostatic drugs. Nevertheless, the extra-haemostatic effects of coagulation,
such as its role in inflammation and immune modulation, are still difficult to characterize,
let alone identify with routine laboratory assays.
Treatment of Trauma-Induced Coagulopathy
Changes in the haemostatic system following a trauma should be viewed in two ways.
First, the urgent need in the early phase of trauma is to treat and avoid bleeding.
The second issue, which is even more difficult to diagnose as well as manage, is the
role and the dynamics of haemostasis in the development of organ dysfunction following
trauma. There are limitations regarding conclusions derived from available evidence,
because data from published studies show considerable heterogeneity in study design,
study population and outcome measures.[75] Evidence for clinically meaningful modification of the haemostatic system in relation
to inflammation and organ dysfunction is still lacking.
Early appropriate damage control contributes to the reduction of the incidence of
TIC. Beyond the mechanical management of bleeding, hyperfibrinolysis during the early
phase of TIC has been the subject of research and treatment strategies. The CRASH-2
trial has shown that early administration of tranexamic acid in adult trauma patients
with, or at risk of, significant bleeding was associated with a reduced risk of death.[76] Guidelines for transfusion of red blood cells and platelets as well as administration
of pro-coagulant factors in the management of bleeding have been developed based on
available evidence.[77] Although such algorithms may help standardize the use of blood products, there are
still several controversial issues to be considered. First, the use of fresh frozen
plasma (FFP) in TIC is declining. While FFP may be a better volume replacement than
a crystalloid solution during trauma management, it is not efficient in correcting
a coagulopathy. Second, platelet count does not necessarily correlate with platelet
function. There is still a lack of appropriate routine laboratory assays to test for
platelet function during trauma, so that treatment recommendations are still mainly
empirical. Third, current management strategies do not address the role of the coagulation
system in inflammation due to lack of appropriate diagnostic tools. For instance,
a very important lesson from the CRASH-2 trial is that administration of tranexamic
acid later than 3 hours after trauma was associated with a poor outcome.[76] This underscores the dynamics of the coagulation system and the role of every coagulation
step in the inflammation and wound healing process. Both hyperfibrinolysis and fibrinolysis
shutdown are part of this adaptive system, so that inadvertent exogenous manipulation
of these steps may be harmful.
Conclusion
TIC is a complex heterogeneous syndrome with a multitude of pathogenetic factors,
which include both adaptive and maladaptive changes in the haemostatic system. Comparing
differences in laboratory parameters between trauma patients and healthy controls
is not an ideal design to differentiate between adaptive and maladaptive changes following
trauma. Based on available data, it is difficult to discern between simple association
and cause–effect relationships. More high-quality research is still required to better
define TIC both in the early and late phases and establish optimal treatment targets.
