Current Understanding of the Pathophysiology of Trauma-Induced Coagulopathy
Hemorrhage is the second leading cause of death following trauma, exceeded only by traumatic brain injury.[1 ] If severe blood loss after trauma remains uncontrolled, it leads to pronounced hemorrhagic shock, which has been identified as an important driver of trauma-induced coagulopathy (TIC). Historically, coagulopathy after severe trauma was assumed to be a result of hemodilution due to fluid replacement therapy, consumption of coagulation factors at the side of injury, and additional confounders, such as hypothermia and acidosis.[2 ] Intensive research over the past 15 years has found that TIC is an endogenous dysregulation of the hemostatic system, primarily driven by tissue trauma, shock-related hypoperfusion, endotheliopathy, altered thrombin generation (TG), and platelet dysfunction, which can be further complicated by exogenous factors, such as hypothermia, acidosis, hypocalcemia, and dilution.[3 ] Early TIC is characterized by a hypocoagulative state and the inability to form sufficient clots, resulting in uncompressible diffuse microvascular bleeding. Studies have revealed that TIC starts early after trauma and can be detected in the most severely injured patients, already at the scene of the accident.[4 ] The presence of a TIC upon emergency room (ER) admission is associated with higher blood loss, increased allogeneic blood product requirements, higher rates of multi-organ failure, and an almost fourfold increase in mortality compared with trauma patients with hemostatic competence.[5 ]
[6 ]
[7 ] When bleeding and shock-related hypoperfusion is controlled and patients survive the initial first 24 hours, a transition from an early hypocoagulable to a later hypercoagulable and prothrombotic state occurs.[8 ] Thus, thromboprophylaxis should be initiated as soon as possible.[9 ] The purpose of this review is to provide current evidence on the potential drivers and mechanisms resulting in “early” TIC.
Definition, Incidence, and Diagnosis of TIC
So far, no universally accepted definition of TIC has been established. The term TIC describes an abnormal hemostatic response following a major injury, which results in an inefficient clot formation process, diffuse microvascular bleeding, and an increased risk of exsanguination.
The diagnosis of TIC still relies on standard coagulation tests, such as prothrombin time and international normalized ratio (INR).[10 ]
[11 ]
[12 ] However, the correlation between INR and TG is poor. Dunbar and Chandler[77 ] demonstrated that TG parameters in severely injured patients with an INR of greater than 1.5—by definition TIC—were upregulated. Moreover, a single parameter, such as INR, cannot display the complexity of coagulation abnormalities related to TIC.
More advanced technologies, such as viscoelastic test (VET) methods, have discovered more trauma patients with abnormal test results compared with prothrombin time or activated partial thromboplastin time.[10 ]
[11 ]
[12 ]
[13 ]
[14 ]
[15 ]
[16 ]
[17 ] This partly explains the wide variation of reported TIC incidences ranging from 25 to 43% of all severely injured patients.[6 ]
[14 ]
In contrast to VET methods, a general accepted “gold standard” for point-of-care platelet function testing has not been established so far.[18 ] Both the applied technologies and the composition of platelet agonists used to activate thrombocytes differ considerably between the different platelet function analyzer.[19 ] Moreover, platelet function analyzers were initially developed to assess the impact of platelet inhibitors, such as aspirin or adenosin diphosphate (ADP) antagonists, on thrombocytes rather than to detect potential bleeding related to trauma-induced platelet dysfunction.[20 ]
Importantly, the laboratory definition of TIC differs from clinically evident coagulopathy with diffuse microvascular bleeding. Chang et al reanalyzed data from the PROHS study and reported that clinically evident coagulopathy was rare (4%) compared with laboratory-defined coagulation abnormalities (39%) but was associated with substantially higher mortality (59 vs. 22%).[21 ]
Shock as a Driver of Trauma-Induced Coagulopathy
An isolated massive tissue injury without shock induces a prothrombotic phenotype of TIC associated with an increased risk of thromboembolic complications.[3 ] In contrast, the bleeding type of TIC requires both shock-related hypoperfusion with a corresponding low-flow state and tissue trauma.[8 ] Frith et al demonstrated that the severity of TIC strongly correlated with the combined degree of both injury and shock.[11 ] A prolongation of the prothrombin time ratio and activated partial thromboplastin time (aPTT) was only detected in shocked patients, defined as an admission base deficit of greater than 6 mmol/L. In contrast, when base deficit remained within normal limits, prothrombin time and aPTT also remained within the reference ranges.[11 ]
Endotheliopathy of Trauma
The vascular endothelium and its anticoagulant intraluminal layer, the glycocalyx, are a huge, often underestimated organ, with a large surface area of approximately 5,000 m2 and a weight of approximately 1 kg.[24 ] The endothelium plays an essential role in coagulation and inflammation, serving as a semipermeable barrier between the fluid phase and the tissue.[25 ] Endotheliopathy of trauma describes a state of endothelial cell damage and glycocalyx shedding with the release of specific serum biomarkers such as soluble thrombomodulin, syndecan-1, heparan sulfate, chondroitin sulfate, hyaluronic acid, and many more.[24 ]
[26 ]
[27 ]
[28 ] Endotheliopathy of trauma is primarily driven by inflammation and shock-related hypoperfusion with the release of high amounts of catecholamines (e.g., adrenalin) and vasoactive hormones such as vasopressin[24 ] ([Fig. 1 ]). In a rat model of hemorrhagic traumatic shock, chemical sympathectomy suppressed the release of inflammatory cytokines, decreased profibrinolytic activation, and was associated with less endothelial damage compared with sham animals.[29 ] In another experimental study, Hofmann et al demonstrated an independent association between shock severity and the intensity of endotheliopathy and sympathoadrenal activation.[27 ] This aligns with the findings in trauma patients, which also showed a strong association between sympathoadrenal activation and the release of markers of endothelial cell and glycocalyx damage.[30 ] Both adrenalin concentration and endotheliopathy were identified as independent predictors of poor outcomes in trauma patients.[31 ]
Fig. 1 Schematic overview of potential drivers of trauma-induced coagulopathy. t-PA, tissue plasminogen activator; PAI-1, plasminogen activator inhibitor 1. Activators; inhibitors.
Importantly, a breakdown of the glycocalyx results in capillary leakage and a significant loss of intravascular volume, which additionally worsens hypovolemia in already shocked patients, further intensifying tissue hypoperfusion and shock severity.[8 ]
Moreover, the release of heparinlike substances, such as heparan sulfate or chondroitin sulfate, as a consequence of glycocalyx shedding was proposed as a potential additional driver of TIC due to an endogenous autoheparinization process.[32 ] Whether and to what extent autoheparinization plays a role as an additional anticoagulant mechanism that increases bleeding tendency is currently under debate. A recent study investigating potential autoheparinization with different VET assays in severely injured trauma patients did not indicate that the release of heparan sulfate plays a significant role in the pathogenesis of TIC.[33 ]
Taken together, endotheliopathy of trauma is driven by shock-related release of adrenalin and vasopressin into the bloodstream, which promotes hypocoagulability, hyperfibrinolysis (HF), increased bleeding risk, transfusion requirements, and mortality.
Hyperfibrinolysis
HF has been identified as a predominant driver of TIC, which is strongly associated with poor outcomes in trauma patients.[34 ]
[35 ]
[36 ]
[37 ]
Two mechanisms have been proposed as potential activators of profibrinolytic pathways after a major injury. Brohi et al suggested that hypovolemic shock stimulates the endothelial synthesis of thrombomodulin, which binds thrombin. This complex, in turn, activates protein C. Activated protein C, the main anticoagulant protein of the body, promotes HF by inhibiting plasminogen activator inhibitor-1 (PAI-1), which is the most important antagonist of the profibrinolytic enzyme tissue plasminogen activator (tPA).[38 ]
Another hypothesis suggests that hypoxemia, in conjunction with high concentrations of adrenalin, vasopressin, and thrombin, powerfully activates endothelial cells. In turn, significant amounts of tPA are released from the Weibel–Palade vesicles into the bloodstream.[39 ] This hypothesis is supported by the finding that HF has also been demonstrated in other nontraumatic low-flow states, such as life-threatening anaphylactic shock or out-of-hospital cardiac arrest.[40 ]
[41 ]
Independent of the suggested mechanism, tPA cleaves plasminogen to plasmin, which dissolves fibrin and—if available in high amounts—fibrinogen. Thus, upregulated plasmin generation promotes premature clot dissolution and hypofibrinogenemia.
With the implementation of VET methods in modern trauma care, HF has been identified as an important contributor to TIC.[35 ]
[36 ]
[42 ]
[43 ]
[44 ] However, there is no uniform definition of HF based on VET results. For the ROTEM/ClotPro devices, HF is defined as a breakdown of greater than 15% of the maximum clot firmness. For TEG, a reduction of greater than 3% 30 minutes (LY30) after reaching the maximum amplitude of the clot is by definition HF.
However, not only the percentage of decreased clot firmness but also the speed of clot dissolution is linked to poor outcomes. Fulminant lysis, defined as a complete breakdown of the clot within 30 minutes (ROTEM) or the so-called diamond of death shape of the clot (TEG), is associated with an almost 100% mortality.[37 ]
[45 ]
[46 ] Thus, the pattern of clot lysis seems to be crucial for clinical outcomes ([Fig. 2 ]). It is essential to note that the absence of lysis signs in VETs does not rule out profibrinolytic activation. Raza et al reported in a cohort of trauma patients that, despite normal maximum lysis (ML) in ROTEM, high plasmin–antiplasmin complexes (>1,500 μg/L) were detected, suggestive of fibrinolytic activation.[47 ] Moreover, currently available VET assays are designed to detect systemic lysis only. Thus, local lysis might take place but remain unnoticed by VET methods.
Fig. 2 Different lysis patterns measured by rotational viscoelastometry. (a ) Fulminant lysis or so-called diamond of death shape. (b ) Intermediate type of clot lysis. (c ) Late lysis. (d ) Physiologic lysis. (e ) Fibrinolytic shutdown.
From an evolutionary point of view, HF counteracts shock-related microvascular stasis, microthrombosis, tissue hypoperfusion, and hypoxemia to maintain blood flow, even at the cost of an increased bleeding rate due to the dissolution of already established clots[48 ] ([Fig. 3 ]).
Fig. 3 Hyperfibrinolysis detected by EXTEM (ROTEM) upon (a ) emergency room admission. (b ) Twenty minutes after admission, no clot formation in the ROTEM analysis could be detected any more. (c ) Plasma (200 µL) collected upon admission and (d ) after 20 minutes was spiked on fibrin plates. Substantial increase in the fibrinolytic area after 20 minutes corresponds to increased lysis observed in EXTEM.[42 ]
Fibrinolytic Shutdown
Lysis and HF are strongly inhibited by PAI-1, which starts its upregulation approximately 2 hours after the initial trauma and may result in complete inhibition of clot lysis, a condition known as “fibrinolytic shutdown.”[49 ]
[50 ] Numerous studies demonstrated that mortality in patients with fibrinolytic shutdown, defined as LY30 < 0.8% in TEG or as ML < 5% in ROTEM, was higher than in patients with physiologic lysis.[51 ]
[52 ]
[53 ]
[54 ]
[55 ]
[56 ]
However, whether VET methods accurately define fibrinolysis phenotypes is still under discussion.[47 ]
[53 ]
[56 ]
[57 ] To diagnose lysis and HF, D-dimers and plasmin–antiplasmin complexes are potentially more sensitive parameters than VETs. Gall et al identified a cohort of trauma patients with high D-dimer levels and increased blood product consumption and mortality despite low ML in ROTEM.[56 ] Cardenas et al analyzed blood samples from trauma patients with TEG and measured the plasmin–antiplasmin complexes and D-dimers. A total of 89% of the shutdown patients had moderate to high fibrinolytic activation by the plasmin–antiplasmin complexes. Thus, low TEG LY30 does not reflect hypercoagulability, but a TIC with moderate fibrinolysis and fibrinogen consumption associated with poor outcomes.[57 ] Similar findings using ROTEM have been reported by David et al, who observed in the ROTEM shutdown group lower fibrinogen concentrations and higher levels of fibrin degradation products than in the patients with physiologic lysis. The authors suggested that fibrinolytic shutdown probably reflects a moderate form of coagulopathy and fibrinolysis rather than hypercoagulopathy.[53 ]
Currently, it remains to be elucidated which is the optimal way to identify fibrinolytic shutdown in major trauma patients, as the measurement of the plasmin–antiplasmin complexes is not feasible in routine clinical practice.[58 ]
In summary, the endothelium reacts uniformly to hypoxic stress and sympathoadrenal hyperactivation, with an early and robust release of tPA and cleavage of plasminogen to plasmin. PAI-1 starts to increase 2 hours after endothelial cell activation, resulting in an endogenous inhibition of lysis.[59 ]
[60 ] This delayed PAI-1 expression promotes a shift toward a hypofibrinolytic state and may lead to microvascular thrombosis and multi-organ failure.
Fibrinogen Deficiency
Fibrinogen has a molecular weight of 350 kDa and is synthesized solely in the liver.[61 ] The circulating levels range between 2 and 4 g/L in a healthy adult but can be upregulated 20-fold, mediated by infection, inflammation, and IL-6 release.[62 ] Thrombin cleaves fibrinogen to fibrin fibers, crosslinked by activated factor XIII, which increases mechanical strength and resistance to premature fibrinolytic degradation.[63 ] Moreover, fibrinogen binds with high affinity to glycoprotein IIb/IIIa receptors expressed on the surface of activated platelets, thereby facilitating further platelet aggregation.[64 ] Thus, fibrinogen plays an essential role in both primary and secondary hemostasis.[65 ]
In a severely injured bleeding patient, fibrinogen is the first coagulation factor to reach critically low levels.[4 ]
[66 ] Moreover, hypofibrinogenemia upon ER admission is associated with higher bleeding rates, increased allogenic blood transfusion requirements, and increased mortality.[67 ]
[68 ]
[69 ] A critical fibrinogen level associated with a tendency toward increased bleeding is assumed to be less than 1.5 g/L.[67 ]
[70 ] Consequently, current guidelines recommend early fibrinogen substitution, particularly when levels decline below 1.5 g/L.[71 ]
Hypofibrinogenemia in trauma patients is driven by blood loss, hemodilution, hyperfibrinogenolysis, and consumption due to clot formation at the site of injury.[65 ] Moreover, experimental studies demonstrated that hypothermia, which is common in severely injured patients, impairs fibrinogen synthesis, and acidosis accelerates fibrinogen breakdown.[72 ]
[73 ] Schlimp et al observed that fibrinogen levels upon ER admission were strongly associated with shock severity in trauma patients. If base deficit exceeded 6 mmol/L, fibrinogen plasma concentrations decreased to less than 200 mg/dL in 81% of the patients and less than 150 mg/dL in 63% of the patients.[74 ] Interestingly, the acute-phase response of fibrinogen is not downregulated by early exogenous fibrinogen substitution during initial trauma care.[75 ]
Taken together, fibrinogen is the most vital and vulnerable coagulation factor, and it reaches critically low levels earlier than other coagulation proteins. Low fibrinogen upon admission is strongly associated with poor outcomes.
Altered Thrombin Generation
Thrombin cleaves fibrinogen to fibrin and activates factor XIII (FXIII), platelets, endothelial cells, and leucocytes. When bound to thrombomodulin, thrombin activates the protein C pathway and becomes an anticoagulant factor.[38 ] Immediately after initial tissue trauma, TG is strongly upregulated to create sufficient clots for quick termination of blood loss.[76 ]
[77 ]
TG can be altered by several trauma-related mechanisms, such as loss and consumption of coagulation factors, dilution, hypothermia, and acidosis ([Fig. 1 ]).[78 ]
[79 ] Studies in severely injured patients demonstrated that factor V, factor VII, and factor IX are predisposed to low levels.[66 ]
[80 ]
[81 ] Nevertheless, experimental and clinical studies have demonstrated that TG remains unaffected.[80 ]
[82 ]
Cardenas et al reported that trauma patients upon ER admission had significantly higher TG parameters than uninjured subjects.[76 ] Only 17% of the patients demonstrated a peak TG of less than 250 nM. However, these patients required more allogeneic blood products, had a fourfold increased risk of massive transfusion, and a threefold increased risk of mortality.[76 ] Coleman et al also observed high-volume blood transfusion in trauma patients with compromised TG.[83 ]
Hypercoagulability was also reported by Schreiber et al in 62% of the investigated trauma patients in the first 24 hours after injury, with a female predilection.[84 ] Hypercoagulability might be related to tissue factor exposure and the additional release of procoagulant microparticles and damage-associated molecular patterns.[85 ] Thus, it is highly questionable whether an initial augmentation of TG should be considered a primary goal of early hemostatic management in major traumas.[86 ]
In summary, major tissue trauma creates an initial procoagulant environment driven by tissue factor exposure and the release of procoagulants, resulting in a substantial upregulation of TG. At a later stage, TG can be altered by shock-related activation of the protein C pathway, dilution and consumption of the coagulation factors, hypothermia, and acidosis.
Platelet Dysfunction
Platelets play a vital role in initial clot formation. Activated platelets adhere to the subendothelial collagen of damaged tissues and provide the surface for the assembly of clotting factors to further amplify TG.[87 ] Moreover, platelets are involved in inflammation and wound healing by recruiting immune cells from the circulation in a P-selectin-dependent mechanism.[88 ]
There is a growing body of evidence that severe trauma not only affects plasmatic coagulation factors but also compromises platelet function. Platelet dysfunction occurs early after initial tissue injury despite a normal platelet count.[15 ]
Numerous studies have demonstrated that even mildly impaired platelet aggregation in response to different platelet agonists, such as thrombin and adenosine diphosphate receptor stimulation, is associated with poor outcomes.[15 ]
[16 ]
[89 ]
[90 ]
[91 ]
[92 ] In a retrospective study, Solomon et al analyzed platelet function in major trauma patients after ER admission. Decreased platelet aggregation assessed by Multiplate was associated with increased mortality.[90 ] Kutcher et al also measured platelet function by Multiplate aggregometry in severely injured patients on admission and during their hospital stay. Despite a normal platelet count, platelet dysfunction was observed in 45% of patients on admission and in 91% during their hospital stay.[15 ]
The exact mechanism that promotes platelet dysfunction following trauma remains to be elucidated. Verni et al spiked healthy platelets with plasma collected from trauma patients and detected a significantly diminished response to multiple platelet agonists. The authors suggested that soluble plasma species may downregulate various platelet activation pathways.[91 ] Another hypothesis suggested that platelets are captured by leukocytes, which was linked to impaired platelet function detected by Multiplate.[17 ] Vulliamy et al observed a ballooning of platelets induced by histone H4, a damage-associated molecular pattern, which is released in massive quantities after severe injury.[93 ]
Taken together, platelet dysfunction occurs early after severe trauma independently of platelet count. This initial inhibition of platelet function has been linked to adverse outcomes. Interestingly, recent studies in major trauma patients could not demonstrate a clear clinical benefit of early estimation of platelet dysfunction by Multiplate or TEG platelet mapping to improve outcome.[94 ]
[95 ]
[96 ]
