The Pathophysiology and Management of Acute Traumatic Coagulopathy

The Activation of PC

Protein C is an important substance in the anticoagulation system. Normally, PC is in a nonactive state and can be converted into activated PC (aPC) by the thrombin–thrombomodulin (T–TM) complex. The aPC inhibits the extrinsic pathway by inactivating factor V and VIII. Brohi et al have proposed that the systemic anticoagulation of ATC was caused by the activation of PC. ²⁰ They found that low PC and high TM levels were associated with poor outcomes among patients with major trauma having hypoperfusion featured by enhanced base deficit. Also, low PC level was associated with prolongation of PT, aPTT, and hyperfibrinolysis with low levels of plasminogen activator inhibitor 1 (PAI-1), suggesting the activation of PC in the ATC, although the aPC levels were not measured. In a prospective cohort study, aPC–PC ratio was used to reflect the activation of PC, and the elevated value was found to be significantly associated with the derepression of the fibrinolysis, as well as the outcomes among patients with hypoperfusion, further supporting the activation of PC. ²¹ In the study, a dynamic PC level was tested, showing that patients with the greatest reduction in the early 12 hours followed by no recovery or more depletion in the coming 12 hours had higher risks of ventilator-associated pneumonia or other complications. Consistent with these findings, monoclonal antibody, which inhibits the anticoagulant function of PC, pretreated mice showed an attenuated coagulopathy and relatively normal histological tissues. However, those pretreated by antibody blocking both the anticoagulant and the cytoprotective functions of aPC had much higher mortality and severe pulmonary thrombosis. ²² Besides, some venous thromboembolism occurred several hours or days after severe injury was inferred due to the exhaustion and impairment of further generation of PC. ²⁰ Thus, PC may have a complicated role in the process of coagulation and response to trauma or shock. However, the exact mechanism of PC pathway activation, such as the trigger of PC activation and related regulatory factor need further study to be verified. Most present studies focus on the fluid phase of the hemostatic system, such as the concentration of several plasma elements, neglecting the local site changes. The activation of PC needs its interaction with 2 receptors, thrombomodulin and the endothelial PC receptor, ²³ so the histological or genetic changes may need to be further validated.

The Role of Hypoperfusion, Glycocalyx, and Catecholamine

It is well known that vascular endothelium has a critical role in the normal hemostasis. Upon vessel damage or trauma, endothelium can contract and provide a binding site for platelet assembling to localize the coagulation as well as release some factors (such as von Willebrand factor [vWF] and adenosine diphosphate) to stimulate platelet. Several studies have shown a well-established association between endothelium dysfunction and the outcome in patients with sepsis, similar to ATC, which is characterized by extensive neurohumoral, inflammatory, and coagulation activation. ^24,25 In the vessel lumen, the endothelium bears the endothelial glycocalyx, a 0.2- to 1-μm thick negatively charged antiadhesive and anticoagulant carbohydrate-rich surface layer that protects the endothelium and maintains the vascular barrier function. ^{26 –28} And a dynamic equilibrium exists between the soluble components of glycocalyx and the plasma component of blood. ²⁹ Previously, Rehm et al have demonstrated that the ischemia/reperfusion injury during the aortic surgery can cause an immediate increase in circulating syndecan 1, a marker of endothelial glycocalyx degradation. ³⁰ In accordance with this, Johansson et al have found that high circulating syndecan 1 is associated with inflammation, coagulopathy, and increased mortality among patients with trauma and speculated that these changes were induced by neurohumoral system activation in the early phase of trauma and would cause downstream derangements. ³¹ It should be noted that heparin sulfate, a major component of glycocalyx, increases simultaneously with sydecan 1 during degradation of glycocalyx. ^30,31 Furthermore, a study has shown endogenous heparinization, detected by TEG, in patients with ATC appeared to be caused by the degradation of endothelial glycocalyx. ³² Taken together, the degradation of glycocalyx seemly has some relationship with the hypocoagulation state, but more convincing evidences are demanded to support this theory.

From an evolutionary prospective, Johansson et al hypothesized that the pro- and anticoagulant state of solid and fluid phase was regulated by trauma and sympathoadrenal response. ³³ Tissue injury and high levels of catecholamine cause endothelium damage, glycocalyx degradation, and exposure of tissue factor and vessel collagen, resulting in a local procoagulant state. In proportional to the endothelium injury, circulating anticoagulant molecules (glycocalyx, tissue type plasminogen activator [tPA], urokinase, soluble thrombomodulin [sTM], and aPC), mainly released from the solid phase (vessel walls and tissues), increase to prevent microvascular thrombi and maintain the oxygen transport. In addition, previous study reported a dose-dependent stimulation of factor VIII, vWF, tPA, and platelets within a 15- to 40-minute infusion of epinephrine, mainly released into circulation through β2 receptor pathway, ³⁴ suggesting a critical role of catecholamine in the coagulation process. However, the hypothesis that catecholamine modulated traumatic coagulopathy needs more experiments to validate.

Reduction in Fibrinogen and Augmented Fibrinolysis

The role of factors consumption in ATC is debated. Several studies have inconsistently identified the elevation of prothrombin fragment 1 + 2, reflecting thrombin generation in patients with ATC; however, the fibrinogen levels detected varies greatly leading to different opinions. ^12,17,20,35 Martini et al have conducted a series of experiments to show deficit of fibrinogen availability in traumatic models. ^{36 –38} Either accelerated fibrinogen breakdown or inhibited fibrinogen synthesis, due to hemorrhage, hypothermia, or acidosis, may have contributed to the change. ³⁹

Fibrinolysis is a key component of this coagulopathy. The magnitude of fibrinolysis, often reflected by d-dimers or fibrin degradation product, is elevated with the severity of injury, ^35,40 suggesting trauma as an independent trigger. As to the mechanism of hyperfibrinolysis in traumatic coagulopathy, there are still some debates over the PC pathway activation (consuming PAI-1) or reduction in thrombin activated fibrinolysis inhibitor (TAFI), ^41,42 since both can augment fibrinolysis and have a binding competition to T–TM. Given the result that PAI-1 has a more related inverse correlation with d-dimer than TAFI among patients with ATC, some researchers speculated that fibrinolysis was due to disinhibition of tPA through the consumption of PAI-1, resulting from PC activation. ³⁵ Similar to the anticoagulation pathology, T–TM was thought to be the mediator of hyperfibrinolysis through PC pathway. ²⁰ However the local T–TM complex level or activity has never been measured, and plasma sTM is often used to reflect endothelial-bound TM, where some arguments about the correlation still exist. ⁴³ The mechanism for rapid increase in plasma sTM or endothelial presentation remains speculative and needs further study. Considering the short time span of coagulant change in patients with ATC (about half an hour), ³⁵ physical endothelial damage or active release rather than transcriptional upregulation is more likely accounting for this change.

Sawamura et al have proposed that, in addition to tPA, neutrophil elastase was also involved in the fibrinolysis. ⁴⁴ Overt high levels of neutrophil elastase as well as decreased α2-antiplasmin were detected among patients with trauma on admission. ⁴⁵ Apart from reducing the 2 plasmin inhibitors, neutrophil elastase may directly degrade several clot factors including fibrinogen, causing a hyperfibrinolytic state. In addition to the primary systemic fibrinolysis (fibrinolytic activation without formation of a fibrin clot) described previously, secondary fibrinolysis also occurred in response to the deposition of fibrin. ⁴⁶

Dysfunction of Platelet

The newly developed cell-based model of coagulation emphasizes the importance of platelet activation through the hemostasis process, especially in the amplification and propagation phases. ^47,48 Platelet can be activated by thrombin previously generated in the TF-bearing cells via protease-activated receptor 1 (PAR1) and PAR4, ^49,50 thus providing a phospholipid membrane for the formation of tenase (FVIIIa/FIXa) and prothrombinase (FVa/FXa). ⁴⁷ Platelet hypofunction, defined as below-normal platelet response to agonist, can be observed in patients with trauma although the platelet count showed little differences. ⁵¹ Hyperfibrinolysis may affect the platelet function, as plasmin could cleave receptors on platelet such as glycoprotein Ib and IIb/IIIa. ⁵² Other combined effects of acidosis and hypothermia may also make contributions. However, the exact underlying mechanism is not clear. Besides, whether the circulating platelet studied truly reflect the “trauma-induced” platelet remains to be seen, as amounts of platelets would replenish into circulation from spleen and liver when blood loss occurs.

Participation of Inflammation

Trauma is a strong trigger of inflammation. Recent studies tend to consider coagulation as an integral part of inflammation, ⁵³ and an increasingly close relationship may exist between the inflammation, coagulation, and fibrinolytic system. ⁵⁴ In patients with severe injury or sepsis, coagulation activation may lead to a spillover release of inflammatory cytokines, which could in turn further activate systemic coagulation. ^46,53 Finally, many studies have described the participation of toll-like receptors 4 (TLR4) in the acute trauma response, such as hemorrhagic shock and ischemia/reperfusion. ⁵⁵ Those cells, which express TLR including monocyte/macrophages, dendritic cells, epithelium, and endothelium, are all involved in injury response, inflammation, and coagulation, ⁵⁶ suggesting a complex interplay between these reactions. Therefore, inflammation may be highly involved in the onset and development of ATC process, but further research is needed to ravel the cause–effect relationship between the 2 responses.

Classic Triad Theory

Hypothermia, acidosis, and dilution are classically considered to be the 3 initiators for traumatic coagulopathy. However, recently these factors are recognized more responsible for the subsequent coagulant derangement rather than the early alteration. Mild hypothermia (35°C-33°C) can be seen in some patients with trauma as a result of environmental exposure, hypoperfusion, and cold fluid administration. ⁵⁷ Previous studies have demonstrated that severe hypothermia (<33°C) could reduce coagulant factor activity (TF, FVIIa complex) and fibrinogen synthesis ⁵⁸ as well as decrease platelet activation, aggregation, and adhesion, which is due to a reduced effect of vWF traction on glycoprotein Ib/IX. ⁵ However, the prevalence of such low temperature is low in patients with trauma. ^10,57 Therefore, hypothermia alone may have a limited impact on hemostasis in patients with ATC. Acidosis can be caused by hypoperfusion. Severe acidosis (pH < 7.1) could prolong PT and aPTT ⁵⁹ and inhibit thrombin generation as well as increase degradation of fibrinogen. ^5,58 However, the coagulant state cannot be reversed even after correcting acidosis by buffer solution, ⁶⁰ suggesting the importance of acidosis causing ATC needs to be reevaluated. Obviously, dilution contributes to coagulopathy in trauma and several studies have shown the close relationship in patients with trauma. ^1,61 In the early stage, fluid administration is often small or limited in some countries, ¹² thus the critical role of hemodilution in the early stage is questioned.

Fresh Frozen Plasma Transfusion

Severe bleeding due to traumatic coagulopathy always requires MT, although the definition of MT is inconsistent (10+ units packed red blood cell [PRBC]/24 hours, 100% blood loss in 24 hours, etc). ⁶⁵ Recently, FFP and PRBCs are widely used in practice to address ATC. High doses of FFP (10-20 mL/kg) are recommended to control the severe traumatic bleeding as soon as possible, yet there are few published evidence-based guidelines. ⁶⁶ And the optimal ratios of FFP and PRBC varies (from 1:3 to 1:1) according to different researchers. ⁶⁵ However, due to lack of prospective randomized controlled trials (RCTs), the true efficiency of these strategies remains to be further elucidated. Despite its frequent usage, the risks of transfusion should be noted, such as immunomodulation, nosocomial infection, even acute respiratory distress syndrome, and multiorgan failure. ¹³

The administration of FFP is under the consideration that the coagulation factors are impaired during ATC, regardless of consumption or loss, aiming at augmenting the concentration and capacity of clotting factors. Based on the theory of aPC pathway, when TM is presented or expressed in excess, the replenished thrombin will activate more PCs, therefore stable clot can hardly be formed. On the other hand, FFP may have a potential effect of restoring the glycocalyx and reducing endothelial inflammation and hyperpermeability, all of which may contribute to the benefits gained after FFP transfusion. ²⁹ In fact, blood loss due to trauma or subsequent complications is inevitable, and the hemostatic condition is out of normal range, regardless of pathologically elevated fibrinolytic molecules or damaged anticoagulant substances. The replenishment of FFP could buffer the “abnormal plasma” toward a physiological and controlled state, thus correcting the ATC.

Platelet Supplement

Another commonly used hemorrhage control therapy is platelet transfusion to redress the potential platelet dysfunction or dilutional thrombocytopenia from MT, although some studies have shown that platelet count is maintained above critical levels. ⁶⁷ The targeted goal of platelet level was set to be >50 × 10⁹/L in polytrauma or >100 × 10⁹/L in the central nervous system injury. ^13,65 Several retrospective studies have shown improved results with high platelet–RBC ratio (eg, >1:5 or >1:2), ⁶⁵ but others have found that platelet transfusion may not be essential. ¹¹ The lack of convincing evidence prompts further studies on the platelet administration in patients with ATC.

Fibrinogen Administration

Although fibrinogen has shown reduction to some extent in several animal models of ATC or retrospective studies, ^68,69 it is still controversial whether fibrinogen has declined to the critical level for adequate hemostasis. ^11,65 One retrospective study of 252 combat-related patients who received MT has shown a better result in high concentrations fibrinogen group (≥0.2 g per unit of PRBC). ⁷⁰ Restoration of fibrinogen could mitigate coagulation disorder and reduce blood loss in the animal model, but the difference was not significant between the high- and the low-dose groups on blood loss. ⁶⁹ Two observational studies have demonstrated improved outcome or reduced transfusion requirements in the fibrinogen-treated group, ^71,72 but the evidence is low, considering the criteria of logistic or screening test difference. Most researchers have suggested supplementation of fibrinogen concentrate (3-4 g) or cryoprecipitate (50 mg/kg) to ensure that fibrinogen level was more than 1 g/L ⁶⁵ or increase fibrin-based clot strength clot firmness at 10 minutes (FIBTEM CA10) to 10 to 12 mm. ¹³

Recombinant Activated Factor VII and Tranexamic Acid

Recombinant activated factor VII (rFVIIa) is another resort to treat trauma-induced coagulopathy, which was first reported in 1999. ⁶⁰ Incipiently, rFVIIa was approved for the treatment of bleeding in patients with hemophilia A or B. Numerous anecdotal reports have described the efficiency of rFVIIa in the treatment of ATC, ⁶⁰ and an off-label usage can be commonly seen (4% or even higher). ⁷³ In a review of 13 trials comprising 1938 patients, Stanworth demonstrated reduced blood requirement in patients receiving rFVIIa (relative risk [RR] = 0.85) with a elevated thromboembolic event (RR =1 .25). ⁷⁴ D’Angelo also reported 2 doses of rFVIIa administration for patients with active ongoing hemorrhage with a high dose (100μg/kg, with shock) and low dose (50μg/kg, without shock), both resulting in favorable endings. ⁶⁰ However, recently 2 large RCTs have shown few benefits and modest reduction in blood usage after rFVIIa application, ^75,76 thus leading us to reconsider the efficiency and risks of the miracle drug. After analyzing data from 35 RCTs containing 4468 patients on the off-label basis, Levi et al found increased arterial thromboembolic events in patients receiving rFVIIa (5.5% vs 3.2%) than placebos, especially high among the elderly patients (>65 or 75 years), although the rates of venous thromboembolic events are comparable. ⁷⁷ Reaching its peak in 2006, the usage of rFVIIa in the US military is dropping in recent years. ⁷⁸

An “old” antifibrinolytic drug, tranexamic acid (TXA), is now highly recommended for the early hemorrhage control after sever trauma. By blocking the lysine-binding sites on plasminogen, TXA can reduce blood transfusion as well as blood loss without obvious complications in patients undergoing elective surgery. ⁷⁹ Similarly, in patients with trauma or at risk of substantial bleeding, a multicenter trial evaluating 20 211 patients, CRASH-2, has shown reduced bleeding-caused mortality (4.9% vs 5.7%, RR = 0.85) in the TXA group treated with 2 g TXA within 8 hours. ⁸⁰ In the following subgroup analysis, ⁸¹ the importance of early administration of TXA was emphasized. Treatment within 3 hours, particularly less than 1 hour, can significantly reduced the risk of death due to bleeding (RR = 0.68 within 1 hour, 0.79 for 1-3 hours). Of note, for the late administration (>3 hours) the harm of TXA would outweigh the benefits. In concert with the results of CRASH-2, a big retrospective observational study including 896 soldiers with combat injury, namely, Military application of tranexamic acid in trauma emergency resuscitation (MATTERs) study, ⁸² showed lower mortality in the TXA group than in the non-TXA group (17.4% vs 23.9%). The benefit was even greater among patients who need MT.

In the CRASH-2 study, however, the correlative laboratory data have not been collected, limiting the insight of the pathophysiology of TXA effect. The exact mechanism underlying the phenomenon is unclear, probably repressing fibrinolysis in the early stage and reducing uncontrolled clot factor consumption, thus lowering transfusion requirements. In fact, the TXA group showed no increase in vascular occlusive events, including pulmonary embolism and deep vein thrombosis, even after attaining a reduction in myocardial infarction (P = .035). The safety maybe was related to the dose and regime, which present the topic for future investigation. Anne Godier has innovatively hypothesized that TXA has an antithrombotic effect, via inhibition of the inflammatory effects of plasmin, on platelet and factor V and VIII. ⁸³ Downregulation of inflammatory response may be achieved, ⁸⁴ for a broad spectrum of proinflammatory responses can be induced by plasmin by binding and activating monocytes, neutrophils, platelets, and endothelial cells via a cytokines-releasing or gene transcription way. ⁵² In the traditional view, inflammation induced by trauma or other chronic disease could activate coagulation, reducing blood loss or even increasing the risk of thrombosis. If inflammation of ATC has been compressed, the stimulator for coagulation could be weakened in some degree. So the importance of inflammation in the ATC, especially on TXA administration, is of a new field worthy of deep exploration.

However, the potential side effect of TXA, such as thrombosis, should be noted, since the coagulant state could change into hypercoagulation in the late phase, and several cases of postoperative convulsive seizures have been reported after cardiac surgery using TXA, referring to the structural similarity between TXA and γ-aminobutyric acid. ⁵² Other antifibrinolytic agents, such as 6-aminocaproic acid, aprotinin and neutrophil elastase inhibitor, may have a positive role in this early situation but need further studies. ⁴⁶