Coagulopathy in trauma patients

INTRODUCTION — Trauma remains a leading cause of death and disability in adults in spite of advances in resuscitation, surgical management, and critical care [1]. Even though improved efficiency of trauma systems military and civilian has reduced the time interval between acute injury and treatment, between 25 and 35 percent of injured civilian trauma patients develop a biochemically evident coagulopathy upon arrival in the emergency department [2-4].

Coagulopathy may be the result of physiologic derangements such as acidosis, hypothermia, or hemodilution related to fluid or blood administration; however, an acute coagulopathy can also occur in severely injured patients independent of, or in addition to, these factors [2,3]. Several terms are used in the literature to refer to this latter condition, including trauma-induced coagulopathy (TIC), acute traumatic coagulopathy (ATC), early coagulopathy of trauma (ECT), and the acute coagulopathy of trauma-shock (ACoTS) [2,3,5-7]. Based on the terminology used by the International Society for Thrombosis and Haemostasis, we will use the term TIC in this review [8].

The etiology, diagnosis, and treatment of coagulopathy associated with trauma will be reviewed here. The general principles of shock management in the trauma patient and the treatment of excessive anticoagulation related to medical treatment are discussed elsewhere. (See "Initial management of moderate to severe hemorrhage in the adult trauma patient" and "Management of warfarin-associated bleeding or supratherapeutic INR".)

IMPACT — Coagulopathy in trauma patients, and specifically trauma-induced coagulopathy as an acute systemic phenomenon, is associated with higher transfusion requirements, longer intensive care unit and hospital stays, more days requiring mechanical ventilation, and a greater incidence of multiorgan dysfunction.

Approximately 25 percent of patients admitted to civilian level I trauma centers receive transfused blood, with approximately 80 percent of these receiving fewer than 10 units. Massive transfusion (traditionally defined as >10 units packed red blood cells over 24 hours) is required in 2 to 3 percent of injured patients [9].

Compared with patients who do not have coagulopathy, those with coagulopathy have a threefold to fourfold greater mortality and are up to eight times more likely to die within the first 24 hours following injury [2-4,10,11]. Most deaths in trauma patients due to in-hospital hemorrhage occur within six hours of admission [12,13].

ETIOLOGIES — The etiology of coagulopathy in the injured patient is multifactorial with overlapping contributing etiologies depending upon the injury and nature of resuscitation. Normal coagulation is a balance between hemostatic and fibrinolytic processes that permit control of bleeding following mild injury while preventing inappropriate intravascular thrombosis. (See "Overview of hemostasis".)

Etiologies that upset the normal balance include classic elements of the "vicious triad": acidosis related to tissue injury and shock, hypothermia from exposure and fluid administration, and hemodilution due to fluid or component blood product administration. Systemic consumption of clotting factors manifesting as disseminated intravascular coagulation may occur early after injury due to inadequate clotting factor repletion in the face of ongoing consumption, or later in the hospital course triggered by secondary insults (eg, sepsis). Distinct from these elements, trauma-induced coagulopathy (TIC) is a multifactorial biochemical response to tissue injury and shock mediated by dysregulated coagulation, altered fibrinolysis, systemic endothelial dysfunction, inflammatory responses to injury, and platelet dysfunction. Injury to brain tissue may predispose to TIC, and approximately one-third of patients with traumatic brain injury (TBI) have a coagulopathy, although whether TBI-associated coagulopathy is fundamentally different from injury-related coagulopathy is not clear [14-16]. (See 'Trauma-induced coagulopathy' below.)

A prospective study of injured patients with blunt trauma and hemorrhagic shock enrolled in the Host Response to Injury Large-Scale Collaborative Program attempted to characterize the relative contributions of multiple risk factors to coagulopathy in 578 of the 1537 patients (37.6 percent). In this study, more than 80 percent of patients with coagulopathy on arrival had risk factors related to elements of the classic vicious triad as well as to TIC. The authors found that, even when adjusting for all of these risk factors in multivariate analysis, arrival coagulopathy remained an independent predictor of multiorgan failure and mortality, implying that additional biochemical factors driving coagulopathy remain to be discovered [17].

Acidosis — Inadequate tissue perfusion in patients with hypovolemic shock due to bleeding leads to metabolic (lactic) acidosis, which can be exacerbated by excessive chloride and component blood administration. (See "Definition, classification, etiology, and pathophysiology of shock in adults".)

Acidosis causes demonstrable clotting dysfunction in experimental models at pH <7.2 by interfering with the assembly of coagulation factor complexes involving calcium and negatively charged phospholipids [18-21]. As an example, the activity of the factor Xa/Va/phospholipid/prothrombin ("prothrombinase") complex is reduced by 50, 70, and 90 percent at a pH of 7.2, 7.0, and 6.8, respectively. However, correction of acidosis alone does not always correct the associated coagulopathy, indicating that tissue injury causes coagulopathy via additional mechanisms [22,23].

Hypothermia — Hypothermia following injury is due to cold exposure at the time of injury, during transport, and during the trauma examination compounded by the administration of cold intravenous fluids. Nearly two-thirds of trauma patients have a temperature below 36°C on presentation; 9 percent of trauma patients have a temperature at or below 33°C [20,24-27]. Hypothermia in injured patients is graded as mild (36 to 34°C), moderate (34 to 32°C), or severe (<32°C) [28].

Patients who require surgery are at a greater risk for hypothermia due to further physical exposure in the operating room, additional fluid administration, and the effects of general anesthesia. Injured patients with hypothermia generally have worse outcomes compared with non-injured patients with hypothermia; however, hypothermia alone is a weak independent predictor of mortality [28,29]. Acidosis and hypothermia are synergistic with increased mortality when both are present, compared with either alone [30].

The effect of hypothermia on clotting includes platelet dysfunction and impaired enzymatic function. Overall thrombin generation in activated in vitro clotting systems is generally preserved at a temperature of 33°C; however, impairment of tissue factor activity, platelet aggregation, and platelet adhesion is evident at temperatures between 33 to 37°C [20,24]. It is important to note that no effects on coagulation tests (either standard or viscoelastic) are seen in hypothermia-induced coagulopathy without special sample handling due to the standard practice of prewarming blood samples to 37°C prior to analysis, which corrects the defect. In other words, the result of coagulation testing reflects clotting characteristics that the patient would have if his/her temperature were 37°C.

Specific measures to correct hypothermia include controlling physical exposure, administration of warmed fluids, and passive rewarming with blankets and forced-air devices. Rapid identification and control of bleeding is vital to preserve normal temperature. Continuous temperature monitoring is essential to ensure that mild hypothermia does not worsen. In the case of moderate or severe hypothermia and coagulopathy, central rewarming may be needed. (See "Perioperative temperature management".)

Resuscitation-associated coagulopathy — Resuscitation-associated coagulopathy (RAC), also now termed iatrogenic coagulopathy, refers to alterations of the coagulation system induced by large volumes of intravenous fluids or unbalanced component blood administration during the management of shock [31]. The age of blood may contribute to RAC.

Trauma resuscitation historically focused on the treatment of hypotension and acidosis with aggressive crystalloid resuscitation followed by packed red blood cells (PRBCs). At that time, treatment of coagulopathy was initiated only in response to abnormal standard coagulation tests. Similarly, platelet transfusion was generally not performed until laboratory evidence of thrombocytopenia was present. Computer modeling [32], in vitro experiments [33], and clinical studies in healthy volunteers confirm that large-volume resuscitation with crystalloid, colloid, and packed red blood cells leads to dilution of plasma clotting proteins [34]. A retrospective study of 8724 injured patients from the German Trauma Registry found a positive correlation between prehospital fluid resuscitation volume and coagulopathy [4]. Coagulopathy was present on admission in more than 50 percent of patients who received >3 L of intravenous fluid prior to arrival, but coagulopathy was also present in 10 percent of patients administered <500 mL and in 32 patients who had received no prehospital fluids, which appears to reflect the frequently overlapping contribution of TIC. (See 'Trauma-induced coagulopathy' below.)

Prolonged storage of packed red blood cell units prior to transfusion has been previously suggested to significantly contribute to RAC. The effects of prolonged storage (referred to as the "storage lesion") include decreased pH, chelation of calcium, low 2,3-diphosphoglycerate levels, and decreased clotting factor concentration. The median duration of storage of a unit of red blood cells in the United States is approximately 15 days, with older units frequently allocated to high-use facilities such as trauma centers [35]. Thus, it was suggested that transfusion of older blood impaired microvascular perfusion and had inflammatory and immunomodulatory effects, which would be particularly relevant for massively transfused trauma patients. While early observational studies suggested that an older storage age of transfused PRBCs was linked to higher morbidity and mortality, the available data suggest similar mortality for transfusion of "fresh" red blood cell units (stored for fewer days) compared with "standard issue" red blood cell units (stored for the usual number of days) [36-41]. Overall, given the logistical and resource challenges related to preferential use of "fresh" red blood cells in the context of a lack of high-quality data to support this practice, standard blood age-related transfusion practices are appropriate for injured patients and are generally not felt to exacerbate underlying coagulopathy in a clinically relevant manner. (See "Practical aspects of red blood cell transfusion in adults: Storage, processing, modifications, and infusion", section on 'Changes during in vitro storage' and "Practical aspects of red blood cell transfusion in adults: Storage, processing, modifications, and infusion", section on 'Donor and component characteristics potentially affecting recipient outcome'.)

Disseminated intravascular coagulation — Disseminated intravascular coagulation (DIC) is a systemic process producing a consumptive coagulopathy in concert with diffuse microvascular thrombosis (table 1). In trauma patients, tissue-injury-induced exposure of tissue factor and activation of the extrinsic coagulation cascade leads to thrombin generation proportional to injury severity [42]. In addition, systemic embolism of tissue-specific thromboplastins from sites of injury (including bone marrow lipid material, amniotic fluid, and brain phospholipids) may predispose patients to DIC [43]. (See "Evaluation and management of disseminated intravascular coagulation (DIC) in adults", section on 'Acute versus chronic DIC'.)

Trauma-induced coagulopathy — TIC is an impairment of hemostasis and activation of fibrinolysis that occurs early after injury and is biochemically evident prior to, and independent of, the development of significant acidosis, hypothermia, or hemodilution. The risk of TIC increases with hypotension, higher injury severity score, worsening base deficit, and head injury [3,10,11]. Once established, TIC is often compounded by the other etiologies of coagulopathy discussed above. (See 'Acidosis' above and 'Hypothermia' above and 'Resuscitation-associated coagulopathy' above.)

Patients with TIC frequently meet criteria for DIC, and some authors have argued that TIC may represent an early, partially compensated stage of DIC [31,44,45]. However, the concept of DIC as a final common pathway for several different phenomena is insufficient to explain the hematologic abnormalities post-injury [10,46,47]. Coagulopathy in the absence of thrombocytopenia and hypofibrinogenemia, as seen in TIC , argues against consumption as a necessary underlying mechanism [22] Although D-dimer levels are frequently elevated and fibrinogen levels depleted in acutely injured patients, indicating intravascular fibrin deposition and active fibrinolysis [48], functional thrombin generation (assayed by the presence of prothrombin fragments and thrombin-antithrombin complexes) remains intact [10,49-51]. Furthermore, TIC occurs only when tissue injury is combined with systemic hypoperfusion. Thus, it is most likely that TIC is mechanistically distinct from DIC but that these frequently overlap. Exploring this distinction is an area of ongoing research.

Dysregulated coagulation — Under normal circumstances, tissue injury leads to thrombin generation, fibrin deposition, and clot formation via the extrinsic pathway (figure 1). The enzymatic pathways making up the coagulation cascade are discussed in detail elsewhere. (See "Overview of hemostasis".)

Initiation of the clotting process is localized to the site of tissue injury. Systemic coagulation due to the escape of thrombin from the injury site is inhibited by circulating antithrombin III, or by the binding of thrombin to constitutively expressed thrombomodulin on nearby undamaged endothelial cells [52]. Protein C, a systemic anticoagulant, is proteolytically converted from an inactive zymogen to activated protein C (aPC) by the complex of thrombin with thrombomodulin. aPC is a serine protease that proteolytically inactivates factors Va and VIIIa and depletes plasminogen inhibitors (figure 2 and table 2) [53,54]. In this manner, aPC can serve a protective function by inhibiting thrombosis during periods of decreased flow.

Initial observations in hypoperfused trauma patients found a correlation between TIC and elevated levels of aPC, reduced levels of non-activated protein C, and elevated soluble thrombomodulin [10,55]. In a multicenter observational study of 165 critically injured trauma patients with serial plasma clotting factor analysis, injury severity and shock were associated with elevation of aPC and reduction of all analyzed clotting factors. Multivariate analysis identified deficits in fibrinogen, thrombin, Factor V, Factor VIII, Factor IX, Factor X, and aPC levels as principal drivers of coagulopathy [56].

Alterations in fibrinolysis — Endothelial release of tissue plasminogen activator (tPA) is stimulated by both hypoperfusion-related hypoxia, as well as the negative feedback mechanism associated with thrombin generation. Consumption of endogenous plasminogen activator inhibitor-1 (PAI-1) by TIC-mediated aPC further destabilizes the fibrinolytic balance, leading to uninhibited tPA-mediated conversion of plasminogen to plasmin [57]. Diversion of thrombin to protein C activation may also reduce activation of thrombin-activatable fibrinolysis inhibitor (TAFI), further enhancing fibrinolytic activity [58]. These mechanisms lead to the relative hyperfibrinolysis seen in trauma patients with TIC, which is reflected in increased levels of tPA, decreased PAI-1, and increased D-dimer [10]. (See 'D-dimer' below.)

Systemic endothelial dysfunction — Tissue injury and shock are also associated with degradation of the endothelial glycocalyx, a protective endothelial layer. Loss of glycocalyx integrity is reflected by the systemic release of syndecan-1, a glycocalyx degradation product, which has been shown to correlate with coagulopathy and mortality [59,60]. Shedding of endogenous heparan sulfates from the glycocalyx can also lead to auto-anticoagulation via increased circulating endogenous heparinoids [61].

The overall degree of glycocalyx damage correlates with post-injury catecholamine levels [62]. Profound activation and consumption of protein C can deplete protein C stores, potentially leading to reduced endothelial protective signaling via the aPC receptors protease-activated receptor-1 (PAR-1) and endothelial protein C receptor (EPCR) independently of aPC's role as an anticoagulant, potentially exacerbating endothelial dysfunction [63]. Suggesting the clinical relevance of this mechanism, one prospective study noted that protein C depletion in trauma patients correlated with elevated markers of endothelial injury and coagulopathy and with a threefold higher risk of mortality [59]. Several ongoing studies are evaluating the relevance of the protein C system to endothelial activation and barrier permeability [64-67].

Inflammatory responses to injury — APC also has anti-inflammatory and cytoprotective effects. In a single-center study of 203 critically injured trauma patients, early coagulopathy was linked to high levels of aPC, and later, protein C depletion as early as six hours after injury [55]. Patients who demonstrated protein C depletion had a significantly increased risk of acute lung injury, ventilator-associated pneumonia, multiorgan failure, and death. [63]The importance of these mechanisms is suggested by separate murine models in trauma and sepsis. In these experiments, selective antibody-mediated inhibition of the anticoagulant function of aPC reduced the rate of coagulopathy, but not mortality, while inhibition of anticoagulant and cytoprotective functions increased mortality after a challenge with trauma/hemorrhagic shock [53], or injection of lipopolysaccharide [68]. Cytoprotective functions of aPC may also play a role in pulmonary capillary endothelial barrier function as suggested by in vitro studies [64,69], and human studies associating persistently low protein C levels in critically injured, mechanically ventilated trauma patients with increased rates of pneumonia [70]. Ongoing studies are evaluating the interactions of the protein C system and innate [71,72] and cellular immunity [64-67,73,74].

Platelet dysfunction — Platelets play a pivotal role in hemostasis after injury [75]. Platelet count at the time of admission has been noted to inversely correlate with transfusion and early mortality in injured patients, even for platelet counts well in the normal range [76]. Such quantitative platelet defects also correlate with progression of intracranial hemorrhage and mortality after traumatic brain injury [77]. Several descriptive studies have also noted qualitative platelet dysfunction in trauma patients, reflecting functional platelet impairment that is independent of platelet count. In a prospective study of impedance aggregometry in 101 trauma patients, primary platelet dysfunction (ie, not due to preinjury aspirin or clopidogrel use) occurred in 46 percent of patients on admission, and 91 percent of patients at some point during their hospital stay [78]. Similar results were seen in studies of brain-injured trauma patients, which identified that aspirin-induced, but not primary trauma-induced, platelet dysfunction improved with platelet transfusion [79,80]. However, other studies question the clinical utility of platelet aggregometry, reporting that aggregometry does not correlate with the need for platelet transfusion [81], and that platelet transfusion does not in fact improve aggregometry results in injured patients [82]. Further studies on the clinical utility of platelet functional testing and on optimal therapy for platelet dysfunction are needed [83].

Emerging effectors of coagulopathy — In addition to the mechanisms driving injury-associated coagulopathy addressed above, areas of emerging research suggest additional factors influencing coagulopathy associated with injury.

●Release of microparticles – Preliminary work suggests that tissue injury may prompt release of thrombin-rich microparticles into the systemic circulation, the local effects of which may contribute to hemostasis, while wider systemic release may lead to a DIC-like phenotype tipping the balance toward coagulopathy. Studies have correlated the presence of cell-derived microparticles in the systemic circulation following injury. One prospective observational study of 180 trauma patients identified elevated levels of endothelial-, erythrocyte-, and leukocyte-derived microparticles into the circulation compared with 65 controls [84]. Lower levels of platelet-derived microparticles and tissue factor-positive microparticles were seen in coagulopathic compared with noncoagulopathic trauma patients. A small study of 16 brain-injured patients also identified elevated levels of endothelial-, platelet-, and leukocyte-derived microparticles compared with controls [85].

●Damage-associated molecular patterns – Multiple studies have also evaluated the association between damage-associated molecular patterns (DAMPs) and coagulation dysfunction, including specific evaluation of soluble receptor for advanced glycation end-products (RAGE) [86], high-mobility group protein B1 (HMBG1) [87], and mitochondrial DNA [88,89]. Whether these or other DAMPs represent true biological mediators or simply correlate with measures of injury burden and coagulopathy is an active area of investigation [90].

DIAGNOSIS — Severely injured trauma patients are routinely screened with standard laboratory evaluation, including complete blood count, serum electrolytes, arterial blood gas analysis, and standard coagulation tests. These laboratory studies provide an assessment of acidosis and hemodilution, indicate the severity of shock (as measured by base deficit and/or lactate levels), guide specific component blood product administration, and serve as a baseline for the assessment of ongoing hemorrhage. (See "Initial management of trauma in adults".)

Readily obtainable coagulation tests (prothrombin time, international normalized ratio, and activated partial thromboplastin time) are the current standard for establishing a definitive diagnosis of coagulopathy. Fibrinogen and D-dimer levels are also available in most clinical laboratories and may serve as surrogate markers of clotting factor consumption and hyperfibrinolysis. (See "Clinical use of coagulation tests".)

Although these assays are commonly relied upon to evaluate bleeding risk in trauma patients, these tests were originally designed to screen for heritable coagulopathies such as hemophilia and subsequently used to monitor anticoagulant therapy [91]. The normal ranges are derived from the general population and correlate poorly with bleeding risk in elective general and vascular surgical operations [92]. In addition, the results from standard laboratory analysis can take up to 30 minutes and may not accurately reflect the patient's evolving coagulation status. The processing delay may result in results that are irrelevant when they do become available, such as in the case of exsanguinating hemorrhage. As a result, there has been a trend toward the use of point-of-care laboratory testing (thromboelastography [TEG]) and clinical scoring systems to guide management of the severely injured patient. (See 'Thromboelastography' below.)

Patients with prolonged clotting times due to known bleeding diatheses or pharmacological anticoagulation are not traditionally defined as having trauma-induced coagulopathy, although TIC physiology may exacerbate any preexisting coagulation disorders. (See "Approach to the adult with a suspected bleeding disorder" and "Management of warfarin-associated bleeding or supratherapeutic INR".)

Coagulation studies

Standard coagulation tests — Clinical studies have used several diagnostic criteria to identify clinically relevant coagulopathy following injury. These include prothrombin time (PT) >18 seconds [49], international normalized ratio (INR) >1.5 [93], activated partial thromboplastin time (PTT) >60 seconds [49], or any of these values at a threshold of 1.5 times the laboratory reference value [94]. The prevalence of prolonged PT is higher, but prolongation of the PTT is more specific.

●In a trauma registry study involving 20,103 patients, the PT and PTT were prolonged in 28 and 8 percent of patients, respectively [3]. The adjusted odds ratios for mortality were 1.35 for PT and 4.26 for PTT prolongation.

●A five-center international retrospective study of 3646 trauma patients identified patients with significantly greater transfusion requirements and increased mortality using a more liberal cutoff >1.2 of prothrombin time ratio (an inter-center standardized version of the INR) [95]. This lower INR value may be a more appropriate cutoff for patients with more severe injury (injury severity scale [ISS] >15) and shock.

Decreased platelet count and decreased platelet function also contribute to coagulopathy and poor outcome following trauma, although little information about platelet function is evident from the platelet count alone.

●TEG, a holistic assessment of clot formation, reflects the contribution of platelet count and function to coagulation and may be modified to investigate platelet function specifically (see 'Thromboelastography' below).

●Other instrumentation available to specifically assess platelet function includes the platelet function analyzer (PFA-100) and the electrical impedance whole blood aggregometer (Multiplate) [96,97]. However, these devices have not been prospectively evaluated in trauma and resuscitation. (See "Platelet function testing".)

It is unknown for certain whether TIC might exist in patients with normal-range values of PT/INR and PTT. There have been anecdotal reports of injured patients with clinically relevant hyperfibrinolysis identified using TEG but without prolonged PT or PTT. A careful study of markers of coagulopathy and inflammation in 80 trauma patients identified that increasing injury severity correlated with elevated markers of endothelial cell and glycocalyx damage, protein C activation, and clotting factor consumption even when INR and PTT values were in the normal range, suggesting that the biochemical derangement of TIC lies on a continuum dependent upon injury and shock severity [51]. Thus, patients who present with the combination of significant injury (ISS >15) and a base deficit (<-6) should be closely monitored and aggressively treated for clinical coagulopathy even in the face of normal standard coagulation tests, pending further clinical studies. Because TIC can occur in patients with normal platelet and fibrinogen levels, specific platelet or fibrinogen levels are not included in current diagnostic criteria; however, when abnormalities are present, thrombocytopenia and hypofibrinogenemia certainly contribute to clinical hemorrhage and should be corrected [2,3,10]. (See 'Treatment' below.)

Thromboelastography — TEG assesses the viscoelastic properties of clot formation in fresh or citrated whole blood in real time. The test synthesizes information obtained from multiple coagulation tests (PT, PTT, thrombin time, fibrinogen level, and platelet count) into a single readout (figure 3) providing information regarding clot initiation, clot strength, and fibrinolysis simultaneously. The bulk of its clinical use has been as a point-of-care adjunct during cardiopulmonary bypass [98], and liver transplantation [99]. (See "Platelet function testing", section on 'Viscoelastic testing (TEG and ROTEM)'.)

As a functional test of clot formation and lysis, it is conceptually well suited to monitor the progress or resolution of coagulopathy after traumatic injury. TEG parameters have been validated against standard laboratory tests [100,101], thrombin-antithrombin complex levels for TEG [102], and euglobin clot lysis times for rotational thromboelastometry (RoTEM) [103]. Specific elements of the clotting cascade can be interrogated by performing tests in the presence of specific clotting activators or inhibitors (table 3) [104].

Representative tracings compared with normal are given for each of the following abnormalities:

●Primary fibrinolysis – (figure 4A-B)

●Secondary hyperfibrinolysis – (figure 4A, 4C)

●Thrombocytopenia – (figure 4A, 4D)

●Clotting factor consumption – (figure 4A, 4E)

●Hypercoagulability – (figure 4A, 4F)

Multiple studies have used TEG to diagnose immediate hypocoagulability and later hypercoagulability following moderate injury despite normal-range standard coagulation tests [105,106]. Studies involving trauma patients have correlated TEG parameters with increased mortality [107-111]. Cutoff values for RoTEM-based parameters correlate with standard laboratory transfusion cutoff values [112]. The use of TEG in trauma as a guide to transfusion is discussed below. (See 'Thromboelastography-based transfusion' below.)

Thromboelastography-based fibrinolytic phenotypes — Demonstrations of alterations in normal fibrinolysis following severe trauma, and the availability of TEG to facilitate their rapid diagnosis (figure 4A-B and figure 4C), have renewed an interest in antifibrinolytic therapy for the treatment of acute hemorrhage [10,113-121]. (See 'Management of fibrinolysis' below.)

Early studies focused on empiric therapy and targeted management of hyperfibrinolysis [114,122], noting that hyperfibrinolysis after injury was associated with increased mortality. In one study, the mortality rate for a group of patients who demonstrated hyperfibrinolysis was significantly greater compared with the group that did not demonstrate hyperfibrinolysis (77 versus 41 percent) [110]. However, studies have identified two distinct injury-related fibrinolytic phenotypes apart from normal physiologic fibrinolysis, based on values of the TEG parameter LY30 (table 4 and table 5):

Hyperfibrinolysis — Excessive fibrinolytic activity leading to coagulopathy was originally noted during the anhepatic phase of liver transplantation [123] and has subsequently been proposed to play a mechanistic role in TIC [49,113]. The physiologic underpinning of hyperfibrinolysis appears to be primarily related to tissue plasminogen activator (tPA) release [124], with an inadequate compensatory increase in antifibrinolytic plasminogen activator inhibitor-1 (PAI-1) [125]. Animal models suggest that hyperfibrinolysis is mechanistically related to the degree of shock, while tissue injury appears to correlate with inhibition of fibrinolysis [126]. A single-center prospective study of critically injured trauma patients identified an increase in risk of massive transfusion (91 versus 30 percent) and death due to hemorrhage (46 versus 5 percent) at an LY30 of 3 percent, which is a much lower target than the manufacturer-provided normal upper bound of 7.5 percent (table 5 and figure 3) [127].

Fibrinolysis shutdown — Near-complete inhibition of fibrinolysis following elective surgery was also identified initially in early studies from the 1960s and 1970s and termed "shutdown" [128,129]. A landmark prospective study of 180 injured patients characterized the incidence and outcomes of fibrinolysis shutdown after injury: fibrinolysis shutdown was the most prevalent phenotype (64 percent), compared with physiologic fibrinolysis (18 percent) and hyperfibrinolysis (18 percent) [130]. An LY30 cutoff of 0.8 percent or less identified fibrinolysis shutdown in this study, derived from an analysis of the receiver-operator curve for mortality. A U-shaped distribution of mortality was identified across these phenotypes, with mortality rates of 17, 3, and 44 percent for patients presenting with shutdown, physiologic fibrinolysis, and hyperfibrinolysis, respectively. While the majority of deaths in the hyperfibrinolysis group were related to exsanguination, 40 percent of deaths in the shutdown group were attributed to multiple organ failure. These initial findings were subsequently recapitulated in a prospective cohort of more than 2500 injured patients from two large centers [131]. Mechanistic study has so far shown that fibrinolysis shutdown is reproducible by the addition of platelet lysate in vitro and that it is attributable to tPA-binding activity based on chromatography [126]; however, the physiologic underpinnings of shutdown remain an active area of investigation.

D-dimer — Elevated levels of D-dimer and other fibrin degradation products have been associated with severity of tissue damage, hyperfibrinolysis, and fibrinogen depletion early after injury [132,133]. In a review of 519 adult trauma patients with an injury severity score ≥16, poor outcome (≥10 units of red cell concentrate transfusion and/or death during the first 24 hours) was optimally identified using cut-off fibrinogen and D-dimer values of 190 mg/dL and 38 mg/L, respectively. Survival was lower for high D-dimer/low fibrinogen group compared with other groups. High D-dimer level on arrival was a strong predictor of early death or requirement for massive transfusion in severe trauma patients, even with high fibrinogen levels. In a cohort study that included 940 patients, D-dimer levels were seven-fold higher among critically injured patients who died compared with those who survived (103,170 versus 13,672 ng/mL) [134].

Fibrinogen levels — Fibrinogen is the terminal substrate for clot formation, and acquired fibrinogen deficiency in the setting of trauma is associated with hemorrhage and mortality [135]. Correction of hypofibrinogenemia with cryoprecipitate or fibrinogen concentrates is advocated based on fibrinogen cut-offs, viscoelastic measures [136], and empirically as part of massive transfusion protocols [137]. Institutional practices vary widely, and strong clinical consensus about triggers for fibrinogen repletion has not yet been established [138].

Factor levels — Although coagulation factors are not commonly assessed in injured patients, coagulation factor depletion due to hemodilution and unbalanced component blood transfusion exacerbates coagulopathy associated with trauma. Fibrinogen (factor I) is the first factor to become depleted, as noted above [139]. Of the other commonly numbered coagulation factors, V and VIII are the most labile and may become selectively depleted during trauma resuscitation, particularly in the setting of low plasma to red blood cell unit transfusion ratios. In a multicenter observational study of 165 critically injured trauma patients with serial plasma clotting factor analysis, multivariate analysis identified deficits in fibrinogen, thrombin, Factor V, Factor VIII, Factor IX, Factor X, and activated protein C levels as age-, injury-, and shock-adjusted predictors of coagulopathy [56].

Clinical scoring systems — Rapid clinical assessment of the trauma patient provides information that helps predict the potential for coagulopathy and empiric need for massive transfusion based upon the mechanism and severity of injuries, hemodynamic status of the patient, and evidence of active hemorrhage (eg, positive focused assessment with sonography for trauma [FAST], or brisk bleeding from a chest tube) [140-142]. Importantly, clinician gestalt alone is an unimpressive predictor of the need for massive transfusion, with sensitivity of 66 percent and specificity of 64 percent (a positive predictive value of only 35 percent) [143].

Several clinical scoring systems have been evaluated for this purpose but are not widely used. These include the Trauma-Associated Severe Hemorrhage (TASH) score [142], McLaughlin score [144], and Assessment of Blood Consumption (ABC) score [140]. A retrospective review compared these clinical scores in 596 patients. Significantly higher TASH (13 versus 6), McLaughlin (3.4 versus 2.4), and ABC (2 versus 1) scores were seen in patients who required massive transfusion compared with patients who did not, but no significant predictive differences were identified between these scores [140]. None of these scoring systems includes coagulation parameter measurements, highlighting the fact that massive hemorrhage is generally due to active bleeding requiring surgical or interventional control, not as a result of coagulopathy.

The utility of any clinical scoring system requires the demonstration that mortality is reduced as a result of a high score triggering a transfusion protocol and that earlier activation of these protocols is associated with a further decrease in mortality. There are no prospective trials that demonstrate such a survival benefit; however, significant reductions in overall blood product use and hospital costs have been reported [145,146].

While a sensitive scoring system designed to identify patients at risk of requiring massive transfusion likely also identifies patients at higher risk of coagulopathy, these scoring systems have not been designed or validated as diagnostic tests for coagulopathy.

TREATMENT — The early identification and treatment of patients with coagulopathy is an important aspect of damage control resuscitation. Damage control resuscitation consists of permissive hypotension; avoidance of excessive crystalloid, hypothermia, and acidosis; rapid surgical correction of anatomic hemorrhage; and early transfusion [147-153]. (See "Overview of damage control surgery and resuscitation in patients sustaining severe injury".)

Empiric transfusion strategies — A diagnosis of trauma-induced coagulopathy predicts significantly higher transfusion requirements in the first 24 hours of hospitalization. In one study, patients with TIC on admission received an average of 10 units of blood, compared with the 2 units received by those with normal coagulation studies [10]. The need for blood transfusion and number of units transfused is a predictor of systemic inflammation, acute respiratory distress, and mortality following traumatic injury [154-156].

Although red blood cell transfusion improves perfusion and oxygen carrying capacity, there is an increasing awareness that traditional transfusion protocols produce or exacerbate resuscitation-associated coagulopathy. Resuscitation protocols continue to vary widely between trauma centers, and ratios of plasma:packed red blood cells range from 1:1 to 1:10 [32,157-159]. With low plasma ratios, treatment of coagulopathy becomes delayed, and ultimately a greater volume of blood is required. The recognition of this problem has led to a widespread reevaluation of transfusion protocols, particularly in patients who demonstrate TIC. Patients with TIC are at risk for massive transfusion, and early observational studies in combat and civilian populations suggested that they may benefit from a resuscitation protocol using plasma, packed red blood cells, and platelets in equal (1:1:1) ratios given early and aggressively, while limiting crystalloids [160,161]. Transfusion ratios of 1:1:1 attempt to mirror the content of whole blood; however, there are no clinical data comparing mortality rates of whole blood versus 1:1:1 transfusion. In an in vitro analysis study, superior viscoelastic maximal clot formation was shown for noncomponent banked whole blood compared with 1:1:1 component "reconstituted" whole blood [162]. Although whole blood transfusion is not currently practical in many civilian trauma settings, its use is associated with improved outcomes after combat injury compared with component blood therapy [163]. Protocols for its potential use in civilian mass casualty and disaster scenarios are under development [164]. Early studies of uncrossmatched group O+ whole blood administration in civilian trauma centers show evidence of safe administration in patients presenting with hemorrhagic shock [165].

Two trials have addressed the validity of "hemostatic" resuscitation using empiric plasma and platelet administration to target blood product transfusion ratios of 1:1:1 RBC:plasma:platelets [166,167].

●The PROMMTT (PRospective, Observational, Multicenter, Major Trauma Transfusion) study was a prospective observational cohort study of 1245 patients receiving at least 1 unit of RBC within six hours of arrival to one of 10 level I trauma centers in the United States, including 905 patients who received at least 3 units of RBCs within 24 hours [166]. Using a multivariate Cox model, an increased ratio of plasma:RBC (hazard ratio [HR] 0.31, 95% CI 0.16-0.58) and platelets:RBC (HR 0.55, 95% CI 0.31-0.98) was independently associated with reduced six-hour mortality, a time period when death from active hemorrhage is predominant. Patients with ratios <1:2 were three- to fourfold more likely to die by six hours than those receiving ratios of >1:1. Importantly, plasma and platelet ratios were not associated with 24-hour or 30-day mortality. In a later subanalysis of 619 patients from this study, whether or not "early" (defined as within either 2.5 hours or within the first 3 to 6 blood product units administered) transfusion of plasma was protective was evaluated. The authors found that "early" transfusion of plasma was associated with reduced 24-hour (odds ratio [OR] 0.47, 95% CI 0.27-0.84) and 30-day (OR 0.44, 95% CI 0.27-0.73) mortality compared with patients who received lower plasma:RBC ratios or who did not received early plasma but "caught up" to ratios approaching 1:1 by 24 hours [168]. Inadequate numbers precluded similar analysis of early platelet transfusion. Overall, the study suggested that the benefit of hemostatic resuscitation is principally clinically relevant in preventing death by hemorrhage within the first six hours and that competing risks from nonhemorrhagic causes of death overshadow mortality differences at later time points.

●The PROPPR (Pragmatic, Randomized Optimal Platelet and Plasma Ratios) trial randomly assigned 680 severely injured patients identified as at risk of requiring massive transfusion from 12 North American level I trauma centers to transfusions of plasma, platelets, and red blood cells in ratios of either 1:1:1 or 1:1:2 [167]. Additional center-specific standard-of-care interventions, including prerandomization blood product and the use of cryoprecipitate and antifibrinolytics, were not controlled, making the separation between groups difficult to evaluate. There were no significant differences in primary outcomes of 24-hour or 30-day mortality between the groups. Similar to the PROMMTT study, death from hemorrhage was significantly less common in the 1:1:1 cohort at three hours after injury; however, no significant difference was seen at any later time point. No differences in the rates of complications were seen between groups (eg, acute respiratory distress syndrome, multiorgan failure, venous thromboembolism, sepsis).

Two other trials have evaluated the prehospital use of empiric plasma transfusion in the setting of ground or air transport of patients with evidence of hemorrhagic shock [169,170]:

●The Control of Major Bleeding After Trauma (COMBAT) trial randomly assigned 144 patients with clinical evidence of hemorrhagic shock to prehospital infusion of 250 mL of plasma versus normal saline [169]. There was no significant difference in the primary outcome of 28-day mortality (15 versus 10 percent, p = 0.37) between plasma versus crystalloid treatment arms. The additional time required to defrost and transfuse plasma in this study was three minutes; plasma transfusion was not associated with an increased rate of adverse events. Importantly, this study was conducted in a single focused urban area with rapid transport times (median transport time <20 minutes).

●The Prehospital Air Medical Plasma (PAMPer) trial randomly assigned 501 patients with clinical evidence of hemorrhagic shock to receive two units of thawed plasma versus standard care during transport to one of nine participating level I trauma centers [170]. The primary outcome of 30-day mortality was significantly improved in the plasma group (23 versus 33 percent), with evidence of separation between groups as early as three hours from enrollment. No difference in adverse events between groups was seen. The transport times in this study (30 to 70 minutes) were significantly longer compared with those in the COMBAT study. Taking advantage of the harmonized inclusion criteria used for enrollment in these two studies, a post hoc analysis identified evidence of increased mortality in the standard of care groups when transport time was longer than 20 minutes [171].

A potential disadvantage of early empiric inclusion of plasma and/or platelets into resuscitation protocols for hemorrhagic shock includes the potential hazards associated with plasma and platelet transfusion (eg, acute respiratory distress syndrome). Although transfusion-related lung injury is a concern, a large retrospective analysis did not find any significant increase in number of ventilator days in patients for whom higher transfusion ratios were used [172]. However, because of these concerns, patients without risk factors for massive transfusion based upon injury severity, shock, and abnormal coagulation should be corrected in response to specific laboratory deficits [173-175]. (See "Treatment of severe hypovolemia or hypovolemic shock in adults".)

Thromboelastography-based transfusion — The rapid turnaround and multifaceted information provided by point-of-care thromboelastography (TEG) appears to provide improved criteria for transfusion in acutely injured patients. TEG-guided "thrombostatic" resuscitation protocols may emerge as the standard. TEG protocols based upon TEG clotting time, clot formation time, amplitude, and clot lysis index (table 4 and figure 5) have been suggested for both standard TEG and rotational TEG based upon observational studies [108,112,127,136,141,176-181]. (See 'Thromboelastography' above.)

A 2015 systematic review did not identify a sufficient number of studies of adequate quality to determine whether TEG-based transfusion is better than existing transfusion practices [182]. Most studies were single-center, observational studies [108,112,136,141,178-180,182-186]. A subsequent trial randomly assigned 111 severely injured patients to a transfusion protocol managed by goal-directed TEG or by conventional coagulation assays (ie, international normalized ratio, fibrinogen, platelet count) [185]. The initial transfusion (4 units of packed red blood cells, 2 units of plasma) was triggered by activation of the massive transfusion protocol (at the treating clinician's discretion) while awaiting results of coagulation tests (standard or TEG). Subsequent transfusions were guided by the assigned coagulation testing. There were no significant differences in injury severity score (ISS) between the groups. The risk of death was significantly higher for the conventional group compared with the TEG group (hazard ratio 2.17, 95% CI 1.03-4.6). Just over one-half of the deaths (16/31) occurred within six hours of arrival in the emergency department. Death rates in the conventional versus TEG groups were 12/55 (21.8 percent) versus 4/56 (7.1 percent) at 6 hours, and 20/55 (36.4 percent) versus 11/56 (19.6 percent) at 28 days. These outcomes may be due to differences in the timing of transfusion and ratio of blood products given. Although there were no differences in the overall volume of transfusion at 24 hours, the standard assay group received more plasma and platelets during the first several hours of resuscitation. Where TEG is available, we suggest TEG-based goal-directed resuscitation for trauma patients requiring massive transfusion. Standard coagulation assays may be performed in parallel to facilitate communication with practitioners unfamiliar with TEG parameters. At centers where TEG is unavailable, empiric plasma-forward transfusion strategies guided by standard coagulation assays remain standard of care.

An example thromboelastogram-guided transfusion protocol from Denver Health Medical Center, providing normal ranges and transfusion cutoffs derived from cut-point analysis in 190 severely injured trauma patients, is given in the table (table 5) [185,187]. Although TEG-based transfusion triggers are preliminary, we feel that TEG is a promising technology for determining the appropriateness of transfusion in the injured population. Examples of the utility of TEG in trauma are given below:

●The Denver group found that TEG "G" values and thrombin generation parameters obtained six hours after arrival were significantly associated with survival, while the INR was not [181].

●A retrospective evaluation of TEG used in 44 combat patients found that maximal amplitude (MA) correlated more strongly with 24-hour transfusion requirements than standard laboratory values [188].

●In another retrospective study of 832 patients requiring massive transfusion (21 percent trauma patients), patients whose transfusion was guided by TEG parameters received more plasma and platelets and had significantly better 30-day survival (32 versus 20 percent) compared with retrospective controls transfused using standard laboratory measurements [183].

Although initial experiences are promising and enthusiasm abounds, care must be taken in the interpretation of TEG results in injured patients, particularly in patients with low clinical suspicion of active hemorrhage. Confounders may exist for which TEG will not be accurate. As an example, in a single-center prospective study of 264 alcohol-intoxicated (out of 415) general trauma patients, alcohol level was associated with prolongation of the R-time and decrease in the alpha angle in adjusted analysis; however, an adjusted analysis showed that alcohol intoxication was a negative predictor of INR-defined coagulopathy and was not associated with transfusion requirements or overall mortality [189].

Management of fibrinolysis — The best studied antifibrinolytic agent in the trauma population is tranexamic acid (TXA), a lysine analog that binds to the kringle domains of plasminogen, preventing rearrangement into its active form. The studies described below identity a subset of injured patients who will likely benefit from early appropriate administration of TXA; the optimal risk/benefit ratio conceptually appears to be those patients with acute life-threatening hemorrhage who can be treated within three hours of injury. A meta-analysis evaluated the incidence of thrombotic events associated with TXA administration in studies of traumatic, other surgical, and medical bleeding, and found no significant increase in thrombotic events associated with any TXA dosing regimen [190]. However, patients who present with the more common "shutdown" fibrinolytic phenotype appear to be at higher risk of thromboembolic complications and long-term organ failure, and thus may remain conceptually at increased risk of harm from TXA treatment. Thus, in clinical settings in which immediate access to TEG allows early determination of the presenting fibrinolytic phenotype, we suggest that an empiric bolus dose of TXA be given as early as possible (provided that it can be administered within three hours of injury) and that TEG may be an appropriate guide to whether additional doses of TXA are needed. Patients who maintain subphysiologic levels of fibrinolysis after initial TXA administration are unlikely to benefit from additional TXA dosing. In clinical settings in which TEG is unavailable, we advocate empiric treatment with TXA for all patients who demonstrate active hemorrhage.

Two large, randomized trials form the principal evidence base for empiric antifibrinolytic therapy with TXA in injured patients with evidence of hemorrhagic shock. These are summarized below.

●The Clinical Randomization of Antifibrinolytic in Significant Hemorrhage (CRASH-2) trial was a prospective, randomized, placebo-controlled trial of empirically administered TXA conducted in over 20,000 trauma patients worldwide with, or at risk of, significant bleeding. The authors identified a significant decrease in both all-cause (14 versus 16 percent) and hemorrhage-related (4.9 versus 5.7 percent) mortality [114]. The wide breadth of clinical settings and patient populations enrolled was a notable strength of this study. However, inclusion criteria were broad, with only one-half of enrolled patients receiving any blood products or requiring emergency surgery for injuries. Further, based on the observed 1.5 percent absolute risk reduction for mortality, 67 patients would need to be empirically treated with TXA to save one life. No differences were seen in the incidence of thromboembolic complications or in head injury-related mortality. Subsequent additional analysis of these data showed that the survival benefit of empiric TXA was only seen in patients treated within three hours of injury; treatment beyond three hours appeared to increase the risk of death due to bleeding [120].

●The Study of Tranexamic Acid During Air and Ground Medical Prehospital Transport (STAAMP) trial was a prospective, randomized, placebo-controlled trial of empiric TXA administration during prehospital transport conducted in 903 trauma patients at risk for hemorrhage [191]. Patients were randomly assigned to 1 g of TXA over 10 minutes versus placebo in the prehospital phase, followed by three differing dosing regimens upon trauma center arrival during the in-hospital phase of the study placebo (abbreviated [1 g TXA over 10 minutes], standard [1 g over eight hours], or repeat bolus [1 g over 10 minutes followed by 1 g over eight hours]). The authors identified no significant difference in 30-day all-cause mortality in unadjusted (8.1 versus 9.9 percent) or adjusted (hazard ratio 0.81) analysis when comparing TXA versus placebo in aggregate. However, prespecified subgroup analysis suggested that prehospital treatment, treatment within one hour of injury, and treatment with the repeat bolus dosing regimen (3 g total) of TXA was associated with significantly lower 30-day all-cause mortality. Further, the prespecified subgroup of patients in severe (systolic blood pressure <70 mmHg) shock showed a significantly reduced 30-day all-cause mortality for TXA treatment compared with placebo (18.5 versus 35.5 percent). There was no association between TXA treatment and arterial or venous thromboembolic events. This study overall suggests that TXA may be of benefit when given early and to patients in severe shock and that optimal dosing regimens including up to 3 g of TXA need further evaluation.

Two additional trials provide evidence of safety with potential benefit for empiric antifibrinolytic therapy in patients with traumatic brain injury. These are summarized below.

●The Clinical Randomization of an Antifibrinolytic in Significant Head Injury (CRASH-3) trial was a prospective, randomized, placebo-controlled trial of empiric in-hospital TXA administration (given as 1 g over 10 minutes, followed by 1 g over eight hours) conducted in over 12,000 patients worldwide with clinical or imaging evidence of traumatic brain injury who could be treated within three hours of injury [192]. The authors reported no significant reduction in brain injury-related mortality at 28 days associated with TXA compared with placebo (18.5 versus 19.8 percent). However, prespecified subgroup analysis suggested that patients with mild-to-moderate traumatic brain injury (TBI; Glasgow Coma Scale [GCS] 9 to 15) had significantly improved mortality (relative risk 0.78). There was no association between TXA treatment and vascular occlusive events or seizures. This trial had methodological issues worth noting, including a change in primary outcome during the trial (from overall mortality to head injury-related mortality at 28 days), a change in inclusion criteria (from treatment within eight hours to treatment within three hours), and under-powering for subgroup analysis (powered for 10,000 patients treated within three hours, but only 9200 were enrolled).

●In a prospective, randomized, placebo-controlled trial, 966 patients with clinical evidence of TBI (GCS ≤12) who could be treated within two hours of injury were randomly assigned to empiric prehospital TXA or placebo (placebo bolus plus placebo infusion) [193]. Patients receiving TXA were randomized to either 1 g of TXA over 10 minutes prehospital followed by 1 g over eight hours in-hospital (bolus-maintenance group) or 2 g of TXA over 10 minutes prehospital followed by placebo infusion in-hospital (bolus-only group). There were no significant differences in favorable neurological outcome (defined as GCS-Extended score >4) between TXA and placebo at six months post-injury (65 versus 62 percent). Although there were no differences in seizure incidence between TXA and placebo overall, the bolus-only group had a higher incidence of seizures compared with either the other groups (5 versus 2 percent). Further, the bolus-only group had a higher incidence of thrombotic complications (10 percent) compared with the bolus-maintenance group (4 percent), although this was similar to the rate seen in the placebo group (9 percent). Interestingly, no difference in arrival TEG parameters was seen between groups, suggesting that empiric TXA in this population does not lead to significant repression of fibrinolysis.

Other antifibrinolytic agents, such as aminocaproic acid and aprotinin, have not been evaluated in patients with traumatic coagulopathy [115]. (See "Thrombotic and hemorrhagic disorders due to abnormal fibrinolysis", section on 'Therapies for hyperfibrinolytic states'.)

Pharmaceutical hemostatic agents — In addition to repletion of coagulation factors by transfusion, several pharmaceutical hemostatic agents are available for the treatment of severe coagulopathy in the injured patient, including recombinant factor VIIa, prothrombin complex concentrate, and desmopressin.

●Recombinant factor VIIa – Recombinant factor VIIa was initially developed and approved for the treatment of hemophilia and congenital factor deficiencies [194]. The recognition of injury-exposed tissue factor binding to activated factor VII as the principle trigger of clot formation after trauma led to interest in using recombinant factor VIIa for the management of traumatic-induced coagulopathy. Recombinant human factor VIIa is an adjunctive treatment for coagulopathy associated with trauma but should be reserved for salvage therapy. When used, it is important to correct acidosis, hypothermia, thrombocytopenia, and hypofibrinogenemia prior to its use. The dosing of recombinant factor VIIa is discussed in detail elsewhere. (See "Recombinant factor VIIa: Administration and adverse effects".)

●Prothrombin complex concentrate – Prothrombin complex concentrate (PCC) is a factor concentrate enriched for factors II, VII, IX, and X originally developed for hemorrhagic complications of hemophilia B [194]. The clinical use of PCC has been well studied in the reversal of warfarin anticoagulation since PCC is preferentially rich in the vitamin K-dependent clotting factors [195]. Some centers have used PCC to correct coagulopathy after trauma, and preliminary studies in animal models of hemorrhagic shock are promising, but PCC has not been thoroughly evaluated in trauma patients [196]. (See "Massive blood transfusion" and "Initial management of moderate to severe hemorrhage in the adult trauma patient".)

●Desmopressin – Desmopressin was developed for the treatment of inherited bleeding diatheses [197] and to counteract uremic bleeding [198]. There is insufficient clinical evidence to support the use of desmopressin in the trauma population except in those patients with preexisting bleeding diatheses [199]. Preliminary animal studies show that desmopressin administration improves but does not completely correct hypothermia or acidosis-induced platelet dysfunction, but clinical validation of these experimental data needs to be performed [200,201]. (See "Uremic platelet dysfunction".)

Monitoring — Reliance upon standard serial laboratory measurements is not compatible with the timely correction of coagulopathy [202]. TEG provides more comprehensive information in real-time and readily identifies hyperfibrinolysis [109]. Serial TEG, when available, should be used to monitor the patient's coagulation status and to guide transfusion and the correction of coagulopathy. (See 'Thromboelastography' above and 'Thromboelastography-based transfusion' above.)

Where TEG is not available, serial PT/INR, PTT, hemoglobin/hematocrit, platelet count, and fibrinogen levels should be obtained on arrival and following transfusion to verify an appropriate response to blood products and/or pharmaceutical hemostatic agents, before and after operative interventions, and as dictated by the patient's clinical course.

Serial measurements of arterial blood gas for pH and base deficit are indicated for monitoring the resolution of acidosis and tissue hypoperfusion in response to resuscitation. During the course of operative intervention and ongoing shock resuscitation, central temperature monitoring (eg, via Foley catheter or esophageal temperature probe) should be performed until normothermia is reestablished.

In the massively resuscitated patient, the early institution of intermittent intra-abdominal pressure transducer is also appropriate to monitor for the development of abdominal compartment syndrome and the need for abdominal decompression. (See "Abdominal compartment syndrome in adults", section on 'Measurement of intra-abdominal pressure'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: General issues of trauma management in adults".)

SUMMARY AND RECOMMENDATIONS

●Coagulopathy is associated with greater transfusion requirements, longer intensive care unit and hospital stays, more days of mechanical ventilation, and an increased incidence of multiorgan failure and mortality. The identification and early correction of coagulopathy is important to decrease fluid and transfusion requirements, decrease complications, and improve survival. (See 'Impact' above.)

●The etiology of coagulopathy in the injured patient is multifactorial with overlapping contributions from acidosis related to tissue injury and shock, hypothermia related to exposure and fluid administration, or hemodilution due to fluid or component blood product administration. In severely injured patients, an additional biochemical process that remains incompletely characterized underlies coagulopathy. Consumption of clotting factors manifesting as disseminated intravascular coagulation (DIC) may contribute upon presentation or later in the hospital course. (See 'Etiologies' above.)

●Trauma-induced coagulopathy (TIC) is an impairment of hemostasis and activation of fibrinolysis that occurs in response to severe injury and is biochemically evident prior to, and independent of, the development of significant acidosis, hypothermia, or hemodilution. TIC is mediated primarily by activation of the thrombomodulin-protein C system. (See 'Trauma-induced coagulopathy' above.)

●Standard coagulation tests including prothrombin time/international normalized ratio (PT/INR), activated partial thromboplastin time (PTT), fibrinogen level, and platelet count are part of the initial laboratory evaluation of trauma patients. In patients without preexisting coagulation defects, a prolonged PT and/or activated PTT greater than 1.5 times normal on admission defines the presence of TIC. Clinically relevant TIC can occur in patients who have normal platelet and fibrinogen levels. (See 'Diagnosis' above.)

●TEG is an important tool for identifying patients with TIC and for real-time monitoring of ongoing resuscitation efforts in injured patients. TEG measures the viscoelastic properties of clot formation providing information on clot initiation, clot strength, and fibrinolysis. For patients requiring massive transfusion, we suggest early TEG-based goal-directed resuscitation and management of fibrinolysis rather than treatment based on standard coagulation assays, when the technology is available (Grade 2C). At centers where TEG is not available, empiric plasma-forward transfusion strategies guided by standard coagulation assays remain standard of care. Additional clinical evaluation is ongoing in an effort to establish guidelines for more widespread use. (See 'Thromboelastography' above.)

●For trauma patients diagnosed with coagulopathy, we suggest plasma-based resuscitation, targeting ratios of packed red blood cells, Fresh Frozen Plasma (FFP) or similar products (eg, PF24), and platelets approaching 1:1:1 over protocols with lower ratios (Grade 2C). Injured patients with TIC are more likely to require massive transfusion and benefit from early matched transfusion. We prefer to use TEG to screen for hyperfibrinolysis, guide ongoing transfusion, and monitor patients rather than empiric treatment with follow-up guided by standard coagulation tests. (See 'Treatment' above and "Initial management of moderate to severe hemorrhage in the adult trauma patient".)

●Pharmaceutical hemostatic agents available as adjuncts for the treatment of severe coagulopathy in the injured patient include recombinant factor VIIa, prothrombin complex concentrate, antifibrinolytic agents (tranexamic acid, aminocaproic acid, aprotinin), and desmopressin. (See 'Pharmaceutical hemostatic agents' above.)