Fredric M. Pieracci, Jeffry L. Kashuk, and Ernest E. Moore
The ponderous literature on the subject of hemostasis could perhaps be considered a classical example of the infinite ability of the human mind for abstract speculation. For several years, the number of working theories of the hemostatic mechanism greatly exceeded and not always respected the confirmed experimental facts. In recent years, however, the revived interest in this field has led to an accumulation of new findings which has been almost too rapid for their orderly incorporation into a logical working pattern. As a result, we have rapidly gone from a state of “orderly ignorance” to one of “confused enlightenment,” from which we have not emerged as yet.
Mario Stefanini, April 19541
The first recorded blood product transfusion to a human being occurred in 1667 in France and involved transfusion of approximately three tablespoons of whole blood from a calf to a man who was suffering from insanity.2 The physician performing the transfusion postulated that the calm temperament of the calf would transfer to the patient via its blood. The procedure was well tolerated, although the patient developed severe flank pain and tar-colored urine following the subsequent three transfusions: the first recorded evidence of immune-mediated hemolysis, albeit unbeknownst to the physician at the time.
Although transfusion medicine has undergone enormous development since this sentinel event, as summarized by Stefanini, important gaps in scientific knowledge persist, and several fundamental issues involving hemotherapy following major trauma remain controversial. There has been an explosion in the science of hemostasis, with a resultant revolution in all aspects of the care of the coagulopathic trauma patient. Our understanding of the mechanisms of coagulation has shifted from that of a simple enzymatic cascade to a cell-based paradigm, in which endothelium, erythrocytes, leukocytes, and platelets interact to coordinate a delicate balance between thrombosis and fibrinolysis. Furthermore, both an early postinjury endogenous coagulopathy associated with traumatic shock and a myriad of secondary factors that exacerbate this condition have been elucidated. Diagnosis of coagulopathy is shifting from the routine use of laboratory tests designed to monitor anticoagulation therapy toward point-of-care testing, which provides essential real-time clinical correlates. Treatment algorithms of traumatic coagulopathy have emphasized early replacement of both clotting factors and platelets with concomitant restraint of crystalloid administration (termed damage control resuscitation), as well as pharmacologic adjuncts that exploit the endogenous coagulation system. On the other hand, documentation of the deleterious effects of overzealous blood component replacement has led to a reevaluation of this strategy, in an attempt to reach a balance between abatement of coagulopathy and minimization of subsequent organ dysfunction. This chapter will attempt to synthesize recent developments in the complex management of the bleeding, coagulopathic trauma patient.
RED BLOOD CELL TRANSFUSION
Red blood cell (RBC) transfusion is lifesaving in the face of critical anemia associated with hemorrhagic shock. However, the optimal target hematocrit during resuscitation remains unknown. Shock is defined broadly as the development of an oxygen debt due to impaired delivery, utilization, or both, with resultant anaerobic metabolism and organ dysfunction. Elimination of this oxygen debt involves optimization of oxygen delivery, which is the product of cardiac output and arterial oxygen content. The arterial oxygen content, in turn, is dependent primarily on the hemoglobin concentration and oxygen saturation. Oxygen consumption, defined as the product of the cardiac output and the difference between the arterial and venous oxygen content, represents a more specific marker of oxygen availability at the cellular level.
During resuscitation, a balance must occur between the competing goals of maximal oxygen content (hematocrit = 100%) and minimal blood viscosity (hematocrit = 0%). Furthermore, irrespective of hematocrit, the oxygen-carrying capacity of transfused allogeneic erythrocytes is impaired due to storage-induced changes in both deformability and hemoglobin oxygen affinity. Accordingly, although many studies have measured an increase in oxygen delivery following transfusion of allogeneic RBCs, almost none have reported an increase in oxygen consumption.3 Finally, beyond a role in oxygen delivery, erythrocytes are integral to hemostasis via their involvement in platelet adhesion and activation, as well as thrombin generation. The hematocrit is thus relevant to hemorrhagic shock as it relates to both oxygen availability and hemostatic integrity.
Early canine models of hemorrhagic shock suggested that oxygen consumption is optimized at a relatively high hematocrit (range 35–42%).4 However, hematocrit variation was achieved via autotransfusion of the animal’s shed whole blood, eliminating the aforementioned limitations of allogeneic erythrocytes, and rendering the results inapplicable to modern resuscitation of hemorrhagic shock. Furthermore, acute normovolemic hemodilution of dogs to a hematocrit of 10% is well tolerated, with little decrement in oxygen delivery secondary to a compensatory increase in cardiac output.5
Retrospective observations among critically ill surgical patients in the 1970s suggested a hematocrit of 30% as optimal for both oxygen-carrying capacity and survival.6 Such studies formed the basis of the traditional recommendation to maintain the hematocrit >30%, although the marked limitations of this retrospective literature were recognized ultimately. As the deleterious effects of RBC transfusion became increasingly evident, renewed interest in the ideal transfusion trigger occurred. The Transfusion Requirements in Critical Care (TRICC) Trial, which compared restrictive (hemoglobin <7.0 g/dL) and liberal (hemoglobin <9.0 g/dL) transfusion triggers among 838 patients, provided the first level I evidence regarding RBC transfusion strategies among the critically ill.7 Although inclusion criteria did not specify ongoing resuscitation, 37% of patients were in shock at the time of enrollment as evidenced by the need for vasoactive drugs. No difference in 30-day mortality was observed between groups. However, in-hospital mortality, as well as mortality among less severely ill patients (Acute Physiology and Chronic Health Evaluation II score <20) and younger patients (age <55 years), was significantly lower in the restrictive transfusion group. Current evidence thus suggests that a hemoglobin concentration of >7 g/dL is at least as well tolerated as a hemoglobin concentration of >9 g/dL among critically ill patients.
It is possible that hemoglobin concentrations below 7 g/dL are safe, particularly in younger patients. However, a hemoglobin concentration of 5 g/dL appears to be the threshold for critical anemia. Whereas hemodilution of healthy volunteers as low as a hemoglobin concentration of 5 g/dL is well tolerated,8 a study of postoperative patients who refused RBC transfusion reported a sharp increase in mortality below this same hemoglobin concentration.9 Such populations differ fundamentally from the multiply injured, exsanguinating patient in need of resuscitation. However, these data are provocative, and future large-scale trials of lower transfusion triggers for the resuscitation of hemorrhagic shock are warranted in light of the accumulating evidence documenting the untoward effects of RBC transfusion.
In addition to oxygen transport, RBCs play an important role in hemostasis. As the hematocrit rises, platelets are displaced laterally toward the vessel wall, placing them in contact with the injured endothelium; this phenomenon is referred to as margination. Platelet adhesion via margination appears optimal at a hematocrit of 40%.10 Erythrocytes are also involved in the biochemical and functional responsiveness of activated platelets. Specifically, RBCs increase platelet recruitment, production of thromboxane B2, and release of both ADP and P-thromboglobulin. Furthermore, RBCs participate in thrombin generation through exposure of procoagulant phospholipids. Interestingly, animal models suggest that a decrease of the platelet count of 50,000 is compensated for by a 10% increase in hematocrit.11Despite these experimental observations, no prospective data exist detailing the relationship between hematocrit, coagulopathy, and survival among critically injured trauma patients.
In summary, prior investigations into the ideal hematocrit for oxygen-carrying capacity during hemorrhagic shock are in large part irrelevant to modern-day resuscitation with allogeneic blood. Banked erythrocytes are subject to a time-dependent diminution of oxygen-carrying capacity, and the effect of blood transfusion on oxygen consumption, regardless of hematocrit, remains questionable. The CRIT trial suggested that patients in shock tolerate a hemoglobin concentration of 7.0 g/dL at least as well as 9.0 g/dL, although this hypothesis was not testing during the initial resuscitation of hemorrhagic shock specifically. Furthermore, the role of erythrocytes in hemostasis must be considered. In practice, clinical circumstance (e.g., ongoing hemorrhage with hemodynamic instability and coagulopathy), as opposed to an isolated laboratory measurement, should inform the decision to transfuse. However, until there is definitive evidence to challenge the CRIT data, a hemoglobin concentration of <7 g/dL should be considered the default transfusion trigger for resuscitation from shock.
POSTINJURY COAGULOPATHY PERSPECTIVE
Uncontrolled hemorrhage is the leading cause of preventable morbidity and mortality following trauma. Hemorrhage is responsible for nearly one half of all trauma deaths, and is the second leading cause of early death, preceded only by central nervous system injury.12 Most hemorrhagic deaths occur within the first 6 hours postinjury, and require tremendous resource mobilization in terms of blood component therapy. Although most life-threatening hemorrhage originates as major vascular injury that is amenable to either surgical or angiographic control, a diffuse coagulopathy frequently supervenes.
Originally described over 60 years ago,1 postinjury hemorrhage that persists despite control of surgical bleeding has been referred to by many names, including medical bleeding, diffuse bleeding diathesis, posttransfusion bleeding disorder, medical oozing, and disseminated intravascular coagulation (DIC). Clinically, the coagulopathy is manifest as nonsurgical bleeding from mucosal lesions, serosal surfaces, and wound and vascular access sites that continues after control of identifiable vascular bleeding. Although postinjury coagulopathy has long been recognized, several authors have struggled to elucidate both its predictors and mechanisms. Reporting on a large cohort of combat casualties, Simmons et al. appropriately identified the relationship between major trauma and coagulopathy, but were unable to predict coagulopathy using a myriad of both clinical and laboratory parameters.13 In 1982, our group described the “bloody vicious cycle,” in which the synergistic effects of acidosis, hypothermia, and coagulopathy combined to create an irreversible clinical deterioration among patients who had received large-volume blood transfusion, eventuating in death by exsanguination despite surgical control of bleeding.14 In the late 1980s, Lucas and coworkers from Wayne State University detailed the relationship among large-volume blood transfusion, decrement in clotting factor concentrations, and the corresponding prolongation of traditional measures of coagulopathy, such as the prothrombin time (PT) and activated partial thromboplastin time (aPTT).15 Development of coagulopathy following massive transfusion (MT), which was postulated to be secondary to both consumption and dilution of clotting factors, was similarly unable to be predicted by either clinical or laboratory parameters.16 Most recently, evidence of an endogenous coagulopathy associated with severe traumatic injury has emerged, which occurs early and is independent of the secondary effects of body temperature, acidosis, and clotting factor consumption or dilution.17
The burden of postinjury coagulopathy on the severely injured trauma patient is enormous. Overt coagulopathy affects at least one in four seriously injured patients and is associated independently with increased mortality.18 In a large series from our institution, over one half of deaths due to exsanguinations occurred after control of surgical bleeding and were thus due to coagulopathy.14 Persistent hemorrhage despite surgical control of bleeding remains the most common reason for abandonment of definitive repair of injuries (damage control surgery). Finally, patients who develop postinjury coagulopathy nearly universally require MT of blood products, placing an incalculable financial burden on both institutional and national health care delivery systems.
CELL-BASED COAGULATION CONSTRUCT
Effective management of postinjury coagulopathy requires an understanding of the coagulation process. Hemostatic integrity involves an intricate balance between hemorrhage and thrombosis, achieved in concert by complex interactions between the anticoagulant, procoagulant, and fibrinolytic systems. The inciting event for thrombosis following injury is the exposure of tissue factor (from both the subendothelium and mononuclear cells) to circulating clotting factors. From this point forward, multiple enzymatic cascades, orchestrated by a myriad of cells, direct the balance of thrombosis and hemorrhage based on both substrate availability and the status of global tissue perfusion (i.e., shock). Whereas clotting factors exist in concentrations sufficient to maintain hemostasis in health, major trauma overwhelms the capacity of the coagulation system, with resultant systemic thrombosis and hemorrhage. For example, an isolated lobar pulmonary contusion may involve a surface area large enough to exhaust the body’s endogenous fibrinogen and platelet reserves. Major proteins involved in the procoagulant, anticoagulant, and fibrinolytic systems are listed in Table 13-1.
TABLE 13-1 Proteins Involved in Coagulation, Anticoagulation, and Fibrinolysis
The coagulation process has been considered traditionally a cascade of proteolytic reactions occurring in isolation. In this classic view of hemostasis (extrinsic and intrinsic pathways), the cell surface serves primarily to provide an anionic phospholipid region for procoagulant complex assembly. Whereas this model is supported by traditional laboratory tests of isolated coagulation in a test tube, it does not correlate with current concepts of hemostasis occurring in vivo.
This antiquated model has been supplanted by the cell-based model (CBM) of coagulation. This model recognizes the important interactions of the cellular and plasma components to clot formation, as opposed to the more simplistic schema of the classic view. The CBM suggests that procoagulant properties result from expression of a variety of cell-based features, originating at the endothelial level, including protein receptors, which activate components of the coagulation system at specific cell surfaces. Furthermore, this model allows for improved understanding and potential mechanistic links with cross-talk between inflammation and coagulation components. In addition, platelet receptors, endothelial cells, proteases, cytokines, and phospholipids have important roles in coagulation. This model also incorporates RBCs and their aforementioned interactions with the hemostatic process. The CBM occurs in three overlapping phases: initiation (which occurs on tissue factor–bearing cells), amplification, and propagation. Amplification and propagation involve platelet and cofactor activation eventuating in the generation of massive amounts of thrombin, known as the thrombin burst. Both amplification and propagation occur on the cell surface of platelets, underscoring the central role of the platelet in the hemostatic process.
In summary, the CBM represents a major paradigm shift from a theory that views coagulation as being controlled by concentrations and kinetics of coagulation proteins to one that considers the process to be driven by diverse cellular interactions. Coagulation factors work as enzyme/cofactor/substrate complexes on the surface of activated cells, and hemostasis requires the interaction of endothelium, plasma proteins, platelets, and RBCs.
HEMOSTASIS MANAGEMENT CONTROVERSIES
Acute Coagulopathy of Trauma
Coagulation disturbances following trauma follow a trimodal pattern, with an immediate hypercoagulable state, followed quickly by a hypocoagulable state, and ending with a return to a hypercoagulable state.19 Conceptualization of the early hypocoagulable state has changed markedly over the last 10 years. Trauma-induced coagulopathy was considered traditionally to be the consequence of clotting factor depletion (via both hemorrhage and consumption), dilution (secondary to massive resuscitation), and dysfunction (due to both acidosis and hypothermia). However, several recent reports have detailed that many trauma patients present with a coagulopathy prior to fluid resuscitation and in the absence of the aforementioned parameters.17,18,20 In a study by Brohi et al., clotting factor concentrations on emergency department entry were correlated with both hypoperfusion (measured by the base deficit) and coagulopathy (measured by both the PT and PTT) for 208 trauma activations from 2003 to 2004.17 Coagulopathy was observed only in the presence of hypoperfusion (base deficit >6) and was not related to clotting factor consumption as measured by prothrombin fragment concentrations. Similarly, in a review of trauma patients from our institution who required at least one transfusion, we noted that early (<1 hour postinjury) fibrinolysis occurred frequently among the most severely injured, and correlated significantly with markers of hypoperfusion, such as presenting systolic blood pressure, arterial pH, and base deficit.21
Such studies provide evidence of what we have termed an “acute endogenous coagulopathy of trauma,” which occurs early after injury, is independent of traditional mechanisms of coagulopathy, and is correlated closely with hypoperfusion. Such a mechanism may have evolved to protect hypoperfused vascular beds from thrombosis in the event of ischemia, but is clearly pathologic in the setting of diffuse tissue injury with resultant hemorrhagic shock. Trauma patients who present with this endogenous coagulopathy incur a 4-fold increase in mortality as compared to those patients who do not develop the coagulopathy.22 Furthermore, these patients are eight times more likely to die in first 24 hours,20 and have an increased incidence of multiple organ failure (MOF), transfusion requirements, intensive care unit (ICU) length of stay, and mortality.22
Although the existence of an endogenous coagulopathy of trauma has been well documented, potential mechanistic links to this process remain elusive. Brohi et al. noted in their study that an increasing base deficit was significantly and directly correlated with thrombomodulin concentration (an auto-anticoagulant protein expressed by the endothelium in response to ischemia [Table 13-1]), and inversely correlated to protein C concentration.17 Moreover, a decreased concentration of protein C was correlated with a prolongation of the PTT, suggesting increased activation of protein C via thrombomodulin upregulation as a possible mechanism. Activated protein C (APC), in turn, both inhibits the coagulation cascade via inhibition of factors Va and VIIIa and promotes fibrinolysis via irreversible inhibition of plasminogen activator inhibitor (PAI). A decreased concentration of protein C also correlated with a decrease in the concentration of PAI, an increase in tissue plasminogen activator (tPA) concentration, and an increase in D-dimers. This final observation suggested that protein C–mediated hyperfibrinolysis via consumption of PAI may contribute to traumatic coagulopathy.
Further associations between the endogenous coagulopathy of trauma and the APC pathway have since been described in both animals23 and humans.22 Using a mouse model of hemorrhagic shock, Chesebro et al. documented an association between coagulopathy and an elevated APC concentration (as opposed to the surrogate protein C concentration utilized in the study of Brohi et al.17).23 Inhibition of APC with mAb1591 prevented coagulopathy associated with traumatic hemorrhage (as measured by the PTT). However, complete inhibition of APC caused universal death at 45 minutes due to thrombosis and perivascular hemorrhage, underscoring the delicate balance between hemorrhage and thrombosis.
Others have argued that the early coagulopathic changes following severe injury simply reflect the traditional concepts of DIC.24 Specifically, the hematologic consequences following injury may be considered to represent a generic coagulopathic response to any insult that induces widespread inflammation (e.g., trauma, infection, ischemia/reperfusion). The release of proinflammatory cytokines, in turn, has two main effects on the coagulation system: (1) release of tissue factor with subsequent clotting factor consumption and massive thrombin generation and (2) hyperfibrinolysis due to upregulation of tPA. In favor of this argument is the long-standing documentation of diffuse intravascular thrombi in multiple, uninjured organs of victims of hemorrhagic shock.25 Furthermore, the cytokine elaboration patterns of both trauma and septic patients are nearly identical, suggesting a potential common pathophysiologic mechanism.26 However, this argument is limited by the aforementioned finding that clotting factor levels are relatively preserved in trauma patients following shock.17 Furthermore, fibrinogen levels are inconsistently depressed in patients with acute traumatic coagulopathy. Moreover, the degree of fibrinolysis, when present, appears substantially higher in the endogenous coagulopathy of trauma as compared to DIC.21 Lastly, DIC occurs classically in the setting of an underlying hypercoagulable state (e.g., malignancy, septic shock) and is associated with an upregulation of PAI-1,27 as opposed to the early hypocoagulable state observed in the bleeding trauma patient, which reflects a predominance of both t-PA upregulation and PAI-1 inhibition.
Our current conceptualization of the acute endogenous coagulopathy of trauma emphasizes the integral role of fibrinolysis. Specifically, diffuse endothelial injury leads to both massive thrombin generation and systemic hypoperfusion. These changes, in turn, result in the widespread release of tPA, leading to fibrinolysis. Both injury and ischemia are well-known stimulants of tPA release,28 and we have observed a strong correlation between hypoperfusion, fibrinolysis, hemorrhage, and mortality among injured patients who require transfusion.21 The various proposed pathways involved in the endogenous coagulopathy of trauma are depicted in Fig. 13-1.
FIGURE 13-1 Proposed pathways for the acute endogenous coagulopathy of trauma. Note the prominent role of fibrinolysis via multiple mechanisms, the necessary thrombin substrate, and the positive feedback cycle that perpetuates the coagulopathy. tPA, tissue plasminogen activator; APC, activated protein C; PAI, plasminogen activator inhibitor.
Regardless of the inciting mechanism, elucidation of an endogenous coagulopathy of trauma has important therapeutic implications. Given that the driving force of early coagulopathy appears mediated initially by hypoperfusion as opposed to clotting factor consumption, replacement of clotting factors at this time would be ineffective. In fact, early clotting factor replacement in the face of ongoing hypoperfusion may serve to exacerbate coagulopathy via generation of additional thrombin substrate for thrombomodulin. For this reason, we have noted the endogenous coagulopathy of trauma to be “fresh frozen plasma (FFP) resistant.” By contrast, the development of a secondary coagulopathy due to the complications of massive resuscitation renders the patient clotting factor deficient and thus “FFP responsive.” Elucidation of the integral role of fibrinolysis also raises the possibility of mitigation of the coagulopathy via early administration of antifibrinolytic drugs (discussed below).
Refinement of the mechanisms underlying the endogenous coagulopathy of trauma represents one future goal within the area of postinjury coagulopathy research. Currently, neither a standardized definition nor diagnostic criteria for the endogenous coagulopathy of trauma exist. Furthermore, little is known about the initiators of both upregulation and eventual downregulation of thrombomodulin during traumatic shock. Overexpression of APC has been inferred in humans from a decreased concentration of protein C rather than direct measurement of the APC concentration. Finally, although a correlation between markers of shock and clotting factor expression profiles has been documented, causality remains to be proven. Despite these limitations, description of the endogenous coagulopathy of trauma represents a major turning point in our understanding of the hemostatic derangements following injury.
Although the endogenous coagulopathy of trauma results in an immediate hypocoagulable state among shocked patients following injury, several secondary conditions may develop, which exacerbate this preexisting coagulopathy. Such conditions are, in large part, due to the complications of massive fluid resuscitation, and include clotting factor dilution, clotting factor consumption, hypothermia, and acidosis. Although these factors were considered traditionally as the driving force of traumatic coagulopathy, recent evidence suggests that their effect may have been overestimated.
The positive interaction between hypothermia, acidosis, and coagulopathy has been termed both the “bloody vicious cycle” and “lethal triad of death,” which we proposed at the 40th annual meeting of the American Association for the Surgery of Trauma in 1981.14 Each of these three factors exacerbates the others, eventuating in uncontrolled hemorrhage and exsanguination. Many causes of hypothermia exist for the trauma patient, including altered central thermoregulation, prolonged exposure to low ambient temperature, decreased heat production due to shock, and resuscitation with inadequately warmed fluids. The enzymatic reactions of the coagulation cascade are temperature dependent and function optimally at 37°C; a temperature <34°C is associated independently with coagulopathy following trauma.29However, both experimental and clinical evidence suggest that the effect of hypothermia is modest at best, with each 1° corresponding to a decrease in clotting factor activity of approximately 10%.30 When defined using an elevation of the seconds, hypothermia did not correlate with coagulopathy among a large cohort of trauma patients.22 Thrombin generation was also not effected by hypothermia . Coagulopathy may be relevant clinically in severe hypothermia (T <32°C),31 but this condition is present in less than 5% of trauma patients. Furthermore, it is unclear if the increased mortality observed in severely hypothermic patients is causal or merely circumstantial. Hypothermia also affects both platelet function32 and fibrinolysis33; however, pronounced platelet dysfunction is only observed below 30°C, and clinical data correlating platelet dysfunction secondary to severe hypothermia and adverse outcomes are lacking. Thus, although severe hypothermia exacerbates coagulopathy, advances in resuscitation of the trauma patient have minimized the risk of this degree of hypothermia, thereby limiting its relevance. By contrast, isolated hypothermia likely has minimal clinical impact on hemostasis within the temperature range commonly seen in trauma patients (33–36°C).
Clotting factor activity is also pH dependent, with 90% inhibition occurring at pH 6.8.34 Coagulopathy secondary to acidosis is apparent clinically below a pH of 7.2. Because hypoperfusion results in anaerobic metabolism and acid production, it is difficult to discern the independent effect of acidosis on hemostatic integrity. A recent study adjusted the pH of blood samples from healthy volunteers using hydrochloric acid.35 Using thrombelastography (TEG), a significant correlation was observed between pH and clot-forming time over the pH range of 6.8–7.4. However, no difference was found in either clotting time (a measurement of the time to initiation of a clot) or maximum clot firmness over the range of pH, with the possible exception of a pH equal to 6.8. Similarly, Brohi et al. found that although thrombin generation increased with increasing injury severity, there was no relationship between degree of acidosis (as measured by the base deficit) and either thrombin generation or factor VII concentration.22 Finally, conflicting evidence exists regarding the ability of correction of acidosis via buffer to reverse coagulation disturbances.35,36 Although the independent effect of acidosis on hemostatic integrity remains unclear, correction of acidosis via resuscitation remains a valuable therapeutic end point in terms of minimizing the aforementioned hypoperfusion-induced endogenous coagulopathy of trauma. Furthermore, maintenance of the arterial during resuscitation of shock (with bicarbonate, if necessary) maximizes the efficacy of both endogenous and exogenous vasoactive drugs.
Finally, although both consumption and dilution of clotting factors have been implicated in postinjury coagulopathy, there is little experimental evidence to support this theory. The amount of thrombin generated is not related to coagulopathy in patients without shock,17 and there is no effect of dilution on coagulopathy either in vitro37 or in healthy volunteers.38
In summary, an endogenous coagulopathy occurs following trauma among patients sustaining shock, and does not appear to be secondary to coagulation factor consumption or dysfunction. Rather, current evidence suggests that it is due to ischemia-induced both anticoagulation and hyperfibrinolysis, and is resistant to clotting factor replacement. Although the hematologic changes observed following severe trauma demonstrate many characteristics of DIC with a fibrinolytic phenotype, clotting factor consumption does not appear integral. During this time frame, therapy should focus on definitive hemorrhage control, timely restoration of tissue perfusion, and point-of-care monitoring in an effort to identify fibrinolysis. Following restoration of tissue perfusion, an “FFP-sensitive” pathway may emerge, which is characterized by coagulopathy due to traditional factors, such as acidosis, hypothermia, consumption, and dilution. Recognition of the transition from the “FFP-resistant” to the “FFP-sensitive” pathway is a critical objective of current research. Fig. 13-2 depicts our “updated” bloody vicious cycle, which incorporates both the acute endogenous coagulopathy of trauma and the aforementioned secondary factors. Finally, a hypercoagulable state supervenes following restoration of tissue perfusion, usually within 72 hours of injury.
FIGURE 13-2 Updated bloody vicious cycle. It incorporates both the early acute endogenous coagulopathy of trauma, which is resistant to clotting factor replacement with fresh frozen plasma (FFP resistant), and a subsequent secondary coagulopathy that may be due to hypothermia, acidosis, clotting factor deficiency (FFP sensitive), or any combination thereof.
Permissive hypotension involves deliberate tolerance of lower mean arterial pressures in the face of uncontrolled hemorrhagic shock in order to minimize further bleeding. This strategy is based on the notion that decreasing perfusion pressure will maximize success of the body’s natural mechanisms for hemostasis, such as arteriolar vasoconstriction, increased blood viscosity, and in situ thrombus formation. Animal models of uncontrolled hemorrhage have revealed that crystalloid resuscitation to either replace three times the lost blood volume39 or maintain 100% of pre-injury cardiac output40 exacerbates bleeding39,40 and increases mortality39 as compared to more limited fluid resuscitation.
Randomized clinical trials (RCTs) that compare fluid management strategies prior to control of hemorrhage among human subjects are limited. In the first large-scale trial, Bickell et al. randomized 598 patients in hemorrhagic shock (systolic blood pressure <90 mm Hg) who had sustained penetrating torso trauma to either crystalloid resuscitation or no resuscitation prior to operative intervention.41Prespecified hemodynamic targets were not used. Mean systolic arterial blood pressure was significantly decreased on arrival to the emergency department for the delayed resuscitation group as compared to the immediate resuscitation group (72 mm Hg vs. 79 mm Hg, respectively, P = .02) with a corresponding increase in survival (70% vs. 62%, respectively, P = .04). A trend toward a decreased incidence of postoperative complications was also observed for the delayed resuscitation group. However, a subsequent subgroup analysis documented that these benefits occurred only among patients who had cardiac injury with tamponade.42
Two recent trials have failed to replicate these findings. Turner et al. randomized 1,306 trauma patients with highly diverse injury patterns and levels of stability to receive early versus delayed or no fluid resuscitation.43 Although no mortality difference was observed (10.4% for the immediate resuscitation group versus 9.8% for the delayed/no resuscitation group), protocol compliance was poor (31% for the early group and 80% for the delayed/no resuscitation group), limiting interpretability. Most recently, Dutton et al. randomized 110 trauma patients presenting in hemorrhagic shock (systolic blood pressure <90 mm Hg) to receive crystalloid resuscitation to a systolic blood pressure of >70 mm Hg versus >100 mm Hg.44 Randomization occurred following presentation to the emergency department. Not all patients required operation, and hemorrhage control was determined at the discretion of the trauma surgeon or anesthesiologist. Although there was a significant difference in mean blood pressure during bleeding between the conventional and low groups (114 mm Hg vs. 110 mm Hg, respectively, P < .01), the mean blood pressure was substantially higher than intended (< 70 mm Hg) for the low group, and the absolute difference between groups was likely insignificant clinically. Mortality was infrequent and did not vary by resuscitation arm (7.3% for each group).
Methodological variability between these trials has precluded a meaningful meta-analysis,45 and may help to explain the discrepant mortality findings. It is clear that the degree of hemorrhagic shock was most pronounced in the study of Bickell et al., as evidenced by the lowest presenting systolic blood pressure as well as the highest mortality. Furthermore, randomization was accomplished in the prehospital setting, and all patients required operative intervention. By contrast, mortality was infrequent in the study of Dutton et al., and the target systolic blood pressure of 70 mm Hg in the “low” group was, on average, not achieved. Thus, at present, it is possible to conclude that limited volume resuscitation prior to operative intervention may be of benefit among patients with penetrating trauma resulting in cardiac injury, although the optimum level of permissive hypotension remains unknown. The benefit of such therapy among a more diverse cohort of patients in hemorrhagic shock, with a low associated risk of death, is not clear. Finally, regardless of therapeutic benefit, reliable achievement of permissive hypotension appears challenging once hospital care has begun.
Preemptive Blood Components
The widespread replacement of whole blood by component therapy in the early 1980s allowed for improved specificity of therapy, increased storage time of individual components, and decreased transmission of infectious disease. However, the relative amounts (if any) of components indicated for resuscitation of the exsanguinating trauma patient were not addressed, and remain debated approximately 30 years later. Traditional doctrine, as espoused by Advanced Trauma Life Support training, calls for 2 L of crystalloid followed by RBCs in the case of persistent hemodynamic instability; clotting factor and platelet replacement are indicated only in the presence of laboratory derangements (PT and platelet count, respectively).46
Although this approach is reasonable for patients who have sustained relatively minor hemorrhage (<20% of circulating blood volume), replacement of lost blood with isolated erythrocytes becomes problematic in the face of ongoing hemorrhagic shock requiring a large volume of blood transfusion. In this case, replacement of shed blood with isolated RBCs will result in a dilutional coagulopathy. Several authors have attempted to quantify the amount of RBCs transfused for which dilutional coagulopathy mandates concomitant component replacement therapy, with definitions ranging from loss of one blood volume to the need for greater than 10 U of RBCs in the first 24 hours following injury. The latter criterion is the most commonly accepted definition of MT, and is the time period on which most studies of empiric component replacement therapy are based. However, because over 80% of blood component therapy transfused to patients who require MT is administered within the first 6 hours of injury,47 we believe this to be a more appropriate time period for analysis. Thus, the focus on preemptive blood components should shift to the first 6 hours postinjury.
The debate regarding preemptive blood components began with platelets during the time of whole blood resuscitation.48 Recognition of the dangers of isolated RBC therapy during MT followed shortly after the widespread institution of component therapy in the early 1980s. Our group and others noted that mortality among massively transfused patients was reduced when increased amounts of both plasma and platelets were administered empirically. Specifically, when introducing the concept of RBC:FFP ratios, we reported increased mortality among a cohort of patients with major vascular trauma associated with RBC:FFP ratios greater than 5:1, with overt coagulopathy observed nearly universally with ratios exceeding 8:1.14 In 2007, Borgman et al. published a series of 254 massively transfused US soldiers in Iraq and Afghanistan, reporting markedly improved survival among those transfused with an RBC:FFP ratio in the range of 1.5:1, as compared to higher ratios.49 This ratio appeared appealing intuitively as it most closely resembled that of whole blood, although a 1:1 formulation is actually both anemic (hematocrit 27%) and clotting factor deficient (65% activity) as compared to fresh whole blood.50 Several subsequent studies, in both the military and civilian settings, have corroborated the findings of Borgman et al.51–53 An association between early, aggressive FFP administration and improved survival has also been documented among trauma patients who underwent sub-MT.54 These data have given rise to the concept of damage control resuscitation, which involves early transfusion of increased amounts of both clotting factors and platelets, in addition to minimization of crystalloid resuscitation in patients who are expected to require MT. Currently, in the absence of routine point-of-care assessment of coagulopathy, many trauma centers advocate preemptive transfusion of RBC:FFP using a target ratio of 1:1 for such patients.
Unfortunately, the literature addressing component transfusion ratios during MT suffers from substantial methodological limitations (Table 13-2). Despite a myriad of retrospective data, mathematical models,55 and expert opinion, there remains no prospective evidence to support an empiric transfusion ratio. A major limitation of the retrospective literature involves survival bias. Specifically, it remains unclear if increased FFP transfusion improves survival or if patients who survive simply live long enough to receive more FFP. Indeed, patients who are bleeding faster get less plasma as the trauma team and blood bank struggle to keep up. Related intimately to the issue of survival bias is that of the time period over which the RBC:FFP ratio is calculated. Although over 80% of RBC transfusions are administered within 6 hours of injury, most studies have reported the cumulative RBC:FFP ratio as calculated at 24 hours. Such a strategy exacerbates survival bias, as the RBC:FFP ratio is known to decrease over time. Accounting for the time-dependent nature of the RBC:FFP transfusion ratio eliminated any association with survival in one recent report.56 Furthermore, when the cumulative RBC:FFP ratio was analyzed at 6 hours as opposed to 24 hours, our group identified a ratio in the range of 2:1 to 3:1, as opposed to 1:1, as associated with the lowest predicted mortality.47
TABLE 13-2 Methodological Limitations of the Literature Addressing FFP:RBC Ratios During Massive Transfusion
The next major limitation involves the lack of a mechanistic link between a lower RBC:FFP ratio and improved survival. The clinical efficacy of FFP remains largely unproven,57 and no study has documented an association between a lower RBC:FFP ratio and fewer total blood products administered. Moreover, differences in laboratory markers of coagulopathy (e.g., PT, TEG) have not been demonstrated between groups of varying RBC:FFP ratios. In fact, a canine model showed no benefit to adding FFP following MT in terms of changes in coagulation protein levels or clotting times.58 The benefit of a 1:1 RBC:FFP ratio has also not been consistent across various mechanisms of injury.59 Finally, early, aggressive FFP administration may serve to exacerbate the endogenous coagulopathy of trauma by generating large amounts of thrombin that, in the setting of ischemia-induced endothelial thrombomodulin overexpression, may perpetuate a hypocoagulable state via continued activation of protein C.
Although many experts advocate an RBC:FFP transfusion ratio of 1:1 during MT, the lowest ratio achieved in most studies approaches 1.5:1. While moving from an RBC:FFP ratio of 2:1 to 1:1 may appear trivial, such a paradigm shift represents a 100% increase in FFP utilization. An increase of this magnitude would place tremendous strain on the marginal FFP donor pool, as well as increase exponentially blood bank labor, likely to the point of nonsustainability in the event of a mass casualty. Finally, unbridled FFP administrated must be viewed with caution in light of the accumulating evidence detailing the immunomodulatory properties of such therapy (discussed below).
The trigger for empiric platelet replacement during MT remains equally controversial. Both thrombocytopenia60 and thrombocytopathy (qualitative platelet dysfunction)61 have long been implicated in the coagulopathy of trauma. Nearly all massively transfused patients develop thrombocytopenia,60 and the bleeding time is universally prolonged in these patients. Interestingly, thrombocytopenia develops following MT even when the lost blood volume is replaced as whole blood, suggesting a mechanism independent of thrombodilution. By contrast, platelet dysfunction following transfusion of less than 10 U of pRBC appears rare.61 This is likely due to the fact that platelets are both stored in the reticuloendothelial system and sequestered on endothelium. These observations have led to the traditional recommendation of empiric platelet transfusion following each 10 U of pRBCs.
However, despite the high prevalence of thrombocytopenia following MT, the platelet count itself has not been found to predict coagulopathy. Harrigan et al. studied platelet function in 22 patients in hemorrhagic shock who had received at least 10 U of RBC (mean 21 U, range 10–80 U) but no platelet replacement therapy.48 Both the platelet count and markers of platelet function, including the bleeding time and aggregation to ADP, were depressed. However, no patient developed clinically apparent coagulopathy. Phillips et al. were similarly unable to predict postoperative coagulopathy based on the lowest platelet count among a cohort of patients who required more than 20 U of RBC in 24 hours.16 One half of patients with a platelet count <50,000 did not appear to have coagulopathy.
An absolute platelet transfusion trigger thus does not exist for the trauma patient requiring MT. Many investigators have similarly attempted to identify an ideal ratio of RBC:platelets for preemptive platelet administration during MT. Only one group has studied the impact of prophylactic platelet administration in a randomized fashion.62 Reed et al. randomized 33 patients who required MT, defined as ≥12 U of modified whole blood within 12 hours of presentation, to receive either 6 U of platelets or 2 U of FFP for every 12 U of modified whole blood transfused. The primary outcome was the development of microvascular, nonmechanical bleeding, determined subjectively by the surgical team. Three patients in each group developed clinical coagulopathy (approximately 18% per group), and the platelet counts were no different between groups at any time point. The authors concluded that prophylactic platelet administration during MT was not warranted. However, apart from small sample size and subjective outcome, this study both employed a relatively low RBC:platelet ratio (12:1) and occurred with modified whole blood, limiting its contemporary relevance.
More recent retrospective investigations into empiric platelet administration during MT are limited by the same problems inherent to the RBC:FFP ratio data. Specifically, several authors have shown that a decreased RBC:platelet ratio is associated with improved survival, with the lowest mortality observed in the range of 1:2 to 1:5.52 These data have been extrapolated to recommend empiric therapy with a ratio of 1:1. However, this so-called 1:1 is actually 1:0.2 as it is based on random donor units of platelets (5.5 × 1010) rather than apheresis units (3.0 × 1011).50
Regardless of the specific ratio, a causal relationship to these data is lacking, and confounding, survival bias, and lack of a mechanistic link render these data difficult to interpret. The overall poor quality of evidence supporting a low RBC:platelet ratio, as well as the morbidity associated with increased platelet transfusion, disfavors this strategy. Our current policy is to transfuse platelets based on point-of-care evidence of dysfunction as determined by TEG, or following the 10th unit of RBC transfusion, regardless of laboratory evidence of platelet dysfunction.
In summary, the literature involving empiric component therapy suffers from several methodological limitations. Currently, the optimal empiric RBC:FFP ratio for resuscitation of patients who require MT appears to be in the range of 1:2 to 1:3. Furthermore, although no absolute numerical trigger for platelet transfusion exists, it appears reasonable to transfuse platelets following the 10th unit of pRBCs. However, whenever possible, component replacement should be both individualized and goal directed, such that overzealous clotting factor and platelet replacement and the complications thereof are minimized.
Lack of an accurate tool to identify and track coagulopathy remains a major limitation of the literature surrounding both postinjury hemostatic derangements and empiric blood component replacement therapy. Classic laboratory tests of coagulation function, such as PT and PTT, were described originally for the assessment of anticoagulation function in hemophiliacs, and are based on the interaction of the coagulation factors in isolation.63 To date, not surprisingly, the performance characteristics of these tests in the trauma patient remain unproven. Furthermore, a prohibitive amount of time (approximately 45 minutes) is required to conduct these assays. Because both the PT and PTT are performed on platelet-poor plasma, they are sensitive only to the earliest initiation of clot formation. However, greater than 95% of thrombin generation occurs after the initial polymerization of fibrinogen. Hence, monitoring of platelet function appears essential for accurate measurement of clot strength. Finally, these tests are performed in an artificial environment, irrespective of the patient’s core body temperature and pH. Measurements of individual clotting proteins, such as protein C and thrombomodulin, are both costly and time consuming. Diagnosis of fibrinolysis is also problematic. The euglobulin lysis time is a complex and time-consuming procedure that can take more than 180 minutes. Other techniques used to identify hyperfibrinolysis, such as plasmin–antiplasmin complex, PAI-1, thrombin activatable fibrinolysis inhibitor, and D-dimers, suffer from the same limitations. Thus, all major aspects of the hemostatic system are measured inadequately using conventional laboratory testing of trauma patients. The limitations of these tests in the trauma setting have been validated repeatedly.64,65
In response to the shortcomings of conventional measurements of coagulopathy, point-of-care, rapid TEG (R-TEG) is emerging as the standard of care for both the diagnosis and treatment of postinjury coagulopathy at many trauma centers. TEG provides a rapid, comprehensive assessment of in vivo coagulation status. The TEG analyzer is composed of two mechanical parts separated by a blood specimen: a plastic cup or cuvette, into which a 0.36-mL blood specimen is pipetted, and a plastic pin attached to a torsion wire and suspended within the specimen (Fig. 13-3). Once the sample within the cuvette is placed on the TEG analyzer, the temperature is adjusted to that of the patient. The cup then oscillates slowly through an angle of 4°45′. Initially, movement of the cuvette does not affect the pin, but as clot develops, resistance from the developing fibrin strands couples the pin to the motion of the cuvette. In turn, the torsion wire generates a signal that is amplified and records the characteristic tracing seen in Fig. 13-3. In earlier iterations, this tracing was recorded on heat-sensitive paper moving at a rate of 2 mm/min. More recent computer technology has allowed for automatic calculation of TEG variables. At our institution, a dynamic TEG tracing is transmitted directly to the operating room or ICU via computer within minutes, enabling prompt interpretation.
FIGURE 13-3 Thrombelastography instrument and tracing. The instrument diagram depicts the cuvette where a whole blood sample is placed, and the pin attached to a torsion wire. Once the assay is initiated, a tracing is produced and an initial linear segment (zone of precoagulation) extends from the beginning of the test to the formation of the first fibrin strand, causing the tracing to split (split point). The progressive divergence of the tracing reflects the formation of the clot. (Image of the TEG® Thromboelastograph® instrument and hemostasis tracing is used by permission of Haemonetics Corporation.)
The various components of the TEG tracing are depicted in Fig. 13-3. The split point (SP, minutes) is a measure of the time to initial clot formation, interpreted from the earliest resistance detected by the TEG analyzer causing the tracing to split; this is the point at which all other platelet-poor plasma clotting assays (e.g., PT and PTT) fail to progress. The reaction time (R, minutes) is defined as the time elapsed from the initiation of the test to the point where the onset of clotting provides enough resistance to produce a 2-mm amplitude reading on the TEG tracing. Of note, in the R-TEG assay (discussed below), due to the acceleration of clotting initiation, the R time is represented by a TEG-derived activated clotting time (TEG-ACT). The R time and TEG-ACT are most representative of the initiation phase of enzymatic clotting factors. Whereas a prolonged R time or TEG-ACT is diagnostic of hypocoagulability, decreased values may suggest hypercoagulability. The coagulation time (K, minutes) is a measurement of the time interval from the R time to the point where fibrin cross-linking provides enough clot resistance to produce a 20-mm amplitude reading. The alpha-angle (α, degrees) is the angle formed by the slope of a tangent line traced from the R to the K time measured in degrees. Both the K time and the alpha-angle denote the rate at which the clot strengthens and are most representative of thrombin’s cleaving of available fibrinogen into fibrin. The maximum amplitude (MA, millimeters) indicates the point at which clot strength reaches its maximum measure in millimeters on the TEG tracing, and reflects the end result of maximal platelet–fibrin interaction via GPIIb–IIIa receptors. The clot strength (G, dynes/cm2) is a calculated measure derived from amplitude (A, mm); . Due to its exponential relationship with A, G appears to be the best measure of overall clot strength. Finally, the estimated percentage lysis (EPL, %) corresponds to the percentage of the clot that has lysed at a given time point (e.g., 5 minutes). The various TEG parameters and their significance are summarized in Table 13-3.
TABLE 13-3 TEG Parameters
Blood coagulation as measured by TEG is initiated by addition of an activating solution consisting of kaolin, phospholipids, and buffered stabilizers, which requires an activation phase of several minutes before coagulation starts. To expedite time to generate results (e.g., in the setting of hemorrhagic shock), clotting initiation can be further prompted by addition of tissue factor. This permits the earliest tracings via R-TEG to be viewed within 10 minutes.
Fig. 13-4 depicts deranged TEG profiles characteristic of specific coagulation abnormalities along with a normal TEG tracing (tracing A). Anticoagulation causes enzymatic inhibition of coagulation factors and produces a prolonged R time due to delayed initiation of clot formation, along with normal fibrinogen (normal K time and alpha-angle) and platelet function (normal MA) (tracing B). Platelet dysfunction is noted by a tracing with a normal R time and a primarily decreased MA, as seen in tracing C. We have defined clinically significant fibrinolysis when EPL values exceed 15%,21 resulting in a characteristic tapering of the TEG tracing immediately after the MA is reached (tracing D). Tracing E is characteristic of hypercoagulability, identifiable by decreased R and K times along with an elevated alpha-angle and MA.
FIGURE 13-4 Characteristic thrombelastographic tracings observed in the trauma patient. (A) Normal tracing; (B) prolonged R time seen with coagulation factor deficiency or inhibition by anticoagulation; (C) decreased maximum amplitude (MA), seen during platelet dysfunction or pharmacologic inhibition; (D) hyperfibrinolysis; (E) decreased R and K times, elevated alphaangle, and increased MA represent a hypercoagulable state. (Image of the TEG® Thromboelastograph® hemostasis tracings is used by permission of Haemonetics Corporation.)
Recent data suggest the superiority of TEG as compared to both aPTT and PT/INR for assessment of the acute coagulopathy of trauma. Kheirabadi et al. showed in a rabbit model that TEG is a more sensitive indicator of dilutional hypothermic coagulopathy than PT.66 A recent clinical study of trauma patients surviving the first 24 hours reported that TEG detected hypercoagulability (by an increased MA and G), whereas the PT and aPTT did not.67 In a retrospective review of penetrating injuries, Plotkin et al. demonstrated hypocoagulation based on delayed propagation of the clot (increased K time and reduced alpha-angle) and decreased clot strength (reduced MA), where MA correlated with blood product use (, ).68 These findings emphasize the limitations of classic coagulation tests and their lack of efficacy particularly in postinjury coagulopathy.
Goal-directed transfusion therapy guided by TEG tailors blood product administration to the pathophysiologic state of the individual patient. At our institution, this approach has become an integral part of resuscitation. Importantly, empiric blood component replacement using an RBC:FFP ratio of approximately 2:1 is initiated prior to attainment of the first TEG tracing; this time period, however, rarely exceeds 10 minutes. An initial hemostatic assessment with R-TEG identifies patients at risk for postinjury coagulopathy on arrival. Following interpretation of the initial tracing, blood component therapy is then tailored to address each deranged phase of clotting in a specific manner, while subsequent reassessment allows the evaluation of response, until a set threshold is reached. This strategy also permits improved communication with the blood bank; based on initial assessment and response to component therapy, more accurate estimations of component requirements can be made using real-time data.
Our current goal-directed approach to coagulopathy is depicted in Fig. 13-5. Reflecting the initiation phase of enzymatic factor activity, a prolonged TEG-ACT value is the earliest indicator of coagulopathy; when the value is above threshold, FFP is administered. K time and alpha-angle are most dependent on the availability of fibrinogen to be cleaved into fibrin while in the presence of thrombin. If indicated by K and alpha-angle, cryoprecipitate is administered, providing a concentrated form of fibrinogen. The MA demonstrates the relationship between fibrin generated during the initial phases of hemostasis and platelets via IIb–IIIa receptor interaction. Platelets are administered based on an , which reflects the platelets’ functional contribution to clot formation. Of note, in this protocol, platelet transfusion is not based on platelet count but on functional contribution to clotting, reflected by the MA. Finally, antifibrinolytics are administered for an . Importantly, interpretation of the various TEG parameters occurs in parallel, rather than in series, such that multiple coagulation derangements are corrected simultaneously.
FIGURE 13-5 Goal-directed hemostatic resuscitation via thrombelastography. Using point-of-care rapid thrombelastography (TEG), all components of the coagulation system of the bleeding trauma patient are both accessed and treated in parallel. Following goal-directed therapy based on the initial TEG tracing, coagulation status is reaccessed using serial rapid TEG determinations, until correction of each coagulation derangement is achieved. ACT, activated clotting time; MA, maximum amplitude; EPL, estimated percentage fibrinolysis; FFP, fresh frozen plasma; CRYO, cryoprecipitate; ACA, α-aminocaproic acid.
In summary, TEG offers the following major advantages over traditional coagulation status testing: (1) rapid, point-of-care testing, (2) assessment of the spectrum of coagulation function, including fibrinolysis and thrombocytopathy, (3) reflection of in vivo clotting activity, and (4) serial graphical representations of goal-directed response to therapy. Implementation of a goal-directed approach to postinjury coagulopathy may reduce transfusion volumes, attain earlier correction of coagulation abnormalities, improve survival in the acute hemorrhagic phase due to improved hemostasis from correction of coagulopathy, and improve outcomes in the later phase due to attenuation of the complications of overzealous blood product administration.
Preliminary validation studies of these potential benefits have been encouraging. Retrospective data and pilot studies support a reduction in MT rates, decreased need for multiple and repeated classic coagulation tests, and decreased morbidity and mortality after implementation of TEG in trauma care.69,70 Our group recently compared pre-TEG to post-TEG outcomes of patients at risk of postinjury coagulopathy admitted to our trauma center.71 The TEG G value was significantly associated with survival , whereas PT/INR and PTT did not discriminate between survivors and nonsurvivors. Further validation studies are ongoing at our institution as well as other trauma centers and will be necessary before supporting widespread application of this technology.
Recombinant Factor VIIa
Treatment of postinjury coagulopathy with supraphysiologic doses of recombinant factor VIIa (rVIIa) is believed to amplify coagulation through generation of a thrombin burst in the presence of both tissue factor and functional platelets, and in the absence of either hypothermia or acidosis. Although the drug was developed originally for use in patients with hemophilia A,72 several case reports of successful off-label use in patients with postinjury coagulopathy raised enthusiasm for a formal indication in this setting. Commonly prescribed doses (range 50–200 μg/kg) for this indication are estimated to be greater than 100 times physiologic.
In 2004, the Maryland Shock Trauma Center reported a case series detailing outcomes of rVIIa use in 81 trauma patients who had undergone MT (defined in this series as 10 U RBC, 8 U FFP, and 1 U platelets).73 A standard dose of 100 μg/kg was used, and 80% of patients received the medication within 24 hours of injury. Coagulopathy, as measured by the PT, improved following administration of rVIIa in all patients, and no thrombotic events were observed. Furthermore, 61 of 81 (75.3%) patients were considered “responders” based on a subjective improvement in hemostasis noted by the trauma team. However, when rVIIa patients were compared to matched coagulopathic registry patients who had not received rVIIa, mortality was equivalent (50% vs. 44%, respectively, ).
Due to continued concerns over both efficacy and thrombotic risk, two parallel randomized controlled trials, one of penetrating and one of blunt trauma patients, were conducted among 32 international institutions (exclusive of the United States) from 2003 to 2004.74 Eligibility criteria mandated transfusion of ≥6 U RBC in 4 hours. Subjects were randomized to receive either placebo or three doses of rVIIa, 200, 100, and 100 μg/kg, with the first dose given after the eighth unit of RBC transfused, and the second and third doses 1 and 3 hours following. The primary end point was the number of RBC transfusions in the 48 hours following the first dose, although no standard transfusion trigger was specified. The trials were powered to detect a clinically relevant reduction in transfusion requirement of 2.6 U. A reduction in RBC transfusion requirement for both blunt (median 7.0 vs. 7.5, ) and penetrating (3.9 vs. 4.2, ) groups was observed among those alive 48 hours after the first dose of the study drug. However, these relatively small differences were eliminated when the analysis was expanded to include patients who had died within the first 48 hours of injury, a subgroup of clear importance. A significant difference in the need for MT, defined as >20 total U RBC, was also eliminated after accounting for these early deaths. Although a significant difference in the incidence of ARDS was noted between the rVIIa and placebo groups among blunt trauma patients (4% vs. 16%, respectively, ) (again only among those alive at 48 hours), no difference in length of stay, ventilator days, or mortality was observed in any analysis.
A subsequent subgroup analysis of 136 patients with coagulopathy reported decreases in both RBC and FFP transfusion requirements at 48 hours among the rVIIa patients as compared to controls (delta = 2.6 U and 600 mL, respectively).75 Importantly, these analyses were not limited to 48-hour survivors. However, mortality remained equivalent. Again, no standardized transfusion practice was followed, and the definition of coagulopathy was determined subjectively as an FFP:RBC transfusion ratio >1:4, or any transfusion of either platelets or cryoprecipitate.
Thus, the only level I evidence regarding use of rVIIa in the trauma setting is hindered by both methodological limitations, such as a lack of power and standardized transfusion policy, and a failure to demonstrate an improvement in either coagulopathy (which was not measured) or survival. Most recently, Knudson et al. compared patients from a multicenter rVIIa registry to patients in a multicenter MT study who had not received rVIIa.76 Patients who had received rVIIa had higher mortality and increased death due to hemorrhage. Thus, at present, the initial enthusiasm for rVIIa in the trauma setting has been curtailed by a lack of data documenting efficacy, prohibitive cost, and continued concerns for increased thrombotic risk.77
If rVIIa therapy is contemplated, several pharmacokinetic properties of the drug warrant consideration. The activity of rVIIa is both pH and temperature dependent, with a 60% decrement at a pH of 7.20 and a 20% decrement at a temperature of 33°C.34 Furthermore, physiologic concentrations of calcium, fibrinogen, and platelets are required, such that replacement of these clotting factors is mandatory prior to rVIIa administration. Dosages range from 50 to 200 μg/kg, with redosing required in the face of continued coagulopathy beyond the half-life of 2 hours.
As the technology for diagnosing coagulopathy improves, further indications for treatment with rVIIa may emerge. For example, our group has observed a postfibrinolysis consumptive coagulopathy, which is characterized by serial TEG tracings documenting diffuse clotting factor deficiency secondary to massive consumption after fibrinolysis. The resulting severe thrombin deficiency may respond to rVIIa, and we have noted anecdotally rapid improvement with normalization of TEG patterns after such treatment.
All current theories of the acute endogenous coagulopathy of trauma implicate hyperfibrinolysis. Our recent report showed that fibrinolysis was identified via point-of-care TEG within 1 hour of injury, and correlated significantly with all traditional parameters of shock, MT, and mortality.21 Furthermore, our identification of lower levels of fibrinolysis in patients with less severe injury supports current theories suggesting a mechanism based on varying concentrations of tPA and PAI, and implicates fibrinolysis as an integral component of the endogenous coagulopathy of trauma.
Antifibrinolytic agents are employed commonly for hyperfibrinolysis associated with elective (predominantly cardiac) surgery, with favorable efficacy and safety profiles.78 Such agents include the competitive plasmin inhibitors α-aminocaproic acid (Amicar) (dosing = 150 mg/kg followed by 15 mg/h) and tranexamic acid (dosing 10 mg/kg, followed by 1 mg/(kg h)), as well as the direct plasmin inhibitor aprotinin (dosing 280 mg, followed by 70 mg/h). In addition to plasmin, aprotinin inhibits tissue kallikrein, trypsin, and factor XIIa. Because use of aprotinin (currently unavailable) is limited due to the need for a test dose, expense, and increased risk of both renal and thrombotic complications, we prefer α-aminocaproic acid.
The evidence for efficacy of antifibrinolytics in the treatment of postinjury coagulopathy is less clear, and no agent currently carries FDA approval for this indication. Two previous trials have attempted to evaluate the benefit of aprotinin in a randomized fashion.79,80 However, small samples, imprecise outcome reporting, and interval advancements in the care of the bleeding trauma patients all limit the applicability of these trials.81 The results of a third RCT addressing the efficacy of antifibrinolytics in the trauma setting, entitled the Clinical Randomization of an Antifibrinolytic in Significant Hemorrhage (CRASH) II study, were published recently.82 The trial randomized 20,127 adult trauma patients with significant hemorrhage (defined as systolic blood pressure <90 mm Hg, pulse >110 beats/min, or both) from 274 collaborating hospitals in 40 countries to receive tranexamic acid (1 g over 10 minutes load followed by an additional gram over 8 hours) or placebo within 8 hours of injury. A small but significant reduction in both all-cause mortality (14.5% vs. 16.0%, respectively, ) and death due to hemorrhage (4.9% vs. 5.7%, respectively, ) was observed for the tranexamic acid as compared to the placebo group. One important limitation of this study involved the lack of point-of-care diagnosis of fibrinolysis as an inclusion criterion. Our data suggest that although approximately two thirds of patients who require MT demonstrate evidence of fibrinolysis, only one third demonstrate clinically significant fibrinolysis .21 Therefore, analysis of all trauma patients in the CRASH-II study may have diluted any survival benefit for the subgroup of patients with significant fibrinolysis. Our current standard is to treat bleeding trauma patients who demonstrate evidence of clinically significant hyperfibrinolysis on a TEG tracing with α-aminocaproic acid.
Factor XIII acts to stabilize the fibrin clot once formed, and a clear relationship between XIII concentration and maximal clot firmness as demonstrated by TEG has been demonstrated.83 Early depletion of the limited endogenous XIII stores has been documented following major hemorrhage, and is believed to contribute to coagulopathy.84 Although prospective data are lacking, empiric therapy with rXIIIa should be considered in postinjury coagulopathy that is refractory to conventional factor replacement therapy.
Prothrombin Complex Concentrate
Prothrombin complex concentrate (PCC) contains the vitamin K–dependent coagulation factors II, VII, IX, and X, as well as the anticoagulants protein C and protein S. As such, its main indication involves rapid reversal of the anticoagulant effects of warfarin in bleeding patients. Importantly, PCC provides these factors in a concentrated form as compared to FFP (approximately 50 mL vs. 300 mL, respectively). The efficacy of PCC for urgent reversal of warfarin reversal in the perioperative period has been well documented85; use of PCC in this setting has been recommended by expert panels in both the United States and Europe.86 Although RCTs in head-injured, warfarinized patients are lacking, it seems reasonable to use PCC in this setting as well, particularly when concern for volume overload is high. Kalina et al. reported decreased time to normalization of the INR and operative intervention using PCC as compared to a combination of vitamin K and FFP in a review of warfarinized patients with intracranial hemorrhage (ICH).87
The role of PCC in the nonwarfarinized patient continues to evolve. Bruce and Nokes reported favorable outcomes following PCC administration for uncontrolled hemorrhage following cardiac surgery, exclusive of warfarin use.88The benefit of PCC over traditional therapies for the treatment of postinjury coagulopathy remains unproven. Several recent animal studies of hemodilutional coagulopathy have reported superior achievement of hemostatic resuscitation using PCC as compared to both FFP89 and rVIIa.90 However, although the reduced time and volume necessary to achieve coagulation factor replacement are appealing, transfusion of PCC likely carries the same immunomodulatory risks as those of FFP. Furthermore, the cost of PCC is currently prohibitive at many trauma centers. Controlled trials in humans are necessary prior to definitive recommendations.
PRE-INJURY ANTITHROMBOTIC AGENTS
Warfarin is an oral anticoagulant that inhibits synthesis of the vitamin K–dependent clotting factors II, VII, IX, and X, as well as the anticoagulant protein C and protein S. It is the most common drug prescribed to achieve chronic anticoagulation; 1% of the world’s population uses warfarin currently.91 As the elderly demographic continues to expand exponentially, warfarin use will only increase.
Several studies have documented an increased likelihood of traumatic ICH, as well as severity of injury and mortality, among trauma patients using warfarin as compared to nonwarfarinized patients.92–94Importantly, the increased likelihood of adverse outcomes is observed only among patients who have achieved therapeutic anticoagulation , negating the possibility of underlying demographic differences between warfarin and nonwarfarin patients.92
Rapid identification of the warfarinized patient at risk for traumatic ICH, as well as reversal of anticoagulation, is paramount to minimizing hemorrhage progression with resultant irreversible neurologic devastation. Most authors advocate empiric transfusion of FFP prior to laboratory documentation of therapeutic anticoagulation in the warfaranized patient at risk for head injury, arguing that the morbidity of a delay in reversal of patients with a supratherapeutic INR outweighs that of unnecessary factor administration to patients with a subtherapeutic INR. Although up to one half of warfaranized trauma patients present with a subtherapeutic INR,92 the morbidity of an expanding ICH renders this argument reasonable. However, a striking amount of variability exists among trauma surgeons as to the INR above which reversal of anticoagulation should be implemented, the rapidity with which the PT is normalized, or the target INR following reversal.95
Ivascu et al. developed a protocol in which warfarin anticoagulated trauma patients suspected to have ICH underwent rapid triage and computed tomography.96 In the case of positive imaging, 2 U of universal donor, prethawed FFP, and 10 mg vitamin K were administered. Protocol subjects were compared to historical controls. The authors observed a decreased time to evaluation, diagnostic imaging, and full anticoagulation reversal . Furthermore, the time to initiation of treatment decreased from 4 to 2 hours. The percentage of patients whose hemorrhage progressed decreased from 40% to 11%, with an associated mortality decrease from 50% to 10%.
Reversal of warfarin-induced anticoagulation using either PCC or rVIIa may be particularly beneficial in patients at high risk for cardiopulmonary complications secondary to large-volume clotting factor concentrate administration. Although case series have documented efficacy without thrombotic morbidity,87,97 prospective, controlled data are lacking.
The risk of warfarin anticoagulation in the non-head-injured trauma patient is less clear. Ott et al. studied 212 blunt trauma patients of age ≥60 years who sustained trauma to the abdomen or thorax in the absence of intracranial injury.98 No outcome differences between warfarinized patients (mean INR 2.1) and nonwarfarinized patients were observed.
Antiplatelet therapies, which consist predominantly of acetylsalicyclic acid (ASA) and clopidogrel, have revolutionized the care of patients with atherosclerotic cardiovascular disease. Although via independent mechanisms, both ASA and clopidogrel inhibit irreversibly platelet aggregation. The average lifespan of a platelet is 7–10 days, such that patients using antiplatelet therapies possess some degree of platelet dysfunction throughout this time period. However, platelet function recovers gradually during this time period, as the bone marrow generates approximately 20,000 platelets per day.
Data comparing the incidence of ICH between patients using antiplatelet therapy and controls are sparse. However, several matched case series of head-injured patients have documented an increased severity of hemorrhage, as well as mortality, among patients using antiplatelet therapy as compared to controls.99–101 These reports are limited primarily by confounding by comorbidity, as patients using antiplatelet therapies tend to be sicker than those who do not.
Despite an apparent increased morbidity associated with pre-injury antiplatelet agents among head-injured patients, little data exist to support routine platelet transfusion in this setting. Neither Ivascu et al.100nor Downey et al.102documented a mortality difference associated with platelet transfusion among head-injured patients using antiplatelet therapies. However, platelet administration was at the discretion of the trauma team. Furthermore, data regarding the rapidity and effectiveness of transfusions were not available. Prospective, randomized data are necessary prior to recommending routine platelet administration in this cohort, preferably incorporating serial point-of-care assessment of platelet mapping, such as is possible with TEG. These studies are particularly timely as both the elderly demographic and prevalence of antiplatelet therapy are increasing exponentially. Pending these data, platelet transfusion may be based currently on expert opinion and clinical circumstance.
Interestingly, the deleterious effects of antiplatelet therapy do not appear to extend to the non-head-injured trauma patient. In a recent report, no mortality difference was noted for patients using antiplatelet therapies as compared to controls among a cohort of trauma patients with no evidence of ICH on CT.98 However, injury severity was relatively low in this group. The effect of pre-injury antiplatelet therapy on trauma patients who require MT remains unknown.
COMPLICATIONS OF BLOOD COMPONENTS
Modern-day transfusion practice has minimized substantially the risk of both infectious disease transmission and blood group compatibility mismatch. Consequently, the current major morbidities of blood product transfusion involve immunomodulation with resultant organ failure, storage-related rheologic changes, and cardiovascular compromise secondary to both volume overload and increased vascular resistance.
The immunomodulatory properties of RBC transfusion were first noted as a correlation between transfusion and graft survival following solid organ transplantation.103 The observation that tumor recurrence was associated with RBC transfusion soon followed.104 It is now appreciated that RBC transfusion both impairs humoral immunity and causes elaboration of proinflammatory cytokines. These phenomena are dependent on both transfusion dose and storage time.
There appear to be at least two important mechanisms resulting in immunomodulation. The first involves passive transfer of antileukocyte and anti-HLA antibodies from alloimmunized donors. Passenger leukocytes accompanying RBCs in storage result in the sustained generation of proinflammatory cytokines. Substantial quantities of additional inflammatory mediators such as activated complement, fibrin degradation products, and arachidonic acid are present in both FFP and platelets. Second, proinflammatory compounds present in donor blood (e.g., phospholipids presumably generated from degradation of the RBC membrane with storage) incite the recipient’s immune system, thereby exacerbating inflammation.
Clinically, this immune activation is manifest as a dose-dependent increased risk of organ failure associated with transfusion. The most common and well-delineated organ system affected is the lung. Several reports of critically ill patients, including trauma patients, have detailed a strong correlation between blood product transfusion and the development of acute lung injury, ARDS, and death, even after adjusting for the fact that patients who receive transfusions are, in general, sicker than those who do not.105,106
Additional deleterious effects of RBC transfusion include storage time–dependent degradation with resultant entrapment in the microcirculation, resulting in obstruction and eventual ischemia.107 Moreover, transfused blood exerts a number of negative effects on cardiodynamics, including increased pulmonary vascular resistance, depletion of endogenous nitric oxide stores, and both regional and systemic vasoconstriction. Finally, large-volume blood product transfusion results in a massive oncotic load, eventuating in organ system edema and the consequences thereof.
Numerous clinical studies have documented an independent association between RBC transfusion and mortality following MT.105,108 Importantly, current evidence suggests that sub-MT also imparts an increased risk of morbidity. In fact, a clear dose–response relationship between blood transfusion and organ dysfunction, as well as death, has been documented repeatedly among trauma patients, such that even a single transfusion worsens outcome.109 In a study of trauma patients who developed ARDS, this dose–response relationship followed an exponential function, with a 60% likelihood of ARDS among patients who received >10 U RBC.110 The risks of transfusion extend beyond the initial resuscitation period. In a recent study of trauma patients who did not receive their first blood transfusion until at least 48 hours after injury, each transfusion was independently associated with an increased likelihood of developing ventilator-associated pneumonia, ARDS, and death.111
Transfusion of both FFP and platelets appears equally deleterious. In a single institution study of 841 critically ill medical patients, RBC transfusion, FFP transfusion, and platelet transfusion each independently increased the likelihood of either ALI or ARDS.112 In a multivariable logistic regression model that adjusted for the probability of transfusion and other ALI risk factors, any RBC transfusion increased the likelihood of lung injury by 39%, FFP transfusion by 248%, and platelet transfusion by 389%. Watson et al. investigated the relationship between platelet, FFP, and cryoprecipitate transfusion among 1,175 blunt trauma patients with hemorrhagic shock.113 Although no mortality differences were noted for either platelets or cryoprecipiate, each unit of FFP increased the risk of MOF by 2.1% and ARDS by 2.5%. Most recently, Inaba et al. documented a significant, independent correlation between units of FFP transfused and complications among a large cohort of trauma patients who had received sub-MT.114 The incidence of ARDS was increased 12-fold in patients who had received >6 U of FFP as compared to those patients who had received ≤6 U.
A major limitation of this retrospective literature involves confounding. Specifically, MT may serve simply as a marker for severe illness rather than a true causal parameter. However, the aforementioned CRIT trial, which remains the only randomized trial of blood transfusion triggers, demonstrated a significantly increased risk of ARDS with a liberal transfusion strategy.7 Furthermore, the persistent adverse relationship between transfusion and organ failure among less ill patients strengthens the argument for causality.
In summary, abundant data now exist documenting the deleterious effects of blood product transfusion. One particularly morbid complication involves inflammation-induced organ failure. These data, in conjunction with the aforementioned limitations of the literature suggesting an empiric RBC:FFP:platelet transfusion ratio of 1:1:1, underscore the importance of minimizing unnecessary transfusions. Consequently, blood product replacement should be both restrictive and goal directed.
Trauma patients who survive their initial injury transition from a hypocoagulable to a hypercoagulable state as early as 24 hours following presentation.115 The etiology of this hypercoagulable state is likely multifactorial, involving endothelial injury, circulatory stasis, platelet activation, decreased levels of endogenous anticoagulants, and impaired fibrinolysis. Hypercoagulability predisposes the trauma patient to venous thromboembolism (VTE); large series among trauma patients have reported an incidence of deep vein thrombosis from 10% to 80%116 and pulmonary embolism from 2% to 22%.117 Pulmonary embolism is the third most common cause of death in trauma patients who survive the first 24 hours.117
Similar to the early coagulopathy of trauma, both diagnosis and treatment of hypercoagulability following injury are limited by lack of accurate laboratory testing. Conventional tests used to monitor coagulation status such as the aPTT and PT/INR are neither able to diagnose hypercoagulability nor delineate the relative contributions of enzymatic and platelet components. Previous prospective trials addressing the benefit of various mechanical and pharmacoprophylactic regimens among injured patients have thus neither documented hypercoagulability nor monitored the efficacy of prophylaxis. Furthermore, current guidelines do not address the contribution of platelets to hypercoagulability, nor do they recommend prescription of antiplatelet drugs as part of standard pharmcoprophylactic regimens.118 This last issue is of particular concern as a growing body of evidence has implicated platelet activation in the development and propagation of VTE.119 In light of these limitations, it is not surprising that most VTEs among trauma patients occur because of prophylaxis failure rather than failure to provide prophylaxis.120
In addition to its usefulness in the management of the coagulopathic, bleeding trauma patient, point-of-care TEG offers many advantages for the treatment of postinjury hypercoagulability. An increased G, decreased R time, increased alpha-angle, and increased MA are all suggestive of hypercoagulability (Fig. 13-4E), and our group recently documented a strong correlation between hypercoagulability as evidenced by the aforementioned TEG parameters and subsequent VTE.69 Furthermore, despite standard chemoprophylaxis, 60% of patients displayed evidence of hypercoagulability. Additional advantages of TEG in this setting include differentiation of enzymatic from platelet hypercoagulability, as well as monitoring of the effect of both enzymatic and platelet inhibitors via platelet mapping.
Despite the many theoretical advantages of TEG-driven chemoprophylaxis protocols, prospective data remain sparse. Our institution is currently conducting an RCT comparing conventional pharmacoprophylaxis to TEG-guided therapy among trauma patients. Outcomes will include the incidence of VTE, as well as the efficacy of both enzymatic and platelet inhibition. We hope that these data will inform future clinical practice guidelines, as well as provide insight into whom, when, and how to prophylaxis adequately against VTE.
Postinjury coagulopathy remains a major cause of both morbidity and mortality. Major advances in the care of the coagulopathy trauma patient include the elucidation of the endogenous coagulopathy of trauma, conceptualization of damage control resuscitation, and the application of point-of-care TEG to the trauma setting. Limitations of the retrospective literature addressing blood component ratios for empiric replacement, as well as continued evidence documenting the adverse effects of blood component transfusion, have led the pendulum to swing away from a “1:1:1” ratio of RBC:FFP:platelets. Current evidence supports an empiric FFP:RBC ratio of 1:2 to 1:3, with early platelet and cryoprecipitate replacement based on goal-directed monitoring. Future directions in the field of postinjury coagulopathy involve continued elucidation of the mechanisms of the endogenous coagulopathy of trauma (including therapeutic targets), rigorous research of hemostatic adjuncts, and dissemination of the technology of TEG to trauma physicians.
1. Stefanini M. Basic mechanisms of hemostasis. Bull N Y Acad Med. 1954;30:239–277.
2. Starr DP. Blood: An Epic History of Medicine and Commerce. 1st ed. New York: Alfred A. Knopf; 1998.
3. Napolitano LM, Kurek S, Luchette FA, et al. Clinical practice guideline: red blood cell transfusion in adult trauma and critical care. Crit Care Med. 2009;37:3124–3157.
4. Crowell JW, Ford RG, Lewis VM. Oxygen transport in hemorrhagic shock as a function of the hematocrit ratio. Am J Physiol. 1959;196: 1033–1038.
5. Takaori M, Safar P. Adaptation to acute severe hemodilution with dextran 75 in dogs. Arch Surg. 1966;92:743–748.
6. Czer LS, Shoemaker WC. Optimal hematocrit value in critically ill postoperative patients. Surg Gynecol Obstet. 1978;147:363–368.
7. Hebert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med. 1999;340:409–417.
8. Weiskopf RB, Viele MK, Feiner J, et al. Human cardiovascular and metabolic response to acute, severe isovolemic anemia. JAMA. 1998;279:217–221.
9. Carson JL, Noveck H, Berlin JA, Gould SA. Mortality and morbidity in patients with very low postoperative Hb levels who decline blood transfusion. Transfusion. 2002;42:812–818.
10. Goldsmith HL. The flow of model particles and blood cells and its relation to thrombogenesis. Prog Hemost Thromb. 1972;1:97–127.
11. Quaknine-Orlando B, Samama CM, Riou B, et al. Role of the hematocrit in a rabbit model of arterial thrombosis and bleeding. Anesthesiology. 1999;90:1454–1461.
12. Sauaia A, Moore FA, Moore EE, et al. Epidemiology of trauma deaths: a reassessment. J Trauma. 1995;38:185–193.
13. Simmons RL, Collins JA, Heisterkamp CA, Mills DE, Andren R, Phillips LL. Coagulation disorders in combat casualties. I. Acute changes after wounding. II. Effects of massive transfusion. 3. Post-resuscitative changes. Ann Surg. 1969;169:455–482.
14. Kashuk JL, Moore EE, Millikan JS, Moore JB. Major abdominal vascular trauma—a unified approach. J Trauma. 1982;22:672–679.
15. Harrigan C, Lucas CE, Ledgerwood AM. The effect of hemorrhagic shock on the clotting cascade in injured patients. J Trauma. 1989;29:1416–1421 [discussion 21–22].
16. Phillips TF, Soulier G, Wilson RF. Outcome of massive transfusion exceeding two blood volumes in trauma and emergency surgery. J Trauma. 1987;27:903–910.
17. Brohi K, Cohen MJ, Ganter MT, Matthay MA, Mackersie RC, Pittet JF. Acute traumatic coagulopathy: initiated by hypoperfusion: modulated through the protein C pathway? Ann Surg. 2007;245:812–818.
18. MacLeod JB, Lynn M, McKenney MG, Cohn SM, Murtha M. Early coagulopathy predicts mortality in trauma. J Trauma. 2003;55:39–44.
19. Cannon WB, Fraser J, Cowell E. The preventive treatment of wound shock. JAMA. 1918;70:618–621.
20. Maegele M, Lefering R, Yucel N, et al. Early coagulopathy in multiple injury: an analysis from the German Trauma Registry on 8724 patients. Injury. 2007;38:298–304.
21. Kashuk JL, Moore EE, Sawyer M, et al. Primary fibrinolysis is integral in the pathogenesis of the acute coagulopathy of trauma. Ann Surg. 2010;252(3):434–442.
22. Brohi K, Cohen MJ, Ganter MT, et al. Acute coagulopathy of trauma: hypoperfusion induces systemic anticoagulation and hyperfibrinolysis. J Trauma. 2008;64:1211–1217 [discussion 7].
23. Chesebro BB, Rahn P, Carles M, et al. Increase in activated protein C mediates acute traumatic coagulopathy in mice. Shock. 2009;32:659–665.
24. Gando S. Acute coagulopathy of trauma shock and coagulopathy of trauma: a rebuttal. You are now going down the wrong path. J Trauma. 2009;67:381–383.
25. Hardaway RM. The significance of coagulative and thrombotic changes after haemorrhage and injury. J Clin Pathol Suppl (R Coll Pathol). 1970; 4:110–120.
26. Gando S, Nakanishi Y, Tedo I. Cytokines and plasminogen activator inhibitor-1 in posttrauma disseminated intravascular coagulation: relationship to multiple organ dysfunction syndrome. Crit Care Med. 1995;23:1835–1842.
27. Levi M. Disseminated intravascular coagulation. Crit Care Med. 2007; 35:2191–2195.
28. Kooistra T, Schrauwen Y, Arts J, Emeis JJ. Regulation of endothelial cell t-PA synthesis and release. Int J Hematol. 1994;59:233–255.
29. Jurkovich GJ, Greiser WB, Luterman A, Curreri PW. Hypothermia in trauma victims: an ominous predictor of survival. J Trauma. 1987;27: 1019–1024.
30. Watts DD, Trask A, Soeken K, Perdue P, Dols S, Kaufmann C. Hypothermic coagulopathy in trauma: effect of varying levels of hypothermia on enzyme speed, platelet function, and fibrinolytic activity. J Trauma. 1998;44:846–854.
31. Wolberg AS, Meng ZH, Monroe DM 3rd, Hoffman M. A systematic evaluation of the effect of temperature on coagulation enzyme activity and platelet function. J Trauma. 2004;56:1221–1228.
32. Valeri CR, Feingold H, Cassidy G, Ragno G, Khuri S, Altschule MD. Hypothermia-induced reversible platelet dysfunction. Ann Surg. 1987; 205:175–181.
33. Yoshihara H, Yamamoto T, Mihara H. Changes in coagulation and fibrinolysis occurring in dogs during hypothermia. Thromb Res. 1985; 37:503–512.
34. Meng ZH, Wolberg AS, Monroe DM 3rd, Hoffman M. The effect of temperature and pH on the activity of factor VIIa: implications for the efficacy of high-dose factor VIIa in hypothermic and acidotic patients. J Trauma. 2003;55:886–891.
35. Engstrom M, Schott U, Romner B, Reinstrup P. Acidosis impairs the coagulation: a thromboelastographic study. J Trauma. 2006;61: 624–628.
36. Martini WZ, Dubick MA, Pusateri AE, Park MS, Ryan KL, Holcomb JB. Does bicarbonate correct coagulation function impaired by acidosis in swine? J Trauma. 2006;61:99–106.
37. Brazil EV, Coats TJ. Sonoclot coagulation analysis of in-vitro haemodilution with resuscitation solutions. J R Soc Med. 2000;93: 507–510.
38. Coats TJ, Brazil E, Heron M, MacCallum PK. Impairment of coagulation by commonly used resuscitation fluids in human volunteers. Emerg Med J. 2006;23:846–849.
39. Bickell WH, Bruttig SP, Millnamow GA, O’Benar J, Wade CE. The detrimental effects of intravenous crystalloid after aortotomy in swine. Surgery. 1991;110:529–536.
40. Owens TM, Watson WC, Prough DS, Uchida T, Kramer GC. Limiting initial resuscitation of uncontrolled hemorrhage reduces internal bleeding and subsequent volume requirements. J Trauma. 1995;39:200–207 [discussion 8–9].
41. Bickell WH, Wall MJ Jr, Pepe PE, et al. Immediate versus delayed fluid resuscitation for hypotensive patients with penetrating torso injuries. N Engl J Med. 1994;331:1105–1109.
42. Wall MJ, Granchi T, Liscum K, Aucar J, Mattox KL. Delayed vs. immediate fluid resuscitation in patients with penetrating trauma: subgroup analysis. J Trauma. 1995;39:173.
43. Turner J, Nicholl J, Webber L, Cox H, Dixon S, Yates D. A randomised controlled trial of prehospital intravenous fluid replacement therapy in serious trauma. Health Technol Assess. 2000;4:1–57.
44. Dutton RP, Mackenzie CF, Scalea TM. Hypotensive resuscitation during active hemorrhage: impact on in-hospital mortality. J Trauma. 2002;52:1141–1146.
45. Kwan I, Bunn F, Roberts I. Timing and volume of fluid administration for patients with bleeding. Cochrane Database Syst Rev. 2003:CD002245.
46. Advanced Trauma Life Support for Doctors. 2nd ed. Chicago, IL: American College of Surgeons; 1997.
47. Kashuk JL, Moore EE, Johnson JL, et al. Postinjury life threatening coagulopathy: is 1:1 fresh frozen plasma:packed red blood cells the answer? J Trauma. 2008;65:261–270 [discussion 70–71].
48. Harrigan C, Lucas CE, Ledgerwood AM, Walz DA, Mammen EF. Serial changes in primary hemostasis after massive transfusion. Surgery. 1985;98:836–844.
49. Borgman MA, Spinella PC, Perkins JG, et al. The ratio of blood products transfused affects mortality in patients receiving massive transfusions at a combat support hospital. J Trauma. 2007;63: 805–813.
50. Armand R, Hess JR. Treating coagulopathy in trauma patients. Transfus Med Rev. 2003;17:223–231.
51. Duchesne JC, Hunt JP, Wahl G, et al. Review of current blood transfusions strategies in a mature level I trauma center: were we wrong for the last 60 years? J Trauma. 2008;65:272–276 [discussion 6–8].
52. Holcomb JB, Wade CE, Michalek JE, et al. Increased plasma and platelet to red blood cell ratios improves outcome in 466 massively transfused civilian trauma patients. Ann Surg. 2008;248:447–458.
53. Teixeira PG, Inaba K, Shulman I, et al. Impact of plasma transfusion in massively transfused trauma patients. J Trauma. 2009;66:693–697.
54. Spinella PC, Perkins JG, Grathwohl KW, et al. Effect of plasma and red blood cell transfusions on survival in patients with combat related traumatic injuries. J Trauma. 2008;64:S69–S77 [discussion S77–78].
55. Hirshberg A, Dugas M, Banez EI, Scott BG, Wall MJ Jr, Mattox KL. Minimizing dilutional coagulopathy in exsanguinating hemorrhage: a computer simulation. J Trauma. 2003;54:454–463.
56. Snyder CW, Weinberg JA, McGwin G Jr, et al. The relationship of blood product ratio to mortality: survival benefit or survival bias? J Trauma. 2009;66:358–362 [discussion 62–64].
57. O’Shaughnessy DF, Atterbury C, Bolton Maggs P, et al. Guidelines for the use of fresh-frozen plasma, cryoprecipitate and cryosupernatant. Br J Haematol. 2004;126:11–28.
58. Martin DJ, Lucas CE, Ledgerwood AM, Hoschner J, McGonigal MD, Grabow D. Fresh frozen plasma supplement to massive red blood cell transfusion. Ann Surg. 1985;202:505–511.
59. Mace JE, White CE, Simmons JW. High plasma to RBC ratio improves survival based on mechanism of combat injury in massively transfused patients. J Am Coll Surg. 2009;209:S47.
60. Faringer PD, Mullins RJ, Johnson RL, Trunkey DD. Blood component supplementation during massive transfusion of AS-1 red cells in trauma patients. J Trauma. 1993;34:481–485 [discussion 5–7].
61. Lim RC Jr, Olcott C 4th, Robinson AJ, Blaisdell FW. Platelet response and coagulation changes following massive blood replacement. J Trauma. 1973;13:577–582.
62. Reed RL 2nd, Ciavarella D, Heimbach DM, et al. Prophylactic platelet administration during massive transfusion. A prospective, randomized, double-blind clinical study. Ann Surg. 1986;203:40–48.
63. Bick RL, Kaplan H. Syndromes of thrombosis and hypercoagulability. Congenital and acquired causes of thrombosis. Med Clin North Am. 1998;82:409–458.
64. Counts RB, Haisch C, Simon TL, Maxwell NG, Heimbach DM, Carrico CJ. Hemostasis in massively transfused trauma patients. Ann Surg. 1979;190:91–99.
65. Lucas CE, Ledgerwood AM. Clinical significance of altered coagulation tests after massive transfusion for trauma. Am Surg. 1981;47:125–130.
66. Kheirabadi BS, Crissey JM, Deguzman R, Holcomb JB. In vivo bleeding time and in vitro thrombelastography measurements are better indicators of dilutional hypothermic coagulopathy than prothrombin time. J Trauma. 2007;62:1352–1359 [discussion 9–61].
67. Park MS, Martini WZ, Dubick MA, et al. Thromboelastography as a better indicator of hypercoagulable state after injury than prothrombin time or activated partial thromboplastin time. J Trauma. 2009;67: 266–275 [discussion 75–76].
68. Plotkin AJ, Wade CE, Jenkins DH, et al. A reduction in clot formation rate and strength assessed by thrombelastography is indicative of transfusion requirements in patients with penetrating injuries. J Trauma. 2008;64:S64–S68.
69. Kashuk JL, Moore EE, Sabel A, et al. Rapid thrombelastography (r-TEG) identifies hypercoagulability and predicts thromboembolic events in surgical patients. Surgery. 2009;146:764–772 [discussion 72–74].
70. Gonzalez EA, Kashuk JL, Moore EE, Silliman CC. Differentiation of enzymatic from platelet hypercoagulability using the novel thrombelastography parameter delta. J Surg Res. 2010;163(1):96–101.
71. Kashuk JL, Moore EE, Wohlauer M. Point of care rapid thrombelastography improves management of life threatening post-injury coagulopathy. Presented at: American Association for the Surgery of Trauma Sixty-Eight Annual Meeting; October 2009; Pittsburgh, PA.
72. Hedner U, Kisiel W. Use of human factor VIIa in the treatment of two hemophilia A patients with high-titer inhibitors. J Clin Invest. 1983; 71:1836–1841.
73. Dutton RP, McCunn M, Hyder M, et al. Factor VIIa for correction of traumatic coagulopathy. J Trauma. 2004;57:709–718 [discussion 18–19].
74. Boffard KD, Riou B, Warren B, et al. Recombinant factor VIIa as adjunctive therapy for bleeding control in severely injured trauma patients: two parallel randomized, placebo-controlled, double-blind clinical trials. J Trauma. 2005;59:8–15 [discussion 15–18].
75. Rizoli SB, Boffard KD, Riou B, et al. Recombinant activated factor VII as an adjunctive therapy for bleeding control in severe trauma patients with coagulopathy: subgroup analysis from two randomized trials. Crit Care. 2006;10:R178.
76. Knudson MM, Cohen MJ, Reidy R, Jaegar S, Wade C, Holcomb JB. Trauma, transfusions, and recombinant activated factor VII: a multi-center analysis of 1,041 patients. Presented at: 68th Annual Meeting of the American Association for the Surgery of Trauma; 2009; Pittsburgh, PA.
77. Thomas GO, Dutton RP, Hemlock B, et al. Thromboembolic complications associated with factor VIIa administration. J Trauma. 2007;62:564–569.
78. Levi M, Cromheecke ME, de Jonge E, et al. Pharmacological strategies to decrease excessive blood loss in cardiac surgery: a meta-analysis of clinically relevant endpoints. Lancet. 1999;354:1940–1947.
79. Auer LM, Marth E, Heppner F, Holasek A. Proteolytic enzyme activity in patients with severe head injury and the effect of a proteinase inhibitor. Acta Neurochir (Wien). 1979;49:207–217.
80. McMichan JC, Rosengarten DS, Philipp E. Prophylaxis of post-traumatic pulmonary insufficiency by protease-inhibitor therapy with aprotinin: a clinical study. Circ Shock. 1982;9:107–116.
81. Coats T, Roberts I, Shakur H. Antifibrinolytic drugs for acute traumatic injury. Cochrane Database Syst Rev. 2004:CD004896.
82. CRASH-2 Trial Collaborators, Shakur H, Roberts I, et al. Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. Lancet. 2010;376:23–32.
83. Schroeder V, Chatterjee T, Kohler HP. Influence of blood coagulation factor XIII and FXIII Val34Leu on plasma clot formation measured by thrombelastography. Thromb Res. 2001;104:467–474.
84. Godje O, Haushofer M, Lamm P, Reichart B. The effect of factor XIII on bleeding in coronary surgery. Thorac Cardiovasc Surg. 1998;46: 263–267.
85. Lorenz R, Kienast J, Otto U, et al. Successful emergency reversal of phenprocoumon anticoagulation with prothrombin complex concentrate: a prospective clinical study. Blood Coagul Fibrinolysis. 2007;18:565–570.
86. Spahn DR, Cerny V, Coats TJ, et al. Management of bleeding following major trauma: a European guideline. Crit Care. 2007;11:R17.
87. Kalina M, Tinkoff G, Gbadebo A, Veneri P, Fulda G. A protocol for the rapid normalization of INR in trauma patients with intracranial hemorrhage on prescribed warfarin therapy. Am Surg. 2008;74: 858–861.
88. Bruce D, Nokes TJ. Prothrombin complex concentrate (Beriplex P/N) in severe bleeding: experience in a large tertiary hospital. Crit Care. 2008;12:R105.
89. Dickneite G, Pragst I. Prothrombin complex concentrate vs fresh frozen plasma for reversal of dilutional coagulopathy in a porcine trauma model. Br J Anaesth. 2009;102:345–354.
90. Dickneite G, Dorr B, Kaspereit F, Tanaka KA. Prothrombin complex concentrate versus recombinant factor VIIa for reversal of hemodilutional coagulopathy in a porcine trauma model. J Trauma. 2010;68: 1151–1157.
91. Tharmarajah P, Pusey J, Keeling D, Willett K. Efficacy of warfarin reversal in orthopedic trauma surgery patients. J Orthop Trauma. 2007;21:26–30.
92. Pieracci FM, Eachempati SR, Shou J, Hydo LJ, Barie PS. Degree of anticoagulation, but not warfarin use itself, predicts adverse outcomes after traumatic brain injury in elderly trauma patients. J Trauma. 2007;63:525–530.
93. Pieracci FM, Eachempati SR, Shou J, Hydo LJ, Barie PS. Use of long-term anticoagulation is associated with traumatic intracranial hemorrhage and subsequent mortality in elderly patients hospitalized after falls: analysis of the New York State Administrative Database. J Trauma. 2007;63:519–524.
94. Karni A, Holtzman R, Bass T, et al. Traumatic head injury in the anticoagulated elderly patient: a lethal combination. Am Surg. 2001;67:1098–1100.
95. Coimbra R, Hoyt DB, Anjaria DJ, Potenza BM, Fortlage D, Hollingsworth-Fridlund P. Reversal of anticoagulation in trauma: a North-American survey on clinical practices among trauma surgeons. J Trauma. 2005;59:375–382.
96. Ivascu FA, Howells GA, Junn FS, Bair HA, Bendick PJ, Janczyk RJ. Rapid warfarin reversal in anticoagulated patients with traumatic intracranial hemorrhage reduces hemorrhage progression and mortality. J Trauma. 2005;59:1131–1137 [discussion 7–9].
97. Deveras RA, Kessler CM. Reversal of warfarin-induced excessive anticoagulation with recombinant human factor VIIa concentrate. Ann Intern Med. 2002;137:884–888.
98. Ott MM, Eriksson E, Vanderkolk W, Christianson D, Davis A, Scholten D. Antiplatelet and anticoagulation therapies do not increase mortality in the absence of traumatic brain injury. J Trauma. 2010;68:560–563.
99. Ohm C, Mina A, Howells G, Bair H, Bendick P. Effects of antiplatelet agents on outcomes for elderly patients with traumatic intracranial hemorrhage. J Trauma. 2005;58:518–522.
100. Ivascu FA, Howells GA, Junn FS, Bair HA, Bendick PJ, Janczyk RJ. Predictors of mortality in trauma patients with intracranial hemorrhage on preinjury aspirin or clopidogrel. J Trauma. 2008;65:785–788.
101. Wong DK, Lurie F, Wong LL. The effects of clopidogrel on elderly traumatic brain injured patients. J Trauma. 2008;65:1303–1308.
102. Downey DM, Monson B, Butler KL, et al. Does platelet administration affect mortality in elderly head-injured patients taking antiplatelet medications? Am Surg. 2009;75:1100–1103.
103. Opelz G, Terasaki PI. Improvement of kidney-graft survival with increased numbers of blood transfusions. N Engl J Med. 1978;299:799–803.
104. Gantt CL. Red blood cells for cancer patients. Lancet. 1981;2:363.
105. Malone DL, Dunne J, Tracy JK, Putnam AT, Scalea TM, Napolitano LM. Blood transfusion, independent of shock severity, is associated with worse outcome in trauma. J Trauma. 2003;54:898–905 [discussion 905–907].
106. Napolitano LM, Corwin HL. Efficacy of red blood cell transfusion in the critically ill. Crit Care Clin. 2004;20:255–268.
107. Simchon S, Jan KM, Chien S. Influence of reduced red cell deformability on regional blood flow. Am J Physiol. 1987;253:H898–H903.
108. Zallen G, Offner PJ, Moore EE, et al. Age of transfused blood is an independent risk factor for postinjury multiple organ failure. Am J Surg. 1999;178:570–572.
109. Shapiro MJ, Gettinger A, Corwin HL, et al. Anemia and blood transfusion in trauma patients admitted to the intensive care unit. J Trauma. 2003;55:269–273 [discussion 73–74].
110. Silverboard H, Aisiku I, Martin GS, Adams M, Rozycki G, Moss M. The role of acute blood transfusion in the development of acute respiratory distress syndrome in patients with severe trauma. J Trauma. 2005;59:717–723.
111. Croce MA, Tolley EA, Claridge JA, Fabian TC. Transfusions result in pulmonary morbidity and death after a moderate degree of injury. J Trauma. 2005;59:19–23 [discussion 23–24].
112. Khan H, Belsher J, Yilmaz M, et al. Fresh-frozen plasma and platelet transfusions are associated with development of acute lung injury in critically ill medical patients. Chest. 2007;131:1308–1314.
113. Watson GA, Sperry JL, Rosengart MR, et al. Fresh frozen plasma is independently associated with a higher risk of multiple organ failure and acute respiratory distress syndrome. J Trauma. 2009;67:221–227 [discussion 8–30].
114. Inaba K, Branco BC, Rhee P, et al. Impact of plasma transfusion in trauma patients who do not require massive transfusion. J Am Coll Surg. 2010;210:957–965.
115. Schreiber MA, Differding J, Thorborg P, Mayberry JC, Mullins RJ. Hypercoagulability is most prevalent early after injury and in female patients. J Trauma. 2005;58:475–480 [discussion 80–81].
116. Attia J, Ray JG, Cook DJ, Douketis J, Ginsberg JS, Geerts WH. Deep vein thrombosis and its prevention in critically ill adults. Arch Intern Med. 2001;161:1268–1279.
117. O’Malley KF, Ross SE. Pulmonary embolism in major trauma patients. J Trauma. 1990;30:748–750.
118. Rogers FB, Cipolle MD, Velmahos G, Rozycki G, Luchette FA. Practice management guidelines for the prevention of venous thromboembolism in trauma patients: the EAST Practice Management Guidelines Work Group. J Trauma. 2002;53:142–164.
119. Kim Y, Nakase H, Nagata K, Sakaki T, Maeda M, Yamamoto K. Observation of arterial and venous thrombus formation by scanning and transmission electron microscopy. Acta Neurochir (Wien). 2004;146:45–51 [discussion 51].
120. Patel R, Cook DJ, Meade MO, et al. Burden of illness in venous thromboembolism in critical care: a multicenter observational study. J Crit Care. 2005;20:341–347.