KEY ISSUES IN ADVANCED BLEEDING CARE IN TRAUMA

Abstract

The incidence of hemostatic abnormalities in the early hours after traumatic incident is high and represents an independent predictor of mortality. Key factors in the development of traumatic coagulopathy include the severity of injury, hypothermia, acidosis, hemorrhagic shock, hemodilution, clotting factor consumption, and fibrinolysis. Assessment of bleeding includes evaluation of the mechanism of injury, vital signs, biochemistry, detection of external and internal bleeding sources, injuries found upon secondary investigation, and response to treatment. Priority in treating the bleeding trauma patient should be given to prevention of further bleeding, hypothermia, acidosis, coagulopathy, and maintenance of tissue oxygenation, achieved by careful physical handling, damage control surgery, analgesia, maintenance of normothermia, correction of coagulopathy, control of blood pH, and serum calcium. Priority during initial treatment is to restore tissue perfusion and achieve hemostasis in vital functions; other nonvital procedures may generally be delayed. This state-of-the-art review aims to address key issues in acute control of bleeding in the trauma patient.

Injury is the leading cause of death worldwide among those aged 5 to 44 years (¹). In the United States, it is the leading cause of death in the 1- to 44-year age group and the third leading cause of death overall (¹). Of the deaths due to injury, 50% occur in the field, 30% occur in the first 24 h, and 20% are late deaths due to multiple organ failure (MOF). Of the early deaths, 30% to 50% are due to exsanguination (¹).

The incidence of early abnormality of hemostasis after trauma is high, and such abnormalities are independent predictors of mortality (^{2, 3}). Key factors in the development of traumatic coagulopathy include the severity of the injury, hypothermia, hypocalcemia, acidosis, bleeding, hemorrhagic shock, hemodilution as a result of fluids given during resuscitation, consumption of coagulation factors, and activation of fibrinolysis, which may combine to form a vicious cycle. In particular, if the lethal triad of hypothermia, acidosis, and coagulopathy are present, surgical control of bleeding alone is unlikely to be successful (^{4, 5}).

Unfortunately, there is no single test for hemorrhage, so assessment of bleeding requires a complex combined evaluation of the mechanism of injury, vital signs (⁶), biochemistry (e.g., lactate) (⁷), injuries found upon secondary investigation, and the response to treatment. Common scoring systems for assessment of severity of injury such as the Revised Trauma Score (⁸), Injury Severity Score (⁹), Glasgow Coma Scale (¹⁰), and Acute Physiology And Chronic Health Evaluation score (¹¹) are not specific enough to be useful for the diagnosis of bleeding. The management and treatment of the bleeding trauma patient require a multidisciplinary approach that includes control of the relevant physiological parameters, rapid control of bleeding, maintenance of tissue perfusion, temperature control, appropriate use of blood components, and pharmacological treatment. This review aims to outline key issues to be considered in the initial treatment of the trauma patient to support rapid control of bleeding.

PHYSIOLOGICAL RESPONSE OF COAGULATION TO TRAUMA

Long-standing theories describe intrinsic and extrinsic coagulation cascades. These theories are a good description of "test tube" coagulation but are out of date and do not describe physiological coagulation. The cell-based model of coagulation is a better description of the complex interactions in vivo and includes cellular and plasma components (¹²). This model involves three stages (Fig. 1): initiation, amplification, and propagation. The initiation phase involves cellular damage, exposure of tissue factor (TF), formation of TF/factor VIIa complex, and the production of a small amount of thrombin. In the amplification phase, the small amount of thrombin amplifies the initial procoagulant signal by enhancing platelet adhesion and fully activating the platelets. During propagation, large-scale thrombin production occurs on the surface of activated platelets, leading to the production of a stable clot (^{13, 14}).

Fig. 1: Three key stages of the hemostasis process: initiation, amplification, and propagation. The initiation phase involves endothelial damage, exposure of TF), formation of TF/factor VIIa complex, and the production of a small amount of thrombin. In the amplification phase, the small amount of thrombin amplifies the initial procoagulant signal by enhancing platelet adhesion and fully activating the platelets. During propagation, large-scale thrombin production occurs on the surface of activated platelets, leading to the production of a stable clot. Modified from Br J Anaesth 2005;95:130.

During trauma, massive TF exposure results in intensive early, prehospital fibrin clot formation. This activation of blood coagulation and the subsequent fibrinolysis (thrombin stimulates the release of tissue plasminogen activator) consume and deplete hemostatic factors. Coagulation factors are further diluted by fluid shift from the extracellular to the vascular space, which is proportional to the grade of shock. Hemostatic function is also strongly influenced by the body temperature. In hypothermia, both platelet function and clotting factors are impaired. Thus, coagulation assays, which are performed at 37°C, may underestimate the extent of in vivo coagulation problems (¹⁵). Furthermore, coagulation disorders are aggravated by acidosis, caused by inadequate tissue oxygen supply. Blood pH should be closely monitored to detect severe acidosis and consequent acidotic coagulopathy. Because metabolic acidosis probably reflects low tissue perfusion and shock, optimized resuscitation is also required. High lactate levels usually reflect the degree of tissue hypoxia, and persistent elevation is associated with poor outcome (¹⁶). Elevated blood lactate levels for more than 24 h may indicate insufficient tissue perfusion and are associated with increased mortality (^{16, 17}). When the lethal triad of coagulopathy, acidosis, and hypothermia are present, mortality rates are high, and a rapid "damage control" strategy is required.

MANAGEMENT OF BLEEDING AND COAGULOPATHY

The general principles of bleeding and coagulopathy management associated with trauma are to achieve rapid control of bleeding and, if necessary, volume resuscitation with fluids and blood products and blood pressure stabilization with vasopressors and inotropic agents. These goals are accompanied by the control of physiological parameters and the monitoring of bleeding and coagulation. Prevention of the lethal triad of coagulopathy, hypothermia, and acidosis involves four aspects as shown in Table 1. In addition to prevention of coagulopathy, hypothermia, and acidosis, physical immobilization and careful handling are required to avoid secondary lesions (i.e., spinal cord injuries), minimize pain, and preserve clot in injured areas. Adequate analgesia is necessary to control pain and avoid hypertension and secondary clot dislodgment. Analgesia also prevents tachycardia secondary to pain being misinterpreted as a sign of hypovolemia. Furthermore, hypocalcemia is frequently present in severely injured patients and may require the administration of intravenous calcium with frequent ionized serum calcium control.

Table 1: Prevention of the lethal triad of coagulopathy, hypothermia, and acidosis

Monitoring of bleeding and coagulopathy

The risk for bleeding is highest in the first few hours after trauma. Basic physiological and clinical variables (heart rate, blood pressure, and urine output) are the most useful parameters for the initial assessment of volemic status (Table 2). Capillary refill and mental status are also useful for the assessor. A heart rate exceeding 100 bpm or a decrease in urine output are probably the earliest signs of hypovolemia and can be detected with blood loss of around 15% (750 mL in a person of 70 kg). Systolic blood pressure below 90 mmHg is a very sensitive sign but generally requires a greater blood loss of approximately 20% to 30% of blood volume.

Table 2: Basic physiological variables for the emergency evaluation of the bleeding trauma patient

During a stay in the intensive care unit, late bleeding is always possible and manifests itself most often as a low-intensity but constant blood loss. Diagnosis is generally made by a progressive hematocrit decrease. Late, sudden, high-intensity bleeding in the ICU occurs less frequently, and there are no "ideal" monitoring systems for its detection. Central venous and arterial pressure and heart rate are useful parameters but may be affected by factors such as fever, vasoactive drugs, diuretics, and positive pressure ventilation. In this setting, other parameters such as continuous cardiac output and pulse pressure variation are more sensitive for detecting hypovolemia (^{18, 19}).

Localization of bleeding

Imaging technologies used in the assessment of bleeding trauma patients include radiography (chest and pelvic), ultrasound, and computed tomography. Radiographic and ultrasound studies are preferable in unstable patients because they can be performed in the emergency department. Computed tomography is highly sensitive but requires both time and patient transportation; therefore, it can cause harm if urgent intervention is delayed. In practice, all three techniques are frequently used, and a judicious balance between hemodynamic status and clinical assessment should determine which imaging technique is appropriate.

Monitoring

Blood coagulation and complete blood count should be determined on admission and at least every 24 h during the first few days. A more frequent analysis of coagulation status is necessary if initial coagulation is abnormal in bleeding patients and in those with liver disease. During massive bleeding, the results of coagulation tests may be obsolete upon receipt or misleading if the patient is hypothermic (¹⁵). In these cases, decisions about the replacement of blood components may have to be based on the clinical evidence or suspicions of coagulopathy until laboratory confirmation is obtained. Near-patient hemostatic assessment such as thromboelastography shows promise in this setting, but requires full validation. Mixed venous oxygen saturation is a marker for the balance between oxygen delivery and oxygen consumption. It requires insertion of a pulmonary artery catheter, rarely indicated in trauma patients. Central venous saturation (through a central venous catheter) has been proposed as an alternative in trauma patients (²⁰), although there may be a clinically relevant difference between central and mixed venous saturation (²¹). Lactate levels in this setting usually reflect the degree of tissue hypoperfusion. Persistent elevation may indicate underresuscitation and is associated with MOF and higher mortality (¹⁶).

Rapid control of bleeding

During the initial management phase ("primary survey"), the presence of hemorrhagic shock must be diagnosed and treated. The basic management principle is to recognize external and internal bleeding sources to stop the bleeding and to replace volume loss (^22-24). The clinical symptoms of shock are the "three windows to the microcirculation":

1. Mental status/level of consciousness (cerebral perfusion): agitation, confusion, and somnolence or lethargy
2. Peripheral perfusion: cold and clammy skin, delayed capillary refilling, and tachycardia
3. Renal perfusion: urine output (<0.5 mL kg⁻¹ h⁻¹) (²²).

These clinical findings help to differentiate whether a patient is "hemodynamically normal" or just "apparently hemodynamically stable" but in compensated shock (²²). Arterial blood gas analysis can further assist because lactate levels and base deficit represent highly sensitive parameters for recognition of "hidden shock" (^{16, 25}). External bleeding is no longer initially controlled by the use of tourniquets because this adds to ischemic tissue damage (²²), but instead, it is controlled by the use of local compression.

Internal bleeding may require immediate surgical control in hemodynamically unstable patients. The aim of "damage control" is survival (^{23, 26-29}). In parallel to volume resuscitation, the main bleeding sources must be screened according to standardized protocols (^{22, 28}). Focused assessment sonography in trauma and anteroposterior radiographs of the chest and pelvis are standard (^{22, 28}).

Damage control surgery for acute bleeding control may include resuscitative thoracotomy and emergency laparotomy (^{27, 30}). A resuscitative thoracotomy with cross-clamping of the descending aorta is indicated in highly unstable patients with severe hemorrhagic shock due to massive intrathoracic bleeding and/or intra-abdominal or retroperitoneal bleeding (^{22, 28, 30}). Laparotomy is indicated in blunt abdominal trauma with severe hemorrhagic shock and evidence of free fluid and in penetrating abdominal trauma with hemorrhagic shock (^{22, 28}).

Unstable pelvic injuries with posterior pelvic ring disruption are associated with massive uncontrolled retroperitoneal bleeding of up to 5000 mL because of lacerations of the presacral and paravesical venous plexus and cancellous bone bleeding. The hallmark of these patients' survival is a rapid recognition and surgical control of hemorrhage because mortality in pelvic fracture-associated hemorrhage is still as high as 50% to 60% (³¹). These patients require immediate closed reduction of the pelvic ring in the emergency department and external fixation, for example, with a pelvic "C-clamp" for the posterior pelvic ring or an external fixator for the anterior pelvic ring (^31-33). If these measures, in combination with fluid therapy, do not result in hemodynamic stability, explorative laparotomy with pelvic "packing" is warranted to achieve surgical hemorrhage control (²⁹). Interventional measures such as angiography and embolization are generally not recommended for the management of patients with pelvic ring disruptions and traumatic hemorrhagic shock because arterial bleeding sources are very rare (less than 10%), and successful embolization is usually performed in less than 2% of all cases (^{31, 34}). Thus, the "damage control" sequence for hemodynamically unstable patients with pelvic ring disruptions includes closed reduction and external fixation with explorative laparotomy, pelvic packing, and provisional closure of the abdomen (^{31, 35}). This strategy has been shown to lower mortality from pelvic bleeding to about 20% to 25% in this highly unstable patient group (^{31, 35}). Change of packing must be performed within 24 to 48 h, and definitive surgery should follow after restoration of the physiological end points of resuscitation, ideally within the "time window of opportunity" between the fifth to tenth day after trauma (^{28, 36}). The aim of abbreviating surgical treatment to a damage control procedure is the early transfer of critical patients to the ICU for restoration of physiological "end points of resuscitation" (^{23, 36, 37}).

Topical hemostatic sealants

For many years, fibrin sealants and fibrin glues have been used as efficacious therapeutic tools for bleeding control as an adjunct to conventional methods of achieving hemostasis (³⁸). The concept of fibrin sealants includes the local application of concentrated plasma derivatives (³⁸) to promote conversion of endogenous fibrinogen into fibrin (³⁹). The reduced intraoperative blood loss and reduced operation time have led to a widespread use of fibrin glues in surgery and endoscopy (⁴⁰). The common application methods for fibrin glues include spray bottles, syringes, and silastic catheters through flexible fiber optic endoscopes (⁴⁰).

In recent years, the list of commercially available topical hemostatic agents has increased dramatically. Aside from fibrin glues, the most commonly used products are cellulose, gelatin, collagen, thrombin, or aldehyde glues (see Oz et al. (³⁹) for an overview). FloSeal represents one of the promising "new generation" topical agents based on the concept of a fusion matrix, whereby a bovine gelatin matrix is mixed with a topical thrombin solution immediately before use (³⁹). Importantly, the hemostatic effect of FloSeal does not rely on the presence of functional platelets or other coagulation factors except for fibrinogen, the terminal protein in the coagulation cascade (³⁹).

Based on positive safety and effectiveness data from different multicenter prospective randomized clinical trials, (^{39, 41, 42}) FloSeal was recently approved for clinical use by the Food and Drug Administration in the United States. Patients undergoing cardiac, vascular, or spinal surgery were randomized to adjunctive hemorrhage control using either FloSeal or a thrombin-based gelatin agent (Gelfoam) as a control substance in instances where conventional intraoperative bleeding control measures failed (^{39, 41-43}). During cardiac surgery, FloSeal was shown to be significantly more efficacious in mediating local bleeding control within 10 min after application compared with the control substance (94% vs. 60% of patients, P = 0.001) (⁴¹). Similar results were obtained in patients undergoing spinal surgery, where a significant bleeding control was achieved as early as 3 min after application of FloSeal, compared with the thrombin-based gelatin agent (97% vs. 71% of patients, P = 0.0001) (⁴²).

Other new-generation topical hemostatic agents include polyethylene glycol-based synthetic sealants, which have important advantages over the biological substances such as FloSeal because of the lack of risk for acquired infectious diseases, such as bovine spongiform encephalitis, and no antigenic properties (⁴⁴). Synthetic sealants (e.g., Focalseal) are also composed of two components, a primer and a sealant (⁴⁴). These components are applied in two steps and then polymerized under the influence of a blue-green light from a xenon light source (⁴⁴). Other synthetic new-generation sealants consist of albumin and glutaraldehyde (⁴⁵) or as hydrogels in sprayable form (⁴⁶).

These new synthetic agents have proven efficacious in experimental models and in clinical trials in cardiac surgery (^{44, 46}). Promising results for hemorrhage control have recently been reported for the topical synthetic glycosaminoglycan polymer fiber poly-N-acetyl glucosamine (p-Glc-NAc) (⁴⁷). In a double-blind randomized prospective study, the local application of a patch containing p-Glc-NAc induced an improved hemostasis at the puncture site in patients undergoing cardiac catheterization (⁴⁸). Furthermore, local p-Glc-NAc application in a splenic hemorrhage model provided superior hemostasis compared with fibrin-based sealants (⁴⁹).

Topical hemostatic sealants therefore may be recommended as first choice for adjunctive local hemostasis in cases where conventional measures of bleeding control fail.

Maintenance of tissue oxygenation

The maintenance of oxygen delivery and tissue oxygenation during the acute treatment of persons with trauma is an important goal. The most important aspects are fluid management, the transfusion of red blood cells (RBCs), the use of inotropes and vasopressors (in selected cases), and the awareness of an abdominal compartment syndrome. However, as discussed below, the right type of fluid, the optimal volume, and the best timing and end points are open to debate.

Type of fluid

The types of fluid used include isotonic and hypertonic crystalloids, colloids (mainly gelatins and starch solutions), and blood products. Factors that influence the choice of fluid include their early hemodynamic effects, effects on hemostasis, oxygen carriage, distribution and capillary endothelial leakage, effect on inflammatory response, safety, pH buffering, method of elimination, practicality, and cost. The advantages of crystalloids include their usefulness in cases of intravascular and interstitial hypovolemia, that they are not associated with anaphylaxis, and that they are inexpensive, safe, and easy to store. Disadvantages of crystalloids include the need to use large volumes, the short duration of effect, the association with edema, and, occasionally, with metabolic disturbances. Colloids have the advantage that they are efficient volume expanders, have rapid activity, improve colloid osmotic pressure, and cause less pulmonary impairment than crystalloids. Disadvantages of colloids include the fact that they are more expensive than crystalloids and carry a risk for anaphylactic response. In addition, several meta-analyses have shown an increased risk for death in patients treated with colloids compared with patients treated with crystalloids, (^50-54) and three of these studies showed that the effect was particularly significant in a trauma subgroup (^{50, 53, 54}). A more recent meta-analysis, however, showed no evidence for reduced risk of death with colloids versus crystalloids (⁵⁵).

Problems in evaluating and comparing the use of different resuscitation fluids include the heterogeneity of populations and therapy strategies, limited quality of analyzed studies, mortality not always being the primary outcome, and different (often short) observation periods. Therefore, it is difficult to reach a definitive conclusion to the advantage of one type of resuscitation fluid over the other. The Saline versus Albumin Fluid Evaluation study compared 4% albumin with 0.9% sodium chloride in 6997 ICU patients and showed no significant differences in outcomes, but a trend toward higher mortality in the trauma subgroup that received albumin (^{56, 57}). Promising results have been obtained with hypertonic solutions. One study showed that use of hypertonic saline was associated with lower intracranial pressure than with isotonic sodium chloride solution in brain-injured patients (⁵⁸), and a meta-analysis comparing hypertonic saline dextran with isotonic sodium chloride solution for resuscitation in hypotension from penetrating torso injuries showed improved survival in the hypertonic saline dextran group when surgery was required (⁵⁹). A clinical trial with brain injury patients found that hypertonic saline reduces intracranial pressure more effectively than dextran solution with 20% mannitol (⁶⁰). However, Cooper et al. (⁶¹) found almost no difference in neurological function 6 months after traumatic brain injury in patients who had received prehospital hypertonic saline resuscitation compared with conventional fluid.

Volume and timing of fluid

Traditional guidelines generally use early and aggressive fluid administration to restore blood volume. Problems with this approach include increased hydrostatic pressure on the wound, dislodgment of blood clots, dilution of coagulation factors, and cooling of the patient (Fig. 2). Some studies have shown increased mortality rates with rapid infusion of fluids compared with standard infusion (⁶²) and with immediate compared with delayed resuscitation (⁶³). The concept of low-volume fluid resuscitation or permissive hypotension avoids the detrimental effects of early aggressive resuscitation while maintaining a level of tissue perfusion that, although decreased from normal, is adequate for short periods (⁶⁴). This approach is contraindicated in traumatic brain injury and spinal injuries, and its effectiveness still needs to be confirmed in randomized clinical trials.

Fig. 2: Blood volume replacement during first 12 to 24 h after trauma. Values of various parameters represent trigger points at which relevant blood components should be transfused. aPTT = activated partial thromboplastin time; Fg = fibrinogen; Hct = hematocrit; PCC = prothrombin complex concentrate; Plt = platelets; PT = prothrombin time; RBCs = red blood cells. Reprinted with permission from Oxford University Press/Br J Anaesth 2005;95:130. Copyright 2005, Board of Management and Trustees.

After penetrating trauma, survival was improved with delayed compared with immediate resuscitation (⁶³), but in patients with blunt trauma, no significant difference was found (⁶⁵). A review concluded that mortality was higher after on-site resuscitation compared with in-hospital resuscitation (⁶⁶). It seems that greater increases in blood pressure are tolerated without exacerbating hemorrhage when they are achieved gradually and with a significant delay after the initial injury (⁶⁷).

RBC transfusion enables the maintenance of oxygen transport in some patients. Early signs of inadequate circulation are relative tachycardia, relative hypotension, a greater than 10% decrease in oxygen consumption, oxygen extraction of greater than 50%, and Pvo₂ of less than 32 mmHg (^68-70). In general, RBC transfusion is recommended to maintain hemoglobin (Hb) above 8 g dL⁻¹ (⁷¹).

Significance of RBC transfusion in trauma

Significant bleeding is inevitable in major trauma, and 40% of trauma-related deaths are attributable to uncontrollable bleeding (⁷²). Injured patients often receive significant amounts of crystalloids, colloids, and allogeneic RBC transfusions (⁵). Blood products are associated with side effects (⁷³) and, in trauma, may be linked to MOF (^{5, 74, 75}) and an increased incidence of infections (^{76, 77}).

This serious complication results in prolonged ICU stays and high mortality. Once MOF has developed, mortality can be as high as 36% (⁷⁴). RBC transfusion has been shown to be an independent risk factor for postinjury MOF (^{74, 75}). In addition, there is a strong dose-response relationship between early RBC transfusion and the development of MOF (⁷⁴). Reducing the number of RBC units transfused may thus decrease the risk and severity of MOF. Furthermore, the age of transfused RBC units has been shown to be an independent risk factor for postinjury MOF (^78-81).

The precise mechanism of RBC transfusion-related MOF has not been completely elucidated; however, recent evidence supports the hypothesis that bioreactive lipids with polymorphonuclear cell priming activity are generated by RBCs during storage (^{80, 82}). Although the initial insult caused by tissue damage and hypoxia primes the inflammatory system, subsequent transfusions of stored RBCs containing bioreactive lipids may activate systemic inflammatory response, resulting in MOF (⁷⁴).

Infection is a common complication in trauma patients, and RBC transfusion has been found to be an independent risk factor for the development of postinjury infections (^{76, 77}). Several mechanisms have been proposed. The exposure of patients to large quantities of foreign antigens is thought to down-regulate the immune system (⁸³). The presence of leukocytes in RBC units may be an additional major contributory factor, particularly because of the priming of neutrophils (^{79, 80, 84}). However, results from clinical trials remain inconclusive to whether leukodepletion reduces the immunosuppressive effect of allogeneic RBC transfusions (^{84, 85}). In addition, stored RBCs are less deformable, and once transfused, they may obstruct capillary blood flow, predisposing tissue to ischemia, and infection (^{83, 84, 86}).

No prospective randomized trial comparing restrictive and liberal transfusion regimens in trauma exists; however, 203 trauma patients from the Transfusion Requirements in Critical Care trial (⁸⁷) were recently reanalyzed (⁸⁸). A restrictive transfusion regimen (Hb transfusion trigger <7.0 g dL⁻¹) resulted in fewer transfusions as compared with the liberal transfusion regimen (Hb transfusion trigger <10 g dL⁻¹) and seemed to be safe. However, no statistically significant benefit of MOF or posttraumatic infections was observed. It should be emphasized that this study was neither designed nor powered to answer these questions with precision. In addition, it cannot be excluded that the number of RBC units transfused merely reflects the severity of injury. Therefore, the observed correlation between numbers of RBC units transfused and MOF may reflect a correlation between the severity of injury and MOF. Adequately powered studies similar to the Transfusion Requirements in Critical Care trial are therefore urgently needed in posttraumatic patients.

In a prospective observational study, 15,534 trauma patients, of whom 1703 received RBC transfusion with a mean of 6.8 ± 6.7 U, were analyzed (⁸⁹). After controlling for severity of shock, RBC transfusion within the first 24 h was associated with increased mortality, admission to the ICU, and length of ICU and hospital stays. Even without a clearly established cause and effect relationship between RBC transfusion and adverse clinical outcome, comparative studies with alternatives to allogeneic RBC transfusions would be more than welcome in trauma care. However, of the multiple alternatives applied during elective surgery (⁹⁰), only cell salvage can be used in patients after major trauma, provided that the surgical field is sterile. Therefore, measures to avoid trauma-associated coagulopathy, and thus reduce the need for RBC transfusion, are crucial in the initial treatment of patients after major trauma (⁵).

SPECIFIC TREATMENT STRATEGIES TO SUPPORT COAGULATION

Strategies to support coagulation include prevention of hypothermia or rewarming, replacement of blood cellular components, replacement of hemostatic factors, and the use of pharmacological agents to reduce bleeding.

Prevention of hypothermia and strategies for rewarming

Hypothermia and coagulopathy in trauma are currently poorly understood; in general, the greater the degree of hypothermia, the greater the risk for uncontrolled bleeding (⁹¹). When hypothermia is associated with severe injury, mortality rates up to 100% have been reported (^{91, 92}). The effects of hypothermia include altered platelet function, impaired coagulation factor function (a 1°C drop in temperature associated with a 10% drop in function), enzyme inhibition, and fibrinolysis (⁹³). Steps to prevent hypothermia and the risk of hypothermia-induced coagulopathy include removing wet clothing, covering the patient to avoid additional heat loss, increasing the ambient temperature, forced air warming, giving warm fluid therapy, and, in extreme cases, extracorporeal rewarming devices (^{94, 95}). Animal studies of controlled hypothermia in hemorrhage have shown some positive results compared with normothermia (^{96, 97}). However, there are no clinical data supporting the therapeutic benefit of controlled hypothermia in trauma patients.

Blood components

Platelets

In trauma, selected indications for platelet transfusion include massive transfusion, disseminated intravascular coagulation (DIC), prophylaxis for surgery (or other invasive procedures), and platelet function disorders (^{98, 99}). In a bleeding patient, or a patient with a coagulopathy such as DIC, a platelet count less than 50 × 10⁹ L⁻¹ is usually an indication for platelet transfusion. Platelets are rarely required when the platelet count is above 100 × 10⁹ L⁻¹ (^{98, 100, 101}). A higher target level of at least 100 × 10⁹ L⁻¹ has only been recommended for those with multiple trauma or injury undergoing surgery or high-risk invasive procedures, (^{98, 100, 101}) although evidence to support this practice is poor (¹⁰²). Platelet transfusion is not indicated to correct a low platelet count in the absence of bleeding (^{5, 98, 100, 101}).

If platelet transfusion is necessary, one platelet apheresis concentrate will increase the platelet count by 20 to 25 × 10⁹ L⁻¹ in most adult patients (^{98, 100, 101}). The platelet count should be checked 10 to 15 min after platelet infusion to ensure the adequacy of therapy. An increase of less than 20 × 10⁹ L⁻¹ after 15 min in a patient without ongoing bleeding may be indicative of human leukocyte antigen (HLA) antibodies, and further non-HLA-matched platelets will not be effective. Transfused platelets become fully functional after 4 h (^{98, 103}).

Fresh-frozen plasma

Fresh-frozen plasma (FFP) is separated within 6 to 8 h of whole blood collection, frozen at −18°C, and stored up to 1 year. The volume of a typical unit is 200 to 250 mL. FFP contains normal levels of all coagulation factors except FVIII (which rapidly decays). The indications for use of FFP in trauma are prolongation of prothrombin time and activated thromboplastin time of more than 1.5 times normal in patients with a bleeding coagulopathy such as diffuse bleeding, massive transfusion, and DIC (^{5, 104, 105}). There is no evidence base for the dose that should be used; however, 15 mL kg⁻¹ is widely accepted. For the acute reversal of the effects of warfarin, the best practice is to use PCC; however, if PCC is not available, the same effect can usually be produced with an FFP dose of 15 mL kg⁻¹ (^{100, 104, 106, 107}).

High infusion rates (¹⁰⁸) of blood products containing citrate can decrease calcium concentrations particularly in patients with hypothermia or liver failure (who are unable to metabolize citrate). Such patients may require monitoring of serum calcium (¹⁰⁹).

Coagulation factors

In bleeding trauma patients, cryoprecipitate or fibrinogen is indicated when despite previous FFP treatment, fibrinogen is less than 1 g L⁻¹ (hypofibrinogenemia) with bleeding in massive transfusion and DIC (¹⁰⁴). One unit of cryoprecipitate per 7 to 10 kg increases the plasma fibrinogen level by 0.5 g L⁻¹. ABO blood group compatibility is preferred because of the 10 to 20 mL of plasma in each unit (^{100, 101, 104}).

Indications for the use of fibrinogen in trauma are the same as for cryoprecipitate, but commercially prepared fibrinogen concentrates are only available in some countries. If it is available, fibrinogen is preferred over cryoprecipitate (^{100, 101, 104}). In the case of low antithrombin levels and concurrent severe bleeding, there are no data to support routine administration of antithrombin concentrates because the fall in antithrombin is usually to a level similar to coagulation factors, resulting in a balanced hemodilution.

Coagulation factor concentrates (Table 3) may be the final step in treating severe bleeding that is unresponsive to standard therapy with FFP and platelets in trauma (^{106, 110}) and include PCC, factor XIII concentrate, and recombinant factor VIIa concentrate (^{106, 111-118}). Recombinant activated factor VII (rFVIIa), an agent widely used in the management of bleeding episodes in hemophilia patients with inhibitors, causes localized activation of coagulation at sites of TF exposure. The enhanced thrombin generation in the presence of fibrinogen leads to the formation of a stable fibrin clot. Recently, the results of a randomized controlled study of rFVIIa for control of bleeding in patients with severe blunt and penetrating trauma have demonstrated efficacy (¹¹⁷). Patients in this study had severe traumatic bleeding (defined as the need for transfusion of 6 U of RBCs within 4 h of admission) and received three infusions of rFVIIa (200, 100, and 100 μg kg⁻¹) or placebo at entry, that is, after transfusion of an eighth RBC unit and 1 and 3 h later-in addition to local standard surgical treatment to manage hemorrhage. In the intention-to-treat population of blunt trauma, the RBC reduction was estimated to be 2 U per patient (P = 0.07), and the incidence of acute respiratory distress syndrome was lowered (P = 0.03). The exclusion of blunt trauma patients who died within the first 48 h from the analyses (an a priori decision to exclude possible unpreventable deaths) significantly decreased the RBC requirement, resulting in an estimated reduction of 2.6 RBC units per patient (P = 0.02). In this same cohort of blunt trauma patients, the need for massive transfusion-defined as more than 20 U of RBCs-was significantly reduced by rFVIIa treatment from 33% to 14% (P = 0.03), a relative risk reduction of 56%. There was also a significant treatment-related reduction in the incidence of acute respiratory distress syndrome and/or MOF (P = 0.05). No such statistically significant benefits were observed in penetrating trauma (¹¹⁷). Adverse event rates between placebo- and rFVIIa-treated groups were similar, with no differences in thromboembolic events observed. An ongoing phase III study, which aims to enroll 1500 patients, investigates if the use of rFVIIa in blunt trauma has a benefit on mortality.

Table 3: Coagulation factor concentrates in trauma

Pharmacological agents to support coagulation

Antifibrinolytic drugs (aprotinin, tranexamic acid, and epsilon aminocaproic acid) block the breakdown of fibrin by plasmin. Given prophylactically after elective surgery, these drugs reduce the rate of RBC transfusion by a relative 30%, save about 1 U of RBCs in patients requiring transfusion, reduce the need for reoperation due to bleeding, and do not have an increased risk of thrombosis (¹¹⁹). It is not known whether antifibrinolytic agents have the same effect in the injured patient (¹²⁰); however, a large randomized trial (Clinical Randomization of Antifibrinolytics in Significant Hemorrhage) is currently underway and aims to assess the effects of tranexamic acid in patients with or at risk for significant bleeding with trauma.

Desmopressin (DDAVP) is a synthetic analog of arginine vasopressin-1-deamino-8-d-arginine vasopressin. Its mechanism of action is to release von Willebrand factor from endothelial storage sites, increase the density of glycoprotein receptors on platelet surfaces, and increase plasma levels of factor VIII and tissue plasminogen activator (^121-123). Although it has been suggested that desmopressin may be effective in reducing hemorrhage after coronary artery bypass grafting in patients receiving aspirin before surgery, results from published studies are inconclusive (^124-126). Desmopressin has been used to minimize perioperative allogeneic blood loss, but there is no convincing evidence that it does so in patients who do not have congenital bleeding disorders (¹²⁷). No data exist for trauma patients.

In patients receiving vitamin K antagonists, the recommendations for serious bleeding at any elevation of the international normalized ratio is to withhold the vitamin K antagonist therapy and to give vitamin K1 supplemented with FFP or PCC (¹²⁸).

CONCLUSIONS

The management of bleeding and coagulopathy after trauma is an area that has not received the attention necessary to answer the many open questions that remain. It is clear that the priority during initial treatment must be the achievement of hemostasis and the maintenance of tissue oxygenation. Coagulation may be supported by controlling temperature and blood pH and by correcting hemostatic deficiencies. Furthermore, nonvital procedures may be undertaken once hemostasis has been achieved and the patient is stable.