CHAPTER 58 CURRENT CONCEPTS IN THE DIAGNOSIS AND MANAGEMENT OF HEMORRHAGIC SHOCK
The precipitating single common factor of hemorrhagic shock is severe acute blood loss, yet the clinical syndrome of hemorrhagic shock is heterogeneous. The diagnosis and management of this condition at first would appear to be simple. However, the very nature of how shock occurs and how the individual compensatory mechanisms respond to both the injury itself and the therapeutic interventions translate into a complex spectrum of diseases. This disease spectrum extends from immediate circulatory collapse to total body ischemia reperfusion injury with associated complex inflammatory and anti-inflammatory responses that in many instances evolve into multiple organ dysfunction. The purpose of this chapter is to provide a review of the current issues and a clinical perspective regarding the diagnosis and management of shock.
EPIDEMIOLOGY OF SEVERE HEMORRHAGIC SHOCK
Injury is the leading cause of death for individuals younger than age 44 years in the United States. Overall, trauma results in approximately 150,000 deaths per year, and severe hypovolemia caused by hemorrhage is a major factor in nearly half of those deaths. The leading causes of death remain head injury and hemorrhage. These findings have been reported both in the civilian literature1 and military experiences—the Vietnam conflict,2,3 and most recently in Baghdad.4 Approximately one-third of trauma deaths occur out of hospital; exsanguination is a major cause of deaths occurring within 4 hours of injury. The distribution of battlefield injuries in the Vietnam War showed that 25% of the deaths occurred as a result of massive exsanguination and were not salvageable. An additional 19% of deaths were deemed salvageable, and these were the result of torso exsanguination (10%) and peripheral exsanguination (19%).3
CLINICAL PERSPECTIVE: LENGTH AND DEPTH OF HYPOTENSION
Severe hemorrhagic shock is characterized by cool, moist, pallid, or cyanotic skin. The patient is tachycardic and hypotensive, and the severity of these clinical manifestations may vary from patient to patient depending on age, underlying cardiovascular disease, and the presence of medications or associated toxic compounds such as drugs or alcohol. Hypotension is the hallmark of shock, is easily measured, and its presence is always a predictor of poor outcome.5–7 The severity of shock, however, can only be partially quantified by the presence of hypotension. Obviously, patients with rapid and massive blood loss will manifest low blood pressure shortly after the injury. Sophisticated devices or monitoring techniques are not required to establish that such patients are close to dying. Prompt and efficient interventions must be instituted immediately, aimed at control of hemorrhage and replacement of the blood volume. Unfortunately, the absence of hypotension after injury does not rule out the presence of shock. In fact, it may mislead the inexperienced clinician to a false sense of security about the need for aggressive resuscitation.8
Injured patients arriving to the trauma center with hypotension or a history of transient hypotension comprise only 6%–9% of the total number of trauma patients.5,6 From a practical viewpoint, patients with hemorrhagic shock can be stratified into three groups. Group 1 comprises patients with exsanguinating hemorrhage. They generally have significant chest injuries to the heart or great vessels, and/or massively disrupted abdominal visceral or significant retroperitoneal bleeding such as pelvic fractures. This group represents one-third of the total number of hypotensive patients or 2% of the total trauma patient population. They arrive alive to our trauma centers only as a direct result of well-organized prehospital and trauma systems. The response to fluid administration in this group is minimal or completely absent, and these patients are termed nonresponders.
Some clinicians suggest that most shock states can be reversed with more volume resuscitation, with correction of hypothermia, and with inotropic agents. Yet, many patients die, if not acutely from irreversible shock, then later from events initiated by severe, prolonged, or unrecognized shock. Clinically, however, circulatory collapse associated with shock is the common pathway of early progressive deterioration precipitated by trauma and hemorrhage. Patients suffering from severe penetrating trauma to the chest or abdomen, with major injuries to the thoracic or intra-abdominal vessels, who do not die at the scene, may undergo aggressive interventions that may include an emergency thoracotomy and massive transfusion. These patients usually require damage control interventions, and manifest severe hypothermia, coagulopathy, and circulatory failure, even after their hemorrhage has been controlled.9 Much less common is the clinical scenario of combat casualty victims or civilian trauma patients from rural areas who may develop circulatory collapse and irreversible shock in a slow protracted fashion because of inadequate resuscitation as a result of lack of blood products, delayed evacuation, or prolonged hypotension before definitive care can be provided. Clearly these are two separate clinical scenarios that in the end may have similar outcomes. Because the spectrum of hypoperfusion, ischemia, and total body reperfusion may differ significantly from one clinical scenario to the next, future therapeutic interventions will need to be more sophisticated and targeted than what we offer today. Fluid resuscitation as currently prescribed represents a rather naive approach to therapy in a complicated pathophysiological process.10
The potential enhancement of early recognition of progressive or refractory shock and reducing the onset of circulatory collapse early in the management of the patient with hemorrhage offers the opportunity of redefining the “golden hour” (i.e., the window of opportunity for efficacious resuscitation).11,12
Discrepancies between Clinical Syndrome of Shock and Animal Models Used to Study Shock
When devising models of hemorrhagic shock, investigators in this field are challenged by two sometimes mutually incompatible goals. First, researchers desire to minimize animal-to-animal variability. Second, researchers seek to achieve a model with clinical relevance. Bleeding in the clinical environment is typically uncontrolled. As hemorrhage continues and arterial pressure decreases, the rate of bleeding tends to slow. Hemostatic mechanisms may come into play when blood pressure is low enough so that bleeding ceases entirely. Resuscitation can actually dislodge the nascent clot by raising blood pressure, and thereby precipitate recurrent hemorrhage.13 Numerous animal models of uncontrolled bleeding have been described.14 These models are generally useful for studying novel approaches to achieve hemostasis, for determining the optimal timing of resuscitation, or for determining the optimal parameters/endpoints for resuscitation.15–17 However, uncontrolled hemorrhage models, while attractive because of their clinical relevance, suffer from a high degree of animal-to-animal variability. Therefore, the use of these models for preclinical efficacy studies of new therapeutic agents requires relatively large sample sizes to obtain statistically meaningful data.
On the other hand, controlled hypotension, as originally described by Wiggers, allows a more standardized grading of injury. When hemorrhagic shock is induced using the Wiggers model, it is initially necessary to periodically withdraw additional aliquots of blood to maintain the target blood pressure because of the body’s normal compensatory responses to acute hypovolemia.18 However, after the shock state has persisted for a period of time, it is no longer possible to withdraw blood to maintain the target blood pressure. On the contrary, it now becomes necessary to reinfuse aliquots of the shed blood to maintain the target blood pressure. This phase has little resemblance to what happens in clinical practice, but in the laboratory it is used to delineate what is known as the “decompensation endpoint,” defined as the transition point from withdrawal to reinfusion of blood. This transition point typically occurs after 90–120 minutes of shock depending on the species. This model is imperfect as well, because variability occurs as to the time of the decompensation endpoint from one animal to the next one. Obviously the Wiggers model allows for a more reproducible injury, one that has little resemblance to the clinical scenario but one that facilitates mechanistic or cause/effect studies and that has been used extensively to delineate many of the cellular and molecular events described after hemorrhage.19,20
Diagnosis of Shock
The phenomenon of ischemia and reperfusion has been extensively studied. Shock and resuscitation are functionally a total body ischemia/reperfusion injury. SIRS (systemic inflammatory response syndrome) and MODS (multiple organ dysfunction syndrome) are the expressions of a conglomerate of events associated with ischemia and reperfusion. Unfortunately, we rely on nonspecific global methods to provide indirect information regarding tissue perfusion. The current challenge is to identify innovative methods for quantifying the magnitude of shock. Some of the questions that should guide future research in this field include: how the interaction among the patient’s physiological reserve, compensatory responses, and the individual organs endurance/response to ischemia may ultimately manifest, and identifying the best methods to accurately measure each of these components. Presently, we scarcely understand these issues and use rather crude methods to quantify the insult. The term tissue perfusion is used liberally to refer to an entity not readily quantifiable. In practical terms, we rely on base deficit and lactate as the global parameters used to gauge tissue perfusion. In the intensive care unit (ICU), we add to the interpretation of these values by hemodynamic parameters that are commonly obtained only after invasive monitoring techniques can be implemented, such as thermodilution cardiac output or continuous SVO2 from either the central or mixed venous circulation.21,22
Assessment of Tissue Perfusion
Base deficit (BD) and lactic acid provide information about the degree of anaerobic metabolism. Lactate is an indirect measure of oxygen debt. Base deficit is defined as the amount of buffer necessary to bring the pH back to 7.40 with pCO2 of 40 torr. Davis et al.23 stratified patients by BD as mild (2–5), moderate (6–14), and severe (>15). The magnitude of the initial BD was found to directly correlate with hypotension and further need for fluid resuscitation. In addition, 65% of patients who had a worsening BD despite resuscitation had ongoing hemorrhage. BD is considered a reliable marker for shock and the need for transfusion in multiple trauma patients.24 Davis et al.25 also demonstrated that BD could remain abnormal despite improvements in mean arterial pressure, cardiac output, mixed venous O2 saturation, and oxygen extraction in an animal model of hemorrhagic shock. Rutherford et al.26 found that BD, age, and head injury had an additive effect on the incidence of mortality. Base deficit is an expedient and sensitive measure of both the degree and the duration of inadequate perfusion. It is useful as a clinical tool and enhances the predictive ability of other trauma scores.
Lactate is the end product of anaerobic glycolysis. As a more direct measurement of tissue hypoxia induced acidosis, it has been shown to have strong prognostic value. Normalization of serum lactate below 2 mmol/l or less within the first 24 hours is associated with 100% survival. Only 78% of patients survived when lactate levels remained elevated during the period of 24–48 hours after shock. The mortality is greater than 85% if lactate remains elevated at 48 hours postshock.27 Sauaia et al. demonstrated an association between an early (12 hours) rise in serum lactate above 2.5 mmol/l and multiorgan failure.28 Manikis et al.29 further found that initial and peak lactate levels, as well as the duration of hyperlactatemia, correlated with the development of MODS after trauma. In summary, the initial lactate level and time to normalization of lactate correlate with risk of MODS and death. However, improved survival using lactate or BD as endpoints for resuscitation has not been shown.30
HEMORRHAGE CONTROL
Local Hemorrhage Control
QuikClot is a new agent designed to induce local hemostasis of large wounds associated with vascular lesions. It is a zeolite mineral granular powder that increases the concentration of clotting factors in the wound. It has been used in models of extremity arterial injury and hepatic laceration with good results. QuikClot has been deployed with the U.S. military forces. Other local hemostatic dressings are currently being tested, such as Chitosan dressings and compressive dressings with a layer of fibrinogen thrombin and calcium chloride embedded in a Dexon mesh.31,32
Systemic Hemorrhage Control
Failure of coagulation in trauma is multifactorial and has been characterized by the combined presence of coagulation abnormalities resembling disseminated intravascular coagulation (DIC), caused by systemic activation of coagulation and fibrinolysis.33 Fibrinolysis is probably caused by release of tissue plasminogen activator from injured tissues33–35 and dilutional coagulopathy caused by aggressive fluid resuscitation.36 This massive transfusion syndrome includes dilution of coagulation factors and impairment of platelet number and function.37 Hypothermia may contribute to the coagulopathy by slowing of enzymatic activities of the coagulation cascade38,39 and impeding platelet function.40
Ideally, the introduction of an effective hemostatic agent that would act only at the site of injury, without induction of systemic activation of coagulation, could improve hemorrhage control and reduce hemorrhage-related mortality and morbidity in both military and civilian trauma victims.41 Systemically administered hemostatic agents include coagulation factors in the form of cryoprecipitate and fresh frozen plasma. A general hemostatic agent may be one that enhances full thrombin generation, and thereby the formation of a stable, tight, fibrin hemostatic plug that is resistant to premature fibrinolysis.
Patients with profuse bleeding resulting from extensive surgery or trauma often develop a complex coagulation pattern that includes reduced plasma levels of fibrinogen, Factor VIII and Factor V, and decreased platelet counts. These patients may have an impaired capacity to generate thrombin. In addition, thrombocytopenia is common, as well as increased fibrinolysis as a result of massive tissue damage. Several of these factors are important for an adequate thrombin formation as well as for the formation of a tight fibrin plug. Thus, low levels of fibrinogen result in the formation of a loose fibrin structure and to decreased activation of Factor XIII, the fibrin-stabilizing factor.42
Factor VIIa
Trauma patients with massive bleeding thus may benefit from intravenous rFVIIa to help generate a thrombin peak, which may be enough to form a firm, stable fibrin hemostatic plug and thereby decrease the bleeding.43 Because thrombin has such a crucial role in providing hemostasis, any agent that enhances thrombin generation in situations with impaired thrombin formation may be characterized as a “general hemostatic agent.” Large animal studies with clinically relevant models have demonstrated the efficacy of rFactor VIIa in induction of clot and improved survival.44–47
An in vitro study reported by Meng et al.48 indicated that rFVIIa may not be efficient in acidosis, but hypothermia has little effect on rFVIIa efficacy. These authors recommended that clinicians who may contemplate using FVIIa in trauma patients take note of the level of acidosis present and consider biochemical correction of acidosis before administration of FVIIa. Because of the profound effect of pH on the prothrombinase (FXa/FVa) complex, correction of acidosis may by itself improve hemostasis.
A preliminary study of FVIIa use reported three patients with high-velocity penetrating trauma, and four suffered blunt trauma. They had all received multiple transfusions and conventional interventions without achieving hemostasis. In all cases, the administration of rFVIIa caused a cessation of the diffuse bleeding and their coagulation parameters normalized.44 Boffard and colleagues49 published a larger multicenter trial demonstrating benefit from the use of rFVIIa. Both animal and human data indicate that there is a need for more research in the field using greater numbers of individuals both in human and animal research, as well as long-term follow-up to understand the issues regarding possible complications related to increased thrombosis.
Fluids
Management priorities in the bleeding trauma patient begin with airway control, ventilation, and oxygenation.50 The goal of resuscitation is to stop the bleeding and replete intravascular blood volume to maximize tissue oxygen delivery. Cardiac output, blood pressure, and oxygenated blood flow to vital organs are important determinants of outcome. Measurement of blood pressure may not be feasible, and it is not a necessary requirement to initiate therapy during the primary survey. In fact, a quick global assessment of adequacy of perfusion can be obtained with examination of the characteristics of the pulse. Obvious signs of shock are sufficient to establish the diagnosis and determine the severity of blood volume deficit. Additional information such as blood pressure measurement, electrocardiographic monitoring, pulse oximetry, and capnography may be useful. Adequate intravenous (IV) access for infusion of normothermic fluids is the next priority while other causes of severe hypotension and shock are being sought, such as tension pneumothorax, pericardial tamponade, or neurogenic shock.
