Chapter 13: Fluid and Blood Resuscitation in Traumatic Shock

Circulatory shock has a high mortality. Severe hemorrhage after injury carries a mortality rate of 30% to 40% and is responsible for almost 50% of deaths occurring within 24 hours of injury.^1,2 Septic shock has a mortality of up to 50%.³ Resuscitation, starting in the prehospital setting and continuing throughout the victim's care in the ED and on into the hospital, has the goal of restoring the necessary level of tissue perfusion and oxygenation for survival while simultaneously limiting further volume loss.

Intravascular volume depletion is a common feature of circulatory shock; crystalloids, colloids, and blood products (packed red blood cells and plasma) are the primary volume expanders to reverse intravascular volume depletion. This chapter focuses on the issues related to fluid and blood resuscitation in traumatic shock, with an emphasis on hemorrhagic shock. Processes that cause loss of plasma fluid and electrolytes (e.g., dehydration, burns, sepsis), often requiring aggressive fluid therapy, are discussed in other chapters (see chapters 150, "Toxic Shock Syndromes," 151, "Sepsis," and 216, "Thermal Burns").

The principal objectives of fluid and blood resuscitation in traumatic shock are: (1) to restore intravascular volume sufficient to maintain oxygen-carrying capacity and tissue perfusion for adequate cellular oxygen delivery, and (2) to prevent or correct derangements in coagulation.

Circulatory shock is a state of impaired oxidative metabolism and homeostasis due to inadequate oxygen delivery to meet metabolic demand, and hypoperfusion leading to inadequate cellular waste removal. Acute shock triggers a complex range of physiologic responses that compensate for intravascular volume loss and maintain perfusion to the most important vascular beds. Shock also produces a global insult to the vascular endothelium that activates the coagulation and inflammatory systems (Figure 13-1). When uncorrected, coagulopathy, additional inflammation, and organ system damage result.

While moderate transient hypoperfusion may be well tolerated, prolonged or severe hypoperfusion leads to accumulation of oxygen debt and progressive cellular and organ dysfunction. If rapid and severe, sudden cardiovascular collapse and death may occur.

Coagulopathy observed in trauma victims, termed trauma-induced coagulopathy, is ascribed to a combination of factors beginning with loss of coagulation factors from hemorrhage, followed by hemodilution from crystalloid resuscitation, and then exacerbated by acidosis (evidenced by a base deficit) and hypothermia that occur during the course of ongoing hemorrhage and resuscitation.⁴ These same principles also apply to other causes of shock.⁵ However, acidosis does not, by itself, have a significant effect on coagulation until the pH decreases below 7.0.⁶

The lethal triad of hypothermia, hemodilution, and acidosis can be viewed a partially iatrogenic based on the type of resuscitation provided. The normal total circulating volume of an adult is approximately 7% of ideal body weight or about 5 L for an average 70-kg adult patient divided into about 3 L of plasma and 2 L of red blood cell volume (Table 13-1). Understanding this distribution explains how trauma-induced coagulopathy can be iatrogenically produced.

Factors exacerbating trauma-induced coagulopathy, including hyperfibrinolysis and clotting factor and platelet consumption or dysfunction, are related to the degree of injury and hemorrhage.⁷ Early trauma-induced coagulopathy is characterized by anticoagulation and hyperfibrinolysis, likely modulated through the protein C pathway. Endothelial damage associated with trauma or sepsis stimulates an increase in thrombomodulin expression on the endothelium that complexes with thrombin and in turn activates protein C (Figure 13-2). Complexed thrombin is now no longer available for its usual hemostatic role of cleaving fibrinogen to form fibrin. Subsequent consumption of plasminogen activator inhibitor-1 by activated protein C contributes to hyperfibrinolysis (Figure 13-2).⁸

Thrombomodulin normally found in the endothelium may be viewed as protective because it is able to sequester thrombin and thus allow for generation of adequate levels of protein C to prevent thrombosis caused by low-flow states. However, on a mass systemic level, this appropriate local response may become pathologic, resulting in clotting inhibition.

The clinical appearance of acute traumatic shock is variable and depends on the cause, rate, volume, and duration of volume loss or bleeding; the presence of other acute disorders; the effects of current medications; and the patient's baseline physiologic status.

The hemodynamic response to acute severe hemorrhage-induced hypovolemia traditionally includes tachycardia, hypotension, and signs of poor peripheral perfusion (cool, pale, clammy extremities with weak peripheral pulses and prolonged capillary refill). Arterial and venous vasoconstriction leads to a narrowing of the pulse pressure. Cerebral hypoperfusion causes alterations in mental status. With increasing blood loss, signs and symptoms become more pronounced. Classification of hemorrhage severity as a percentage of blood volume loss based on vital signs is not accurate and should not be used to guide ED resuscitation.⁹

Patients with excellent baseline physiologic status (e.g., young athletes) may have such a robust compensatory response to hemorrhage that they appear stable and do not manifest tachycardia or hypotension, even with large blood volume loss. Signs of peripheral hypoperfusion and subtle mental status alterations may be the only clues that the severity of hemorrhage is greater than predicted based on hemodynamic parameters. Elderly patients may not develop tachycardia due to underlying heart disease or medications such as β-adrenergic blockers. Bradycardia or lack of tachycardia may occur in about 30% of patients with intra-abdominal hemorrhage from increased vagal tone in response to hemoperitoneum. In a pregnant trauma patient, compression of the inferior vena cava by the gravid uterus can decrease central venous return and worsen hypotension and tachycardia in the setting of less severe hemorrhage.

Vital signs offer little value unless they are in extreme low ranges. Arterial blood pressure does not adequately reflect cardiac output or regional perfusion. Clinical evidence of peripheral hypoperfusion is useful but is not a quantitative measurement. Metabolic information (discussed below), assessment of mechanisms of injury and illness, and appropriate imaging studies offer the best chance for early recognition of severe hemorrhage and for guiding an appropriate response to treatment.

There is a biphasic relationship between oxygen consumption and oxygen delivery (Figure 13-3). While this relationship exists for the body as a whole, it also exists for each individual organ system and is largely mediated by the microcirculation. Oxygen debt develops in tissues or organs when oxygen delivery does not meet the metabolic demands and represents the amount of extra oxygen that is needed to metabolize the accumulated products of anaerobic metabolism once perfusion is restored. Oxygen debt is the accumulation of multiple oxygen deficits over time and is a measure of whole-body ischemia. Oxygen debt is the only physiologic measure that has clearly been linked to both mortality and even morbidity in the form of multiple organ failure after shock. The degree of oxygen debt incurred after injury has also been clearly linked to inflammation.¹⁰

Measurements of lactate and base deficit are in fairly ubiquitous use in trauma centers and EDs and are now even being used as point-of-care testing in the prehospital setting as a triage tool. Using lactate and/or base deficit as a resuscitation trigger and its clearance/normalization as an endpoint is prudent and is associated with improvements in survival. However, elevated lactate and base deficit is a late finding, which is first preceded by increases in oxygen extraction as indicated by a decrease in indicators of oxygen consumption such as mixed or central venous hemoglobin oxygen saturation (Figure 13-2). Furthermore, improvements in oxygen delivery resulting in lactate clearance do not indicate that oxygen extraction is normal.

Although lactate is also a measure of oxygen debt, its resolution during resuscitation does not ensure the important repayment and resolution of oxygen debt. Lactate also suffers from the intermittent and time-dependent nature of its measure and change. To counter these, continuous measures of oxygen consumption surrogates are available in the form of invasive oximetric catheters measuring central venous oxygen saturation (ScvO₂) and tissue oxygen saturation (StO₂) noninvasively from target tissues such as skeletal muscle using near-infrared spectroscopy. StO₂ measurements with infrared spectroscopy are based on the fact that tissue blood volume is 70% to 80% venous blood with the remaining 10% to 20% a combination of arterial and capillary blood. Thus aggregate measures of hemoglobin oxygenation in the tissue are heavily dominated by the venous blood and thus the postextraction compartment. Although not an exact mirror of ScvO₂, ensuring StO₂ values above 50% appears to be reasonable from both a diagnostic and treatment standpoint to begin and sustain hemostatic resuscitation. A target StO₂ or ScvO₂ between 60% and 70% should be maintained to help ensure resolution of tissue hypoxia.

Resuscitation begins in the prehospital setting and continues in the ED. The priority for prehospital care is treatment of life-threatening conditions and rapid transport to an appropriate facility. For the hemorrhaging patient, this entails assuring adequate ventilation and oxygenation (including securing an airway if necessary), controlling external bleeding (if present), and protecting the spinal cord (if potentially vulnerable).

In the ED, restore intravascular volume to reverse or limit systemic and regional hypoperfusion, maintain oxygen-carrying capacity so that tissue oxygen delivery meets demand, and limit ongoing blood loss and prevent the development of coagulopathy. From the moment resuscitation begins, prevent the development of hypothermia by keeping the patient warm and administering warmed IV fluids and blood products.¹¹ Endogenous hypothermia occurs when heat production from cellular respiration is decreased by hypoperfusion and inadequate tissue oxygen delivery. Major causes of exogenous hypothermia are exposure and the use of below-body-temperature fluid and blood resuscitation. Apply external warming devices early to prevent external heat escape. Warming devices that allow for the rapid warming of infused fluid and blood should be used for all patients in whom large-volume resuscitation is undertaken.¹¹

AIRWAY CONTROL, VENTILATION, AND OXYGENATION

If spontaneous ventilation is not adequate, intubate and ventilate to achieve an arterial hemoglobin oxygen saturation of ≥94%. Identify and treat potential respiratory conditions such as pneumothorax, tension pneumothorax, hemothroax, or upper airway obstruction.

VASCULAR ACCESS AND MONITORING

Establish adequate IV access concurrent with airway management. Large-bore (14- to 16-gauge in adults) peripheral lines may be adequate if two or more can be secured. Central venous access may be necessary. Intraosseous lines are suitable for resuscitation.

Institute continuous ECG heart rate monitoring, continuous pulse oximetry, and, if possible, continuous end-tidal carbon dioxide monitoring. Monitor arterial blood pressure, mental status, and peripheral perfusion frequently. Bedside US is useful to identify intraperitoneal bleeding, assess cardiac function and volume status, and assist in central venous cannulation.^12,13

HEMOSTATIC-HYPOTENSIVE RESUSCITATION

Hemostatic resuscitation begins in the field. In traumatic shock, the goal is deliberate hypotensive resuscitation, so that intravascular volume expansion is limited and blood pressure is not normalized. Blood pressure goals are systolic blood pressures of 80 to 90 mm Hg (10.7 to 12.0 kPa) in trauma and 90 to 95 mm Hg (12.0 to 12.7 kPa) in head injury. Normalization of blood pressure and the raising of hydrostatic pressure can increase hemorrhage. This is particularly evident in noncompressible hemorrhage where raising blood pressure could potentially "pop the clot" and reestablish active hemorrhage.

Hypotensive resuscitation targets fluid resuscitation only when systolic blood pressure falls below 70 to 80 mm Hg (9.3 to 10.7 kPa) and/or there is evidence of decreasing mental status and, thus cerebral hypoperfusion. This practice has been widely adopted in combat casualty care (see chapter 302, "Military Medicine") and is partially supported by civilian studies.¹⁴ Preliminary results of traumatic injury comparing low versus traditional mean arterial pressure guided resuscitation indicate that a lower mean arterial pressure group sustains less blood loss, requires fewer transfusions, and has improved early and 30-day survival.¹⁵

Hypotensive resuscitation is problematic for a number of reasons, including the inability to obtain frequent and accurate blood pressures and, in the civilian world, its application to the elderly and those with hypertension or cerebral vascular disease. Hypotensive resuscitation should not be used in patients with myocardial disease, cerebral ischemia, or traumatic brain injury. Obviously there is not an unlimited time span that patients can tolerate hypotensive resuscitation. From a cellular standpoint, patients will at some time incur irreversible damage from prolonged tissue hypoxia if adequate tissue oxygenation is not restored.

ISOTONIC CRYSTALLOID SOLUTIONS

Isotonic crystalloids, normal saline, and lactated Ringer's solution are the most commonly used resuscitation fluids in the United States. Concerns about each fluid are: (1) infusion of large volumes of either normal saline or lactated Ringer's solution can cause increased neutrophil activation, (2) lactated Ringer's solution can increase cytokine release and may increase lactic acidosis when given in large volumes, and (3) normal saline can exacerbate intracellular potassium depletion and cause hyperchloremic acidosis.¹⁶

Crystalloid solutions are isotonic but hypo-oncotic, because they lack the large protein molecules present in the plasma. Low oncotic pressure results in substantial shifts of crystalloid to the extravascular space corresponding to the relative size of the intravascular and interstitial fluid compartments (Table 13-2). This was the physiologic basis for the 3:1 ratio for crystalloid to blood. For every amount of blood lost, three times that amount of isotonic crystalloid is required to restore intravascular volume because, at best, about 30% of the infused fluid stays intravascular. Based on this rule, a loss of 1 L of blood (about 15% to 20% of total circulating blood volume) would require about 3 L of isotonic crystalloid to restore normovolemia, assuming no ongoing blood loss. Bearing this in mind, the recommendations for initial fluid resuscitation have been to administer 2 to 3 L of crystalloid solution in acute hemorrhage and assess the response before initiating blood transfusion. In light of the new data reflecting improved outcomes with plasma-based resuscitation in traumatic hemorrhage, no more than 2 L of crystalloid solution should be administered before giving serious consideration to using blood products, especially if ongoing hemorrhage is expected.

Common modifications of isotonic fluids include use of acetate instead of lactate in Ringer's solution, Hartmann's solution, and Plasma-Lyte (Table 13-3). Plasma-Lyte is the shared name for a family of isotonic crystalloid solutions available worldwide, marketed with variation in composition according to regional preferences. Solutions containing lactate or acetate are considered balanced crystalloids because they are buffered and have a lower chloride concentration compared to normal saline. There is evidence of low confidence that balanced crystalloids are associated with reduced mortality compared with normal saline when used for fluid resuscitation in sepsis.¹⁷

COLLOID SOLUTIONS

Colloid solutions contain larger molecular weight particles that have oncotic pressures similar to normal plasma proteins. Therefore, colloids theoretically have several advantages during resuscitation. They would be expected to remain in the intravascular space, replacing plasma proteins lost due to hemorrhage, and more effectively restore circulating blood volume than crystalloid solutions. An argument favoring the use of colloids has been the concern that extravascular shift, or third-spacing, of infused crystalloid solutions has potential adverse effects, including pulmonary interstitial edema with impaired oxygen diffusion and intra-abdominal edema with diminished bowel perfusion. However, pathologic conditions, such as hemorrhagic shock and sepsis, lead to increased vascular permeability that can allow for eventual extravascular leakage of these larger colloid molecules.¹⁸

The colloids used as resuscitation fluids are a heterogeneous group of agents with widely varying characteristics and effects. These agents have no proven consistent benefit, and there is evidence of harm in some patients with critical illness.^19,20

Hetastarch is a high-molecular-weight polysaccharide that has no proven benefit in critically ill patients.²¹ Currently, for logistical reasons of weight and size, hetastarch is the recommended fluid of choice for the far forward resuscitation of battlefield casualties.²² It is used as a part of a hypotensive resuscitation strategy with no more than 1000 mL to be given. The administration of 1000 mL of hetastarch is theoretically equal to the administration of 3 L of isotonic crystalloid.

HYPERTONIC SOLUTIONS

Hypertonic saline (7.5% saline) has been proposed as a potential crystalloid alternative that would limit the tissue edema effects that are often of concern with isotonic crystalloid solutions. Hypertonic saline has anti-inflammatory and immunomodulatory effects, with demonstrable decrease in lung and intestinal injury in animal models of hemorrhagic shock.¹⁸ The intravascular shift of fluid from the extravascular space may be potentially beneficial in head trauma patients by limiting cerebral edema, lowering intracranial pressure, and improving cerebral perfusion. The addition of dextran to hypertonic saline was aimed at sustaining the hemodynamic effect of the hypertonic saline. The volume of hypertonic saline solution that can be given during resuscitation is limited due to the potential for hypernatremia. Similar to hetastarch, an approach using smaller volumes (250 mL) of hypertonic saline (7.5%) has also been advocated. Randomized controlled trials have not demonstrated a clinically significant difference in outcome when hypertonic saline was compared with conventional isotonic fluid.²³ However, these studies did not generally incorporate hypotensive resuscitation as a part of the overall strategy, and in the prehospital trail, transport times to trauma centers were short.²⁴

PACKED RED BLOOD CELLS

Packed red blood cells (PRBCs) are the most commonly transfused blood product (see chapter 238). Using only PRBCs in the patient with traumatic shock and ongoing bleeding (without plasma and platelets) will do little to promote hemostasis and may not restore tissue oxygenation. If hemorrhage is definitively controlled, do not transfuse if the hemoglobin concentration is > 10 g/dL (> 100 g/L). Consensus recommendations for transfusion are a hemoglobin concentration between 6 and 7 g/dL (60 to 70 g/L) for those without cardiopulmonary, cerebral, or peripheral vascular disease.²⁵ For a hemoglobin between 6 and 10 g/dL (60 to 100 g/L), use clinical judgement for transfusion.

When possible, typed and cross-matched blood is preferable. However, if time and the patient's clinical status do not permit full cross-matching, type-specific blood is the next option, followed by low-titer O-negative blood. In U.S. blood banks, whole blood is not stocked, and only PRBCs are available (see the chapter 302 for further discussion of whole-blood transfusion).

PRBC transfusion obviously restores lost hemoglobin. Using current preservatives, PRBCs can be stored for up to 45 days, and the average age of a unit of blood administered in the United States is about 21 days. Stored red blood cells can lose deformability, limiting their ability to pass normally through capillary beds or even resulting in capillary plugging. The oxygen dissociation curve is altered by loss of 2,3-diphosphoglycerate in the erythrocytes of stored PRBCs, impeding the off-loading of oxygen at the tissue level.²⁶ Despite these changes, no consistent harm has been identified with use of "older" stored blood.^27,28

FRESH FROZEN PLASMA

Fresh frozen plasma (FFP) is plasma obtained after the separation of whole blood from red blood cells and platelets and then frozen within 8 hours. A unit of FFP has a volume of 200 to 250 mL and contains all the coagulation factors present in fresh blood. Kept frozen, FFP can be stored for up to a year after the unit was collected. It takes between 15 and 20 minutes to thaw a unit of FFP in a 37°C water bath, which can limit availability in a massive transfusion situation. However, some major trauma centers and their respective blood banks keep thawed FFP (kept at 1°C to 6°C) available for immediate use. FFP may be kept thawed at this temperature for up to 5 days with little degradation of plasma proteins. FFP ABO compatibility is required, but because there are no red cells in FFP, Rh compatibility is less important, and universal donor FFP is typically AB+ (see chapter 238, "Transfusion Therapy").

PLATELETS

Platelets are collected from whole-blood donations or from single donors using apheresis techniques and can only be stored for up to 5 days. Six units of pooled random-donor platelet concentrate or one apheresis-collected single-donor platelet concentrate in an adult will increase platelet count up to 50,000/mm³ (50 × 10⁹/L).

MASSIVE TRANSFUSION PROTOCOLS

Massive transfusion is generally defined as the requirement for >10 units of PRBCs within the first 24 hours of injury. Massive transfusion is not a substitute for definitive surgical hemostasis but enhances the ability to achieve surgical hemostasis and to limit complications.^29,30 Draw sufficient specimens from the patient early in anticipated massive transfusion because once the patient has received close to one blood volume of transfused products, new blood specimens will contain so much donor blood that cross-matching of subsequent units is difficult. Initially use isotonic solutions to begin resuscitation in predetermined aliquots (250 to 500 mL) while assessing the likelihood of ongoing active hemorrhage and the need for, type, and timing of hemostasis and for hypotensive resuscitation.^31,32 Variables predictive of need for massive transfusion include penetrating mechanism of injury, positive FAST examination, blood pressure <90 mm Hg (< 12.0 kPa), and pulse rate >120 beats/min.^33,34 If hemorrhage is clinically significant and immediate hemostasis is not achievable, transition from crystalloid-based resuscitation to a plasma-based massive transfusion protocol.

An estimated 10% of military trauma patients and 3% to 5% of civilian trauma patients receive massive transfusion.¹ Soldiers receiving high ratios of plasma to PRBC (a plasma:PRBC ratio of about 1:1.4) have significantly improved survival rates than those receiving lower ratios (between 1:1.8 and 1:2.5). Additional evidence supporting this also comes from combat casualty care data suggesting that early administration of fresh whole blood (something not currently practical in civilian trauma) offers survival advantages.^18,35 These survival observations using high plasma-to-PRBC ratios were confirmed by civilian studies, which also appear to indicate less incidence of trauma-induced coagulopathy.^33,36

High plasma-to-PRBC ratio resuscitation appears to also offer survival benefit independent of coagulopathy, and plasma may enhance cell survival by endothelial repair and reducing vascular permeability. It may also be that plasma is simply a superior fluid for perfusion and tissue oxygenation restoration. In addition, many trauma centers now include early platelet administration in predetermined ratios to plasma and PRBC in their resuscitation protocols based on findings that platelet function can rapidly decrease soon after trauma.^1,34

Studies routinely using FFP and platelets with PRBCs during massive transfusion have yielded mixed results on reducing mortality.^34,37 The best ratio of PRBCs to platelets to FFP during massive transfusion is controversial.³⁸ Some experts advocate a 1:1:1 ratio, although lower ratios of platelets and FFP have been used without evidence of inferiority.³⁹ Likewise, the benefits with adjunctive agents such as calcium chloride, prothrombin complex concentrates, and fibrinogen concentrate included in some protocols are unknown. Tranexamic acid (an antifibrinolytic) is advocated early in the resuscitation process of major trauma victims based on the early fibrinolytic phase of trauma-induced coagulopathy and on clinical studies suggesting improved survival.⁴⁰ An institution-specific massive transfusion protocol is recommended to guide the clinician when ordering blood components and to facilitate release from the blood bank (Figure 13-4). Tranexamic acid, an antifibrinolytic agent, is not currently a component of massive transfusion protocols in the United States. Its use is discussed in the chapters 254, "Trauma in Adults" and 302, "Military Medicine." If given, it must be administered as soon as possible and within 3 hours of injury. The dose is 1 gram IV bolus in 100 mL normal saline.

Calcium

PRBCs and FFP contain citrate that can complex calcium, producing life-threatening hypocalcemia.⁴¹ Most massive transfusion protocols include the administration of calcium and monitoring of ionized calcium. Calcium chloride is preferred over calcium gluconate because a well-perfused liver is required to liberate more free calcium from calcium gluconate. Maintain ionized calcium levels at or above 0.9 mmol/L.⁴²

Thromboelastography and Thromboelastometry

The blood coagulation system consists of nearly 80 very tightly coupled biochemical reactions. Conventional coagulation testing such as the prothrombin time and activated partial thromboplastin time do not take into account the cellular components of the clot such as red cells and platelets. Because the blood clot itself consists of a complex three-dimensional network of cross-linked fibers made of fibrin, platelets, and red cells entrapped within this mesh, newer viscoelastic tests such as thromboelastography or thromboelastometry are being developed to measure the physical and dynamic characteristics of clotting. Thromboelastography and thromboelastometry can detect trauma-induced coagulopathy and can be used to guide massive transfusion protocols better than traditional prothrombin time and activated partial thromboplastin time assays.⁴³ However, impact on clinical care is uncertain at this time.⁴³

TRANSFUSION-RELATED ACUTE LUNG INJURY

Acute lung injury is the leading cause of transfusion-related fatalities. It develops within 6 hours of blood transfusion and is felt to be related to an inflammatory reaction.⁴⁴ Care is supportive.

TRANSFUSION-ASSOCIATED CIRCULATORY OVERLOAD

Circulatory overload occurs when the recipient's circulatory system is overwhelmed by the rate of transfusion or the volume of products transfused.⁴⁵ Clinically, it is characterized by the acute onset of dyspnea, hypertension, tachypnea and tachycardia, and marked elevation in brain natriuretic peptide levels. Care often involves rapid diuresis.