Circulatory Changes

Organ Perfusion and Function in Shock

Hormones and Metabolism

As mentioned before, a severe decrease in cardiac output resulting in a decrease in arterial blood pressure during hypovolemic shock results in activation of the sympathetic nervous system through the baroreceptor reflex and liberation of norepinephrine from nerve endings and epinephrine from adrenal medulla so that circulating levels of these catecholamines increase.11–13,74,81,99 The insulin secretion by the pancreas is inhibited, and glucagon secretion is enhanced by high circulating norepinephrine levels.⁹⁹ The renin-angiotensin-aldosterone system is activated, and the pituitary secretion of vasopressin/antidiuretic hormone and opioids increases.^{11–13,23,99} The pituitary response to stress further includes an increase in adrenocorticotropic hormone (ACTH) with resultant corticosteroid release by the adrenal cortex, unless limited by the so-called relative adrenal insufficiency following hypoperfusion-induced adrenal damage.^12,99,152 These factors may be essential for survival because prior adrenalectomy decreases survival of animals subjected to hypovolemic shock, and steroid repletion is protective in this respect.¹² This protective effect can be attributed to, among others, less overactivation of the sympathetic nervous system and increased sensitivity of the heart and vasculature to circulating levels of catecholamines.¹²

Finally, the secretion of atrial natriuretic peptides by the myocardium declines in response to hypovolemia and diminished wall stress of the atria. These factors, among others, result in tachycardia and a diminished renal excretion of water and salt to restore circulating blood volume. Endogenous opioids could play a role in maintaining shock by their vasodilating and myocardial depressant properties, however.11,13,23 Administration of the opioid antagonist naloxone and its derivatives augment arterial blood pressure in hypovolemic shock.^{13,23,143,225} Similarly, thyrotropin-releasing hormone depresses the opioid system and increases arterial blood pressure, cardiac function, and survival during hypovolemic shock in animals.²³ Thyroid hormone may have a similar effect.²²⁶

During trauma, hypovolemic shock, and cellular ischemia, intermediary metabolism undergoes profound changes, partly caused by an altered hormonal milieu.⁹⁹ The early hyperglycemic response to traumatic/hypovolemic shock is the combined result of enhanced glycogenolysis, caused by the hormonal response to stress and elevated epinephrine, cortisol, and glucagon levels; increased gluconeogenesis in the liver, partly mediated by glucagon; and peripheral resistance to the action of insulin, the secretion of which may be diminished shortly after onset of shock but may be enhanced later after shock.^24,47,99,129 This resistance is most likely the result of an altered hormonal milieu—the increase in circulating epinephrine and cortisol levels. During the late, irreversible stage of hypovolemic shock, however, hypoglycemia supervenes, at least in animal models, because glycogen stores may be depleted and the capacity for gluconeogenesis by the liver may decrease because of ischemia.^{12,24,47,111,130}

Increased gluconeogenesis in the liver, and to a lesser extent in the kidneys, follows increased efflux of amino acids such as alanine and glutamine from the muscle to the liver because of breakdown of muscle protein.99,222 The latter is evidenced by increased urinary losses of nitrogen and a negative nitrogen balance.^99,222 Amino acid metabolic changes may contribute to the immunodepression of trauma. Lactate produced in muscle also can be converted to glucose in the liver.⁹⁹ Finally, fatty acid metabolism undergoes profound changes, with depressed lipolysis, ketogenesis, and combustion of fatty acids during shock and an increase in the resuscitation phase.^99,222 Some investigators regard a deranged intermediary metabolism of primary importance for the eventual outcome of shock, whereas others merely consider these changes a result of the shock process itself.²²²

Inflammatory and Immunologic Changes

Activation of the xanthine-oxidase system and formation of uric acid from the ATP breakdown products hypoxanthine and xanthine during reperfusion could liberate ROS, which damage vascular endothelium and parenchymal cell membranes through peroxidation of lipids.* The release of ROS during ischemia-reperfusion may activate macrophages and attract neutrophils, partly mediated by release of cytokines via activated nuclear factor-κB (NF-κB).^194,228 The interaction of ROS fueled by oxygen and NO may further play a role in inflammation and vascular tone after perfusion.^90,228 ROS scavengers may inhibit formation of toxic peroxynitrite via NO, and ROS and may inhibit breakage of DNA single strands and activation of poly(ADP-ribose) polymerase, which contributes to cellular injury.^90,92,221 Some time after hypovolemic shock and resuscitation, iNOS may become active particularly in the gut and liver; circulating NO breakdown products may increase and inhibition of the excessive NO release may ameliorate hemodynamic changes, organ inflammation, and neutrophil accumulation and function, partly via less peroxynitrite formation, unless inhibition leads to a decrease in cardiac output.^{90,92,155,220} Increased iNOS-derived NO also may be prevented and treated by corticosteroids or ACTH fragments.^93,229

Proinflammatory mediators may be expressed locally in a variety of organs in response to hemorrhagic shock, including heart and lungs, and this is partly under control of α-sympathoadrenergic and neuroimmune stimuli, toll-like receptor 4, NF-κB, hypoxia-inducible factor, glycogen synthase kinase-3β, and other factors involved in cell signaling.228,230–233 During and after hemorrhage, hypovolemic shock, and resuscitation, macrophages, including lung macrophages and Kupffer cells in the liver, may release cytokines, including TNF-α, IL-1, IL-6, and IL-8. This inflammatory response is attenuated when reperfusion takes place in hypoxic, rather than normoxic, conditions.²³⁴ The response can be ameliorated by blockade of NF-κB, administration of the macrophage-inhibitor pentoxifylline, or ATP-MgCl₂ increasing hepatic blood flow.*

Ischemia per se and the immune consequences of gut barrier injury may play a role in Kupffer cell responses. The reperfused gut, together with the liver, may be a source of systemically released cytokines, as suggested by animal experiments and observations in humans after trauma, and translocated endotoxin may play a role.^† During reperfusion after resuscitation, cytokines may induce and amplify the inflammatory response to ischemia and may induce further local and remote organ damage with circulatory changes.^{166,187,236,240,241} Spillover of mediators into the mesenteric lymph or portal and systemic circulations during reperfusion of prior ischemic gut may have deleterious effects on remote organs by inducing neutrophil activation and adherence, which may contribute to a lung vascular injury with increased permeability.^{185,186,188,240,241} Circulating levels of proinflammatory cytokines may be of predictive value for remote organ damage, including ARDS, after trauma in patients.^230,237,238 Endotoxin binding or antibodies and cytokine antibodies may ameliorate remote tissue damage after bleeding, hypovolemic shock, and resuscitation.^185,186

Trauma and shock/resuscitation have also been shown to activate the complement and the arachidonic acid systems.151,242–245 Complement activation may yield potent vasodilating and leukoattractant substances and contribute to remote inflammatory organ damage (ARDS). Ischemia may generate phospholipase A₂, catalyzing arachidonic acid metabolism into prostaglandins via the cyclooxygenase pathway, releasing thromboxane A₂ and prostacyclin, and into leukotrienes via the lipoxygenase pathway.^10,151,243 Thromboxane A₂, released from platelets, neutrophils, and cell membranes, has potent vasoconstricting properties and promotes aggregation of platelets and neutrophils, whereas prostacyclin has vasodilating properties and inhibits platelet and neutrophil aggregation.²⁴³ Leukotrienes have vasoconstricting properties, increase capillary permeability, and attract neutrophils.¹⁵¹ Vasoconstricting prostaglandins may be involved in tissue damage during ischemia-reperfusion, and vasodilating prostaglandins may be involved in the vasodilated state of terminal hypovolemic shock.^10,242 Another lipid mediator that may be released is platelet-activating factor, but the precise action of this mediator is unclear.^149,170

The interplay of these factors may result in endothelial activation throughout the body and an inflammatory reaction, ultimately involving attraction, activation, and endothelial adherence of neutrophils, as shown in animal models of hypovolemic shock after bleeding and ischemia-reperfusion.* Neutrophils release vasoconstricting, platelet-aggregating, and damaging thromboxane A₂ and may inhibit vasodilating prostacyclin, via secreted ROS and proteases such as elastase.^{117,228,238,244,246} Neutrophil aggregation and secreted activation products also may also play a role in the reperfusion injury by impairing resumption of small vessel blood flow, even in the presence of a seemingly adequate cardiac output and arterial blood pressure.^† In humans, the activation of neutrophils after trauma, with increased adhesion molecule expression and propensity for degranulation, is associated with morbidity after trauma, such as development of MOF and predisposition to sepsis.^{115–117,233,237,247} After initial leukopenia (neutropenia) following trapping of leukocytes in the microcirculation, activation of the pituitary-adrenal axis and release of corticosteroids and catecholamines during hypovolemic shock result in an increase of circulating neutrophils following demargination and release from bone marrow, together with eosinopenia and lymphocytopenia.^{46,112–115,216} A tertiary decrease of circulating neutrophils in patients with a downhill course may be explained by microcirculatory sequestration.¹¹⁵ The hemodynamics, organ function, and survival of rats with hypovolemic shock/resuscitation are improved if rats are made neutropenic before the challenge, and this may relate to improved regional and capillary blood flow.^112,243 A monoclonal antibody against or antagonists of neutrophil-endothelial adhesion molecules decrease reperfusion injury in lungs, liver, stomach, and intestines after hypovolemic shock or ruptured aortic aneurysm and may improve survival, at least in animal models.^{114,117,118,187,228} This does not impair host defense against subsequent bacterial infections.¹¹⁴

However, neutrophils may later become downregulated after initial stimulation by circulating proinflammatory and anti-inflammatory mediators.111,229,248 Neutrophil dysfunction is evidenced by a diminished potential to migrate and to digest and kill bacteria, perhaps in the presence of an inhibited respiratory burst.^{111,188,247,248} In hemorrhaged mice, the infusion of granulocyte colony-stimulating factor or IL-6 after hemorrhage may partly prevent neutrophil defects and protect against death from subsequent pulmonary sepsis.²⁴⁸ Also, the opsonization function of macrophages, that is, the reticuloendothelial system, is depressed so that removal from the circulation of fibrin, cell aggregates, and bacteria by the liver is at least transiently impaired.* This may relate to the appearance after hypovolemic/traumatic shock of substances in blood that depress reticuloendothelial system function or to a decrease of the α₂-glycoprotein fibronectin in plasma, a substance that aids the reticuloendothelial system in opsonization.^{4,167,236,249,251} This deficiency may contribute to development of MOF and might be reversed by infusion of plasma cryoprecipitate.²⁵¹ Hypovolemic shock and gut-derived factors may blunt the increase in bone marrow cytopoiesis after soft tissue trauma and endotoxin and contribute to susceptibility to sepsis.^252,253

Hemorrhagic/hypovolemic shock and subsequent resuscitation depress the immune system by suppressing the function of not only neutrophils but also lymphocytes and macrophages; this depresses humoral and cellular immune responses, decreasing antigen presentation and delayed hypersensitivity to skin test antigens and increasing susceptibility to sepsis.^† Part of this may be mediated via neuroimmune modulation and resulting efferent sympathetic, adrenergic, and vagal stimulation.^37,93 In patients, the immune defect correlates with the extent and severity of trauma and the degree of blood resuscitation required, but animal experiments document that hemorrhage/resuscitation per se depresses immune function, although trauma and blood transfusions may only be synergistic in this respect.^{236,250,254,255,258} Priming of immune cells may explain in part the increased sensitivity to endotoxin and sepsis after hypovolemic shock, although other authors have described that prior hypovolemic shock and priming decreased the immune response and increased the tolerance to endotoxin or sepsis.^{114,184,187,259} Hemorrhage decreases the capability of lymphocytes to proliferate and to produce lymphokines (IL-2) in response to mitogens, an effect that seems dependent on an energy or NO deficit or on Ca²⁺ influx in these cells after ischemia because the defect can be overcome by administration of Ca²⁺ influx blockers.^{236,250,254,255,258} Increased macrophage production of cytokines during hypovolemic shock and resuscitation may be followed by decreased ability of the cells to release mediators such as TNF-α and to express HLA-DR, upon challenges, and to process and present antigens to lymphocytes. This may relate to a cellular energy deficit, accumulation of Ca²⁺, and enhanced prostaglandin E₂ synthesis.^‡ The immunodepression after hypovolemic shock and predisposition to sepsis may finally include the release by Kupffer cells, among others, of anti-inflammatory mediators, such as IL-10 and soluble receptors (receptor antagonists) for previously released proinflammatory cytokines, and this may relate to sepsis-induced MOF and increased risks of morbidity and mortality in trauma patients.^250,260 Otherwise, the immunologic consequences of trauma, hemorrhage, and hypovolemic shock depend on numerous additional factors, including gender and other genetic influences.^230,247,250 Men may exhibit more immunodepression after trauma/hemorrhage than women. The clinical implication may be that men are more susceptible than women to microbial infections after trauma.²⁶¹

Circulating coagulation factors and platelet counts may decrease after hypovolemic shock and resuscitation, whereas fibrin products may increase. This is the consequence of coagulation activation and fibrinolysis inhibition by endothelial activation, tissue injury, and inflammatory responses, even though dilution after fluid resuscitation may heavily contribute.181,237,262 Disseminated intravascular coagulation (DIC) and fibrin deposits, if insufficiently removed by the fibrinolytic system, are believed to contribute to a decrease in plasma coagulation factors and to widespread microvascular organ dysfunction.^181,237,263 Proinflammatory responses, some resuscitation fluids, hypothermia, and acidosis may contribute to DIC and the coagulation defect of severe hemorrhagic/traumatic shock.^237,264

Reperfusion and Irreversible Shock

Reperfusion of various organs, including the heart, gut, skeletal muscle, brain, kidneys, and liver, after a transient episode of ischemia, as occurs during hypovolemic shock, results in the so-called reperfusion injury, which limits the possibility for resumption of microvascular tissue blood flow and function of organs, particularly of the liver, even if cardiac output and arterial blood pressure have been restored to normal value.* Redistribution of blood flow during hypovolemia may be only partly attenuated by reperfusion.

Reperfusion after a certain period of shock and diminished oxygen uptake results in an increase in oxygen uptake above baseline levels, provided that oxygen delivery and cellular function are adequate.42,62,73,99 This repayment of the oxygen debt is largely determined by the increased demands for oxygen to resynthesize ATP from adenosine and phosphates and to rebuild the lost energy stores. This repayment is determined by the extent to which mitochondria are damaged during ischemia and the availability of substrates to resynthesize high-energy phosphates and restore cellular contents of these compounds because the substrates needed for synthesis may have been washed out, necessitating de novo synthesis.^24,142,143 Resuscitation may not completely restore energy levels, the activity of the Na⁺/K⁺ pump, and the membrane potential of skeletal muscle and liver necessary to remove accumulated fluid and Na⁺ in the cell.^125,127,142

Reperfusion not only results in resumption of oxygen delivery but also of Ca²⁺ to the tissues. This Ca²⁺ may be taken up by cells and may contribute to the reperfusion injury by damaging cell organelles, inhibiting mitochondrial respiration, and activating proteases and prostaglandin synthesis.127,129,148,149 Reperfusion injury of heart, gut, kidneys, and liver after resuscitation from hypovolemic shock in animals may be prevented in part by administration of Ca²⁺ influx blockers independently of their vasodilating effects, suggesting that Ca²⁺ overload is partly responsible for the reperfusion injury.^{127,129,148,149} Finally, endothelial damage and swelling and cellular aggregation may hamper the regional regulation of blood flow during resuscitation from hypovolemic shock.^{79,95,110,122} Neutrophil-mediated endothelial injury may increase capillary permeability and contribute to fluid losses during resuscitation.^{112,178,179,228,267} Conversely, the intravenous administration of energy in the form of ATP-MgCl₂ or adenosine-regulating compounds may help tissues to recover from ischemia and resume function, independently of the vasodilating effects of the compounds, by providing energy, improving the microcirculation, and reducing cell swelling to promote survival.* Nevertheless, the ability of organs or the whole body to increase oxygen uptake during reperfusion above normal may be associated with survival in experimental animals with hypovolemic shock and in hypovolemic patients after trauma or major surgery, whereas inability may be associated with ultimate demise.^9,73,82 Also, ischemic preconditioning may protect against hemorrhagic shock and reperfusion-induced tissue injury.¹⁵⁶

If shock syndrome with hypotension and subnormal oxygen uptake persists after optimal fluid repletion and attempts at reperfusion with inotropic and vasoactive drugs, the condition can be regarded as irreversible and terminal.4,9 The term irreversible shock has been mainly used in animal experiments, however, in which reinfusion of the shed blood after a certain period is unable to reverse the shock syndrome.^4,268 Various factors may play a role.⁴ First, vascular decompensation may contribute to a further decrease in blood pressures and may include diminished constrictive reactivity, dilation of arterioles, and insensitivity to circulating or exogenous catecholamines.^97,108 The decline in vascular resistance may be partly caused by metabolic vasodilation in ischemic and acidotic tissues, overcoming vasoconstrictive influences.^4,10,108 Other factors that may be involved include dysfunction of vascular smooth muscle after induction of iNOS and resultant increased production of vasodilating NO in the vessel wall, acidosis and activation of low ATP-activated K⁺ channels, histamine release, and prostaglandin-induced neurotransmission failure.^† Circulating levels of NO breakdown products, nitrate and nitrite, may be elevated already early after hemorrhage in animals and trauma in humans, although other authors described low levels in humans.²⁶⁹ iNOS upregulation and NO production may be prevented by NO blockers, corticosteroids, or ACTH fragments.^92,93,229 Finally, central cerebral or humoral mechanisms may contribute to the irreversible hemorrhagic shock, and this may relate to endogenous opioids, thyrotropin-releasing hormone, or macrophage-derived cannabinoids.^13,23,270

The decrease in arterial vascular resistance may be particularly pronounced in the tissues, showing most intense vasoconstriction during hypovolemic shock, including gut and skeletal muscle, offsetting the redistribution of blood flow during hypovolemic shock and increasing blood flow to these organs at the expense of blood flow to vital tissues.10,22 In contrast, venous compliance and resistance increase, leading to peripheral pooling of blood and a decrease in venous return to the heart.^4,166 The latter changes may be particularly pronounced in the splanchnic region.¹⁶⁶ During prolonged or irreversible hypovolemic shock, capillary hydrostatic pressure may increase after arteriolar vasodilation and venular constriction, resulting in a decrease in the precapillary-to-postcapillary resistance ratio and promoting fluid filtration into the interstitium.^4,5 Capillary permeability also may increase, resulting in a high capillary hydraulic conductance and a decrease in the reflection coefficient for plasma proteins. Increased permeability for proteins increases capillary filtration for a given intravascular hydrostatic pressure and promotes the formation of edema.^178,267 The increase in permeability may be the consequence of endothelial damage and loss of protective glycocalyx by ischemia-reperfusion, possibly involving ROS and proinflammatory mediators.²⁷¹ It may contribute further to a decline in circulating blood volume.^5,267 Cells may swell, and this may diminish circulating blood volume further.^{4,125,127–129} Expansion of the cellular and interstitial fluid volume at the expense of the intravascular volume is manifested by a preterminal increase of the hematocrit.^4,5,126

Irreversible hypovolemic shock may contribute to MOF and death of patients.2,207 An inflammatory response to ischemic tissue and patchy necrosis/apoptosis may contribute to organ damage and dysfunction and thereby to the irreversibility of hypovolemic shock.^4,114,228 Reperfusion injury may aggravate organ damage and contribute to irreversible shock.^114,228,266 The pump function of the heart may diminish after a decrease in systolic contractility and compliance, and this may contribute to irreversibility of shock during resuscitation.¹⁵⁸ Myocardial dysfunction may contribute to the development of pulmonary alveolar edema if aggressive fluid infusion in attempts to increase cardiac output results in an elevated PCWP.⁴ Diminished function of the heart may hamper restoration of oxygen delivery and uptake to the tissues during resuscitation.^9,45,157,158 Damage of the gut mucosa may cause translocation of luminal bacteria and endotoxins from gut lumen to systemic circulation, at least in experimental animals, and the resultant sepsis may contribute to the irreversibility of hypovolemic shock.*

Causes

One of the most frequent causes of hypovolemic shock is blood loss after trauma (see Box 26.1), including blood loss during or after major surgery.² Ruptured aortic aneurysm and gastrointestinal hemorrhage are other frequent causes of hypovolemic shock. Upper gastrointestinal bleeding can be caused by peptic ulcer disease, reflux esophagitis, variceal bleeding, erosive gastritis (stress ulcer), or aortoduodenal fistula after vascular surgery. Lower gastrointestinal bleeding can result from diverticular disease, carcinomas, or polyps in the colon. Sometimes, massive hemoptysis resulting from a tumor, tuberculosis, fungal infection, or bronchiectasis can be the cause of hypovolemic shock. Hematuria as a result of a tumor or trauma is a rare cause of hypovolemic shock. During multiple trauma, blood loss is essential in causing hypovolemic shock, but trauma itself can activate various mediator systems, with resultant release of vasoactive substances that contribute to the development of shock. In contrast to pure hypovolemic shock, cardiac output can be elevated, and peripheral vascular resistance is often decreased in cases of multiple trauma.²⁵

In trauma patients, external blood loss can be accompanied by internal, invisible blood loss after renovascular trauma or major fractures (e.g., fractures of pelvis or femur). After blunt abdominal trauma, splenic or hepatic ruptures and perforations of hollow viscera are possible. Blunt chest trauma can be accompanied by an aortic rupture, tension pneumothorax, hemothorax or hemopericardium, and tamponade. Femoral artery injury (after puncture) can lead to massive retroperitoneal hematoma. Nonmechanical causes of hypovolemic shock include uncontrolled diabetes mellitus and acute adrenocortical insufficiency, causing severe renal fluid losses. Acute and severe vomiting following obstruction of the gastric outlet or gut, diarrhea, and burn wounds result in loss of plasma water.

Signs and Symptoms

Hypovolemic shock warrants an early diagnosis, avoiding delay in initial treatment. As soon as possible after admission of the patient, fluid resuscitation should begin via a large-bore catheter in a peripheral vein or a percutaneously inserted central venous catheter. During initial fluid resuscitation a history is taken and a brief but thorough physical examination is performed. The latter serves to establish rapidly the cause and severity of shock. Extensive manipulation of a fractured spine or extremity should be avoided. The history of a patient in hypovolemic shock is mainly determined by its cause. The patient may complain of thirst, diaphoresis, and shortness of breath. The patient’s mental state is usually normal unless shock is severe and the patient becomes apathetic or confused. With less severe cases, the patient is anxious, and with more severe cases, the patient is apathetic.

For a clinical diagnosis of shock, hypotension and clinical signs of organ ischemia should be present. Arterial blood pressure and clinical signs are relatively insensitive for small blood losses (Table 26.1).²⁰⁸ This sensitivity can be improved by using the shock index, calculated from heart rate divided by systolic blood pressure.¹⁸ The clinician can recognize shock from a decrease in systolic blood pressure to less than 90 mm Hg or a decrease of more than 40 mm Hg below preshock levels, with a reduction in pulse pressure. Hypotension may be so severe that blood pressure is unrecordable noninvasively. There can be a large gradient between invasively and noninvasively measured arterial blood pressure and between central and radial artery pressure during shock and drug-induced vasoconstriction.²⁷² Hypotension may become particularly marked when the patient sits or stands versus when the patient is supine (orthostatic hypotension).^8,17,18 Postural dizziness, tachycardia, and hypotension are reliable and early signs of hypovolemia, whereas dryness of mucous membranes and axillae, decreased turgor, supine hypotension, and other signs have less diagnostic value.^17,18

Tachycardia may be absent in case of prior use of beta blockers. Elderly patients may have atrial fibrillation and a high ventricular response. Occasionally, bradycardia is present, particularly when vagally mediated fainting supervenes.¹⁶ The peripheral veins are collapsed, and the jugular venous pressure is low. Conversely, an elevated jugular venous pressure should warn the clinician of associated obstruction of the circulation, following pneumothorax, pericardial tamponade, and others, or of pump failure following myocardial contusion or infarction. The respiratory cycle–induced changes in stroke volume and in systolic arterial blood pressure and CVP are exaggerated.²⁰ Although dependent on tidal volume and respiratory compliance, these variations in a mechanically ventilated patient may constitute fair indices of hypovolemia, so high variations may predict an increase in cardiac output upon fluid loading.¹⁹ Noninvasive, pulse contour–based techniques are suitable for these purposes. Fluid responsiveness also can be predicted by an increase in blood pressure and decrease in stroke volume and pressure variations during leg raising or similar maneuvers. The body temperature may decrease, particularly in elderly patients. The gradient between the ambient and toe temperature may be a fair index of peripheral blood flow and a measure for the severity of hypovolemic shock because a reduction in skin blood flow (cold, clammy skin) is an early and ominous sign of shock in view of selective cutaneous vasoconstriction.¹²³ Other signs of hypovolemic shock include tachypnea, oliguria/anuria, diaphoresis, cold and clammy skin with diminished capillary refill, and peripheral cyanosis.

The clinical diagnosis of hypovolemic shock is not difficult in the presence of hypotension and visible loss of blood volume, as occurs during trauma (e.g., fractures), gastrointestinal or pulmonary hemorrhage, burn wounds, and diarrhea. Internal hemorrhage after a ruptured aortic aneurysm, blunt abdominal trauma, or hemothorax is difficult to diagnose except when the history of the patient and obvious physical signs, including dullness on thoracic percussion and abdominal distention and tenderness, point to potential internal bleeding. In the case of upper gastrointestinal blood loss, one should look for signs of chronic liver disease, including palmar erythema, spider nevi, and portal hypertension (ascites), because they could predict variceal bleeding as a cause of hypovolemic shock. Brown discoloration of the palms of the hands and mucosal membranes may point to adrenocortical insufficiency, and a smell of acetone in expiratory breath may point to uncontrolled (ketoacidotic) diabetes mellitus.

General

The diagnostic workup of a patient with hypovolemic shock should not hamper initial resuscitation. After the history and physical examination, the necessity for further diagnostic procedures depends on the underlying cause of shock.

If trauma and external blood loss are the cause of shock, control of external bleeding, crossmatching of blood, and infusion of fluids and blood components have a higher priority than further diagnostic procedures. Treat first what kills first. Blunt chest trauma can be complicated by aortic rupture, tension pneumothorax, hemothorax or hemopericardium, and tamponade. A chest radiograph can be useful to diagnose these conditions. After blunt abdominal trauma, splenic or hepatic ruptures are possible, and an abdominal tap and analysis of the fluid can be performed to exclude or establish intra-abdominal bleeding or hollow-organ perforation.273,274 This diagnostic procedure has been largely replaced, however, by imaging, if time permits, with help of so-called focused assessment sonography for trauma (FAST) or computed tomography of the abdomen in the emergency department. This helps in selecting patients for explorative laparotomy in order to avoid negative surgery or for percutaneous coiling.^273,274 The abdominal viscera show characteristic lesions in hemorrhagic shock with low filling of large veins, decreased perfusion of some organs, wall thickening, submucosal edema, and enhancement of the gut.²⁷⁵ A ruptured abdominal aortic aneurysm can be diagnosed via ultrasonography or angiography if the patient’s condition allows the use of such an invasive, time-consuming procedure. The usefulness of emergency aortic clamping or balloon tamponade for massive abdominal hemorrhage is controversial.²⁷⁶ In the case of gastrointestinal hemorrhage, diagnostic procedures also are performed after initial resuscitation, including gastroscopy for upper gastrointestinal bleeding, sigmoidoscopy for lower gastrointestinal bleeding, and angiography. Introduction of a nasogastric tube (and early intubation) can be useful to aspirate blood, diagnose bleeding, prevent aspiration during vomiting, and follow the course of bleeding.

Laboratory Investigations

At admission of a patient with suspected hypovolemic shock, blood samples should be taken to determine the hemoglobin/hematocrit and leukocyte and platelet counts; electrolyte, creatinine, and lactate concentrations; arterial blood gases and pH; and blood typing (crossmatching). Immediately after hemorrhage, the hemoglobin content and hematocrit of blood are normal, but they decrease in time with refilling of the plasma compartment, as does the protein content.* A high hemoglobin content and hematocrit can be encountered during pure loss of plasma (water), as occurs during burn wounds or severe diarrhea. Acute hypovolemic shock may be accompanied by slight leukopenia followed by leukocytosis.^{46,111,112,248,253} If coagulation disorders are suspected (therapy with anticoagulants, liver disease, bleeding tendency), platelet counts and coagulation tests should be performed. Transient thrombocytopenia may ensue if shock is severe and massive amounts of whole blood are lost and rapidly replaced by erythrocyte concentrates or nonsanguineous fluids (i.e., through dilution). Isolated thrombocytopenia without DIC may thus occur.

The concentrations of electrolytes (sodium, potassium, chloride) in blood are essentially normal unless the concentrations in the fluid lost deviate from those in plasma (hypertonic and hypotonic dehydration) and resuscitation fluids and shock is accompanied by severe metabolic acidosis. In the latter example, potassium leaves the cell, potentially leading to hyperkalemia.111,125 More often, however, less severe forms of shock are accompanied by hypokalemia because of adrenergic receptor–stimulated Na⁺/K⁺-ATPase. Saline fluid loading or overloading can result in hyperchloremic metabolic acidosis. Adrenocortical insufficiency may result in hyponatremia, hyperkalemia, and hyperchloremic acidosis, caused by changes in urinary excretion induced by mineralocorticoid deficiency. In a patient with liver disease, the corresponding abnormalities can be found in laboratory studies. In the case of uncontrolled diabetes mellitus, hyperglycemia and glucosuria are observed. As previously mentioned, the glucose concentrations in blood can be elevated in early shock and, occasionally, depressed in late shock. Finally, the concentration of unbound Ca²⁺ in blood may diminish during hypovolemic shock after cellular uptake and polytransfusion of red blood cell concentrates if they contain calcium-binding citrate as an anticoagulant.^127,150

During hypovolemic shock, metabolic acidosis, often associated with an elevated lactate level in blood, is common and of prognostic significance, although the decrease in bicarbonate and base excess may not parallel the increase in lactate.27,31–33,38,41 The pH can be subnormal after lactic acidosis and a decrease in the bicarbonate content, even if ameliorated by hyperventilation and a decrease in PCO₂.^{27,32,33,49,50} The lactate level in blood can be determined rapidly and followed frequently (every 2 hours). The lactate level and its course during treatment also is of prognostic significance during shock because during successful treatment the lactate level decreases and the bicarbonate concentration and pH increase, whereas an unchanged or even increased lactate level during resuscitation is usually associated with morbidity, including sepsis, MOF, and death.^27,31,32,56 An elevated anion gap, the difference between the sodium on the one hand and the sum of the bicarbonate and chloride concentrations in blood on the other hand, can be a first sign of lactic acidosis, although, as previously mentioned, elevated lactate levels may not be associated with acidosis in the absence of an oxygen debt.²⁷ The serum creatinine concentration is initially normal. The urea content increases following prerenal renal insufficiency, catabolism, or breakdown of blood in the gut during gastrointestinal hemorrhage. In the urine, the osmolarity is increased.

The sodium content is low, together with a low fractional excretion of sodium (FE_Na),¹³⁸ calculated as the quotient of urinary (U) and plasma (P) sodium (Na) and creatinine (creat) concentrations:

In case of acute renal injury (acute tubular necrosis), the urinary sodium content and fractional excretion are increased.¹³⁸ This increased sodium also occurs during adrenocortical insufficiency. Prior diuretic therapy may invalidate this diagnostic tool, however, whereas the fractional excretion of urea may not be affected by diuretics.¹⁰⁴ Tubular injury may be tracked from increased urinary excretion of biomarkers, which may help predict the need for renal replacement therapy.¹⁰⁴

Miscellaneous abnormalities may include elevated levels of nitrate and nitrite, the stable breakdown products of NO.²⁶⁹ Transient elevations of bilirubin, alkaline phosphatase, γ-glutamyltransferase, and transaminases in blood may be severe and denote ischemic liver damage.^222,223 Elevations of creatinine kinase may be caused by skeletal muscle, cardiac, gut, or, less likely, brain damage. Elevated troponin concentrations may specifically indicate cardiac injury.^163,173

Monitoring

Noninvasive monitoring of arterial blood pressure to judge the course of shock and its response to treatment suffices for some patients with hypovolemic shock. Nevertheless, there may be substantial differences between the invasive and noninvasive readings of arterial blood pressure, favoring arterial catheterization and invasive monitoring. Urinary output should be measured hourly in patients with shock to judge the adequacy of treatment because transition of oliguria to a diuresis exceeding 40 mL/hour is an indicator of adequate renal perfusion. The gradient between toe and body temperature and capillary refill time can be used as noninvasive indices of peripheral perfusion.¹²³

Unless hypovolemic shock is rapidly reversed by initial infusion of fluids, there is often a need for hemodynamic and respiratory monitoring in the intensive care unit for a patient with hypovolemic shock. The goal of monitoring is to document the course of shock and its reaction to treatment. Complications can be diagnosed in an early phase so that action can be rapidly undertaken, if necessary. Respiratory monitoring is meant to detect, at an early stage, respiratory insufficiency and muscle fatigue, which are caused by an imbalance in oxygen supply to demand and which may necessitate intubation and mechanical ventilatory support.⁵⁰

Arterial blood pressure can be monitored invasively via a catheter in the radial, axillary, or femoral artery, introduced percutaneously using the Seldinger technique, under aseptic conditions. Percutaneous insertion of a double-lumen or triple-lumen central venous catheter may be useful but does not allow for more rapid fluid infusion than through two peripheral cannulas. The internal jugular, subclavian, or femoral vein may be used for that purpose. This also permits monitoring of CVP and oxygen saturation and a measure of total body oxygen supply-to-demand ratio and predictor of fluid responsiveness. Pressures in the lesser circulation (pulmonary arterial pressure and PCWP) can be measured with the help of a balloon-tipped pulmonary artery catheter inserted percutaneously and advanced under pressure monitoring until the inflated balloon wedges in a pulmonary artery side branch. This catheter also allows for thermodilution measurement of cardiac output and obtaining mixed venous blood for blood gas analysis.73,279 Together with arterial blood measurement of oxygen variables, this allows the calculation of oxygen delivery, extraction, and uptake.^9,25,73 These calculations may contribute to judging the severity of shock and its response to treatment.^6,9,25,73

The CVP reflects the filling pressure of the right ventricle, and the PCWP reflects the left atrial pressure and, in the absence of mitral valve disease, the filling pressure of the left ventricle.¹⁹ Under certain circumstances, however, including ventilation with positive end-expiratory pressure or measurement above the level of the left atrium, when the measured pressure is more influenced by alveolar than by venous pressure, the CVP and PCWP may overestimate true (i.e., transmural) right and left atrial pressures. The response of filling pressure and cardiac output to fluid loading, as measured with the central venous or pulmonary artery catheter, is an index of myocardial function and can be useful to assess fluid responsiveness, particularly in case of preexistent and prognostically unfavorable cardiac disease.* Measurement of PCWP is important if function or compliance of the left ventricle is altered (e.g., in case of preexistent heart disease), when the CVP may underestimate PCWP.^55,71 Conversely, the CVP may overestimate PCWP in cases of severe pulmonary hypertension and right ventricular failure. It has been suggested that changes in CVP during fluid loading do not predict changes in PCWP.⁷¹ The intensity and speed of therapy can be guided by the response of filling pressures and cardiac output, as measured with the use of the central venous or pulmonary artery catheter.^{9,19,31,42,55}

These measurements also can help to time, choose, and dose concomitant therapy with inotropic or vasopressor agents. Together with the plasma colloid osmotic pressure, the PCWP determines filtration of fluid across pulmonary capillaries according to the Starling equation. Monitoring of the PCWP during infusion of fluids may prevent pulmonary edema because infusion can be guided by the filling pressures of the heart. Taken together, data obtained with the pulmonary artery catheter are useful if the hypovolemic origin of shock is not immediately apparent in complicated cases, as in patients with preexistent cardiac disease.9,25,73 Data obtained with the catheter are of diagnostic value in complicated forms of shock because hypovolemic shock is characterized by low filling pressures and cardiac output and a high peripheral vascular resistance. These characteristics may serve to differentiate from other types of shock. The indications for insertion may also include a high risk for shock in patients undergoing major surgery and shock of unknown origin when clinical judgment fails to recognize severe hypovolemia.

Difficulties during treatment also may constitute indications for pulmonary artery catheterization, including hypovolemic shock unresponsive to liberal fluid repletion in the absence of a low jugular venous pressure or CVP and hypovolemic shock together with preexistent cardiac disease unresponsive to fluid repletion if a large discrepancy between CVP and PCWP is suspected and if vasoactive drugs are considered. Monitoring the PCWP may help to lessen the risk for pulmonary edema during fluid loading. Contraindications for pulmonary artery catheterization include those for central venous catheterization. The complications of the technique are discussed elsewhere. Although the use of pulmonary artery catheters is hotly debated because of lack of direct evidence that they help to increase survival, there are some indications that therapy guided by variables collected with the catheter improves the outcome of selected critically ill patients after trauma or surgery.9,25,73,280 Nevertheless, the exact hemodynamic and metabolic resuscitation goals are difficult to define, so the usefulness of the pulmonary artery catheter is difficult to prove.^{9,26,42,67,280}

As an alternative to invasively inserted catheters, various less invasive or even noninvasive (pulse contour–based) systems have been developed that circumvent some of the problems associated with filling pressures as preload indicators and predictors of responsiveness of cardiac output to fluid loading.19,20,281 Among others, the transpulmonary thermodilution technique with detection of thermal changes after central venous injection of cold dextrose 5% in water in the iliac artery allows for calculation of cardiac output, global end-diastolic volume, and extravascular lung water—measures of cardiac preload, pulmonary fluid filtration, and edema.²⁸¹ Assessment of cardiac volumes can be helpful to judge function, similar to echocardiography.^157,281 The latter technique also evaluates filling or injury of large vessels, suspected cardiac contusion, and pericardial tamponade.²⁸² The diameter (changes) of the large veins can be used as an indicator of filling status. The use of pulse-contour techniques for beat-to-beat evaluation of arterial pressure curve–derived stroke volume, as well as pulse pressure and stroke volume variations invoked by the respiratory cycle to guide fluid treatment in mechanically ventilated patients, remains controversial.^20,281 The esophageal Doppler flow probe with which flow time, stroke volume, and cardiac output can be estimated is somewhat operator dependent.

Developments in monitoring the circulation of a patient in hypovolemic shock further include continuous monitoring of central venous or mixed venous oxygen saturation with the help of the fiber-optic technique introduced via catheters, allowing for the continuous evaluation of the oxygen supply-to-demand ratio; right ventricular end-diastolic volume monitoring as an index of filling status; and measurement of tissue blood flow, PO₂, PCO₂, and oxygenation by electrodes and optic techniques.* Venous O₂ saturations may indeed help to guide fluid resuscitation in clinical practice because a low saturation increases upon adequate fluid challenges in fluid-responsive patients. Tissue PO₂ decreases and PCO₂ increases during regional perfusion failure, and these events (in skin, conjunctiva, muscle, or bladder) are probably early signs of hypovolemia following redistribution of blood flow before hypotension ensues and have be used as guides to treatment.* The adequacy of gastrointestinal blood flow can be judged noninvasively with the help of a tonometer balloon catheter in the stomach (or gut) or with help of a sensor sublingually or buccally, in which fluid or air is instilled, and the PCO₂ is measured to calculate the mucosal-to-blood PCO₂ gradient, as explained previously.^† Fluid treatment guided by the adequacy of gastrointestinal mucosal perfusion as judged by tonometry could improve the outcome of hemorrhaged trauma patients compared with resuscitation based on standard hemodynamic variables alone.^{40,123,209,211}

Monitoring the end-tidal CO₂ fraction, determined by and directly related to the blood flow–dependent tissue CO₂ production, and the gradient to arterial PCO₂, determined by blood flow–dependent dead-space ventilation, can help to judge the response to resuscitation.49,65 Hydrostatic pressure measurements in the urinary bladder may reflect the measurements in the abdominal compartment and may help to identify intra-abdominal hypertension and abdominal compartment syndrome developing during extensive fluid resuscitation.²⁸⁵ This may impair gut and renal perfusion with subsequent dysfunction and warrant decompression laparotomy.²⁸⁵

General

Treatment of shock cannot be delayed, so in practice diagnosis and treatment are done simultaneously. Treatment of hypovolemic shock is aimed at the restoration of the circulation and treatment of the underlying cause. Box 26.2 describes some general guidelines, but we do not specifically address burn wound shock requiring a special approach. The main therapeutic goal in hypovolemic shock is to restore circulating blood volume and to optimize oxygen delivery so that oxygen uptake plateaus (see Fig. 26.1) and meets tissue needs.^72,286 Optimization of cardiac output, stroke work, and tissue oxygenation and maintenance of arterial blood pressure are physiologically reasonable resuscitation targets for patients.⁴² Optimization does not imply maximization above levels adequate for tissue needs.^26,67 Studies by some investigators have suggested that supranormal rather than normal oxygen delivery and consumption may be associated with survival from severe trauma or hemorrhage, including a ruptured aortic aneurysm, and that therapeutic targeting at these values (with oxygen delivery >600 mL/minute/m² and oxygen consumption >170 mL/minute/m²) improves the outcome of severe trauma in humans.^{9,25,40,42,83} These concepts are highly controversial, however. In any case, resuscitation based on blood pressure alone probably does not fully restore tissue oxygenation, particularly in the case of myocardial dysfunction after shock and resuscitation.^42,55

Adequate resuscitation from hypovolemic shock should result in an increase in oxygen delivery and uptake, a decrease in lactate levels, and amelioration of metabolic acidosis.* Concomitantly with the increase in cardiac output, arterial blood pressure increases, but normal levels may not be necessary to aim at during resuscitation.^71,72,172 Indeed, successful resuscitation in terms of a restored cardiac output and arterial blood pressure may poorly reflect effective recovery of tissue perfusion, even though reversal of oliguria is considered a favorable return of renal perfusion.^† Animal experiments have suggested that early circulatory optimization also ameliorates inflammatory changes after trauma/hemorrhage. In attempts to restore oxygen delivery, hypoxemia should be prevented or corrected, although recent evidence suggests that hypoxic resuscitation ameliorates proinflammatory responses to reperfusion as compared to normoxic resuscitation.²³⁴ The arterial oxygen saturation, important for the oxygen content of delivered blood, usually exceeds 90% if the PO₂ is greater than 60 mm Hg. If arterial PO₂ is less than 60 mm Hg, supplemental oxygen can be given through a nasal cannula or mask, but hyperoxemia should be avoided. The patient should be intubated and artificially ventilated in case of impending respiratory insufficiency from whatever cause. Prophylactic positive end-expiratory pressure in attempts to prevent the development of ARDS in at-risk patients, including patients with traumatic/hemorrhagic shock, is useless.

For the initial stabilization of trauma patients in shock with severe abdominal trauma or bleeding from lower extremities, passive leg raising or Trendelenburg positioning has been used, but a large benefit on preload-augmented tissue oxygen delivery has been doubted.8,19,20,287 The pneumatic antishock garment can be applied as a temporary, immediately lifesaving procedure to (1) stop the hemorrhage and splint pelvic and lower extremity fractures for transportation of the patient after inflation, (2) mobilize blood volume by exerting external pressure on the leg and abdomen, and (3) redirect flow toward vital organs such as the heart and brain.^288–290 Use of the garment and hypertonic saline may act synergistically.²⁸⁸ The disadvantages of the procedure include aggravation of pressure-dependent and uncontrolled bleeding and ischemia of intra-abdominal organs. Deflation should be gradual to prevent lethal hypotension after return into the circulation of vasoactive substances and lactic acid from ischemic tissue.^288,289 The use of the pneumatic antishock garment has greatly declined in recent decades. Spontaneous breathing against an inspiratory threshold or decreasing expiratory pressure during mechanical ventilation may increase venous return and has been studied experimentally.²⁹¹ Measures to establish a way to the bloodstream for infusing fluids include insertion of a large-bore catheter in a peripheral vein or, if cannulation of peripheral veins seems impossible because of collapse, a catheter in a central (i.e., jugular/subclavian) vein. The latter catheter also allows monitoring of CVP. The intraosseous route for administration of hypertonic fluids is particularly useful in traumatized children with hypovolemic shock.²⁹² To this end, a marrow screw is inserted in the sternum or tibia. Fluids can be given by gravity or pressure.

Further treatment of hypovolemic shock depends on the underlying cause. Immediate surgery is warranted in case of extensive trauma and ongoing internal or external blood loss. Massive intra-abdominal bleeding after blunt or penetrating injury may warrant attempts for control by drug therapy with vasopressin, aortic clamping, or preferably, coiling during a radiologic examination prior to laparotomy for insufficient control. Gastrointestinal hemorrhage can be stopped by conservative treatment, including histamine-2 receptor blockade and endoscopic electrocoagulation or laser coagulation of bleeding peptic ulcers, or coiling of bleeding vessel during angiography. Bleeding varices can be treated by continuous infusion of vasopressin or somatostatin, a Sengstaken-Blakemore balloon tube, endoscopic injection sclerotherapy or banding, or transjugular intrahepatic portosystemic shunt. Surgery is rarely necessary in the institutional presence of experienced intervention radiologists. Apart from fluids, uncontrolled diabetes mellitus necessitates continuous intravenous infusion of insulin (about 6 U/hour). Acute adrenocortical insufficiency warrants administration of steroids (hydrocortisone 100 mg two to four times daily).

Resuscitation Strategies

The speed with which shock can be reversed depends on the delay from onset to treatment and the severity of shock, as estimated from the clinical condition and hemodynamic status. The speed of fluid infusion can be guided by the clinical condition of the patient, heart rate and blood pressure, diuresis, and determinations every 2 hours of the arterial blood lactate level and acid-base balance.³¹ Repeated assessment of jugular venous pressure, auscultation of the lungs, and arterial blood gases are indicated to prevent overhydration and pulmonary edema. If a central venous or pulmonary artery catheter is in place, a fluid challenge protocol can be used (Table 26.2).³¹ Monitoring of central venous or PCWP or preload volumes allows for rapid infusion of solutions and evaluation of the response of oxygen delivery and uptake to resuscitation.^9,25,31,73 When in doubt, predicting fluid responsiveness in the patient, for instance with cardiac failure on mechanical ventilation, by passive leg raising and other tests that reversibly challenge the circulation can be helpful to further guide fluid therapy and avoid harmful fluid overloading.^19,20

The usefulness of immediate and vigorous resuscitation in the course of uncontrolled bleeding (e.g., after penetrating vascular trauma) has been challenged in recent years.279,293–295 Studies have shown that infusion of substantial amounts of nonsanguineous fluids and increasing pressure-dependent bleeding lead to dilution of blood components, coagulation disturbances, and increased mortality risk unless the bleeding is controlled before resuscitation.^{279,290,294,295} The clinical implication is that resuscitation perhaps should not take place, at least not vigorously, at the scene of the accident when bleeding cannot be controlled and transport to the hospital where the bleeding can be controlled is practicable. This may apply not only to trauma patients but also to patients with a ruptured abdominal aneurysm. Authors have proposed slow (low volume) rather than rapid (early and large volume) infusion rates and controlled hypotensive resuscitation (e.g., with help of vasodilators) at least as long as bleeding is uncontrolled.^{290,293,294,296–298} The value of this type of resuscitation in humans with hyperoncotic/hypertonic solutions, including acetate with vasodilating and buffering properties, is debatable.^210,279,298 Also, the levels of hypotension that are safe remain unclear.²⁹⁹ Conversely, early vasopressor therapy by vasopressin, for instance, may benefit patient outcomes more than large-volume infusions.³⁰⁰ The treatment of massive bleeding by blood products requires an institutional protocol. Closed-loop control of fluid therapy may become clinically applicable in the near future.³⁰¹

Fluids

During hypovolemic shock, the repletion of intravascular volume is of primary importance to restore cardiac output and oxygen transport to the tissues before repletion of the interstitial and intracellular fluids.* Available fluids are described in Tables 26.3 to 26.5. Electrolyte solutions (the crystalloid fluids) are shown in Table 26.3, and high-molecular-weight solutions (the colloid fluids) are shown in Tables 26.4 and 26.5. In the first group, the hypertonic fluids (i.e., fluids with a higher osmolarity than plasma) are presented. Among the colloid solutions are solutions with a higher colloid osmotic pressure than plasma—the hyperoncotic solutions. Hypertonic and hyperoncotic solutions are plasma expanders because they are able to mobilize cellular and interstitial fluid during resuscitation and to expand plasma volume rapidly.

Table 26.6 shows how the volume of the various compartments is replenished during fluid resuscitation. Because infusion of glucose 5% hardly increases intravascular volume, this solution has no place in resuscitation from hypovolemic shock. Lactated Ringer’s solution containing K⁺ should not be used during renal insufficiency because of the danger of inducing hyperkalemia. Infusion of lactate-containing solutions during lactic acidosis may be controversial because the capacity of the liver to regenerate bicarbonate from lactate may be impaired.³³ Nevertheless, infusion of lactate-containing solutions such as lactated Ringer’s for resuscitation from hypovolemic shock generally does not substantially increase lactate levels, worsen acidosis, or adversely affect outcome.³⁰⁵ However, racemic Ringer’s lactate may be proinflammatory and proapoptotic as compared to normal saline and only L-lactate containing solutions are therefore currently recommended.¹⁹¹ Adverse effects of overzealous fluid administration of any type include promotion of pulmonary edema and ascites formation with subsequent aggravation of abdominal compartment syndrome.^285,286

The natural colloids consist of albumin and plasma solutions. Albumin is an effective colloidal solution for intravascular volume repletion, and its intravascular half-life is approximately 16 hours.* For the artificial colloids, the volume-expanding effect and the duration of action generally increase with increasing in vivo molecular weight (Table 26.7). About one third of the hyperoncotic fluids with high molecular weight (dextran 70 and hydroxyethylstarch) may still be in the circulation after 24 hours, whereas dextran 40 is retained for approximately 3 hours. Dextrans and gelatins are perhaps equally effective in restoring circulating volume.^309,310 The latter substances might increase tissue blood flow in the microcirculation. Dextrans are currently less often used than gelatins and starches and institutional and geographic habits largely determine the choice among fluids.

The starch compounds have gained wide interest for the resuscitation of hypovolemic shock.* These colloids are at least as effective as albumin but less expensive.^{135,136,172,308,312} Because the molecular range of the starch compounds varies enormously, the pharmacokinetics are complex.³¹¹ Nevertheless, about 40% of the infused hetastarch (molecular weight >200) still remains in the circulation after 24 hours because 30% of the infused substance may have a half-life of 67 hours. Ninety percent of smaller starch is cleared in 24 hours. The duration of the volume-expanding effect of the starches depends not only on molecular range and concentration but also on the so-called substitution grade, which is the number of hydroxyethyl groups per glucose unit, and the substitution type, the ratio of C2 to C6 hydroxyethylation.³¹¹ A high molecular weight, high substitution grade, and high C2/C6 ratio retard breakdown by plasma amylase and prolong intravascular retention. The residual starch compounds are partly excreted by urine and partly taken up by the reticuloendothelial system. Accumulation may also occur in dendritic cells of the skin and the liver and in the renal tubules, with subsequent adverse effects. Starch compounds may increase the amylase level in blood and may confound the diagnosis of acute pancreatitis. Experimental evidence shows that some starch compounds, particularly in the 100,000 to 300,000 D molecular weight range, have the advantage in sealing the capillary endothelium in case of increased permeability after ischemia or trauma, diminishing fluid and protein filtration and preventing edema.^{136,179,311,312}

In the resuscitation of hypovolemic shock (e.g., after trauma or burns), the use of hypertonic solutions with sodium concentrations greater than 0.9% also has gained wide interest.* The solutions essentially consist of hypertonic sodium chloride, to which colloids have often been added. The combinations include NaCl 7.5% with dextran 70 (6%/10%), NaCl 7.2% with dextran 60 (10%), and NaCl 7.5% with hydroxyethylstarch 6%.^{33,62,122,314–317} Hypertonic solutions usually result, at a much lower infusion volume than isotonic solutions (small volume resuscitation), in a rapid hemodynamic improvement, that is, an increase in cardiac output and in oxygen delivery and uptake and arterial blood pressure in experimental animals and patients with traumatic/hypovolemic shock.^† Infusion of rapidly acting hypertonic saline, particularly if combined with hyperoncotic colloids, increases survival in bleeding animals compared with infusion of either component or other isotonic or hypertonic (nonelectrolyte) solutions.³¹⁶ Clinical trials also have shown some value of hypertonic solutions in the initial treatment of hypovolemic shock after burns and trauma with uncontrolled bleeding.^{119,286,314,316,317} The use of hypertonic saline solutions, however, warrants close monitoring of plasma sodium levels to prevent excessive hypernatremia and hyperosmolarity.^313–316

The hypertonic fluids primarily act through resorption of interstitial and cellular fluid volume and expansion of the plasma volume.7,131,314 It has been calculated that only 4 mL/kg 7.5% saline solution can increase circulating plasma volume by 8 to 12 mL/kg body weight. A hyperosmolarity-induced increase in cardiac contractility may also contribute to the increase in cardiac output, although this effect has been doubted.³¹⁸ Other potential mechanisms include activation of pituitary and pulmonary osmoreceptors, leading to release of vasopressin and vagal afferent-mediated venoconstriction, and hyperosmolarity-induced arterial vasodilation.^7,62,313,314 Infusion of hypertonic sodium combined with hyperoncotic colloid solutions more rapidly and completely increases cardiac output and arterial blood pressure, and the effects last longer than those produced with infusion of hypertonic or colloid solutions alone.^313–317 During infusion of hypertonic saline, particularly if combined with hyperoncotic colloid solutions, the distribution of peripheral oxygen delivery is reversed to a more favorable pattern with preferential perfusion of vital organs, including gut and kidney.^62,313 The increase in oxygen uptake in bled dogs was less rapid and complete during resuscitation with hypertonic saline plus hydroxyethylstarch, however, than during infusion of relatively large volumes of the latter.^33,62

The hypertonic solutions may also ameliorate immunodepression, the translocation of bacteria, and susceptibility to sepsis after hypovolemic shock in rodents.201,319 Hypertonic solutions (plus dextrans) restore capillary blood flow and organ function during resuscitation better than iso-osmotic fluids because of their ability, among others, to reduce endothelial cell swelling, adhesion molecule expression, neutrophil activation and adherence, and cellular apoptosis compared with normotonic crystalloids such as racemic Ringer’s lactate.^191,320 Hypertonic solutions may also prevent lung injury after hemorrhage/resuscitation, probably via these mechanisms.^{95,119,122,319}

Fluid Controversies

The choice between available fluids should be guided by the estimated extent and type of fluid losses; their composition and localization; and the properties of infusion fluids, their distribution over body compartments, and, perhaps, the associated costs, which are high for albumin, intermediate for artificial colloids, and low for crystalloid solutions.131,302–304 Nevertheless, the use of various solutions for resuscitation from hypovolemic shock is hotly debated, partly because the importance of the colloid osmotic pressure for resuscitation and prevention of pulmonary edema is uncertain.³⁰² Also, the relative merits and detriments (i.e., safety) of natural and artificial colloids remain unclear.^302,321

Capillary filtration depends on the pericapillary hydrostatic and the colloid osmotic pressure gradient, according to the Starling equation. If at a given permeability an imbalance in pressures augments capillary filtration of fluids, a decrease in interstitial colloid osmotic pressure, an increase in interstitial hydrostatic pressure (which also depends on the compliance of the interstitium), and increased lymph flow can either alone or in combination partially prevent gross accumulation of interstitial fluid (edema). The colloid osmotic pressure of plasma is primarily determined by the plasma albumin content and normally measures about 24 mm Hg.31,126,137,322 The pressure can be estimated from albumin and protein concentrations in plasma, but infusion of artificial colloid solutions invalidates this calculation, so proper assessment of plasma colloid osmotic pressure necessitates direct measurement.^137,323 Because of a decrease in circulating plasma protein levels, hypovolemic shock results in a decrease in plasma colloid osmotic pressure.* During hypoproteinemia and a reduced plasma colloid osmotic pressure, fluid filtration for a given hydrostatic pressure increases until the pericapillary colloid osmotic pressure gradient decreases and a new steady state, often at increased lymph flow, has been achieved.^133,135,136 Evoking safety mechanisms such as a reduced interstitial colloid osmotic pressure and increased lymph flow may keep the interstitium relatively dry, and these mechanisms may be more effective in the lung than in the systemic circulation.¹³³ During hypoproteinemia, the hydrostatic pressure needed to invoke pulmonary edema decreases, however, because of more rapid exhaustion of safety mechanisms.^135,324 Conversely, increased lung water caused by an elevated hydrostatic pressure can be ameliorated by colloid infusion.³²⁴

For a given increase in hydrostatic pressure, the infusion of crystalloids decreases the plasma colloid osmotic pressure and tends to enhance, if insufficiently compensated by a decrease in the pericapillary colloid osmotic pressure gradient, pulmonary and systemic fluid filtration and interstitial fluid expansion more than infusion of albumin/colloids, which maintain plasma colloid osmotic pressure.* Crystalloid solutions replenish not only the intravascular but also the interstitial space by increased filtration, whereas colloid fluids tend to primarily fill the former compartment, at least initially.^131,278 Widening of the intravascular-to-interstitial colloid osmotic pressure gradient may prevent increased fluid filtration, but the effect may be transient when some colloids have been filtered along with fluids into the interstitium and a new steady state of perimicrovascular pressure and draining lymph flow has been established.¹³⁶ The mechanism may form the basis for the well-known observation that colloid solutions yield a twofold to threefold greater expansion of the intravascular space than crystalloids and that the latter have greater tendency for edema formation for a given amount of fluid infused, so less colloid than crystalloid is probably needed for resuscitation to similar hemodynamic end points in hypovolemic shock.^† In some clinical trials, colloids proved to be superior to crystalloids in resuscitation from hypovolemic shock,^137,172,309 in terms of both the speed and the extent of correction of the hemodynamic abnormalities. Conversely, this may also explain the observations of some investigators that during resuscitation from hypovolemic shock pulmonary edema can be prevented in part if the intravascular filtration pressure (i.e., the gradient between plasma colloid osmotic and PCWP) is kept greater than approximately 6 mm Hg, with an elevated risk for pulmonary edema, particularly in case of increased permeability, if the gradient is less than approximately 3 mm Hg, and that resuscitation with colloids less often induces evidence for pulmonary edema than infusion of crystalloids during hypovolemic shock.^‡

If the permeability for proteins increases and the reflection coefficient decreases, the hydraulic conductance of the capillary membrane also increases.¹³³ Increased permeability for proteins increases capillary fluid filtration for a given intravascular hydrostatic pressure and promotes the formation of edema.³⁰⁸ During increased permeability, the filtration of fluids and expansion of the interstitial fluid space depend more than normally on hydrostatic pressures and less on colloid osmotic pressures because the colloid osmotic pressure gradient is decreased.¹³³ The differences between the types of solutions in fluid filtration and formation of edema in the lung and peripheral tissues diminish.³²⁶ This may explain in part why some clinical studies did not find a predictive value of the colloid osmotic pressure-PCWP gradient for pulmonary edema and lack of a difference between fluid types for formation of pulmonary edema and impaired gas exchange during resuscitation from hypovolemic shock.^302,307 Moreover, an increase in CVP increases the back-pressure for lymph flow. Careful animal studies on hypovolemic shock combined with a lung vascular injury showed, however, that colloids are more effective than crystalloids in restoring the circulation and that the former increased lung water less than the latter unless permeability was severely increased.^308,326 This can be explained by the fact that even in case of increased permeability the reflection coefficient is not zero, and that the pericapillary colloid pressure gradient still exerts some influence on the transcapillary movement of fluids.

Clinical studies on the colloid/crystalloid controversy may be difficult to interpret because of differences in patient populations and end points between fluid types.302,321 Lack of similar end points used for resuscitation may partly explain why infusion of colloid solutions increased the risk for pulmonary failure compared with infusion of crystalloids because colloids, owing to their greater intravascular volume-repleting effect, tend to increase hydrostatic filtration pressure in the lung more rapidly than crystalloid fluids even though colloid osmotic pressure is maintained or increases during infusion of the former and decreases with the latter.³⁰⁶ The importance of a difference in hydrostatic pressure for the risk of pulmonary edema would be accentuated in case of increased permeability.³⁰⁶ Finally, pulmonary mechanics, gas exchange, and radiographic changes used to evaluate the effects of fluid infusions in many studies may not accurately reflect changes in lung water.^{52,172,306–308}

There are safety concerns with artificial colloids, particularly in the presence of other risk factors for organ damage.304,321 Potential disadvantages of (artificial) colloid over crystalloid solutions include inhibition of the coagulation system; the risk for anaphylactoid reactions; inhibition of renal salt and water excretion; renal injury; and perhaps, at least for albumin, depression of myocardial function, possibly owing to binding of Ca²⁺, although this has not been seen in all studies.^{171,172,321,327} Of all artificial colloids, dextrans affect coagulation most adversely, independently of hemodilution, by interfering with coagulation factors and diminishing thrombocyte and red blood cell aggregation.³²⁷ The gelatins may also have some intrinsic effects on coagulation, whereas the anticoagulant effects of hydroxyethyl starches probably relate, in addition to hemodilution, to less endothelial release of von Willebrand factor.^290,311,327 Anaphylactoid reactions to artificial colloids are extremely rare and vary from slight fever and skin reactions to life-threatening anaphylactic shock. Starch compounds elicit these reactions less often than dextrans and gelatins.³¹¹ Large-molecular-weight starches may accumulate in subcutaneous tissues and may cause pruritus, even for weeks after administration.³¹¹ However, resuscitation of trauma patients with hydroxyxethyl starch may result in less endothelial damage, renal injury, and pulmonary dysfunction than resuscitation with gelatins.¹⁷⁹ Nevertheless, the renal damaging effect of starch is probably greater than that of gelatins in patients with prior risk factors for acute kidney injury.³²¹ Hence, colloids may contribute to the development of acute kidney injury and failure, particularly in the case of overadministration, which may otherwise be more frequent with artificial colloids than with albumin/plasma solutions, which can be monitored by measurements of plasma albumin concentrations. As opposed to albumin/plasma, infusion of colloid solutions is often bound to a maximum (see Table 26.5). Side effects may be more frequent with artificial colloids than with albumin/plasma infusion even though the latter carries a very low risk of anaphylactoid reactions and disease transmission. Crystalloid and artificial colloid solutions may activate neutrophil-endothelial interactions and depress macrophage and immune functions more than albumin does.^{191,321,328,329} In contrast to lactated Ringer’s solution, Ringer’s ethyl pyruvate solution has favorable anti-inflammatory and cell-protecting actions.^29,37

If used in the resuscitation from hypovolemic shock, artificial colloids may still be preferred over natural colloids because the latter are more expensive and less available, even though albumin and plasma solutions are effective volume expanders. There is some evidence that the type of fluids infused during hypovolemia may influence the extent and speed with which oxygen uptake is restored: infusion of colloid (plasma/albumin) solutions in hypovolemic postoperative and trauma patients may increase uptake of oxygen for a given increment in plasma volume and oxygen delivery more rapidly than infusion of crystalloids and may thereby improve outcomes.9,25,330 This is thought to result in part from increased diffusion distances for oxygen in the tissues subsequent to tissue edema, evoked by massive crystalloid infusion.^9,25,307 Administration of nonbuffered (unbalanced) crystalloid solutions such as normal saline carries the risk of hyperchloremic metabolic acidosis, which can be avoided in part by infusion of buffered (balanced) solutions such as Ringer’s lactate.³²⁵ Nevertheless, meta-analyses suggest, at least in some groups, a slightly increased mortality risk after resuscitation with artificial colloids.^{302,321,331,332} In the SAFE study comparing albumin and saline for resuscitation in the intensive care unit, a slight but nonsignificant increase in mortality rate was observed in the (neuro)trauma subgroup treated by albumin infusions, although animal studies suggest some beneficial and anti-inflammatory effects of albumin resuscitation from hemorrhagic shock.³³³

Blood Products and Substitutes

Resuscitation with blood components may restore tissue oxygen delivery and energy metabolism more rapidly, completely, and persistently than resuscitation with crystalloid during hemorrhage and hypovolemic shock, although this is controversial.162,305 Nevertheless, it has been suggested that infusion of sodium salts may be essential, and that addition of saline to blood improves survival from hypovolemic shock after hemorrhage because of correction of both the intravascular and the interstitial volume deficits.³¹³

In the treatment of hypovolemic shock following ongoing hemorrhage, infusion of red blood cells in the form of erythrocyte concentrates, or packed red blood cells, remains crucial.277,286,334 This is achieved by autotransfusion from uncontaminated areas during surgery, if possible; by infusion of blood group O Rh-negative donor blood in emergency situations; or by infusion of typed and stored/anticoagulated donor blood. The position of the oxyhemoglobin dissociation curve of old, stored blood is shifted to the left.^60,78 Although this theoretically may impair the delivery and uptake of oxygen in the tissues, the effects of these changes in animal experiments are usually limited and clinical repercussions are unclear.^60,76 Transfusion of substantial amounts of erythrocyte concentrates is preferentially accomplished through a microfilter to avoid alloimmunization and infusion of neutrophils and other cellular aggregates, which develop in time during storage of blood and which may lodge in the lung, promote pulmonary injury, and impair gas exchange, leading to TRALI.^175–177 Today, prior leukocyte-reduced red blood cell concentrates are often used, but it is controversial whether this is associated with less risk. Excluding multiparous women who may have become sensitized to allogenic leukocytes and may carry leukocyte antibodies contributing to TRALI is another strategy applied in some countries. Also, humoral mediators released in stored blood or during infusion might be responsible in part for pulmonary vascular injury after massive transfusion of blood.^180,190 Nevertheless, massive transfusion may remain a risk factor for bacterial sepsis, ARDS, and MOF, independently from bleeding and severity of hypovolemic shock.* Finally, transfusion of blood components and plasma carries a small risk of transmitting infectious diseases and depressing immune function.^305,329

Because loss of blood also leads to loss of coagulation factors and platelets, and blood concentrations are diluted further during nonsanguineous fluid resuscitation, replenishing plasma levels by infusion of fresh frozen plasma and platelets is usually required to help stop ongoing bleeding.²⁶² Fresh frozen plasma should not be used solely for the treatment of hypovolemia, even though plasma may diminish endothelial hyperpermeability through restoration of glycocalyx.^262,271 The strategy of blood products infusion has undergone some changes in the last decade, in which studies suggest optimal hemostasis and outcome when fresh frozen plasma units are infused at a 1 : 1 ratio with packed red blood cell concentrates and random donor platelet units at a 1 : 3 to 1 : 5 ratio.^334,337,338 If, during resuscitation, the clotting times are prolonged by a factor of 1.5 or more, more fresh frozen plasma can be given, and if platelet counts decrease to less than 50 to 100 ×10⁹/L, platelets can be transfused, particularly in case of intracranial or life-threatening bleeding. The value of prophylactic administration of coagulation factors in a polytransfused, traumatized patient to prevent further bleeding after initial hemostasis is unclear, however. Supplementation of Ca²⁺ may be necessary only if more than 12 to 20 units of packed red blood cells, anticoagulated with Ca²⁺-binding citrate, have been given if rapidly transfused and particularly if liver function is impaired.¹⁵⁰ Further treatment is guided by ionized Ca²⁺ determinations in plasma. Fibrinogen concentrates are increasingly used with increasing evidence that fibrinogen plays an important role in coagulation and that primary hyperfibrinolysis is common in trauma patients, as revealed by thrombelastometry.^334,337 Antifibrinolytic drugs, such as tranexaminic acid, given prior to fibrinogen concentrates are useful adjuncts.³³⁹ The exact place of recombinant factor VIIa, a potent procoagulant, in the treatment of refractory bleeding has not been settled yet.^286,340 The factor stops bleeding and saves blood transfusion, but cost-effectiveness is unclear. Adverse effects include a tendency for thromboembolic events.³⁴⁰ The bleeding tendency of trauma is also aggravated by hypothermia and acidosis, but it is unclear whether aggressive treatment of hypothermia or acidosis substantially ameliorates coagulation disturbances and to what extent DIC contributes.^181,264,337 The importance of the latter also remains somewhat unclear, and treatment may consist of infusion of antithrombin III concentrates or, in case of severe bleeding, of fresh frozen plasma and platelets.¹⁸¹

To overcome some of the problems associated with donor or autologous red blood cell transfusions, investigators have intensively searched for safe and effective hemoglobin substitutes applicable in humans.230,286,341–343 These substitutes include chemical oxygen carriers, hemoglobin modifications, and liposome/vesicle-encapsulated hemoglobin.³⁴⁴ The chemical hemoglobin modifications have been designed to prevent or limit the renal toxicity of free hemoglobin. They include polymerized, modified, cross-linked, and recombinant hemoglobins.^34,345 The use of hemoglobin substitutes has been under clinical investigation. Some nonrecombinant substitutes seemed to increase arterial and, particularly, pulmonary arterial blood pressure more than accounted for by fluid loading, but some compounds have more adverse effects than others.³⁴⁶ This may relate to the property of hemoglobin to scavenge NO or release endothelin and platelet-activating factor, or combinations, and the use of these solutions is therefore still not without hazard.³⁴⁶ Diaspirin cross-linked hemoglobin may beneficially influence intracranial hemodynamics during resuscitation from hypovolemic shock.²⁹⁶ A clinical trial on diaspirin-cross-linked hemoglobin in trauma failed to improve survival over resuscitation with saline, however.³⁴¹ The use of hemoglobin substitutes such as perfluorocarbons has not yet reached the stage of widespread, routine clinical practice, although they may effectively carry oxygen in humans and may improve resuscitability from hypovolemic shock following bleeding in animals compared with nonhemoglobin-based solutions.^{341–343,347} Further research is ongoing.

Acidosis and Optimal Hematocrit

The underlying idea for partial correction of metabolic acidosis is that acidosis is detrimental for, among others, myocardial function by increasing pulmonary artery pressure and right ventricular afterload, impairing catecholamine sensitivity, and diminishing adrenergic receptors and intracellular Ca²⁺ transport necessary for contraction, even if masked by increased sympathetic activity.36,168 Metabolic acidosis may increase the tendency for life-threatening ventricular arrhythmias and may lessen defibrillation thresholds and vascular tone.²⁷ The need for treatment of metabolic (lactic) acidosis (e.g., by intravenous administration of buffer solutions) remains unclear.* Administration of sodium bicarbonate may carry the risk for aggravation of intracellular acidosis in the tissues because bicarbonate releases CO₂ during buffering and CO₂ more rapidly traverses the cell membrane than the bicarbonate ion.* Alkali therapy with sodium bicarbonate carries the risks of shifting the oxyhemoglobin dissociation curve to the left and impairing tissue oxygenation, a decrease in ionized Ca²⁺, and causing hypernatremia and osmolarity, although the consequences of these theoretical drawbacks are unclear.^27,36,44 Albeit not beyond doubt, experimental and clinical studies suggest that the administration of buffers such as sodium bicarbonate is not harmful, even though the hemodynamic and metabolic effects of the solution may not surpass those obtained by saline infusion.^† In many institutions, small doses of alkali buffers such as sodium bicarbonate (50 to 100 mL of a 4.2%, 0.5 mmol/mL solution) are still given to treat metabolic (lactic) acidosis if arterial pH is less than 7.2 and acidosis persists despite optimal cardiovascular resuscitation.^27,36 During sodium bicarbonate infusion, the patient should be hyperventilated to prevent hypercapnia in arterial blood and bicarbonate doses should be guided by the arterial blood acid-base status to prevent alkalosis and diminished oxygen release after overadministration.^27,44,348 Prevention of hypercapnia may obviate increased CO₂ diffusion and aggravation of intracellular acidosis.^27,36 The value of buffers, including bicarbonate/carbonate, that do not generate CO₂ and prevent aggravation of intracellular acidosis is still controversial.^{27,35,348,349} Dichloroacetate is a stimulator of pyruvate dehydrogenase, and the drug may ameliorate lactate accumulation and postresuscitation organ dysfunction, but there is probably no benefit for patient outcome.^27,35,36

The hematocrit is the main determinant of blood viscosity, and the latter determines, together with the geometric features of the vascular bed, the blood flow in the microvasculature.¹²⁰ Experimental studies suggest that during normovolemic hemodilution normal oxygen delivery is achieved at a range of hematocrit values from 12% to 65% for the heart; 30% to 65% for the brain; 30% to 55% for liver, intestine, and kidney; and 30% to 60% for the whole body because adaptations in vessel diameter and changes in blood flow in this hematocrit range are able to compensate for changes in oxygen content, maintaining a normal oxygen delivery.^{85,120,265,350} The optimal hematocrit for the whole body may not conform to the regional optimal hematocrit.

Because blood is a non-newtonian fluid, so that blood viscosity depends not only on hematocrit but also on blood flow velocity (shear stress), there may be differences along the vascular profile in blood viscosity, with propensity for red blood cell aggregation in postcapillary venules, where flow velocity is lower than in arterioles, particularly during hypovolemic shock.4,120 Increased blood viscosity in postcapillary venules may contribute to impaired tissue perfusion during hypovolemic shock.⁴ Conversely, the volume status, myocardial function, and vascular tone contribute to blood viscosity so that, for example, the optimal hematocrit for oxygen delivery is lower during hypovolemia than hypervolemia.^21,120,350 Finally, red blood cell deformability is decreased in hemorrhagic shock, contributing to increased viscosity.²²⁴

Taken together, it is hard to generally define the optimal hematocrit for oxygen delivery to the body during hypovolemic shock, even though hematocrit-induced changes in the rheologic properties of blood may contribute to hemodynamic changes in critically ill patients.21,120,350,351 Most, but not all, authors believe, however, that mild hemodilution (hematocrit approximately 0.30) may benefit delivery and uptake of oxygen in the tissues and promote survival of critically ill patients with hypovolemia, whereas severe hemodilution or hemoconcentration may be detrimental.* A low hematocrit in the course of hypovolemic shock after major surgery may warrant red blood cell replacement, whereas a high hematocrit may necessitate infusion of nonsanguineous fluids.^9,352 Mild hemodilution may benefit resumption of red blood cell flow and oxygen uptake after prior ischemia.

Vasoactive Drugs

Generally, catecholamines do not have a place in the treatment of hypovolemic shock unless they are used to bridge a period in which infusion fluids are not yet available, or if adequate fluid resuscitation has proved insufficient to reverse hypotension (irreversible shock) and to increase oxygen delivery to the point that tissue needs are met.^† Persistent hypotension despite normovolemia can be caused by a low cardiac output following myocardial dysfunction or by peripheral vasodilation. Data obtained with advanced hemodynamic monitoring may help to identify these abnormalities, which can be important for choosing among the available vasopressor and inotropic drugs, which have widely differing receptor affinities and hemodynamic effects.^9,45

Treatment with the drugs is best guided by the prevailing hemodynamic profile and aims at optimization of the circulation toward values associated with survival.9,42,45 Drugs are given as a continuous intravenous infusion, preferably via a central vein. The initial dose is low, and often combinations of drugs are used. The use of catecholamines should be judicious and carefully guided by hemodynamic parameters to reach predefined hemodynamic goals.⁹ β-Adrenergic drugs increase cardiac output by inotropic (β₁) or vasodilating (β₂) properties.^9,353 Dopaminergic compounds may preferentially increase splanchnic and renal perfusion, glomerular filtration, and diuresis.^66,284 Dobutamine, having vasodilating β₂ properties, may exert greater effects on delivery and uptake of oxygen than dopamine at a lower PCWP.^9,353 A decrease in the arterial blood pressure concomitantly with a decreased wedge pressure after dobutamine infusion may warrant additional fluid repletion.⁹ Drugs with α-adrenergic activity, such as norepinephrine, increase arterial blood pressure, but this increase may not lead to a decrease in cardiac output because they may increase venous return to the heart by decreasing venous compliance.⁷ The vascular reactivity to vasoconstrictors may diminish in the late phase of shock, but norepinephrine remains the agent of first choice in the (bridging to definitive) treatment of hypovolemic shock after fluid loading.³⁵⁴ The use of adrenergic drugs is not without hazards. They may enhance the metabolic demands of the body so that the oxygen supply-to-demand ratio is not favorably influenced even if oxygen delivery is enhanced.³⁵⁵ Particularly, epinephrine may increase lactic acid levels independently of oxygen balance.³⁸ Low-dose dopamine has been shown, at least in bled dogs, to impede oxygen extraction by the gut during a decrease in oxygen supply, probably associated with transmural distribution of blood flow.⁶⁶ Finally, vasoconstricting vasopressin and methylene blue, a guanylate cyclase inhibitor, have been tried to overcome intractable hypotension in this phase.^96,300 Vasopressin has also been used in the initial management of uncontrolled hemorrhagic shock to safeguard arterial pressure for vital organ (e.g., cerebral) perfusion without overzealous fluid administration that may dilute coagulation factors and promote further bleeding.³⁰⁰

Brain Injury and Resuscitation

Hypovolemia and a decreased mean arterial blood pressure are considered as major threats for cerebral perfusion in brain injury. The latter may create intracranial hypertension following edema, bleeding, and contusion so that perfusion is more dependent on pressure than normal. Small volume resuscitation from hypovolemia with hypertonic (and hyperoncotic) solutions could increase mean arterial blood pressure at a small increase in plasma volume and could, by virtue of hypertonicity, decrease cerebral edema and intracranial pressure. The solutions are highly suitable for treatment of multiple trauma that includes the brain.314,315 Conversely, too much normotonic, certainly hypotonic, and perhaps albumin solutions may aggravate cerebral edema, but too little fluids and under-resuscitation with resulting hypotension and hypoperfusion may do the same. Vasopressor drugs such as vasopressin may be useful adjuncts in the initial treatment of hemorrhagic hypotension plus brain injury.

In Practice

In practice, the different types of fluids, including isotonic and hypertonic crystalloid and iso-oncotic or hyperoncotic colloid solutions, are often combined in the resuscitation from hypovolemic shock (see Box 26.2). For resuscitation of hypovolemic shock following hemorrhage, typed blood is often not immediately available, even though a blood sample for crossmatching has been sent to the blood bank as soon as possible after admission. If shock is severe and warrants immediate infusion of blood, type O Rh-negative erythrocyte concentrates can be safely used. In the absence of blood, resuscitation should begin with nonsanguineous fluids. During hypovolemic shock, initial resuscitation is often begun with hypertonic or isotonic (balanced) crystalloids, supplemented with colloid solutions, and finally accomplished through infusion of erythrocyte concentrates and plasma (Fig. 26.2).

In the case of uncontrolled diabetes mellitus, profound diarrhea, and acute adrenocortical insufficiency with loss of plasma water and electrolytes, the infusion of crystalloid solutions usually suffices. These solutions restore intravascular, interstitial, and intracellular (in case of diabetes mellitus) fluids. Changes in the electrolyte concentrations in blood have to be corrected through adaptation of the type and composition of the infusion fluid; in the case of hypokalemia (and in the presence of diuresis), potassium should be supplemented.

Supportive Care

A vomiting patient in hypovolemic shock should be protected against aspiration of gastric contents by early intubation. The value of specific measures for prevention of hemorrhagic gastric mucosal stress ulceration remains controversial.²⁰⁷ After resuscitation from shock, attention also should be paid to the nutritional status of the patient.²²² It should be judged whether enteral or parenteral nutrition is necessary to improve nitrogen balance and energy intake.³⁵⁶ Although enteral feeding during hypovolemic shock and after resuscitation may increase metabolic demands of the gut, luminal application of nutrients such as glutamine may induce an increase in mucosal blood flow, ameliorate damage, and diminish the likelihood for translocation of endotoxins and bacteria and of septic complications.^148,206 In addition, there is some evidence that early enteral feeding favorably influences organ function after hemorrhage and reperfusion, in contrast to (early) parenteral feeding, which may have adverse effects even though earlier meeting caloric requirements.¹⁴⁸ The value of selective decontamination of the digestive tract or luminal absorption of endotoxin to prevent sepsis and its harmful sequelae originating from the gut is still controversial in multiple trauma, although such measures may inhibit the cytokine response to hypovolemic shock in animals.³⁵⁷

When treating pain in a patient with extensive trauma, morphinomimetics are cautiously applied because the drugs may have adverse circulatory effects during hypovolemic shock and half-life may be prolonged.11,358 Because many resuscitated patients after trauma or hemorrhage exhibit hypothermia, partly caused by exhausted energy reserves and infusion of substantial amounts of room temperature infusion fluids, and because hypothermia may denote more severe illness, rewarming infusion fluids may be necessary during resuscitation, and this may prevent some organ dysfunction and perhaps promote survival despite the increase in oxygen demand with an elevation in body temperature.^56,217,264 In contrast, there also may exist some protective effect of mild hypothermia during bleeding and resuscitation, particularly when accompanied by brain injury, at least in experiments.

Miscellaneous Therapies

A wide array of experimental drugs has been tried in animal experiments to improve the hemodynamics, ameliorate inflammation, and increase survival rates from hypovolemic shock and resuscitation.²³⁰ Although many experimental drugs have shown some benefit in animal models of hypovolemic shock, in terms of hemodynamics during shock and after resuscitation and ultimate survival, there are now clinical trials ongoing or showing a benefit of such interventions in humans, including treatment with immune-enhancing factors.^{24,93,118,230,359} Blockers of NO synthesis (L-arginine analogs) and NO donors, inhibitors of activated K⁺ channels (oral antidiabetic, sulfonylurea drugs), Na⁺/H⁺ exchanger (amiloride, benzamide), and poly(ADP-ribose) polymerase have been tried in animal experiments to overcome vascular unresponsiveness, inflammation, and organ dysfunction early or late after development of hypovolemic shock.* ATP-MgCl₂ may provide energy to cells, improve the microcirculation, reduce cell swelling, protect tissues from injury, and promote organ function and survival during hypovolemic shock and resuscitation.^{24,128,143,144} Ca²⁺-entry blockers have been used to prevent intracellular accumulation of Ca²⁺ and further damage of ischemic cells during resuscitation from hypovolemic shock.^† Opiate antagonists or inhibitors such as naloxone, ACTH, and thyrotropin-releasing hormone been shown to increase arterial blood pressure, decrease inflammation, and improve survival.^‡ Sedatives and analgesic drugs such as dexmedetomidine and ketamine may have anti-inflammatory properties and are tissue protective.³⁶⁰ Other vasoactive agents, including thyroid hormone, glucagon, and angiotensin inhibitors, with a preferential effect on splanchnic blood flow also may have beneficial effects in hypovolemic shock.^105,225,226

Experimental models indicate that pretreatment with xanthine oxidase inhibitors (allopurinol) or scavengers of ROS or antioxidants including lazaroids may ameliorate microvascular hemodynamics and membrane injury of organs such as the heart and gut and improve survival after resuscitation from hypovolemic shock, although some of these compounds may have greater effects than others.* A study in trauma patients with elevated lipid peroxidation products in plasma showed that superoxide dismutase ameliorated the inflammatory response and MOF.²³⁹ It has been suggested that corticosteroids prevent lysosomal disruption and release of toxic proteases, prevent NO synthesis, and ameliorate the hemodynamic changes and promote survival during hypovolemic shock in animals.^{92,167,229,268} Cortisol treatment may increase the vascular sensitivity for catecholamines, particularly in patients in whom adrenal cortisol secretion is low relative to severity of disease, the so-called relative adrenal insufficiency. It may also decrease respiratory infections in the hospital course.³⁶¹ Intravenously administered ACTH fragments may have an adrenal-independent central opioid-inhibiting effect, which may help to prevent vascular decompensation and treat hypotension, even clinically.⁹³ Drugs such as pentoxifylline and complement inhibitors may prevent neutrophil-mediated endothelial injury, dysfunction, and downregulation of NO synthesis after bleeding and resuscitation, diminishing endothelium-dependent vasodilation.^118,215 Pentoxifylline may ameliorate not only macrophage cytokine generation, but also adhesion molecule expression and neutrophil activation and aggregation and may improve red blood cell deformability. Administration of pentoxifylline or other methylxanthines may ameliorate reperfusion injury, at least in the rat gut and liver, and survival may be enhanced.^{79,166,206,230,319}

Heparin and nonanticoagulant heparin sulfate or other analogues may have anti-inflammatory effects and may improve the microcirculation, and administration may partly protect various tissues, including the liver and gut, against reperfusion injury after hypovolemic shock by bleeding.250,363 Protease inhibitors, such as aprotinin, may also have beneficial effects. Administration of female hormones or inducers such as dehydroepiandrosterone or testosterone depletion after trauma/hemorrhage and resuscitation partly protects animals from microcirculatory organ dysfunction and immunosuppression, and a wide variety of mechanisms has been implicated.²¹⁸ The potential of other immunologic and hormonal agents to treat immunosuppression has also been shown.^218,250,362 Further drug developments include anticytokine strategies and tissue protective agents interfering with cell stress, apoptosis, or necrosis, such as erythropoietin.²¹⁶ Mesenchymal stem cells are under investigation.