Circulatory Shock
PATHOGENESIS AND PATHOPHYSIOLOGY OF SHOCK
Introduction
The syndrome of shock in humans is often the final pathway through which a variety of pathologic processes lead to cardiovascular failure and death. As such, it is perhaps the most common and important problem with which critical care physicians contend. The importance of shock as a medical problem can be appreciated by the prominence of its three dominant forms. Cardiogenic shock related to pump failure is a major component of the mortality associated with cardiovascular disease, the leading cause of death in the United States with almost 800,000 deaths annually.1 Similarly, hypovolemic shock remains a major contributor to early mortality from trauma, the most common cause of death in those between the ages of 1 and 45 (approximately 200,000 cases annually.1,2 Finally, despite improving medical and surgical therapy, overall mortality coded as septicemia has increased from the 13th to the 11th most frequent cause of death in the United States.1,3 Most current estimates suggest that there are more than 100,000 cases of septic shock annually in the United States alone.4,5 In addition, all forms of shock increase the probability of other major comorbidities such as serious infection, acute respiratory distress syndrome (ARDS), and multiple organ dysfunction syndrome (MODS).
History
Despite recognition of a posttraumatic syndrome by Greek physicians such as Hippocrates and Galen, the origin of the term shock is generally credited to the French surgeon Henri Francois Le Dran who, in his 1737 “A Treatise of Reflections Drawn from Experience with Gunshot Wounds,” coined the term choc to indicate a severe impact or jolt.6 An inappropriate translation by the English physician Clarke, in 1743, led to the introduction of the word shock to the English language to indicate the sudden deterioration of a patient’s condition with major trauma.6 It was Edwin A. Moses,7 however, who began to popularize the term, using it in his 1867 “A Practical Treatise on Shock after Operations and Injuries.” He defined it as “a peculiar effect on the animal system, produced by violent injuries from any cause, or from violent mental emotions.” Prior to this definition, the rarely used term shock referred in a nonspecific sense to the immediate and devastating effects of trauma, not a specific posttrauma syndrome. Although not entirely accurate by today’s standards, his definition was one of the first to separate the syndrome involving the body’s response to massive trauma from the immediate, direct manifestations of trauma itself.
By the late 1800s, two theories of traumatic shock physiology dominated. The first, based on observations by Bernard, Charcot, Goltz, and others, was proposed by Fischer in 1870.8–10 He suggested that traumatic shock was caused by generalized “vasomotor paralysis” resulting in splanchnic blood pooling. The corollary was that total circulating blood volume is preserved in shock. The second dominant theory, articulated by Mapother in 1879, suggested that decreased cardiac output in traumatic shock is caused by intravascular volume loss due to extrusion of plasma through the vessel wall from the intravascular space to the interstitium.11 He proposed that this was a consequence of the failure of “vasodilator nerves” in traumatic shock and subsequent generalized arteriolar vasoconstriction. With the 1899 publication of “An Experimental Research into Surgical Shock” (perhaps the first experimental studies of shock), George W. Crile provided scientific data supporting a variation of the vasomotor paralysis theory.12 After documenting the importance of decreased central venous pressure and venous return in experimental shock due to hemorrhage and demonstrating the potential for intravascular volume replacement as therapy, he proposed that traumatic shock was caused by exhaustion of the overstimulated “vasomotor center” and subsequent generalized relaxation of large vessels (veins) leading to decreased ventricular filling and cardiac output.
Further advances in shock research were substantially driven by military concerns. During World War I, Walter B. Cannon and other physiologists/physicians studied the early clinical response to battlefield trauma. Their work eventually led to the publication of the classic monograph “Traumatic Shock” in 1923.13 Cannon and his colleagues were the first to relate trauma-associated hypotension in a large group of patients to a fall in blood volume, loss of bicarbonate, and accumulation of organic acids. Others, using dye dilution techniques, demonstrated that severity of shock was directly related to the decrease in intravascular volume.14 Clinical data from war casualties also suggested the importance of reduced blood flow (independent of blood pressure) in shock.15 The observation that blood in the capillaries of victims of massive trauma was hemoconcentrated compared to venous blood would lead to the practice of resuscitating trauma patients with dried pooled plasma rather than whole blood in the early part of World War II.16
Although work originating from the battlefields of World War I clearly linked traumatic shock associated with substantial, obvious bleeding to a loss of circulating blood volume, the origin of traumatic shock in the absence of defined hemorrhage was unclear. The accepted explanation for this phenomenon remained a variation of the vasomotor paralysis theory of shock. It was postulated that nonhemorrhagic, posttraumatic shock (“wound shock”) was caused by the liberation of “wound toxins” (histamine or other substances), which resulted in neurogenic vasodilation and peripheral blood pooling. However, after the war, Blalock and others demonstrated in animal models that nonhemorrhagic traumatic shock was due to the loss of blood and fluids into injured tissue rather than circulating toxins resulting in stasis of blood within the circulation.17
Additional advances occurred during World War II. Using injured subjects from the European front, Henry Beecher confirmed that hemorrhage and fluid loss leading to metabolic acidosis was a major cause of shock.18 In the first use of indicator dye techniques in humans for studying blood flow, Cournand and Richard, in 1943, demonstrated that cardiac output was typically reduced in shock.19 They also reinforced Blalock’s findings regarding nonhemorrhagic “wound shock” in trauma patients by demonstrating that circulating blood volume was reduced in such patients through loss of fluid into damaged tissues. The importance of maintaining intravascular volume in traumatic and hemorrhagic shock was supported by the well-known cardiovascular physiologist, Carl J. Wiggers, who published a landmark series of studies20 in the 1940s using a standardized animal model, which showed that prolonged hypovolemic shock resulted in a resuscitation-resistant state that he termed irreversible shock. He defined it as a condition resulting from “a depression of many functions but in which reduction of the effective circulating blood volume is of basic importance and in which impairment of the circulation steadily progresses until it eventuates in a state of irreversible circulatory failure.” Aggressive fluid support became the standard of resuscitation for trauma and shock.
Subsequently, the Korean War fueled the research that demonstrated the relationship of acute tubular necrosis (ATN) and acute renal failure (ARF) to circulatory shock.21 In addition, studies of battlefield casualties clearly demonstrated the relationship between early resuscitation and survival.21 During the Vietnamese conflict, with the widespread use of ventilator technology, the dominant research concern became postshock infection and “shock lung” (ARDS), a concern that has evolved to the present interest in shock-related MODS.
Definitions and Categorization of Shock
The definition of shock has evolved in parallel with our understanding of the phenomenon. As noted, until the late 1800s, the term shock was used to indicate the immediate response to massive trauma, without regard to a specific posttrauma syndrome. The definition consisted of descriptions of its obvious clinical signs. In 1895, John Collins Warren22 referred to shock as “a momentary pause in the act of death,” which was characterized by an “imperceptible” or “weak, threadlike” peripheral pulse and a “cold, clammy sweat.”
Subsequently, with the introduction of noninvasive blood pressure monitoring devices, most clinical definitions of shock added the requirement for arterial hypotension. In 1930, Blalock8 included arterial hypotension as one of the required manifestations of shock when he defined it as “peripheral circulatory failure resulting from a discrepancy in the size of the vascular bed and the volume of the intravascular fluid.” Simeone,23 as recently as 1964, suggested that shock exists when “the cardiac output is insufficient to fill the arterial tree with blood under sufficient pressure to provide organs and tissues with adequate blood flow.”
Current technology, which allows for the assessment of perfusion independent of arterial pressure, has shown that hypotension does not define shock. The emphasis in defining shock is now on tissue perfusion in relation to cellular function. According to Fink,24 shock is “a syndrome precipitated by a systemic derangement of perfusion leading to widespread cellular hypoxia and vital organ dysfunction.” Cerra25 has emphasized supply/demand mismatch in his definition: “a disordered response of organisms to an inappropriate balance of substrate supply and demand at a cellular level.”
Effective tissue perfusion, as opposed to tissue perfusion per se, is an important issue. Effective tissue perfusion may be reduced by either a global reduction of systemic perfusion (cardiac output) or by increased ineffective tissue perfusion due to a maldistribution of blood flow or a defect of substrate utilization at the subcellular level (Box 21.1).
Classification
Although hypovolemic shock associated with trauma was the first form of shock to be recognized and studied, by the early 1900s it became broadly recognized that other clinical conditions could result in a similar constellation of signs and symptoms. Sepsis as a distinct cause of shock was initially proposed by Laennec (1831) and subsequently supported by Boise (1897).26,27 In 1934, Fishberg and colleagues introduced the concept of primary cardiogenic shock due to myocardial infarction.28 Later the same year, Blalock developed the precursor of the most commonly used classification systems of the present.29 He subdivided shock into four etiologic categories: hematogenic or oligemic (hypovolemic), cardiogenic, neurogenic (e.g., shock after spinal injury), and vasogenic (primarily septic shock). Shubin and Weil, in 1967, proposed the additional etiologic categories of hypersensitivity (i.e., anaphylactic), bacteremic (i.e., septic), obstructive, and endocrinologic shock.30 However, as the hemodynamic profiles of the different forms of shock were uncovered, a classification based on cardiovascular characteristics, initially proposed in 1972 by Hinshaw and Cox,31 came to be accepted by most clinicians. The categories include (1) hypovolemic shock, due to a decreased circulating blood volume in relation to the total vascular capacity and characterized by a reduction of diastolic filling pressures and volumes; (2) cardiogenic shock related to cardiac pump failure due to loss of myocardial contractility/functional myocardium or structural/mechanical failure of the cardiac anatomy and characterized by elevations of diastolic filling pressures and volumes; (3) extracardiac obstructive shock involving obstruction to flow in the cardiovascular circuit and characterized by either impairment of diastolic filling or excessive afterload; and (4) distributive shock caused by loss of vasomotor control resulting in arteriolar and venular dilation and (after resuscitation with fluids) characterized by increased cardiac output with decreased systemic vascular resistance. We have adapted these categories into an etiologic/physiologic classification of shock that is summarized in Figure 21.1 and Box 21.2. This figure and box represent our current understanding of the causes and typical hemodynamic features of different forms of shock.
Despite this hemodynamic-based categorization system, it is important to note the mixed nature of most forms of clinical shock. Septic shock is nominally considered a form of distributive shock. However, prior to resuscitation with fluids, a substantial hypovolemic component may exist due to venodilatation and third-spacing. In addition, depression of the myocardium in human septic shock is well documented (see Fig. 21.1).32–34 Similarly, hemorrhagic shock in experimental models has been linked to both myocardial depression35,36 and vascular dysfunction (see Fig. 21.1).37,38 Cardiogenic shock typically presents with increased ventricular filling pressures. However, many patients have been aggressively diuresed prior to the onset of shock and may have a relative hypovolemic component. In addition, systemic vascular resistance (SVR) is only inconsistently increased in cardiogenic shock, suggesting that an inflammatory element may exist under some circumstances. Finally, shock from any cause may cause a deterioration of the coronary perfusion pressure, the difference between mean arterial pressure (MAP) and the higher of left ventricular diastolic pressure or the right atrial pressure, resulting in some degree of myocardial ischemia and myocardial dysfunction.39 Thus, although four categories of shock exist based on hemodynamic profile, clinical shock states tend to combine components of each.
Hypovolemic Shock
Hypovolemic shock may be related to dehydration, internal or external hemorrhage, gastrointestinal fluid losses (diarrhea or vomiting), urinary losses due to either diuretics or kidney dysfunction, or loss of intravascular volume to the interstitium due to decrease of vascular permeability (in response to sepsis or trauma). In addition, venodilatation due to a number of causes (sepsis, spinal injury, various drugs and toxins) may result in a relative hypovolemic state (see Box. 21.2, Fig. 21.1). Hemodynamically, hypovolemic shock is characterized by a fall in ventricular preload resulting in decreased ventricular diastolic pressures and volumes (Table 21.1). Cardiac index (CI) and stroke volume index (SVI) are typically reduced. In addition to hypotension, a decreased pulse pressure may be noted. Due to a decreased output and unchanged or increased metabolic demand, mixed venous oxygen saturation (MVO2) may be decreased and the arteriovenous oxygen content difference widened. Clinical characteristics include pale, cool, clammy skin (often mottled); tachycardia (or if severe shock, bradycardia)7,40; tachypnea; flat, nondistended peripheral veins; decreased jugular venous pulse; decreased urine output; and altered mental status.
Table 21.1
Hemodynamic Profiles of Shock*
, mixed venous oxygen saturation; ↑↑ or ↓, mild to moderate increase or decrease; ↑↑ or ↓ ↓, moderate to severe increase or decrease.*The hemodynamic profiles summarized in this table refer to patients with the diagnosis listed in the left column who are also in shock (mean arterial blood pressure < 60-65 mm Hg).
Modified from Parrillo JE: Septic shock: Clinical manifestations, pathogenesis, hemodynamics, and management in a critical care unit. In Parrillo JE, Ayers SM (eds): Major Issues in Critical Care Medicine. Baltimore, Williams & Wilkins, 1984.
A number of factors may influence the development and hemodynamic characteristics of hypovolemic shock in humans. Studies in animals and humans have demonstrated a clear relationship between the degree of circulating blood volume loss and clinical response.41–44 Acute loss of 10% of the circulating blood volume is well tolerated, with tachycardia the only obvious sign. CI may be minimally decreased despite a compensatory increase in myocardial contractility. SVR typically increases slightly, particularly if sympathetic stimulation augments mean arterial pressure (MAP). Compensatory mechanisms begin to fail with a 20% to 25% volume loss. Mild to moderate hypotension and decreased CI may be present. Orthostasis (with a blood pressure decrease of 10 mm Hg and increased heart rate of 20 to 30 beats/minute) may become apparent. There is a marked increase in SVR and serum lactate may begin to rise. With decreases of the circulating volume of 40% or more, marked hypotension with clinical signs of shock is noted. CI and tissue perfusion may fall to less than half normal. Lactic acidosis is usually present at this stage and predicts a poor outcome.45,46 The case fatality rate can exceed 50% in hemorrhagic shock associated with trauma.47
The rate of loss of intravascular volume and the preexisting cardiac reserve is of substantial importance in the development of hypovolemic shock. As an example, whereas an acute blood loss of 1 L in a healthy adult may result in mild to moderate hypotension with a reduced pulmonary artery occlusion pressure (PAOP) and central venous pressure (CVP),42 the same loss over a longer period of time may be well tolerated due to compensatory responses such as tachycardia, increased myocardial contractility, increased red blood cell 2,3-diphosphoglycerate (2,3 DPG), and increased fluid retention. On the other hand, a similar slow loss may lead to substantial hemodynamic compromise in a person with a limited cardiac reserve, even while the person’s PAOP and CVP remain elevated.
Hypovolemic shock represents more than a simple mechanical response to loss of circulating volume. It is a dynamic process involving competing adaptive (compensatory) and maladaptive responses at each stage of development. Thus, although intravascular volume replacement is always a necessary component of resuscitation from hypovolemia or hypovolemic shock, the biologic responses to the insult may progress to the point where such resuscitation is insufficient to reverse the progression of the shock syndrome. Patients who have sustained a greater than 40% loss of blood volume for 2 hours or more may be unable to be effectively resuscitated.37,41,44 A series of inflammatory mediator, cardiovascular, and organ responses to shock are initiated, which supersede the importance of the initial insult in driving further injury.
Cardiogenic Shock
Cardiogenic shock results from the failure of the heart as a pump (see Box 21.2, Fig. 21.1). It is the most common cause of in-hospital mortality in patients with Q-wave myocardial infarction.48,49 Hemodynamically, cardiogenic shock is characterized by increased ventricular preload (increased ventricular volumes, pulmonary wedge pressure [PWP] and CVP) (see Table 21.1). Otherwise hemodynamic characteristics are similar to those for hypovolemic shock (see Table 21.1). In particular, both involve reduced CI, SVI, and ventricular stroke work indices with increased SVR. Due to inadequate tissue perfusion, the MVO2 is substantially reduced and the arteriovenous oxygen content difference increased. The degree of lactic acidosis may predict mortality.50 Clinically, the specific signs of shock are similar. However, signs of congestive heart failure (volume overload) are typically present in cardiogenic shock. The jugular and peripheral veins may be distended. An S3 and evidence of pulmonary edema are usually found.
Cardiogenic shock is most commonly due to ischemic myocardial injury with a total of 40% of the myocardium nonfunctional.49,51–53 Such damage may involve a single large myocardial infarction or may involve accumulation of damage from multiple infarctions. In addition, viable but dysfunctional “stunned” myocardium may temporarily contribute to cardiogenic shock postinfarction. Cardiogenic shock usually involves an anterior myocardial infarction with left main or proximal left anterior descending artery occlusion. Historically, the incidence of cardiogenic shock due to Q-wave infarction has ranged from 8% to 20%.48,54–56 Although several large studies demonstrate lower incidence rates (4% to 7%) when patients receive thrombolytic interventions,55,57–60 retrospective community studies suggest no overall decrease in the incidence of postinfarction cardiogenic shock or cardiogenic shock mortality (70% to 90%) in the first decades following the introduction of this therapy.48 Further, no trials have demonstrated that thrombolytic therapy reduces mortality rates in patients with established cardiogenic shock.60,61 In contrast, several major studies suggest that mortality of infarction-related cardiogenic shock may be improved by emergent angioplasty.56,62–64 Accordingly, data suggest a reduction in the incidence of acute infarction-related cardiogenic shock to <2% in 2003 in association with widespread use of emergent percutaneous coronary intervention.62 This intervention has also been associated with a reduction of cardiogenic shock mortality risk from 60% to 84% to 43% to 47% in two large analyses.56,62
Mortality is better for cardiogenic shock due to surgically remediable cardiac lesions. Mitral valve failure may be associated with rupture or dysfunction of chordae or papillary muscles due to myocardial ischemia or infarction, endocarditis, blunt chest trauma, or prosthetic valve deterioration and is characterized by “v” waves of greater than 10 mm Hg on a PAOP tracing. Ischemic papillary muscle rupture frequently occurs 3 to 7 days after an infarct in left anterior descending coronary artery territory and may be preceded by the onset of a mitral regurgitant murmur.65 Mortality is high in the absence of surgical therapy.65 Acute aortic valve failure is most commonly due to endocarditis but may involve mechanical failure of prosthetic valves, or aortic dissection. Ventricular septal defects caused by myocardial infarction may also result in the abrupt onset of cardiogenic shock and can be diagnosed by a 5% step up in hemoglobin oxygen saturation between the right atrium and the pulmonary artery (due to left-to-right shunting of blood through the septum).66 As with ischemic papillary muscle rupture, rupture of the intraventricular septum is most frequently seen with occlusions of the left anterior descending artery, a few days after infarction.66
The pathophysiology of cardiogenic shock due to a right ventricular infarction and failure is different from other forms of cardiogenic shock. Although some degree of right ventricular involvement is seen in half of inferior myocardial infarctions, only the largest 10% to 20% result in right ventricular failure and cardiogenic shock.67 These infarctions usually involve part of the left ventricular wall as well. Isolated infarctions of the right ventricle are rare.67,68
Because therapy of this form of shock requires fluid resuscitation and inotropes (rather than vasopressors), differentiation from other causes of cardiogenic shock is crucial. Conditions compromising right ventricular function such as cardiac tamponade, restrictive cardiomyopathy, constrictive pericarditis, and pulmonary embolus are also included in the differential diagnosis. Each of these conditions may present with some of the typical clinical and hemodynamic findings of right ventricular infarction including Kussmaul’s sign, and pulsus paradoxus with elevation and equalization of CVP, right ventricular systolic pressure, pulmonary artery diastolic pressure, and PAOP. Prognosis in this form of cardiogenic shock is distinctly better than that of cardiogenic shock due to left ventricular infarction69,70; however, an inferior infarction with right ventricular injury has a substantially worse prognosis than such an infarction without significant right-sided involvement.71
As with hypovolemic shock, a number of interactions may complicate the development of cardiogenic shock. Optimal cardiac performance in patients with impaired myocardial contractility may occur at substantially higher than normal PAOP (i.e., 20 to 24 mm Hg). Yet patients who develop cardiogenic shock are frequently initially treated with diuretics and may have a degree of hypovolemia (relative to their optimal requirements). Thus, patients should not be diagnosed with cardiogenic shock unless hypotension (MAP < 65 mm Hg) and reduced cardiac output (CI < 2.2 L/min/m2) coexist with an elevated ventricular filling pressure.72 Cautious fluid challenge may be required (in the absence of overt pulmonary edema) to increase the filling pressures to an optimal range. Other interactions include increased right ventricular ischemia due to decreased right coronary perfusion pressure (MAP decreased while right ventricular end-diastolic pressure is increased) and increased right ventricular afterload due to pulmonary hypertension. Right ventricular ischemia may also lead to right ventricular dilatation, septal shift, and impairment of left ventricular function.
Other causes of cardiogenic shock include acute myocarditis, end-stage cardiomyopathy, brady- or tachyarrhythmias, hypertrophic cardiomyopathy with obstruction, and traumatic myocardial contusion (see Box 21.2).
Obstructive Shock
Extracardiac obstructive shock results from an obstruction to flow in the cardiovascular circuit (see Box 21.2, Fig. 21.1). Pericardial tamponade and constrictive pericarditis directly impair diastolic filling of the right ventricle. Tension pneumothorax and intrathoracic tumors indirectly impair right ventricular filling by obstructing venous return. Massive pulmonary emboli (two or more lobar arteries with >50% of the vascular bed occluded), nonembolic acute pulmonary hypertension, large systemic emboli (e.g., saddle embolus), and aortic dissection may result in shock due to increased ventricular afterload.
The characteristic hemodynamic/metabolic patterns are, in most ways, similar to other low output shock states (see Table 21.1). CI, SVI, and stroke work indices are usually decreased. Because tissue perfusion is decreased, the MVO2 is low, the arteriovenous oxygen content difference increased, and serum lactate frequently elevated. Other hemodynamic parameters are dependent on the site of the obstruction. Tension pneumothorax and mediastinal tumors may obstruct the great thoracic veins, resulting in a hemodynamic pattern (decreased CI and elevated SVR) similar to hypovolemia (although distended jugular and peripheral veins may be seen). Cardiac tamponade typically causes increased and equalized right and left heart ventricular diastolic pressures, pulmonary artery diastolic pressure, CVP, and PAOP. In constrictive pericarditis, right and left ventricular diastolic pressures are elevated and within 5 mm Hg of each other. Mean right and left atrial pressures may or may not be equal as well. Massive pulmonary embolus will result in right ventricular failure with elevated pulmonary artery and right heart pressures whereas PAOP remains normal. A systemic saddle embolus or aortic occlusion due to dissection causes peripheral hypotension and signs of left ventricular failure including an elevated PAOP. Clinical signs are similarly dependent on the site of the obstruction.
As with other forms of shock, the time course of development of the insult has a substantial impact on the clinical response. Ischemic rupture of the left ventricular free wall (usually 3 to 7 days after myocardial infarction) leads to immediate cardiac tamponade and shock with as little as 150 mL blood in the pericardium.73–75 Survival requires emergency surgery.74,75 Similar situations may develop with bleeding into the pericardium after blunt chest trauma or thrombolytic therapy. Pericardial tamponade due to malignant or inflammatory pericardial effusions usually develop much more slowly. Although shock may still develop, it usually requires substantially more pericardial fluid (1 to 2 L) to cause critical failure of right ventricular diastolic filling.73 No large reliable studies examining mortality rates with and without therapy in these conditions are available due to the small numbers of cases.
A similar time course–dependent risk is seen with major pulmonary emboli. In those without preexisting cardiopulmonary disease, a massive embolus involving two or more lobar arteries and 50% to 60% of the vascular bed76,77 may result in obstructive shock. However, if recurrent smaller pulmonary emboli result in right ventricular hypertrophy, a substantially larger total occlusion of the pulmonary vascular bed may be required to cause right ventricular decompensation. Analyses have suggested that the presence of shock due to pulmonary embolus (regardless of underlying chronic cardiopulmonary dysfunction) indicates a three- to sevenfold increase in mortality risk with the majority of deaths occurring within an hour of presentation.78,79 An analysis of more than 70,000 unstable (hemodynamic instability or ventilator-requiring) patients with pulmonary embolus in the national inpatient sample shows that mortality in untreated patients is approximately 47%.80 Systemic thrombolysis is associated with a substantial reduction in mortality to 15%. Where available, catheter-directed therapy may be even more efficacious with a lower risk of serious hemorrhage.81 Shock due to pulmonary embolism is an indication for urgent thrombolytic or catheter-directed intervention.
Distributive Shock
Hemodynamically, distributive shock is characterized by an overall decrease in SVR (see Table 21.1). However, resistance in any specific organ bed or tissue may be decreased, increased, or unchanged. Initially, CI may be depressed and ventricular filling pressures decreased. After fluid resuscitation, when filling pressures are normalized or increased, CI is usually elevated. Due to hypotension, left and right ventricular stroke work indices are normally decreased. MVO2 is increased above normal. Concomitantly, arteriovenous oxygen content difference is narrowed despite the fact that oxygen demand is usually increased (particularly in sepsis). The basis of this phenomenon may be that because total body perfusion (CI) is increased, perfusion is not effective in that either it does not reach the necessary tissues or the tissues cannot utilize the substrates presented. As a reflection of this inadequate “effective” tissue perfusion, lactic acidosis may ensue. Clinical characteristics of resuscitated distributive shock include, in contrast to the other forms of shock, warm, well-perfused extremities, a decreased diastolic blood pressure, and an increased pulse pressure. Nonspecific signs of shock include tachycardia, tachypnea, decreased urine output, and altered mentation. In addition, evidence of the primary insult may exist (urticaria for anaphylaxis, spinal injury for neurogenic shock, and evidence of infection in septic shock).
Septic shock (shock due to infection) and sepsis-associated multiple organ failure are the most common causes of death in ICUs of the industrialized world. As many as 800,000 cases of sepsis are admitted every year to American hospitals (comparable to the incidence of first myocardial infarctions) with half of those developing septic shock and about half of those (200,000) dying.82 Since the 1970s there has been a progressive increase in the incidence of and total deaths from sepsis and septic shock.5 The total toll of septic deaths is comparable to deaths from myocardial infarction and far exceeds the impact of illnesses such as AIDS or breast cancer.82,83
Septic shock is caused by the systemic activation of the inflammatory cascade. Numerous mediators including cytokines, kinins, complement, coagulation factors, and eicosanoids are activated or systemically released, resulting in profound disturbances of cardiovascular and organ system function84 (Table 21.2). These mediators, particularly tumor necrosis factor α (TNFα), interleukin-1β (IL-1β), platelet activating factor (PAF), and prostaglandins are thought to mediate reduced peripheral vascular resistance seen in septic shock.
Table 21.2
Inflammatory Mediators in Sepsis and Septic Shock
Adapted from Bone RC: The pathogenesis of sepsis. Ann Intern Med 1991;115:457.
Loss of vascular autoregulatory control may explain some of the typical metabolic findings of sepsis and septic shock. An early theory postulated the existence of microanatomic shunts between the arterial and venous circulations. During sepsis, these shunts were said to result in decreased SVR and increased MVO2.85 However, although microanatomic shunting has been noted in localized areas of inflammation, systemic evidence of this phenomenon in sepsis and septic shock is lacking.85–89 “Functional” shunting due to defects of microcirculatory regulation in sepsis has also been suggested.90,91 Overperfusion of tissues with low metabolic requirements would increase MVO2 and narrow the arteriovenous oxygen content difference. Relative vasoconstriction of vessels supplying more metabolically active tissues would result in tissue hypoxia and lactate production due to anaerobic metabolism. Observations that some capillary beds may be occluded by platelet microaggregates, leukocytes, fibrin deposits, and endothelial damage support this theory.86,90,92 Additional support comes from studies that demonstrate evidence of oxygen supply–dependent oxygen consumption in sepsis.93–97 A third theory suggests that circulating mediators cause an intracellular metabolic defect involving substrate utilization, which results in bioenergetic failure (decreased high-energy phosphate production) and lactate production.98,99 Increased mixed venous oxygen saturation could then be explained by perfusion, which is increased in excess of tissue oxygen utilization capability. However, animal studies using nuclear magnetic resonance (NMR) spectroscopy demonstrate that high-energy phosphates are not depleted in septic animals as is expected in all of these theories.100–102 According to these and other studies, cellular ischemia is not the dominant factor in metabolic dysfunction in sepsis.100–106 Rather, circulating mediators may result in cellular dysfunction, aerobic glycolysis, and lactate production in the absence of global ischemia.101 This position is weakened by data suggesting that increased lactate in septic shock is also associated with decreased pH (which would not be expected in aerobic glycolysis)101 and, to some extent, by studies that support the existence of oxygen supply–dependent oxygen consumption in sepsis.94–97
The trigger for systemic activation of the inflammatory cascade is the presence of gram-negative bacilli in 50% to 75% of cases of septic shock. Gram-positive bacteria account for most of the remainder, but infection with fungi, protozoa, and viruses can also result in septic shock.107–109 Investigations suggest a surprising commonality of signaling mechanisms in septic shock via Toll-like receptors from a broad range of etiologic agents.110–114 Despite aggressive supportive care and antibiotic treatment, mortality is 50% overall and may exceed 70% for gram-negative septic shock.107 Of those succumbing to septic shock, approximately 75% are early deaths (within 1 week of shock), primarily due to hyperdynamic circulatory failure.115 Late mortality is usually due to MODS.115
More than any other form of shock, distributive and, particularly, septic shock involves substantial elements of the hemodynamic characteristics of other shock categories (see Fig. 21.1, Table 21.1). As noted, all forms of distributive shock involve decreased mean peripheral vascular resistance. Prior to fluid resuscitation, distributive shock also involves a relative hypovolemic component. The first element of this relative hypovolemia is an increase of the vascular capacitance due to venodilatation. This phenomenon has been directly supported in animal models of sepsis116–120 and is reinforced by the fact that clinical hypodynamic septic shock (low cardiac output) can usually be converted to hyperdynamic shock (high cardiac output) with adequate fluid resuscitation.115,121,122 Relaxation of vascular smooth muscle is attributed to a number of the mediators known to circulate during sepsis. These same mediators also contribute to the second cause of hypovolemia in sepsis, third-spacing of fluid to the interstitium due to a loss of endothelial integrity. In addition, a number of studies have demonstrated that human septic shock is characterized by myocardial depression (biventricular dilatation and decreased ejection fraction).32–34 Circulating substances such as TNFα, IL-1β, platelet activating factor (PAF), leukotrienes, and, most recently, interleukin-6 (IL-6) have been implicated in this process.123–130
Anaphylactic shock is a form of distributive shock caused by the release of mediators from tissue mast cells and circulating basophils. Anaphylaxis, an immediate hypersensitivity reaction, is mediated by the interaction of IgE antibodies on the surface of mast cells and basophils with the appropriate antigen. Antigen binding results in the release of the primary mediators of anaphylaxis contained in the basophilic granules of mast cells and basophils. These include histamine, serotonin, eosinophil chemotactic factor, and various proteolytic enzymes.131 Subsequently, a number of secondary lipid mediators are synthesized and released including PAF, bradykinin prostaglandins, and leukotrienes (slow-reacting substance of anaphylaxis).131 An anaphylactoid reaction (clinically indistinguishable from anaphylaxis) results from the direct, nonimmunologic release of mediators from mast cells and basophils and can also result in shock.
Anaphylaxis is triggered by insect envenomations (Hymenoptera bees, hornets, and wasps) and certain drugs, especially antibiotics (beta-lactams, cephalosporins, sulfonamides, vancomycin).131 In addition, less frequently, heterologous serum (e.g., tetanus antitoxin, snake antitoxin, antilymphocyte antisera), blood transfusion, immunoglobulin (particularly in IgA-deficient patients), and egg-based vaccine products have been implicated.131 Anaphylactoid reactions can be caused by a wide range of medical agents including ionic contrast media, protamine, opiates, polysaccharide volume expanders such as dextran and hydroxyethyl starch, muscle relaxants, and anesthetics.131
The hemodynamic features of anaphylactic shock are very similar to those for septic shock and include elements of hypovolemia (due to interstitial edema and venodilatation) and myocardial depression.132–136 Cardiac output and ventricular filling pressures may be reduced until patients are fluid resuscitated.136,137 In addition to typical findings of shock, patients may demonstrate urticaria, angioedema, laryngeal edema, and severe bronchospasm.
Adrenal crisis (see also Chapter 59) is an uncommon cause of shock, which can be difficult to diagnose as it occurs in patients with other active disease processes and the clinical features may mimic infection. It is a life-threatening emergency that requires prompt diagnosis and management.
Adrenal crisis is caused by a deficiency of adrenal production of mineralocorticoids and glucocorticoids. It may occur de novo in patients with critical illness or may occur against a background of occult adrenal insufficiency. In the critical care setting, the most common cause of de novo acute adrenal insufficiency is bilateral adrenal hemorrhage in association with overwhelming infections (classically meningococcal, but frequently gram-negative bacteria), human immunodeficiency virus infection, or anticoagulation.138,139 In addition, fungal infections such as histoplasmosis, blastomycosis, and coccidioidomycosis and malignant infiltration of the adrenals may cause acute adrenal insufficiency in ICU patients.139 In some patients, steroid production remains adequate for the baseline state despite adrenal disease. Once stressed, however, the adrenal response is inadequate, leading to decompensation and adrenal crisis. Stressors may be relatively innocuous or may be severe. A febrile illness, infection, trauma, surgery, dehydration, or any other intercurrent illness may trigger the crisis. Abrupt cessation of glucocorticoid therapy or replacement may also result in adrenal crisis.
Symptoms are generally nonspecific and may include anorexia, nausea, vomiting, diarrhea, abdominal pain, myalgia, joint pains, headache, weakness, confusion, and agitation or delirium.139,140 Fever (often out of proportion to any minor infection) is almost always present, and hypotension, initially due to hypovolemia, is frequent.139 The initial hemodynamic pattern may resemble hypovolemic shock (if shock is due only to adrenal crisis). With volume resuscitation, a high output, vasopressor-refractory shock may become apparent.141,142
An unrecognized “relative” adrenal insufficiency has been implicated in the pathogenesis of human septic shock.143–147
