Myocardial infarction (MI) is defined by myocyte necrosis. If large in its territory or critical in its location, this injury can lead to loss of contractile function or other structural complications (see Chapter 26 ) that result in insufficient cardiac output and consequent heart failure or cardiogenic shock. Despite significant advancements in the acute management of MI, incident heart failure still occurs in 10% to 30% of patients during the initial hospitalization for acute MI and is associated with at least a two-fold higher adjusted risk of in-hospital death compared to patients without heart failure. Cardiogenic shock occurs in aproximately 5% to 7% of patients presenting with an acute MI and carries a substantial 40% to 60% mortality rate.
This chapter reviews the current understanding of the epidemiology, pathophysiology, evaluation, and management of heart failure and cardiogenic shock after MI. The mechanical complications of MI are detailed in Chapter 26 . The implementation of mechanically assisted circulatory support for cardiogenic shock after MI is discussed in Chapter 27 .
Defining and Classifying Heart Failure and Cardiogenic Shock
Heart failure after MI is a clinical syndrome typically defined by evidence of pulmonary venous or central venous congestion. Cardiogenic shock is characterized by congestion and inadequate tissue or end-organ perfusion secondary to cardiac insufficiency. This reduction in perfusion results in decreased oxygen and nutrient delivery to tissues, which, if severe or protracted, can lead to multiorgan dysfunction and death. Generally accepted criteria for shock include (1) frank or relative hypotension defined by a systolic blood pressure below 80 or 90 mm Hg or by a reduction in mean arterial pressure of 30 mm Hg; (2) inadequate cardiac index defined as below 1.8 L/min/m 2 without mechanical or pharmacologic support or less than 2.2 L/min/m 2 with support; (3) elevated end-diastolic pressures on the right (greater than 10 to 15 mm Hg) and/or left (greater than 18 mm Hg) side of the heart; and (4) evidence of end-organ hypoperfusion. End-organ hypoperfusion may manifest as altered mental status, decreased urine output, acute kidney injury, cool or mottled extremities, acute liver injury, or lactic acidosis.
Heart failure and cardiogenic shock can be categorized by severity using one of several classification systems ( Table 25-1 ). The Killip classification system, derived specifically in patients with acute MI, is defined by physical examination findings consistent with heart failure. Killip class I is characterized by an absence of heart failure; class II is consistent with mild to moderate heart failure (S 3 gallop, pulmonary rales, or jugular venous distention); class III heart failure includes overt pulmonary edema; and class IV is defined as cardiogenic shock. The Killip classification has a strong graded relationship with mortality after acute MI, in that patients with class II or III heart failure have a 4-fold increased risk of in-hospital death, and those with cardiogenic shock, a 10-fold increased risk. The New York Heart Association (NYHA) system describes a functional or symptomatic status, whereby classes I to IV are defined by the spectrum of no symptomatic limitation of physical activity (NYHA class I), slight limitation (NYHA class II), marked limitation without symptoms at rest (NYHA class III) and symptoms with minimal exertion or symptoms at rest (NYHA class IV), respectively. NYHA class also is a marker of heart failure severity and therefore is associated with survival. Finally, investigators with the Interagency Registry of Mechanically Assisted Circulatory Support ( INTERMACS ) developed a classification system based on combination of signs, symptoms and level of therapeutic support, with the goal of refining the description of NYHA class III-IV and shock patients to best define populations that would benefit from advanced therapies such as pacing, cardiac transplantation, and mechanical circulatory support (MCS). The INTERMACS profiles ( Table 25-1 ) range from NYHA class III (INTERMACS 7), exertion-limited, exertion-intolerant, symptoms at rest, stable but inotrope-dependent, progressive decline on inotropes, to critical cardiogenic shock (INTERMACS 1).
|Class I||No evidence of heart failure|
|Class II||Mild to moderate heart failure, including S 3 gallop, rales auscultated over less than one half of the posterior lung fields, or jugular venous distention|
|Class III||Overt pulmonary edema|
|Class IV||Cardiogenic shock|
|Class I||No limitation of physical activity|
|Class II||Slight limitation of physical activity; comfortable at rest, but ordinary physical activity results in symptoms of heart failure|
|Class III||Marked limitation of physical activity; comfortable at rest, but less than ordinary activity causes symptoms of heart failure|
|Class IV||Unable to carry on any physical activity without symptoms of heart failure or symptoms of heart failure at rest|
|Profile 1||Critical cardiogenic shock (“crash and burn”); life-threatening hypotension despite rapidly escalating inotropic support|
|Profile 2||Progressive decline (“sliding on inotropes”); declining function despite intravenous inotropic support|
|Profile 3||Stable but inotrope-dependent (“dependent stability”); stable blood pressure, organ function, and symptoms on continuous intravenous inotropic support or temporary circulatory support or both with repeated failure to wean|
|Profile 4||Resting symptoms; stabilized close to normal volume status but experiences daily symptoms at rest or during activities of daily living (ADLs)|
|Profile 5||Exertion intolerant (“housebound”); comfortable at rest or with ADLs, symptoms with any further activity|
|Profile 6||Exertion limited (“walking wounded”); fatigues after first few minutes of anything beyond minor activity|
|Profile 7||Advanced NYHA III|
Epidemiology and Outcomes
Left ventricular failure is the predominant mechanism of cardiac insufficiency in approximately 75% of patients in whom shock develops after MI ( Figure 25-1 ). Mechanical causes of heart failure and shock (see Chapter 26 ), including ventricular septal rupture, severe mitral regurgitation, and free wall rupture with tamponade, account for 12% of cases, and predominant right ventricular failure, another 3%.
Incidence and Risk Factors
Depending on the characteristics of the study population and heart failure definitions applied, the incidence of new-onset heart failure after acute MI is estimated at 10% to 30%. For example, in four major trials of fibrinolysis, 13% of the 61,041 patients presenting with ST-elevation MI (STEMI) had heart failure without shock on admission, and 29% had heart failure that emerged some time during admission. Of 13,707 patients in the Global Registry of Acute Coronary Events (GRACE) registry without previous heart failure or cardiogenic shock, 13% of patients presenting with an acute coronary syndrome had Killip class II or III heart failure, and heart failure developed in an additional 5% during the initial hospitalization.
The available data are conflicting regarding the trends in incident heart failure without shock after acute MI, with some evidence to support a decrease in early heart failure due to improved reperfusion strategies, but an increase in chronic heart failure rates due to increased survival of patients with substantial left ventricular damage. For example, investigators with the GRACE registry found that with increasing rates of percutaneous coronary intervention (PCI) and evidence-based pharmacotherapies, rates of incident heart failure decreased by 9% in patients with STEMI and by 6.9% in patients with non–ST-elevation MI (NSTEMI) between 1999 and 2006. In this registry, predictors of incident heart failure in the setting of acute coronary syndromes included older age, prior MI, atherosclerotic disease in non-coronary vascular beds, presentation with an MI (versus unstable angina), diabetes, and increased heart rate. Other studies have identified left ventricular systolic dysfunction and infarct size as predictors of new-onset heart failure after MI.
The incidence of cardiogenic shock complicating acute MI has been declining, a trend that also may relate to increased use of more efficacious reperfusion strategies (see Chapter 13 ). For example, in a registry of 13,663 residents of Worcester, Massachusetts, hospitalized with acute MI, the incidence of cardiogenic shock decreased from 7.3% in 1975 to approximately 5% in 2005 ( Figure 25-2 ). Similarly, an analysis of data for 7,531 patients from three French registries reported that the incidence decreased from 6.9% in 1995 to 5.7% in 2005; a period over which the rate of PCI increased dramatically.
Cardiogenic shock occurs more frequently in the setting of STEMI than NSTEMI, with estimated rates of 5% to 8% and 2% to 3%, respectively. Other risk factors for the development of cardiogenic shock, identified in French and Danish registries, include older age, female sex, hypertension, diabetes, previous MI, heart failure, anterior MI, left bundle branch block, and reduced left ventricular systolic function.
Outcomes with Heart Failure and Cardiogenic Shock
As noted, heart failure without cardiogenic shock continues to be a common complication of acute MI and portends a worse prognosis. In the previously described GRACE registry of 13,707 patients with ACS without shock or previous heart failure, presentation with heart failure or development of heart failure after admission identified patients with markedly higher in-hospital mortality than that for patients without heart failure (12.0% versus 17.8% versus 2.8%; P < .0001), representing a greater than two-fold increased risk after adjustment for other predictors of mortality. Analysis of data for 3,343 patients with STEMI in the Harmonizing Outcomes with Revascularization and Stents in Acute Myocardial Infarction (HORIZONS-AMI) trial revealed that development of new-onset heart failure was associated not only with higher mortality but also with increased risk for recurrent MI, stent thrombosis, and need for coronary revascularization.
Cardiogenic shock is a relatively infrequent complication of acute MI, but the mortality rate associated with the condition is staggeringly high. Although in-hospital mortality rates have decreased, from 70% to 80% in the 1970s to approximately 40% to 60% currently, cardiogenic shock remains the major cause of death among patients hospitalized with acute MI (see Figure 25-2 ). The improvement in mortality over time is certainly multifactorial. During this period, multiple medical therapies that reduce the rate of recurrent cardiovascular events have been incorporated into standard practice, including lipid-lowering therapies, β-adrenergic blockade (beta-blockade), inhibition of the renin-angiotensin-aldosterone system (RAAS), and use of potent antiplatelet agents (see Chapter 13 ). In addition, the options for and prioritization of reperfusion strategies for acute coronary syndromes have advanced significantly.
Mortality Risk Prediction
Clinical predictors of higher mortality rates among patients with shock include delayed or inadequate revascularization, older age, anoxic brain injury, lower left ventricular ejection fraction, low systolic blood pressure, vasopressor requirements, renal dysfunction, elevated lactate levels, and complicated coronary disease, such as left main artery or three-vessel disease. Of interest, outcomes are relatively similar in patients with cardiogenic shock after STEMI and in those after NSTEMI, and also for cardiogenic shock due to predominant left ventricular failure and that due to right ventricular failure. A depressed cardiac power output, defined as the product of mean arterial pressure and cardiac output, was found to be the strongest, most independent hemodynamic predictor of in-hospital mortality in the Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock? (SHOCK) trial. Biomarkers of inflammation, such as interleukin-6 (IL-6), and of microcirculatory dysfunction or vascular leakage (angiopoietins) also have been associated with poorer prognosis and higher mortality in these patients but are not currently used for clinical risk stratification.
The pathophysiology of cardiogenic shock is complex and dynamic, involving adaptive and maladaptive compensatory mechanisms coordinated by multiple organ systems.
The multiple myocardial and vascular parameters that determine stroke volume and work are represented in the left ventricular pressure-volume loops ( Figure 25-3A ). The left ventricular end-diastolic pressure is a function of the end-diastolic volume and myocardial compliance, represented by the end-diastolic pressure-volume relationship (EDPVR). The difference between the end-diastolic and end-systolic volume represents the stroke volume and area contained in the pressure-volume loop, the stroke work. The timing of the events in the cycle is determined by both myocardial and vascular characteristics. For example, the load-independent left ventricular contractility (E max ), described by the end-systolic pressure-volume relationship (ESPVR) across the range of loading conditions, and the effective arterial elastance (E a ) define the timing of aortic valve closure and thus of end systole. E a is a measure of arterial loading, approximated as the ratio of end-systolic pressure and stroke volume and is influenced by peripheral resistance, vascular compliance and impedance, and systolic and diastolic time intervals. Stroke work or left ventricular pump efficiency is maximal when the ratio of effective arterial elastance and left ventricular contractility (E a /E max ) approaches 1.
In the setting of an acute MI, left ventricular contractility is reduced through myocardial necrosis and stunning, thereby shifting the ESPVR down, and reducing stroke volume (see Figure 25-3B ). In many cases, myocardial compliance decreases as well, shifting the EDPVR up, leading to an increase in left ventricular end-diastolic pressure for a given volume. If severe, these acute alterations in systolic and diastolic function can lead to decreased cardiac output and stroke work and increased pulmonary congestion, resulting in clinical heart failure.
In the extreme case of these perturbations ( Figure 25-3C ), left ventricular contractile function is severely reduced and left ventricular end-diastolic volume and pressures are significantly increased, culminating in cardiogenic shock. Without appropriate interventions, cardiogenic shock begets worsened cardiogenic shock and death ( Figure 25-4 ). That is, the shock state, defined by inadequate cardiac output and congestion, results in a cascade of events that reinforce and exacerbate the underlying pathology of ischemia and progressive left ventricular dysfunction (see also Figure 27-1 ).
Impact of Altered Hemodynamics
The primary insult, myocyte necrosis in the setting of an acute MI, results in reduced myocardial contractility and systolic dysfunction. The resultant reduction in cardiac output leads to hypotension and peripheral hypoperfusion. Hypotension can promote further ischemia and myocardial depression through decreased coronary perfusion, which can be exacerbated by atherosclerotic lesions in non-culprit coronary vessels ( Figure 25-4 ). In an attempt to maintain perfusion to vital organs, the body releases endogenous catecholamines (e.g., norepinephrine) and other vasopressors (e.g., angiotensin II). These catecholamines and vasoconstrictors may mediate improved blood pressure through increased systemic vascular resistance (SVR) as well as myocardial contractility; however, this increase in blood pressure may come at the cost of increased myocardial oxygen demand resulting from increased afterload and heart rate, promoting further ischemia. Hypoperfusion also activates the neurohormonal cascade, resulting in sodium and fluid retention, thereby increasing perfusion through increased intravascular volume. In the setting of diastolic dysfunction, however, this compensatory response can lead to a greater elevation of left ventricular end-diastolic pressures, translating to more pulmonary edema and hypoxemia with further ischemia and progressive left ventricular dysfunction.
Vasodilatory and Inflammatory Response in Cardiogenic Shock
In addition to the critical myocardial hemodynamic changes, cardiogenic shock can be complicated by development of a systemic inflammatory response syndrome (SIRS). The development of SIRS is more common with a longer duration of shock and often is not associated with superimposed infection. For example, approximately 20% of patients enrolled in the SHOCK trial had suspected sepsis. Only three quarters of those patients ultimately had positive blood cultures, however, and the vasodilatory state generally preceded bacteremia by several days, suggesting an earlier noninfectious inflammatory process.
Several mechanisms may be contributors to this vasodilatory state, including (1) the development of a vasopressin deficiency, (2) activation of ATP-sensitive potassium (K ATP ) channels on vascular smooth muscle, and (3) release of inflammatory cytokines. Vasopressin, or antidiuretic hormone, promotes vasoconstriction, free water absorption, decreased plasma osmolality, and increased blood volume. It is secreted in response to increased plasma osmolality, angiotensin II, cardiac wall stress, and adrenergic stimuli. Vasopressin levels are notably lower in vasodilatory shock after cardiopulmonary bypass or ventricular assist device placement, possibly as a result of depletion of neurohypophyseal stores. Patients in this setting tend to respond with brisk increases in arterial pressure with vasopressin administration, even when refractory to catecholamines.
K ATP channels on vascular smooth muscle allow for efflux of potassium, resulting in cellular hyperpolarization, decreased calcium influx, and vasodilation. A number of factors present during shock contribute to K ATP channel activation, including acidosis (e.g., hydrogen ion and lactate), decreased energy stores, reduced vasopressin levels, and increased atrial natriuretic peptide and nitric oxide levels.
Inflammatory cytokines are elevated in the setting of cardiogenic shock complicating acute MI—a phenomenon that is even more pronounced in the subset of patients with SIRS. These cytokines, particularly IL-6 and tumor necrosis factor-alpha (TNF-α), are believed to promote progressive shock through a number of mechanisms. IL-6 and TNF-α stimulate expression of inducible nitric oxide synthase (iNOS). Increased iNOS results in higher levels of nitric oxide, which in turn cause inappropriate vasodilation through induction of soluble guanylate cyclase and increased cyclic guanosine monophosphate (cGMP), as well as activation of K ATP channels in vascular smooth muscle. This cascade of inflammatory activation drives progressive hypotension with worsened peripheral and coronary hypoperfusion and, as a result, further myocardial depression and worsened shock. Nitric oxide exerts additional negative effects by promoting further release of inflammatory cytokines, induction of DNA damage through generation of peroxynitrite, suppression of mitochondrial respiration, and reduced contractility as a result of decreased calcium release from the sarcoplasmic reticulum. IL-6 also may contribute to myocardial depression through downregulation of sarcoplasmic reticulum calcium ATPase (SERCA2) and myosin expression, upregulation of the IL-6 receptor, and generation of oxygen free radicals secondary to increased xanthine oxidase activity.
The inflammatory cascade in cardiogenic shock is thought to be driven by myocyte necrosis, as well as hypoxemia and hypoperfusion of other tissues, most notably the gut. Intestinal ischemia–reperfusion is believed to lead to increased intestinal permeability, bacterial translocation and endotoxin release, and, in a subset, development of bacteremia and sepsis. Endotoxin, a lipopolysaccharide (LPS) found in the cell walls of gram-negative bacteria, binds to Toll-like receptors on macrophages, resulting in production of inflammatory cytokines, including TNF and the interleukins IL-1β and IL-6. The link between intestinal hypoperfusion and endotoxin is supported by a study demonstrating a correlation between biomarkers of intestinal injury, including urinary intestinal fatty acid–binding protein (IFABP) and plasma citrulline, and endotoxin levels in 21 patients with out-of-hospital cardiac arrest (OHCA). Furthermore, those patients with post-arrest shock demonstrated increasing levels of endotoxin over time, and those without shock had decreasing levels in the days after the arrest. In this setting, high endotoxin levels after OHCA were associated with increased severity and duration of shock, as indicated by the mean dose of vasopressor and days requiring vasopressor therapy.
Studies attempting to target systemic inflammation and inappropriate vasodilation in cardiogenic shock are described later in the chapter.
Presentation and Initial Evaluation
Presentation of Heart Failure and Cardiogenic Shock After Myocardial Infarction
Cardiogenic shock secondary to acute MI typically develops during the initial hospitalization, with only 15% of cases documented as present at the time of hospital arrival. For patients who develop shock after hospitalization, the onset tends to occur early, usually within 24 hours, of STEMI and later, on the order of 3 to 4 days, for NSTEMI.
Nohria and colleagues described a classification system for heart failure states based on the presence or absence of congestion, described as “wet” or “dry,” and on the presence or absence of hypoperfusion, described as “cold” or “warm.” The physical examination in patients with heart failure without cardiogenic shock reveals congestion with preserved perfusion (“warm and wet”) and the examination in cardiogenic shock indicates both hypoperfusion and congestion (“cold and wet”). On physical examination, congestion is identified by the presence of jugular venous distention, hepatojugular reflux or a square-wave blood pressure response to the Valsalva maneuver, orthopnea, peripheral edema, or a third heart sound (S 3 gallop). Hypoperfusion is identified by hypotension, a narrow pulse pressure, cool extremities (except in patients who develop a low-SVR state), or altered mental status. A narrow pulse pressure, defined by a ratio of the pulse pressure (systolic minus diastolic pressure) to systolic pressure of 25% or less, has been shown to have a 91% sensitivity and 83% specificity for a cardiac index below 2.2 L/min/m 2 .
Findings on the history and physical examination assist in the differentiation among types of shock, including cardiogenic, hypovolemic, distributive, and mixed (e.g., cardiogenic and distributive). Once a cardiogenic component is suspected, it is prudent to consider a broad group of possible etiologic disorders. Multiple causes of cardiogenic shock may coexist, and the treatment may vary accordingly. For example, in a patient hospitalized after an anterior STEMI, cardiogenic shock may develop secondary to left ventricular dysfunction, as well as from mechanical complications, such as ventricular septal rupture. Potential causes of cardiogenic shock, including those possible after MI, are listed in Table 25-2 . These etiologic factors include complications of MI, such as left, right, and biventricular pump dysfunction in the setting of the initial ischemic event, as well as mechanical complications (see Chapter 26 ), including acute mitral regurgitation, ventricular septal rupture, free wall rupture, and cardiac dysrhythmia (see Chapter 28 ). The differential diagnosis also includes entities not related to epicardial coronary disease, such as stress cardiomyopathy, inflammatory myocarditis, pericarditis with tamponade, native or prosthetic valvular dysfunction, and massive pulmonary embolism (see Chapter 6 ).
Approach to Evaluation of a Patient with Shock
Evaluating a patient with shock requires integration of data from multiple sources, including the clinical history, physical examination, laboratory data, electrocardiography, imaging, and invasive hemodynamic assessments.
Clinical History and Physical Examination
The clinical history and physical examination are critical elements in the evaluation of patients with shock. The history can provide important information regarding etiology, duration, and progression. The physical examination is invaluable for rapidly assessing the likely type of shock. For example, in patients with hypovolemic shock, the physical examination will expose manifestations of decreased preload, including low jugular venous pressure, dry mucous membranes, decreased skin turgor, and possibly cool, mottled extremities (reflective of high SVR). Distributive shock is characterized by an inappropriate decrease in SVR, which typically results in an augmentation of heart rate and compensatory increase in cardiac output. Preload may be low or normal in the case of distributive shock. Although many of the physical examination findings may be similar to those in hypovolemic shock (e.g., tachycardia, low jugular venous pressure), a distinguishing feature of distributive shock is warm or hyperemic extremities, reflecting a low-SVR state. Finally, as described previously, the physical examination in cardiogenic shock often reveals congestion, as manifested by pulmonary or peripheral edema, ascites, an S 3 gallop, and jugular venous distention, as well as hypoperfusion with cool extremities (elevated SVR), altered mental status, or oliguria.
Depending on the severity of shock, laboratory evaluation may demonstrate evidence of anaerobic metabolism with a lactic acidosis and organ hypoperfusion, including acute kidney and liver injury. Biomarkers of myonecrosis, including troponin, are important in the diagnosis and prognostication in patients with acute MI (see Chapter 6 and Chapter 7 ), and also may be elevated in patients with shock outside of the setting of acute MI. Additionally, troponin has been associated with worse clinical outcomes and higher mortality in patients with acute decompensated heart failure. B-type natriuretic peptide (BNP) and the amino-terminal cleavage product (NT-proBNP) are released by cardiomyocytes in the setting of increased myocardial wall stress. These biomarkers are useful to support clinical judgment for diagnosis or exclusion of acute decompensated heart failure, as well as for risk stratification. Natriuretic peptides can be elevated in other cardiac and noncardiac conditions and may be falsely decreased in the setting of obesity.
Several investigational biomarkers have shown promise for risk stratification in cardiogenic shock. Consistent with the inflammatory response to myocardial necrosis and prolonged cardiogenic shock, investigators found that higher levels of interleukins, including IL-6, predicted occurrence of death in 40 patients with acute MI and cardiogenic shock in the IABP-SHOCK trial. Similarly, markers of vascular integrity, the angiopoietins, were independently associated with mortality in cardiogenic shock. As described earlier, the presence of endotoxin probably reflects end-organ (intestinal) hypoperfusion, with higher levels indicating more severe shock. Endotoxin is known to drive production of procalcitonin, a precursor to the hormone calcitonin, which is used as a relatively specific biomarker of systemic bacterial infections and sepsis. Of interest, in a study of 29 patients with acute cardiogenic shock without bacteremia, Brunkhorst and colleagues found that in 18 of 20 patients who survived for more than 12 hours, pyrexia and an increase in procalcitonin level developed in the absence of positive cultures, supporting the hypothesis that the development of SIRS in cardiogenic shock may be related to exposure to bacterial endotoxin. Further studies are needed to determine the optimal combination of biomarkers for risk stratification and, possibly, to monitor response to therapy in cardiogenic shock.
Electrocardiogram, Imaging, and Other Assessments
An electrocardiogram (ECG) should be obtained immediately at the time of presentation, to assist with diagnosis, prognosis, and therapeutic decision making (see Chapter 6 ). The extent of ST-segment deviation on the ECG, location of infarction, and QRS duration correlate with risk of adverse outcomes, including the risk for onset of severe heart failure or shock.
Noninvasive imaging with echocardiography is recommended in patients with acute MI and should be performed urgently in such patients with cardiogenic shock. Echocardiography is a widely available and rapidly applied modality that can identify left and right ventricular dysfunction, myocardial wall motion abnormalities, pericardial tamponade, severe valvular disease, papillary muscle rupture, and ventricular septal rupture (see Chapter 31 ). In addition, echocardiography can be used to estimate left ventricular filling pressures using several different techniques. For example, the ratio of the peak transmitral pulsed Doppler inflow velocity (E), a marker of early diastolic flow, to the tissue Doppler-derived mitral annular velocity (e′), a marker of myocardial relaxation, is well correlated with invasively measured pulmonary capillary wedge pressure ( r = 0.87; P < .001) ( Figure 25-5 ). Specifically, an E/e′ ratio of 12 or higher using the lateral annulus, or an E/e′ ratio of 15 or higher using the septal annulus, are correlated with a wedge pressure of 15 mm Hg or more, and an E/e′ less than 8 at either annular location is correlated with normal left ventricular filling pressures. Doppler echocardiography also has been studied as a method to estimate cardiac output in critically ill or perioperative patients, using small probes placed in the esophagus after appropriate sedation and mechanical ventilation have been instituted. With this technique, the cardiac output is based on the diameter and velocity of flow in the aorta and an estimation of the proportion of output delivered to the descending aorta.