Pulmonary Edema

62

PULMONARY EDEMA MICHAEL A. MATTHAY, MD • JOHN F. MURRAY, MD

INTRODUCTION PATHOPHYSIOLOGY OF PULMONARY EDEMA Increased Pressure Edema Increased Permeability Edema

DIAGNOSIS Clinical Assessment Measurement of Lung Water Measurement of Barrier Function TREATMENT Emergency Therapy Increased Pressure Edema Increased Permeability Edema

INTRODUCTION Pulmonary edema—defined as excessive extravascular water in the lungs—is a common and serious clinical problem. Pulmonary edema can be life-threatening, but effective therapy is available to rescue patients from the deleterious consequences of disturbed lung fluid balance, which usually can be identified and, in many instances, corrected. Because rational and effective therapy depends on understanding basic principles of normal and abnormal liquid, solute, and protein transport in the lungs, this chapter begins with a brief overview of the major factors that govern fluid and protein filtration in healthy lungs before focusing on the pathophysiology of pulmonary edema. Next, the chapter discusses diagnosis, treatment, and resolution of pulmonary edema. Chapters 6 and 9 also provide additional information about the regulation of fluid balance in the lungs, and Chapter 100 includes details about the onset and management of acute lung injury and acute respiratory distress syndrome, as currently defined and subsequently discussed.

PATHOPHYSIOLOGY OF PULMONARY EDEMA Pulmonary edema results when fluid is filtered into the lungs faster than it can be removed from them. Accumulation of fluid has serious consequences on lung function because gas exchange is greatly impaired in fluid-filled alveoli. Lung structure relevant to the forces governing fluid and protein movement in healthy lungs and lungs with pulmonary edema has been the subject of classic and more recent reviews.1-6 There is always a net outward flux of fluid and protein crossing from the vascular space into the interstitium in the lungs, first, because the prevailing driving forces normally cause filtration out of the bloodstream and, second, because the microvascular endothelium is a permeable barrier that varies in its leakiness. Lung lymph flow, which represents the flow of fluid leaking across the microvascular barrier, normally is less than 0.01% of total lung blood flow. The 1096

OUTCOME Resolution of Pulmonary Edema Overview

term microvascular bed (or barrier), is used throughout this chapter to refer to sites of fluid exchange. In addition to the vast interconnecting network of capillaries embedded in the alveolar walls, fluid is exchanged across capillaries in the interstitium at alveolar wall junctions (corner vessels) and across small interstitial arteries and veins. The essential factors that govern fluid exchange in the lungs are expressed in the Starling equation for the microvascular barrier: Jv = LpS [(Pc − Pi) − σd (πc − πi)] where Jv is the net fluid-filtration rate (volume flow) across the microvascular barrier; Lp is the hydraulic conductivity (“permeability”) of the microvascular barrier to fluid filtration (a measure of how easy it is for water to cross the barrier); S is the surface area of the barrier; Pc is the pulmonary capillary (microvascular) hydrostatic pressure; Pi is the interstitial (“perimicrovascular”) hydrostatic pressure; πc is the capillary (microvascular) plasma colloid osmotic (or oncotic) pressure; πi is the interstitial (perimicrovascular) fluid osmotic pressure; and σd is the average osmotic reflection coefficient of the barrier (a measure of how effective the barrier is in hindering the passage of solutes from one side of the barrier to the other). The microvascular hydrostatic pressure is the principal force that causes fluid filtration in the lungs. If blood was not flowing through the lungs, the opposing hydrostatic and osmotic forces on either side of the microvascular barrier would be equal, their sum would be zero, and there would be no filtration. The pumping action of the heart causes blood to flow through the lungs and generates the microvascular hydrostatic pressure that establishes the steady-state values of the other driving pressures that cause filtration of fluid.5,7 According to the Starling equation, the difference between the prevailing transmural hydrostatic pressures (Pc − Pi) and the colloid osmotic pressures (πc − πi) provides the “driving force” for fluid filtration. The actual amount of filtrate that forms at any given driving force is determined by the integrity of the barrier to filtration, which is reflected in the conductivity (Lp) and reflection (σd) coefficients. The equation predicts the development of two fundamentally different kinds of pulmonary edema: (1) increased pressure

62  •  Pulmonary Edema 1097

Table 62-1  Safety Factors That Protect the Lungs Against Interstitial and Alveolar Edema Accumulation 1. 2. 3. 4. 5. 6. 7.

Lung lymphatic system Resorption into blood vessels Drainage into the mediastinum Drainage into the pleural space Extremely low alveolar epithelial barrier permeability Low alveolar surface tension (surfactant) Active transport by alveolar and distal airway epithelial cells

pulmonary edema—when the net result of the driving forces increases, fluid filtration is forced across the barrier at a rate that exceeds removal by lymphatic drainage, and (2) increased permeability pulmonary edema—when the normal barriers to fluid filtration are damaged, typically by some form of injury, conductance of liquid and protein in the lungs is allowed to increase. A third type of pulmonary edema is caused by impaired lymphatic drainage of filtered fluid, but this has less clinical relevance than the other two types. The lymphatic drainage of the lungs provides a vital means for removing filtered fluid and proteins from the perimicrovascular interstitial space, as discussed later.5,7 Because the healthy microvascular barrier is permeable, the alveolar barrier must serve as the principal protection against the accumulation of pulmonary edema. Fluid and protein do not normally move into alveoli because the alveolar epithelial barrier has a low permeability even to small molecules (similar to cell membrane permeability); in addition, any fluid that is filtered is continuously being pumped back into the interstitium by alveolar epithelial cells,2 drained away from the alveolar walls through the interstitium, and removed by lymphatic vessels and the lung microcirculation. The several factors (Table 62-1) that normally protect the lungs against edema have been called safety factors. Under normal conditions, the lymphatic system pumps filtered fluid and protein out of the lungs as rapidly as they are formed, even when filtration of fluid and protein from the bloodstream into the interstitium is increased. Increases in fluid and protein filtration across the microvascular barrier also can be drained away from the alveolar walls down the prevailing pressure gradient into the loose peribronchovascular connective tissue or can be resorbed directly into blood vessels.4 The lung lymphatics can increase their pumping capacity manifold, particularly when the microvascular wall has been injured.5 When the usual driving forces are upset by higher hydrostatic pressure, the increase in filtration of water across the microvascular barrier is much larger than that in protein flux because the microvascular barrier has a low protein conductance. This results in dilution (“wash-down”) of interstitial protein concentration and, thereby, an increase in the balance of the protein osmotic pressure opposing the higher hydrostatic pressure (because plasma protein concentration remains high). Furthermore, the interstitial gel also becomes hydrated and the exclusion volume for protein decreases, either because of swelling or because its composition changes as hyaluronan is washed out from the interstitium, reducing the concentration of protein by expanding the available volume.

Figure 62-1  Frontal chest radiograph in a 55-year-old man with increased pressure edema due to heart failure. The heart is enlarged and bilateral perihilar linear and ground-glass opacity are present. This distribution is often referred to as a “butterfly” pattern and is commonly seen with chronic volume overload. (Courtesy Michael Gotway, MD.)

The protein osmotic pressure safety factors work only when the microvascular barrier is normal—as in increased pressure edema. In contrast, if the endothelial barrier is injured and its functional integrity is compromised—as in increased permeability edema—barrier conductance increases and the osmotic reflection coefficient decreases, making this safety factor much less effective or even completely ineffective. The compliance of the interstitial space also protects the lungs against edema. Increases in interstitial volume result in only small elevations of interstitial pressure until the interstitial volume is large. This maintains the hydrostatic driving pressure across the alveolar barrier suitably low. When interstitial pressure within the lungs rises to greater than pleural pressure, fluid flows across the visceral pleura into the pleural space, where its effects on lung function are relatively minor. Pleural fluid is drained by lymphatics in the parietal pleural and, even when pleural liquid accumulates, does not flow back from the pleural space into the lungs. Fluid that does accumulate in the alveoli is pumped out by active ion transport.2 There are several mechanisms that can up-regulate the rate of alveolar fluid clearance (see Chapter 9). In summary, pulmonary edema results from increases in either driving pressures (increased pressure edema) or barrier conductance (increased permeability edema) or both combined. What distinguishes between the two types is barrier permeability, which is normal in increased pressure edema but leaky in increased permeability edema. Fluid flow into the lungs is driven across the barrier in both types of edema according to the prevailing pressures.

INCREASED PRESSURE EDEMA Increased pressure pulmonary edema (Fig. 62-1) is caused by an increase in the net sum of driving forces for fluid filtration into the lungs. The essential feature of this edema is

1098 PART 3  •  Clinical Respiratory Medicine

that the barriers to fluid and protein flow into the lungs are functionally intact. Increased pressure edema is often called cardiogenic, high-pressure, or hydrostatic pulmonary edema.

Pathophysiology The flow of fluid and protein into the lungs increases when the sum of driving pressures is elevated. If the rate of fluid accumulation exceeds the rate at which it can be removed, increased pressure edema results. Because the barriers limiting fluid and protein flow into the lungs are intact, the lungs are protected against edema by the prevailing (normal) safety factors. Especially important is the ability to protect against increases in the principal driving pressure, lung microvascular hydrostatic pressure. Because of the low protein conductance of the microvascular barrier, fluid flow increases much more than protein flow when microvascular hydrostatic pressure rises. Interstitial protein concentration is diluted both by this higher fluid flow relative to protein flow and by a diminished exclusion volume for proteins as the interstitial gel is hydrated and swells. Lower interstitial protein concentration, owing to washout of interstitial proteins, results in lower perimicrovascular protein osmotic pressure and thereby a greater protein osmotic pressure difference across the barrier, which opposes any rise in hydrostatic pressure. In experimental animals, slightly less than 50% of an increase in hydrostatic pressure is offset by the increase in osmotic pressure difference.7,8 Increased pressure edema may be gradual in onset and progression because any elevation in microvascular hydrostatic pressure is attenuated by a rise in protein osmotic pressure difference across the microvascular barrier, owing to a decline in interstitial protein osmotic pressure. The consequences of increased pressure edema on lung mechanics and gas exchange (Table 62-2) depend on how much edema accumulates.9 Deliberate dehydration of healthy subjects increases lung volumes and improves tests of ventilatory function.10 Early in the development of pulmonary edema, increases in hydrostatic pressure result in enlargement of the intrapulmonary blood volume, as Table 62-2  Effects of Vascular Congestion, Interstitial Edema, and Alveolar Flooding on Pulmonary Function and Lung Mechanics VASCULAR CONGESTION Increased diffusing capacity Increased arterial PO2 Decreased compliance Bronchoconstriction INTERSTITIAL EDEMA Increased closing volume Decreased maximal expiratory flow Increased ventilation-perfusion mismatching Decreased arterial PO2 ALVEOLAR FLOODING Increased closing volume (air trapping) Increased vascular resistance Decreased lung volume (especially vital and inspiratory capacities) Decreased compliance Decreased diffusing capacity Right-to-left shunting of blood (severely compromised gas exchange)

vessels, including capillaries, are both recruited and distended, which causes the diffusing capacity of the lung (DLCO) to increase above normal; similarly, arterial oxygen pressure (PO2) may rise because ventilated units are better perfused when vascular pressures increase.11 The small reversible changes in airflow resistance and dynamic compliance, unaffected by vagotomy, in congested lungs appear to be due to reflex bronchoconstriction responses, but only when baseline bronchial tone is normal.12 When interstitial edema is present, closing volume may be increased and maximum expiratory airflows may be decreased. These changes were originally thought to be due to a decrease in caliber of small airways caused by compression by rising volume and pressure in the peribronchovascular connective tissue spaces. This effect would have to be in airways larger than bronchioles, because bronchioles and smaller airways do not have loose connective tissue sheaths1 and their diameter is a function of lung volume, not transpulmonary pressure. Arterial PO2 often falls as a result of ventilation-perfusion mismatching, but gas exchange is not seriously compromised until the alveoli are flooded. With alveolar flooding, lung volumes are diminished.10 This is most marked in measurements of vital capacity, with inspiratory capacity affected more than expiratory capacity. Airways may close at higher than normal distending pressures, resulting in trapping of larger volumes of gas in the lungs.11 Lung compliance is reduced when alveolar edema is present because lung volume decreases. Gas exchange becomes severely compromised when flooded alveoli remain perfused, causing right-to-left shunts,12 and there is an increase in wasted ventilation (ventilation of units in which there is decreased or absent perfusion). Light and electron microscopic examination of human and animal lung tissue in increased pressure edema shows alveolar edema and hemorrhage; thickened interstitial compartments (especially large peribronchovascular fluid cuffs) with separated, dispersed collagen fibrils; and increased capillary surface area and volume.13 More intercellular vesicles can be seen, but otherwise there are no detectable changes in the ultrastructure of the vascular endothelium, and the gap widths at intracellular junctions are not different from those in normal lungs. Long-standing pulmonary edema (e.g., in patients with chronic mitral stenosis) can be associated with basement membrane thickening and increased distance between the alveoli and the capillaries; increases in fibroblasts, histiocytes, and bulky strands of collagenous fibers can be seen in the interstitium. Dogs with pacing-induced chronic congestive heart failure developed a significant increase in the threshold for highvascular-pressure edema formation—about a 50% reduction in the amount of water and protein cleared across the lung microvascular endothelial barrier at high pulmonary vascular pressures—compared with control animals. Morphometric analysis of the alveolar-capillary barrier showed that endothelial, interstitial, and epithelial thicknesses were all increased compared with controls, indicating that remodeling may confer an increase in the resistance to development of high-pressure–induced alveolar edema. Alveolar type II cells may be more numerous than in normal lungs, and alveolar macrophages proliferate.14 Sites of chronic severe increased pressure edema may also become

62  •  Pulmonary Edema 1099

Table 62-3  Mechanisms of Increased Pressure Pulmonary Edema INCREASED LUNG MICROVASCULAR HYDROSTATIC PRESSURE Left ventricular dysfunction Mechanical obstruction of left atrial outflow Volume overload Pulmonary venous hypertension Overperfusion Increased lymphatic outflow pressure DECREASED PERIMICROVASCULAR HYDROSTATIC PRESSURE Inspiratory airway obstruction Increased alveolar surface tension

organized and fibrotic, may calcify, and may even result in bone formation.15

Mechanisms By far the most common cause of increased pressure edema is elevated lung microvascular hydrostatic pressure. The influence of driving pressures would be greater than usual if either perimicrovascular hydrostatic pressure or protein osmotic pressure difference across the microvascular barrier was decreased. At the alveolar barrier, an increase in interstitial hydrostatic pressure, a decrease in alveolar hydrostatic pressure, or a decrease in osmotic pressure difference across the barrier could result in a greater sum of driving pressures. The possibilities are listed in Table 62-3. Increased Microvascular Hydrostatic Pressure.  Congestive heart failure is the most common cause of increased pressure edema. That is why increased pressure edema is often called “cardiogenic,” even though the heart is not always primarily involved. Elevated pressures in the pulmonary microvasculature are usually due to left-sided heart failure, with elevated left atrial pressures transmitted retrograde into the pulmonary circulation. Common causes are left ventricular dysfunction (e.g., caused by acute myocardial infarction, severe coronary insufficiency, tachyarrhythmias, bradyarrhythmias, cardiomyopathies, constrictive pericarditis, aortic stenosis or regurgitation, mitral regurgitation, coarctation of the aorta, rupture of chordae tendineae or intraventricular septum, systemic hypertension) or mechanical obstruction of the left atrial outflow tract (e.g., mitral stenosis, left atrial myxoma). Left atrial and pulmonary microvascular pressures can also be elevated by severe fluid volume overloading in a patient with a normal or diseased heart. An unusual cause of increased microvascular hydrostatic pressure is pulmonary venous hypertension in the absence of left ventricular or atrial disease, which can arise if the pulmonary veins are contracted (e.g., by possible muscular sphincters), compressed, or obstructed (e.g., because of veno-occlusive disease or mediastinal fibrosis). Bronchial venous hypertension, in contrast, does not appear to significantly increase fluid filtration in the lungs.16 Increases in fluid filtration also can be associated with increases in vascular pressure proximal to the filtration sites in the lungs. For example, pulmonary hypertension, combined with depressed left ventricular function, has been implicated in the pathogenesis of cocaine-induced pulmo-

nary edema. Whether such increases lead to pulmonary edema depends on what happens to microvascular pressure. If high right-sided pressures are caused by increased resistance proximal to the main site of filtration in the lungs—as found in hypoxic pulmonary vasoconstriction of small arterial vessels,17 primary pulmonary hypertension, and pulmonary artery or valvular stenosis—pulmonary edema does not develop. Conversely, if the lung vascular bed is only partially constricted or obstructed, or if the vascular surface area is greatly decreased (e.g., by lung resection), higher flow in perfused vessels can lead to increased pressure edema,18 because microvascular pressures at the fluid exchange site are elevated in the overperfused lung. For example, pulmonary edema resulted in about 15% of patients after pneumonectomy and seemed to be exacerbated by administration of fresh frozen plasma,19 in part because intravascular volume is presumably increased by such transfusions. Any increase in blood flow through the lungs increases the pulmonary microvascular pressure at the fluid exchange sites even when pulmonary venous pressure remains constant. The mechanism of an uncommon cause of pulmonary edema, high-altitude pulmonary edema,20 which is also discussed in Chapter 77, may also be related, in part, to elevated pulmonary vascular pressures. As noted earlier, overperfusion of a restricted pulmonary vascular bed, even in the absence of hypoxia, causes increased pressure edema, not increased permeability edema.18 This can explain why some cases of high-altitude pulmonary edema are correctly classified as increased pressure pulmonary edema. Evidence from climbers studied at high altitude suggests that high intravascular pressures cause physical damage to vascular walls (so-called stress failure). In experimental animals, stress failure has been demonstrated after extreme, but sometimes transient, increases in pulmonary vascular pressures.21 Such structural failures need not happen in large numbers to explain an increase in permeability edema, because edema would form readily and in a quantity driven by the prevailing elevated vascular pressure.22 The suggestion has been made that high-altitude pulmonary edema results from the stress failure of overdistended, relatively thin-walled pulmonary arteries rather than from microvascular rupture,23 which might help explain why the prevailing vasoconstriction does not seem to offer much protection to downstream vessels, why there are no reports of gradual progression through an indolent prodrome of increased pressure edema before stress failure results, and why highaltitude pulmonary edema is first detected radiographically in the central lung fields surrounding large vessels rather than in the lung bases and periphery. Alternative mechanisms for the increased permeability in high-altitude pulmonary edema have been proposed. However, inflammatory responses in high-altitude pulmonary edema may be a consequence rather than a cause of the edema.24 The rapid reversibility of high-altitude pulmonary edema with descent to lower elevation, oxygen therapy, or pharmacologic reduction of pulmonary vascular pressure is not characteristic of increased permeability pulmonary edema with coexisting inflammation. Neurogenic pulmonary edema25 also may be related in part to elevated pulmonary vascular pressures (Fig. 62-2). Measurements of edema fluid protein concentration relative to

1100 PART 3  •  Clinical Respiratory Medicine

plasma protein concentration in 12 patients with neurogenic pulmonary edema have been reported. Seven of the patients had ratios typical of increased pressure pulmonary edema, and the other five had ratios consistent either with increased permeability pulmonary edema or with late sampling during the resolution phase of increased pressure pulmonary edema, because edema fluid protein concentration

Figure 62-2  Frontal chest radiograph in a patient with subarachnoid hemorrhage and both increased intracranial pressure and neurogenic pulmonary edema. Multifocal bilateral consolidation and ground-glass opacity, somewhat upper lobe predominant, is present. The patient’s volume status was normal and no clinical evidence of infection was present. (Courtesy Michael Gotway, MD.)

rises as fluid is reabsorbed from the alveoli at a faster rate than protein. Decreased Perimicrovascular Hydrostatic Pressure.  The sum of driving pressures would increase if perimicrovascular hydrostatic pressure was greatly diminished, thereby resulting in an increase in fluid and protein filtration at the microvascular barrier in the lungs. Pulmonary edema has been described in circumstances in which this might happen. The best clinical example may be postobstructive pulmonary edema as a consequence of upper airway obstruction or its release, which can be caused, for example, by laryngospasm, endotracheal tube obstruction, foreign body aspiration, epiglottitis, croup, severe acute asthma, airway compression by tumors, strangulation, or hanging. High negative intrathoracic pressures generated by inspiratory attempts against the occluded airway are transmitted to the interstitium, promoting fluid movement into the interstitium. Mechanical effects on the cardiovascular system likely contribute to this kind of edema. High negative intrathoracic pressure causes increases in cardiac preload and afterload and in pulmonary blood flow, all of which increase the microvascular pressure that drives fluid out into the interstitium. In three patients with upper airway obstruction and pulmonary edema, all had low edema fluid protein concentration relative to plasma protein concentration (ratios of 0.44, 0.31, and 0.52), indicating increased pressure pulmonary edema.26 Aspiration of air or fluid from the pleural space with consequent reexpansion of a collapsed lung could result in a decrease in perimicrovascular hydrostatic pressure as the lung expands to fill the thorax. So-called reexpansion pulmonary edema has been reported both in experimental animals27 and in patients28 after lung reexpansion (Fig. 62-3), but the high edema fluid protein concentration measured in three patients29 indicated that reexpansion may result in an increased permeability edema rather than an increased

Figure 62-3  Frontal chest radiograph in a patient with increased pressure pulmonary edema showing peribronchial cuffing. Central vascular enlargement and indistinctness are present, associated with airway thickening best seen in end-on bronchi, such as the anterior segmental upper lobe bronchi. These findings are shown to advantage on the magnified, inset image (arrows). (Courtesy Michael Gotway, MD.)

62  •  Pulmonary Edema 1101

pressure edema. However, another study of protein concentration in pulmonary edema fluid from patients with reexpansion edema indicates that hydrostatic mechanisms predominate.30 The increased permeability hypothesis was supported in studies of rabbits with experimental reexpansion edema.31 Reperfusion injury is one of the causes of the pulmonary edema seen in transplanted lungs and appears to be predominantly an increased permeability edema.32 If high alveolar surface tension was transmitted to the interstitium, perimicrovascular hydrostatic pressure would also be lowered, thereby increasing filtration across the microvascular barrier. Such an effect has been suggested by experimental findings in dog lungs.33 The effect of alveolar surface tension on lung fluid balance is discussed later. Decreased Transmural Protein Osmotic Pressure Difference.  The sum of driving pressures would be increased if the protein osmotic pressure difference opposing the hydrostatic pressure difference across the microvascular barrier was decreased, either by lowering plasma protein concentration or by raising interstitial protein concentration, resulting in an increase in the sum of driving pressures for fluid and protein flow into the lungs. This theoretical mechanism of increased pressure edema has been the subject of study in experimental animals, with contradictory results.34 When plasma protein osmotic pressure is low, the ability of the osmotic pressure gradient to widen in response to increased hydrostatic pressure is diminished, and edema accumulates at hydrostatic driving pressures lower than those needed to cause edema when protein concentration is normal. Alveolar Barrier Function.  If interstitial hydrostatic pressure was raised or if alveolar hydrostatic pressure or the osmotic pressure difference across the alveolar barrier was lowered, driving pressure for fluid and protein flow across the alveolar barrier would be elevated, resulting in increased pressure edema. Interstitial hydrostatic pressure rises as interstitial edema accumulates in the lungs.4,35 Increased interstitial hydrostatic pressure would raise the sum of driving pressures across the alveolar barrier and could drive edema formation across the alveolar or airway epithelium. Greater pressure filtration across the alveolar barrier would also result from an increase in the sum of driving pressures if alveolar hydrostatic pressure was lowered. This is complicated by the interrelation between alveolar and interstitial hydrostatic pressures.4,36 A drop in alveolar hydrostatic pressure, which increases the sum of driving pressures across the alveolar barrier, also results in a lowering of interstitial hydrostatic pressure, which decreases the sum of driving pressures across the alveolar barrier but increases the sum of driving pressures across the microvascular barrier. Administration of a detergent aerosol resulted in loss of surfactant activity, higher alveolar surface tension, lower static compliance, atelectasis, and pulmonary edema; low protein concentration in alveolar edema fluid and lefthilar afferent lymph relative to plasma protein concentration indicated that this was an increased pressure type of pulmonary edema.37 Because increased pressure edema can impair surface activity of dog lung extracts and isolated rabbit lungs, it is possible that changes in alveolar surface

tension may accelerate edema formation. However, the notion that changes in alveolar surface tension in edema lead to a self-perpetuating vicious circle of edema formation is not borne out clinically. The only kind of clinical pulmonary edema caused by transmural osmotic pressure differences is near-drowning.38 Seawater is three times more hyperosmotic (1000 mOsm) than plasma, so the volume of fluid in the air spaces after saltwater aspiration increases threefold to reach osmotic equilibrium, thus markedly increasing the alveolar edema already present owing to the volume of aspirated seawater itself in the alveoli. Osmotic equilibrium is reached in minutes as water is drawn from neighboring blood vessels into the alveoli by osmotic pressure.39 Alveolar barrier function is not significantly compromised (unless perhaps the patient aspirates gastric contents or the seawater is contaminated or rich in particulate matter), and the alveolar edema is cleared rapidly (50% to 60% of excess alveolar fluid is cleared in 4 hours). Freshwater near-drowning proceeds in the opposite fashion: osmotic equilibrium is reached rapidly by flow of water out of the alveoli into the interstitium and bloodstream. Rapid water flux and hypotonicity can cause severe hemodilution, with hemolysis and fibrinolysis, as well as severe distortion of pulmonary ultrastructure, including damage to type I and type II cells, endothelial cell swelling, basement membrane detachment, and cell disruption. Both the alveolar epithelium and microvascular endothelium can thus be injured by the hypotonic fluid, leading to increased permeability pulmonary edema rather than a normal barrier type of pulmonary edema.

INCREASED PERMEABILITY EDEMA Increased permeability pulmonary edema (Fig. 62-4) is caused by an increase in liquid and protein conductance across the barriers in the lungs. The essential feature of this edema is that the integrity of the barriers to fluid and protein flow into the lung interstitium and alveoli is altered from lung parenchymal damage. Increased permeability

Figure 62-4  Frontal chest radiograph in a patient with increased permeability edema. Bilateral consolidation, more prominent on the left, is present. The heart is not enlarged and no pleural effusions are present. No features typical of volume overload are evident. (Courtesy Michael Gotway, MD.)

1102 PART 3  •  Clinical Respiratory Medicine

edema is sometimes called noncardiogenic pulmonary edema, and the resulting clinical syndromes in humans—when explicitly defined (see later)—are referred to as acute lung injury (ALI) or, when severe, the acute respiratory distress syndrome (ARDS).40

Pathophysiology If the rate of fluid and protein accumulation from lung endothelial and epithelial barrier injury exceeds the rate at which it can be removed, increased permeability edema results. Because the barriers limiting fluid and protein flow into the lungs do not function normally when the lungs are injured, the lungs are not protected against edema by the usual safety factors. Although increases in fluid and protein filtration across the barriers are removed by lymphatics and drained away from the alveolar walls as in increased pressure edema, much more fluid and protein are filtered at any given sum of driving pressures because the barriers to their flow are much less restrictive than normal. Edema formation in injured lungs becomes extremely sensitive to driving pressures.22 Driving pressures are often increased when the lungs are injured because of the vasoconstrictive effects of inflammatory mediators such as thromboxanes, which may shift the main site of resistance to postcapillary venules, thus increasing hydrostatic pressure at the microvascular fluid exchange sites,67 or because of effects on the heart as well as on the circulation. For example, elevated left atrial pressure, pulmonary venoconstriction, and an increase in cardiac output in sepsis can increase hydrostatic pressure at the microvascular fluid exchange sites.41 Because the endothelial-epithelial barrier becomes leaky, protective protein osmotic pressure differences are lost across them, driving pressure is unopposed by protein osmotic pressure, and even normal hydrostatic pressure results in significant fluid and protein extravasation into interstitial and alveolar spaces. The ability of the lymphatics to pump the excess filtrate away is increased when the lungs are injured. Maximal lung lymph flow increases more when the microvascular wall has been injured than when hydrostatic pressure alone is increased, but even this augmented lymphatic pumping capability is taxed at low driving pressures. If the epithelial barrier is injured, edema may accumulate readily in the alveoli, because most of the resistance to fluid and protein flow into the alveoli resides in the epithelial barrier.42 Increased permeability edema is often rapid in onset and progression because injured barriers offer much less resistance to flow and because hydrostatic driving pressure is unopposed by increases in osmotic pressure difference. Clinically, patients with increased permeability edema usually have a low intravascular hydrostatic pressure, commonly measured as a low or normal pulmonary capillary wedge pressure. In some cases, this reflects the low intravascular pressures associated with the underlying disease process (e.g., sepsis). The consequences of increased permeability edema on lung mechanics and gas exchange depend on how much edema accumulates and how severe the causative lung injury is.43 As with increased pressure edema, the major effects on pulmonary mechanics follow alveolar flooding. In experimental lung injury, functional residual capacity decreases as a consequence of alveolar flooding, and this loss of ventilated units accounted for virtually all the

observed decrease in static lung compliance.44 Computed tomography (CT) has provided new insights into structurefunction relationships in human ALI.45 In its early stage, when alveolar edema predominates, the lungs are characterized by a homogeneous alteration of vascular permeability, and edema accumulates evenly in all lung regions with a nongravitational distribution. Increased lung weight due to edema causes collapse of lung regions along the vertical axis through the transmission of hydrostatic forces (compression atelectasis, caused by the weight of edema). Thus lung volume is lost mainly in the dependent lung, where the superimposed weight from above is greatest. Measurements of pulmonary mechanics in mechanically ventilated patients with diffuse parenchymal lung damage showed decreased static lung compliance as a consequence of loss of ventilated lung. In addition, airflow resistance was increased as a result of decreased lung volume.43 Bronchospasm may add to the increase in airflow resistance and can be substantially reversed by bronchodilator inhalation.46 Chest wall compliance was reduced, probably because of alterations of intrinsic mechanical properties of the chest wall by abdominal distention, chest wall edema, and pleural effusion.43 Different respiratory mechanical abnormalities and responses to positive end-expiratory pressure (PEEP) during mechanical ventilation were reported in patients with severe increased permeability edema originating from pulmonary disease (pneumonia with consolidation) or from extrapulmonary disease (associated with pulmonary edema and alveolar collapse).47 Although the effects of surface forces on decreased lung compliance in patients with diffuse lung damage were thought to be small, results of experiments in isolated rabbit lungs indicated that increased permeability edema may result in more severe mechanical changes than equivalent degrees of increased pressure edema. Surfactant is thromboplastic, and coagulation may compound surfactant depletion when plasma proteins enter the air spaces. The injured lung may release substances that interfere with the normal low surface tension in the alveoli,48 and activated polymorphonuclear leukocytes (PMNs) impair surfactant function in vitro and degrade the major surfactant apoproteins by proteolysis and oxidant-radical–mediated mechanisms.49 Human lung surfactant obtained by bronchoalveolar lavage (BAL) of patients at risk for diffuse lung damage and patients with established injury is abnormal in chemical composition and functional activity.50 Abnormalities also could be caused by interactions between surfactant and edema proteins, because plasma proteins (especially fibrin monomers but also fibrinogen and albumin) interfere with surfactant function. Proteinaceous edema fluid has been associated with surfactant inhibition in various experimental models.51 The role of surfactant in the development and treatment of diffuse parenchymal lung damage and the potential role for surfactant therapy are discussed in more detail later in this chapter. Gas exchange is often severely compromised in increased permeability edema, owing both to intrapulmonary shunting of blood and to ventilation-perfusion inequalities.52 Patients with early lung injury typically have a marked increase in their pulmonary dead-space fraction, indicating that many ventilated lung units are not well perfused, although intrapulmonary shunting may also contribute to

62  •  Pulmonary Edema 1103

the elevated dead space53; this finding explains why minute ventilation rises to twice normal (12–16 L/min) at the onset of severe increased permeability edema. An elevated pulmonary dead-space fraction also has been reported in pediatric patients with widespread lung injury,54 the mechanism for which may in part be explained by an increase in procoagulant and antifibrinolytic pathways in parenchymal damage. The physiologic abnormalities associated with increased permeability edema are dominated by early alveolar flooding and depend on the severity and the duration of injury as well as its cause. Manifestations may resolve or worsen, but typically evolve in three pathologic patterns: exudative, proliferative, and fibrotic, usually in sequence. The earliest changes are marked by widespread alveolar and interstitial edema and hemorrhage. Injury to alveolar ducts may be particularly severe. Hyaline membranes, composed of precipitated plasma proteins, fibrin, and necrotic debris, can be seen. The alveolar epithelium may be more extensively damaged than the vascular endothelium, even if the underlying insult is blood-borne. Widespread, local areas of alveolar destruction, particular of type I alveolar epithelial cells, alternate with normal-appearing alveoli. The injured alveolar epithelium is swollen, disorganized, discontinuous, and often lifted off the exposed, but usually intact, basement membranes, which are covered by hyaline membranes. Type II cells are nearly always less severely damaged than type I cells, because their thin squamous cytoplasmic extensions, distant to the nucleus covering the thin side of the alveolar-capillary barrier, are frequently most gravely affected. The interstitium is widened by edema (especially in peribronchovascular cuffs) and may have leukocytes, platelets, red blood cells, fibrin, and debris (especially near the alveolar walls). The microvascular endothelium is often relatively preserved, usually showing little other than irregular focal thickening as a result of cytoplasmic swelling or vacuoles and greater numbers of luminal leukocytes, although frank swelling of endothelial cells may be seen on ultrastructural histology. The exudative phase is followed by a proliferative phase, which begins within the 5 to 7 days after the onset of injury.55 The relative contributions of the original insult, repair processes, and effects of therapies to this and subsequent phases are not well known, but some of the abnormalities after the initial exudative phase were related to the effects of traditional modes of mechanical ventilation that used tidal volumes between 12 and 15 mL/kg predicted body weight.56 In the proliferative phase, some of the edema fluid has been reabsorbed from the air spaces. Fibrin may be prominent in alveoli and interstitium, and there is infiltration with inflammatory cells and fibroblasts. The alveolar epithelium is often cuboidal, made up largely of proliferating type II cells. The air-blood barrier can be thickened by interstitial and epithelial enlargement. The pulmonary vascular bed may be partially or completely disrupted, and structural alterations may reduce its surface area. A final stage may follow, often about 10 to 14 days after the initial insult, in which fibrotic changes of the alveolar ducts, alveoli, and interstitium predominate: alveoli may be obliterated, alveolar walls coalesced, and functional lung units lost. Less commonly, the lungs show emphysema-like bullous changes.55 Pulmonary function test results in

Table 62-4  Clinical Disorders Associated with Increased Permeability Edema Infections Aspiration Trauma Hemodynamic disturbances Drugs, medications Hematologic disorders Neurologic disorders Miscellaneous disorders

5-year survivors of severe increased permeability edema usually return to normal or near-normal values, but exercise limitation and both physical and psychological quality of life are apt to remain compromised.57,58

Mechanisms The major types of clinical conditions that have been associated with increased permeability edema are listed in Table 62-4. The most common causes are pneumonia, sepsis, gastric aspiration, and major trauma. The lungs are injured via either the airways or the bloodstream. The exact mechanisms by which diffuse lung injury leads to increased permeability edema have been the subject of intense investigation in humans, animal models, and cellular systems.56 Human studies have provided descriptive data about events in the air spaces before and after the onset of lung injury. Studies using BAL in patients before and after the onset of diffuse lung damage have shown that there is a major acute inflammatory response that begins before lung injury is clinically recognized, peaks during the first 1 to 3 days of clinical involvement, and then resolves slowly over the next 7 to 14 days in patients who remain intubated.57,59 These studies have shown the complexity of the evolving inflammatory responses, which are characterized by the accumulation of acute response cytokines and their inhibitors, oxidants, proteinases and antiproteinases, lipid mediators, growth factors, and collagen precursors involved in the repair process.57,60-66 Extensive efforts have been made to find single biologic markers that predict the onset or the outcome of diffuse parenchymal lung injury, but these have met with only limited success.67,68 Hypotheses about mechanisms of lung injury have been tested in animal models and in vitro studies, and several reviews have summarized the findings.56,69 The existing animal models do not completely reproduce all of the various aspects of different injuries in humans, in part because human injuries typically evolve over a longer period of time than can be studied in the laboratory. In addition, the lungs of humans are exposed not only to the initial injurious insult but also to the therapies that are used for treatment, such as mechanical ventilation. Experiments with isolated cells have been useful to test specific concepts, but the complexity and redundancy of intact biologic systems are not reproduced in simplified experimental systems. Most experimental work purposely limits a study to a single causative agent; however, this turns the reality of clinical complexity into the simplicity of a single experimental pathway. Increased permeability edema in humans is likely to be caused by interactions between a number of different pathways acting in parallel or in series.

1104 PART 3  •  Clinical Respiratory Medicine

Studies in isolated organs and small animals in which hemodynamic variables are not measured can be difficult to evaluate because indices of lung injury, usually measured by the appearance of markers in lungs, lavage fluid, or perfusate, are not determined solely by the barrier function of the microvasculature. For example, when the vascular endothelium is injured, movement of fluid and protein from the vascular space into the lungs is extremely sensitive to hydrostatic driving pressures and surface area for filtration,22 and the effects of experimental interventions may be caused by changes in these parameters and not solely by changes in microvascular barrier function. The effects of microvascular driving pressures and surface area can be difficult to evaluate even in large, instrumented animals. Data from experimental animal models suggest that there are at least two broad categories of mechanisms of increased permeability edema: those that are direct (i.e., not requiring intermediary mechanisms, with injury a direct result of contact between an offending substance and lung tissue) and those that are indirect (i.e., requiring the participation of intermediary mechanisms, such as host defenses). These categories overlap because, once the lungs are injured, inflammatory responses may compound the primary mechanism of injury. Three major hypotheses about the mechanism of increased permeability pulmonary edema have been proposed, and are interrelated. A recent review provides specific information on animal models of experimental lung damage and an American Thoracic Society Consensus Conference provided further recommendations.69

DIAGNOSIS The diagnosis of moderately and, especially, severely advanced pulmonary edema, particularly when caused by heart failure, is usually fairly easy. Unraveling the cause of other kinds of pulmonary edema may not be so straightforward, particularly in patients with increased permeability edema.

term ALI could be “applied to a wide spectrum of this continuum of pathologic process so as to acknowledge and define it”71; moreover, the term ARDS continued to represent the most severe end of the ALI spectrum of damage. To satisfy clinical and epidemiologic needs, three different definitions of lung damage have been widely used, but each has its shortcomings: the Acute Lung Injury Score,70 the American-European Consensus Conference (AECC) definition,71 and a recent revision of the AECC definition termed the “Berlin Definition of ARDS.”72 The Lung Injury Score (Table 62-5) provides an assessment of the severity of lung injury, taking into account supportive therapy, such as mechanical ventilation with PEEP and oxygen supplementation.70 This score is important because not all lung injuries are of equal severity and severity changes over time. In this scoring system, ARDS defines only the most severe injuries (those that yield a score >2.5); milder lung injuries, termed mild-to-moderate, may have a better prognosis and may differ from ARDS in other important aspects. This scoring system has been used widely in clinical research and clinical trials. The American-European Consensus Conference Definition71 (Table 62-6) has four elements: (1) timing (onset

Table 62-5  Components and Individual Values of the Lung Injury Score 1.  CHEST RADIOGRAPH SCORE No alveolar consolidation Alveolar consolidation confined to 1 quadrant Alveolar consolidation confined to 2 quadrants Alveolar consolidation confined to 3 quadrants Alveolar consolidation in all 4 quadrants

VALUE 0 1 2 3 4

2.  HYPOXEMIA SCORE PaO2/FIO2 ≥ 300 PaO2/FIO2 225–299 PaO2/FIO2 175–224 PaO2/FIO2 100–174 PaO2/FIO2 < 100

0 1 2 3 4

3.  PEEP SCORE (WHEN VENTILATED)

CLINICAL ASSESSMENT Definitions Newcomers interested in the history of catastrophic lung disease should remember the clinical anecdotes that began to surface as far back as the 1950s, 1960s, and 1970s showing that little by little survival from what had been known as “shock lung,” “Danang lung,” and “adult respiratory distress syndrome” represented a gigantic medicaltechnical-operational breakthrough. Previously, virtually all such egregiously wounded or injured casualties died. (The name adult changed to acute respiratory distress syndrome, but the familiar acronym ARDS remained.) Subsequently, the number of survival miracles steadily increased, but mortality remained high. Conceptually, ALI comprise a continuum of lung damage ranging from trivial to severe or lethal. For decades and by general agreement, the term “ARDS” has defined the most grievously afflicted victims of the syndrome. In 1988, an “expanded definition” with numerical grading of ARDS was proposed,70 and 6 years later it was recognized that the

PEEP ≤ 5 cm H2O PEEP 6–8 cm H2O PEEP 9–11 cm H2O PEEP 12–14 cm H2O PEEP ≥ 15 cm H2O

0 1 2 3 4

4.  RESPIRATORY SYSTEM COMPLIANCE SCORE (WHEN AVAILABLE) Compliance ≥ 80 mL/cm H2O Compliance 60–79 mL/cm H2O Compliance 40–59 mL/cm H2O Compliance 20–39 mL/cm H2O Compliance ≤ 19 mL/cm H2O The final value is obtained by dividing the aggregate sum by the number of components that were used. No lung injury Mild to moderate lung injury Severe lung injury (ARDS)

0 1 2 3 4

SCORE 0 0.1–2.5 >2.5

ARDS, acute respiratory distress syndrome; PaO2/FIo2, arterial oxygen tension to inspired oxygen concentration ratio; PEEP, positive end-expiratory pressure. From Murray JF, Matthay MA, Luce JM, et al: An expanded definition of the adult respiratory distress syndrome. Am Rev Respir Dis 138:720–723, 1988.

62  •  Pulmonary Edema 1105

Table 62-6  American-European Consensus Conference Definition of Acute Lung Injury and Acute Respiratory Distress Syndrome TIMING Acute OXYGENATION (REGARDLESS OF PEEP LEVEL) PaO2/FIO2 ≤ 300 = acute lung injury PaO2/FIO2 ≤ 200 = acute respiratory distress syndrome CHEST RADIOGRAPH Bilateral opacities on frontal view PULMONARY ARTERY WEDGE PRESSURE <18 mm Hg when measured, or No clinical evidence of left atrial hypertension

Table 62-7  Berlin Definition of ARDS Mild

PaO2/ FIO2

200–300 mm Hg

Moderate Severe ARDS

PaO2 /FIO2 PaO2/ FIO2

100–199 mm Hg <100 mm Hg

must be “acute”); (2) oxygenation (arterial PO2/FIO2 [fractional concentration of oxygen in inspired gas] <300 mm Hg, regardless of the level of PEEP, for the diagnosis of ALI to be made, and arterial PO2/FIO2 ≤ 200 mm Hg, regardless of PEEP level, for the diagnosis of ARDS to be made); (3) chest radiograph (bilateral opacities seen on frontal view); and (4) pulmonary artery wedge pressure (<18 mm Hg when measured, or no clinical evidence of left atrial hypertension). The third explication, the Berlin Definition,72 simply subdivides ARDS into three categories of severity: mild, moderate, and severe, based on the arterial PO2/FIO2 ratio (Table 62-7). Initially, so-called ancillary variables—severity of chest radiographic findings, minimum PEEP level, respiratory system compliance, and standardized minute volume— were considered for inclusion, but finally discarded. Of interest, the term “acute lung injury” was abandoned. Mortality increases successively as arterial PO2/FIO2 ratio worsens in mild, moderate, and severe ARDS. Another recent method of defining and grading the severity of clinically meaningful lung damage, first in a single center73 and then in 21 others,74 is what has been called the Lung Injury Prediction Score, or LIPS, whose goal is to improve the early identification of patients at high risk for development of serious early or impending lung injuries. Both risk factors (predisposing conditions, such as pneumonia, severe sepsis, trauma, and aspiration) and risk modifiers (e.g., alcohol abuse, hypoalbuminemia, and use of supplementary oxygen), have each been graded numerically to yield a combined LIPS value, which has an impressively high negative predictive value (0.96 to 0.98) but a much lower positive predictive value (0.14 to 0.23).75 A more recent study showed that diffuse alveolar damage was observed in fewer than half the patients having clinical criteria for ARDS, but was more frequent (69%) in those with ARDS lasting more than 72 hours.76 The current working definitions all have useful attributes, but fail to link any particular clinical feature to any

particular change in the structure or function of the relevant barriers in the lungs or to the degree of pulmonary edema. What defines the ALI-ARDS continuum in a meaningful way is the altered barrier permeability to protein in the lungs, the structural damage to the lung microvascular endothelial and alveolar epithelial barriers, and the consequent excess lung water content. In the future as research studies progress, these clinical definitions may need to be supplemented with biologic markers (see next section) and pathologic findings (when available) to help categorize ALI and ARDS into more specific disease entities.

Symptoms and Signs The clinical manifestations of pulmonary edema vary with its severity and depend on the underlying pathophysiology and the extent to which excess edema fluid has accumulated in the lungs. Characteristic symptoms comprise dyspnea, cough, and tachypnea. Wheezing, when audible, may present a problem in differential diagnosis, but patients with typical asthma generally do not have other symptoms and signs of congestive heart failure or pulmonary edema.77 Once alveoli have flooded, the diagnosis of pulmonary edema is not subtle. Patients with alveolar edema usually have severe respiratory distress with tachypnea and cough that is often productive of frothy and sometimes bloodtinged edema fluid. Crackles and rhonchi are heard over the lung fields and wheezing may be present. The patient may be cyanotic if alveolar flooding has seriously compromised gas exchange. Development of pulmonary edema is often slow and progressive in increased pressure pulmonary edema because the alveoli are protected by the normal safety factors (see Table 62-1). In contrast, in increased permeability edema, alveolar flooding and symptoms of respiratory distress often happen rapidly. Edema that develops suddenly (or unexpectedly) sometimes is called “flash pulmonary edema,” which usually pertains to the rapid development of high-pressure edema, after protective safety factors have been surmounted. Because pulmonary edema is always a sign of an underlying pathologic process, its cause must be identified so that effective therapy can be directed at the underlying problem producing abnormal transvascular fluid and solute flow into the lungs. Increased pressure edema is most often caused by cardiac failure from systolic dysfunction with impaired myocardial contractility and thus is usually accompanied by a history of heart disease; manifestations include signs and symptoms of any of the many causes of chronic and acute congestive heart failure, such as coronary insufficiency, hypertension, valvular heart disease, and severe volume expansion. Elevated jugular venous pressure, cardiac enlargement, gallop rhythms, heart murmurs, arrhythmias, large tender liver, and peripheral edema almost always suggest an underlying abnormality of cardiac function. However, pulmonary edema may be the only manifestation of silent myocardial infarction or diastolic dysfunction of the left ventricle.78 History and physical examination may also be helpful in differentiating between increased pressure and increased permeability pulmonary edema, because most patients in the latter category usually do not have signs or symptoms of underlying cardiac disease. The cause of increased

1106 PART 3  •  Clinical Respiratory Medicine

permeability edema may be suggested by a history of exposure (e.g., to toxic gases or chemicals, near-drowning, drug ingestion, trauma), the clinical setting (e.g., sepsis, pneumonia, emesis, seizures, pancreatitis), or the physical findings (e.g., chest trauma, long bone fractures, coma, shock). Because infections, including the sepsis syndrome, are the leading causes of increased permeability edema in patients, a thorough search must be made for signs and symptoms of infection. Pulmonary and intra-abdominal sources are the most common sites of involvement, and all patients should be examined carefully, with special attention being paid to abdominal, rectal, and pelvic examinations.

Diagnostic Studies Laboratory and other diagnostic studies are often helpful but, by the time many of the results are found to be abnormal, the diagnosis is usually obvious. Appropriate cultures for microorganisms and toxicology screens of blood and urine are useful in identifying underlying causes of increased permeability edema. Examination of sputum or tracheal aspirate, bronchoscopy with protected-specimen brushing, or mini-BAL are all useful in diagnosing pneumonia in ventilated patients, even those who are being treated with antimicrobial drugs.79-82 Lung biopsy can sometimes provide a specific diagnosis in critically ill patients,83 but the results often are not helpful because lung injuries caused by diverse underlying conditions have similar histologic appearances and because specific therapies may not be available. In the special case of pulmonary edema suspected to be caused by saltwater near-drowning, measurement of plasma magnesium level can help determine whether a patient has aspirated or swallowed seawater, or both. Severe hypermagnesemia has been reported following aspiration of ordinary seawater and, especially, highly concentrated saltwater from the Dead Sea.84 Chest Radiographs The plain chest radiograph is the most practical laboratory study available for the detection of pulmonary edema.85,86 Disadvantages are that chest radiographs are insensitive to small changes in lung water and are only semiquantitative.1 An additional limitation is that chest radiographs are not consistently helpful in distinguishing increased pressure edema from increased permeability edema.85,87 These disadvantages are offset by the advantages that chest radiographs are noninvasive, inexpensive, easily repeatable, readily available, and free of serious side effects (apart from a small amount of radiation). Before alveolar flooding, plain chest radiographs typically show distended vascular shadows (particularly in the upper lung fields), enlargement and loss of definition of hilar structures, development of septal lines (Kerley lines) (Fig. 62-5; Video 62-1, loss of peribronchial and perivascular definition or cuffing) (Fig. 62-6), and perihilar haze indicating the presence of interstitial pulmonary edema. Acinar shadows, often confluent and creating irregular, patchy increases in lung density that obscure vascular markings, indicate the presence of alveolar edema. Air bronchograms may be observed in severe edema. Because the radiographic signs of interstitial and alveolar edema are determined by gas and blood volumes and their distribution in the lungs

Figure 62-5  Frontal chest radiograph in a patient with increased pressure pulmonary edema from heart failure showing alveolar edema and bilateral consolidation. Note presence of air bronchograms. (Courtesy Michael Gotway, MD.)

Figure 62-6  Frontal chest radiograph in a 56-year-old woman with increased pressure pulmonary edema from chronic heart failure showing interlobular septal thickening. Thin linear opacities are best seen in the inferolateral thorax, particularly on the right, representing thickened interlobular septae, or Kerley B lines. See the chest CT scan for this patient in Video 61-1. (Courtesy Michael Gotway, MD.)

in addition to the presence of edema, the recognition and quantitation of edema are not precise, and the radiographic appearance of edema is strongly influenced by the lung volume at the time the film is made. The chest radiograph score is an integral part of the Lung Injury Score and the revised Berlin Definition, but the interpretation of chest radiographs is not well standardized and significant interobserver variations have been reported.88 One recent approach for scoring the chest radiograph and accounting for atelectasis correlated well with lung weight in lungs that were studied from brain-dead potential organ donors.89

62  •  Pulmonary Edema 1107

Arterial Blood Studies The arterial PO2, arterial PCO2, and pH are the most informative laboratory indicators of overall pulmonary function in patients with pulmonary edema. Arterial blood studies are not sensitive to early edema. Interstitial pulmonary edema does not usually affect oxygen uptake in the lungs beyond modest hypoxemia caused by ventilation-perfusion mismatching. In contrast, alveolar flooding seriously compromises gas exchange, resulting in right-to-left shunting of blood from ongoing perfusion of alveoli that cannot be ventilated because they are fluid filled or collapsed. In two studies of groups of patients with increased permeability edema from lung injury, oxygenation appeared to depend more on the vasoconstrictive ability of the pulmonary circulation—owing to the ability to reduce perfusion of damaged and edematous areas of the lungs—than on the amount of edema present.90,91 In patients hospitalized for acute cardiogenic pulmonary edema, arterial PCO2 may be low, especially in the early stages, when tachypnea results in alveolar hyperventilation; arterial PCO2 may also be within the normal range or elevated, the latter indicating alveolar hypoventilation, which can be caused by underlying lung disease, increased metabolic production of carbon dioxide (perhaps related to increased work of breathing), increased wasted ventilation (ventilation of poorly perfused alveoli), or mechanical impairment caused by weak respiratory muscles.92 An elevation in the pulmonary dead space fraction in the first 24 hours after development of severe increased permeability edema identifies patients with a higher risk for survival, particularly if the dead space fraction is greater than 0.60.53 When pulmonary edema is severe or the lungs have been injured, many patients develop metabolic acidosis as a result of tissue hypoxia, increased work of breathing, intrinsic lung lactate production, or all of these.93 Attempts to correct acidosis with parenteral bicarbonate administration usually are not necessary; rather, the underlying cause must be identified and treated appropriately. Maintenance of a satisfactory systemic blood pressure is crucial. Respiratory acidosis caused by alveolar hypoventilation can be treated either by noninvasive ventilation or by invasive mechanical ventilation with endotracheal intubation. Metabolic acidosis can be partially corrected by alleviating hypoxemia and improving cardiac function; the possibility of underlying disease amenable to surgery (e.g., intestinal ischemia or infarction, perforation of a viscus) or pancreatitis should be considered. Measurement of Pulmonary Edema Fluid Protein Concentration When florid pulmonary edema is present, measurement of protein concentrations in both simultaneously collected edema fluid (suctioned through an endotracheal tube) and plasma provides a rapid, noninvasive method for distinguishing increased pressure edema from increased permeability edema.94 Because the microvascular barrier is functionally intact in increased pressure edema, plasma proteins remain largely confined to the intravascular space, and edema fluid protein concentration is low relative to plasma protein concentration (the ratio of edema fluid to plasma protein concentration is generally <0.65). In con-

trast, in increased permeability edema, when the microvascular barrier is injured, plasma proteins leak in high concentrations into the vascular space, which raises edema fluid protein concentration high relative to plasma protein concentration (ratio of edema fluid to plasma protein concentration generally >0.75). Intermediate values (between 0.65 and 0.75) may indicate that both types of edema are present and suggest the relative contributions of each. Measurement of the ratio of edema fluid to plasma protein concentrations has been shown to be a simple method for separating the two different pathophysiologic types in numerous reported series of patients with pulmonary edema.95-98 Three studies indicated that an increasing protein concentration in serial measurements of edema fluid in increased permeability edema was a good sign, reflecting an intact epithelial barrier and net removal of edema fluid from the alveoli.97,98 There is new evidence that the edema fluid to plasma protein ratio has prognostic value as well as diagnostic value.99 Such measurements need to be correlated with the patient’s clinical condition, because an increasing protein concentration in edema fluid over time might also mean that increased permeability edema was complicating what had been increased pressure edema or that the lungs were injured more severely or more extensively as time passed. Edema fluid can be collected by inserting a standard 14to 18-gauge catheter through an endotracheal tube and advancing it into a wedged position in the distal air spaces (similar to the procedure for wedging a fiberoptic bronchoscope). Gentle suction is applied as the catheter is slowly withdrawn, and fluid is collected in a small trap. Several attempts may be needed; if no fluid can be suctioned, the clinician should try changing the patient’s position. Samples grossly contaminated with airway secretions, which have a very low protein concentration, less than 1 g/dL, such as mucus, pus, and debris, should be discarded. The protein concentration in the edema fluid and the plasma can be measured by the clinical laboratory or estimated quickly at the bedside from the protein scale of a handheld refractometer. Standard BAL using fiberoptic bronchoscopy or miniBAL (with a wedged suction catheter) have been used as research and diagnostic tools that may also yield useful information about the cellular biochemical and microbial composition of the air space,100 but lavage is not useful as a method to measure alveolar protein concentration, because the instilled saline dilutes alveolar fluid by approximately 50- to 100-fold, depending on the method.67

MEASUREMENT OF LUNG WATER AND BARRIER FUNCTION In theory, measurement of the quantity of water or edema fluid in the lungs could be useful in detecting early pulmonary edema and in assessing its clinical course and response to treatment; however, no optimal technique is available. Methods in use or under investigation focus on either measurement of lung density or equilibration of tracers with water in the lungs.1,101 Interest in such measurements assumes that accurate knowledge about lung water content would be useful in diagnosis and would be beneficial in the treatment of patients with pulmonary edema. Recent work

1108 PART 3  •  Clinical Respiratory Medicine

with a single thermal indicator for measuring extravascular lung water has shown promise based on some studies, although the actual clinical utility has not been convincingly proven.102

Barrier Function Clinical distinction between increased pressure edema and increased permeability edema is difficult,103,104 because transvascular fluid flow into the lungs can be abnormal long before lung water content increases greatly. In theory, the two types of edema can be separated on the basis of differences in barrier function. Detection of increased transvascular fluid flow into the lungs and measurement of barrier integrity might be more helpful than measurement of lung water content in edema. A simple and practical method exists to evaluate barrier integrity (edema fluid protein concentration measurements, discussed earlier), and several methods have been studied to detect early edema and changes in barrier function. However, not one is in routine clinical use. Biologic Markers of Lung Injury Potential biologic markers of imminent increased permeability edema from various kinds of lung injury have been the subject of many studies and comprehensive reviews.67,105,106 Naturally, considerable interest revolved around finding a simple blood, urine, or BAL test that could identify patients who are destined to develop or who are already in the earliest stages of increased permeability edema, or that could predict the outcome of patients with injured lungs. To be of clinical use, such a marker would have to be both practical and inexpensive to measure as well as sensitive and specific for the detection of lung injury. An extensive search has been made for a reliable biologic marker for the detection of early or impending lung injury comparable to the decisive clinical role played by troponin measurements for diagnosing acute myocardial infarction. The long-sought diagnostic nirvana for increased permeability edema remains distant. Several studies from multicenter clinical trials have reported the independent predictive value of some plasma markers for mortality and other clinical outcomes in patients with increased permeability edema from various lung injuries. The biologic markers of greatest predictive value are surfactant protein D,107 interleukin-6 and interleukin-8,108 von Willebrand factor antigen,109 soluble tumor necrosis factor-α receptors I and II,110 intercellular adhesion molecule-1,111 protein C and plasminogen activation inhibitor-1,112,113 and the receptor for advanced glycation end products.114,115 Because increased permeability edema follows a wide variety of insults that range in severity, and because many abnormalities detected in increased permeability edema are found in other severe illnesses of diverse etiology that do not involve the lungs, it seems unlikely that any single marker will be found that unequivocally identifies the risk or the presence of severe lung injury. Increased attention is being given to the sensitivity and specificity of combinations of markers. In support of this approach, one recent study found that combinations of three plasma biomarkers had a statistically significant better prognostic value for predicting death than the use of standard clinical predictors alone.68

Currently, the latest intriguing biologic marker of increased permeability edema, angiopoietin-2, an endothelial growth factor and potent regulator of vascular permeability, plays a role in several different clinical conditions, including malignancies, liver failure, malaria, chronic kidney disease, heart failure, and sepsis. Heightened interest in angiopoietin-2 followed the observation that levels in patients with sepsis were “markedly elevated within the first hour of clinical care,” correlated with disease severity, rose further in patients who died, and were predictors of shock or death.116 Moreover, these observations have been broadened by using angiopoietin-2 values to predict the development of increased permeability edema from lung injuries in a wide spectrum of critically ill patients; combining angiopoietin-2 levels with the Lung Injury Prediction Score (described previously) improved the resulting receiver operating curve characteristics compared with each curve separatelely.75 Clearly, additional studies of larger numbers of patients and outcomes are needed to bolster these observations.

TREATMENT Treatment of pulmonary edema often requires vigorous life-saving measures, followed by specific therapy directed at the factors that led to accumulation of water in the extravascular spaces of the lungs. Rational therapy also requires an accurate diagnosis and an understanding of the nature of the underlying disease state and of the strategies that might prove useful in limiting further edema accumulation and that favor fluid removal from the lungs.117,118

EMERGENCY THERAPY Patients with alveolar edema are frequently severely ill and require immediate treatment for acute respiratory failure. The basic principles of treatment of hypoxemic respiratory failure are discussed in Chapter 100. Essential requirements for patients with pulmonary edema include preservation of the airway, and provision of satisfactory alveolar ventilation and arterial blood oxygenation. Maintenance of arterial blood pressure is indispensable. Reliable means of monitoring and sustaining oxygen saturation and blood pressure in rescue operations and ambulance travel are now widely used. Ventilatory management in patients with florid alveolar flooding and severe gas exchange abnormalities, regardless of cause, requires emergent endotracheal intubation, high inhaled oxygen concentrations, and PEEP. After stabilization, obligatory lung-protective strategies are introduced in patients with increased permeability edema (see later discussion); in contrast, increased attention is being directed at noninvasive ventilatory methods (discussed later) that hasten improvement in respiratory distress and metabolic disturbances in patients with increased pressure edema, nearly always from congestive heart failure, but have no effect on short-term mortality.

INCREASED PRESSURE EDEMA In increased pressure pulmonary edema, the common goal of therapy is to reduce the transudation of fluid into the

62  •  Pulmonary Edema 1109

lungs. Because the rate of edema formation increases exponentially with increases in pulmonary vascular pressures,7 hydrostatic pressure control is crucial for successful therapy. Therapy must be clearly goal oriented, responsive to the underlying pathophysiology, and frequently reassessed until the patient is stable.

General Principles The acute recognition and management of increased pressure pulmonary edema, exemplified by congestive heart failure, has been the subject of comprehensive reviews.5,6,8,119,120 With increased pressure edema, the goal of therapy is to reduce the hydrostatic pressure causing edema formation in the lungs. The major objective is to achieve a net negative fluid balance without adversely affecting myocardial performance. The work of the heart must be reduced as much as possible by restricting physical activity and preventing pain and anxiety, which act to increase cardiac work by increasing sympathetic tone. As the heart fails, cardiac performance is reflexly preserved by progressive increases in vascular volumes that act to increase cardiac stroke volume and work (the FrankStarling mechanism). This compensatory increase in preload of the left ventricle results in pulmonary venous hypertension and raises the driving pressure for fluid filtration out of the pulmonary microcirculation. As heart failure worsens, cardiac output falls, pulmonary and systemic venous pressures rise, systemic vascular resistance increases, and edema, in the lungs and in the periphery, becomes the major manifestation of compromised cardiac function. In patients with increased pressure edema, a reduction of vascular volume and an increase in cardiac output decrease the driving pressure for edema formation. Because the normal safety factors protecting the lungs from edema driven by high filtration pressures are intact, the pressure need be lowered only toward normal. At pulmonary capillary wedge pressures less than 20 mm Hg, fluid filtration in the lungs usually should not be sufficient to cause pulmonary edema. Patients with severe cardiac failure may tolerate higher pressures—because baseline barrier permeability is lowered in the lungs and lymphatic removal capability is increased—and may require such pressures to maintain cardiac output. Therapy is directed at reducing the work the heart must perform and at increasing the heart’s efficiency for the work it must do including, in some patients with severe hydrostatic pulmonary edema, the use of positivepressure mechanical ventilation. The resolution of alveolar edema in these patients is not simply a function of lowering lung vascular pressures; other mechanisms that augment alveolar fluid clearance are important.120 Most patients with acute pulmonary edema caused by cardiac failure have systolic dysfunction from weakened contractility, but approximately 30% of patients have diastolic dysfunction. Cardiac contraction is normal, but relaxation is impaired. Because the ventricle does not relax normally, end-diastolic pressure is increased, thereby increasing hydrostatic pressure in the lung microcirculation. Acute diastolic dysfunction producing pulmonary edema is now recognized as a common manifestation of acute myocardial ischemia or uncontrolled hypertension. Other causes include diabetes mellitus, aortic stenosis,

infiltrative cardiomyopathies, endocardial fibroelastosis, hypothermia, septic shock, elevated thoracic pressures from mechanical ventilation, and pericardial effusion. Causal or aggravating conditions should be corrected (e.g., revascularization for coronary artery disease, control of systemic hypertension). The goal of therapy in acute diastolic dysfunction is to lower elevated filling pressures (by the cautious use of diuretics and nitrates) without significantly reducing cardiac output. These patients are prone to develop hypotension in response to diuretics and nitrates because adequate cardiac output depends on elevated filling pressures in the heart. Because systolic function is normal, positive inotropic agents do not help and can actually aggravate ischemia. In the setting of acute increased pressure pulmonary edema, it is especially important to identify and treat correctable causes of heart failure. Acute myocardial infarction, ongoing myocardial ischemia, arrhythmias, valvular lesions, systemic hypertension, ventricular septal rupture, rheumatic or other inflammatory myocarditis, digitalis intoxication, pulmonary embolism, infection, thyrotoxicosis, or severe anemia may have caused the heart to fail and must be corrected. Cardiac patients presenting with pulmonary edema may not complain of chest pain, but most of them have significant coronary artery disease, and pulmonary edema may be the only manifestation of silent myocardial ischemia.

Morphine Sulfate The sovereign emergency therapy for acute cardiogenic pulmonary edema has long been morphine sulfate, 5 to 10 mg, or its equivalent, given intravenously slowly over several minutes, taking care to avoid hypotension. Morphine is an extremely useful drug in the treatment of heart failure, because it is a potent vasodilator as well as a central nervous system sedative and because it does not depress myocardial contractility. Its vasodilating effects can substantially reduce pulmonary capillary pressure and may improve depressed cardiac output. The work of the heart is lessened by the vasodilating, bradycardic, and sedative effects of morphine. Cautiously administered morphine usually does not cause respiratory failure or aggravate existing carbon dioxide retention associated with acute pulmonary edema, but the patient must be closely watched if he or she is not intubated and receiving assisted ventilation. Hypotension following morphine administration indicates that too much drug was given or that the intravascular volume is lower than suspected; prompt administration of naloxone (Narcan) should quickly correct the disturbance. Decreasing Venous Return If sophisticated medical care is not immediately available, two useful age-old remedies for increased pressure pulmonary edema can be lifesaving. First, rotating tourniquets can reduce intrathoracic blood volume—and thereby pulmonary perfusion pressure—by trapping blood in the extremities, away from the pulmonary circulation. Tourniquets, or blood pressure cuffs inflated to less than systolic pressure, are applied to three of the four extremities and rotated every 15 minutes. The purpose is to decrease venous return, not to stop all blood flow to the extremities. Care must be taken that the tourniquets are rotated and that

1110 PART 3  •  Clinical Respiratory Medicine

venous return from any extremity is not obstructed for more than 45 minutes at a time. A study of patients with left ventricular dysfunction following myocardial infarction showed trapping by tourniquets of considerable blood volume in the periphery, but variable and sometimes unfavorable effects on left ventricular function.121 Second, removal of 100 to 500 mL of bloody by phlebotomy122 can be used to reduce blood volume in acute pulmonary edema from congestive heart failure when the patient is not in shock and when drugs and supportive care are not immediately available.

Ventilatory Strategies Most patients hospitalized with increased pressure edema suffer from acute or chronic heart failure and require oxygen treatment to ensure satisfactory arterial blood oxygenation. Depending on the prevailing severity of oxygen deficiency, patients with mild abnormalities may require only supplementary oxygen administered through a nasal catheter or simple mask; those with moderate oxygen deficiency may be helped with noninvasive methods of oxygen administration. Finally, profound gas exchange abnormalities require endotracheal intubation with high oxygen concentrations and elevated end expiratory pressures. Today, after many studies, noninvasive ventilation—by either continuous positive airway pressure or noninvasive intermittent positive-pressure ventilation—has proved beneficial for the treatment of acute cardiogenic pulmonary edema; both methods cause more rapid improvement in respiratory distress and metabolic abnormalities than standard oxygen treatment.123 The two types of noninvasive ventilation are similar in their clinical benefits, but have no effect on short-term mortality.124 Properly used noninvasive ventilation spares many patients from the ordeal of endotracheal intubation, but when cardiogenic pulmonary edema is severe and worsening, intubation is immediately indicated.125,126 Right Heart Catheterization In 1970, Drs. HJC Swan and William Ganz127 introduced the balloon flotation catheter into clinical medicine for measurements of right atrial pressure, right ventricular pressure, pulmonary artery pressure, and pulmonary capillary wedge pressure, and for assessment of cardiac output and oxygen saturation values in the right heart chambers. Pulmonary artery catheters are reasonably easy to use but training is essential. Naturally, residents and fellows, as well as seasoned attending physicians in cardiology, critical care, surgery, and anesthesiology, joined the rush to insert SwanGanz catheters and learn about their benefits. The good news is that these same eager young trainees and veteran doctors did indeed learn a gigantic amount of cardiovascular and respiratory physiology that most doctors knew little or nothing about before the advent of pulmonary artery catheterization.128 As knowledge concerning hemodynamics became everybody’s business, new conceptual models for the management of acute myocardial infarction, heart failure, and cardiogenic shock based on readily obtained hemodynamic data produced new clinical insights and research rewards. Similar approaches explored the mysteries of sepsis, septic shock, and postoperative complications, and how to differentiate increased pressure from

increased permeability edema. Novel hemodynamic concepts were enhanced by the use of new and increasingly potent diuretics and vasoactive and inotropic drugs. The bad news, of course, was that the stampede “to swan” led to considerable overuse and abuse of pulmonary arterial catheterization.129-133 Technical competence and professional know-how sometimes suffered, and would-be experts did not always know how to interpret the information. But the Swan-Ganz bandwagon kept rolling on and did not slow down until well into the 1990s, when authorities began to question whether or not the unrestricted use of balloon flotation catheters in patients with acute myocardial infarction was warranted. Although the outcomes of early clinical trials were not definitive, they raised considerable doubts, and it was finally recognized that pulmonary artery catheterization had no place in the routine management of acute myocardial infarction.134 Subsequently, the prevailing certainty that monitoring pulmonary capillary wedge pressure, and cardiac output, rather than central venous pressure, would optimize both control of volume status and regulation of vasopressor and inotropic treatment was challenged and put to the test. A large NHLBI-sponsored clinical trial of 1000 patients concluded that “[pulmonary artery catheterization]guided therapy did not improve survival or organ perfusion and complications were higher than [central venous catheterization]-guided therapy.”135 The latest (2013) information on the use of balloon flotation catheters for adult patients in intensive care comes from the Cochrane Central Register of Controlled Trials.136 The authors concluded “that the use of a [pulmonary artery catheter] did not alter the mortality, general [intensive care unit] or hospital [length of stay], or cost for adult patients in intensive care.” As usual, experience and clinical judgment is essential. It must be said and emphasized—now that the old controversies have been largely clarified—that several incontrovertible indications for Swan-Ganz catheterization endure unchallenged and remain commonly employed, particularly in the diagnosis and monitoring of cardiovascular diseases.128 Alternative methods of hemodynamic monitoring, such as the Pulse Index Continuous Cardiac Output device,137 are being designed and evaluated, but there remains an important role for balloon flotation catheters for critically ill adult patients.

Specific Pharmacologic Therapy After emergency treatment has been instituted and the patient’s condition stabilized, three major therapeutic options—vasodilators, diuretics, and inotropic agents— need to be considered and administered when appropriate. The goal of therapy with these agents is to lower the hydrostatic pressure at the filtration sites in the pulmonary vascular bed while maintaining adequate systemic delivery of oxygen. The choice of therapy is dictated by the patient’s condition and the cause of the pulmonary edema. Vasodilators.  Depending on their specific therapeutic indications, vasodilators are used for hypertension, congestive heart failure, and angina. Several medications are available, usually in combination and usually as part of a long-term treatment regimen. Vasodilators are useful

62  •  Pulmonary Edema 1111

pharmacologic agents for the treatment of acute increased pressure pulmonary edema, because their effects happen rapidly, usually in minutes. Through dilation of veins, vascular capacitance is increased and blood is redistributed peripherally, thereby lowering the driving pressure for fluid filtration in the lungs; through dilation of arteries, systemic vascular resistance (cardiac afterload) falls, cardiac output and stroke volume increase, and the heart works more efficiently. In addition to morphine, three classes of vasodilators may be useful in pulmonary edema: venodilators (e.g., nitrates), arteriolar dilators (e.g., phentolamine, hydralazine), and mixed dilators (e.g., nitroprusside). The most common side effects of vasodilators such as dizziness, especially when standing, are related to low blood pressure. Tolerance to vasodilators may require revision of long-standing therapy. Diuretics.  Patients with symptoms of pulmonary edema, especially from increased vascular pressure, usually benefit from administration of diuretic agents, a standard means of initial therapy.138 These drugs may exert a modest immediate effect by increasing venous capacitance and decreasing the relative perfusion of flooded alveoli (acting as vasodilators), but their principal mechanism of action is to increase sodium and water excretion by the kidneys.139 The resultant diuresis causes a decrease in left ventricular volume and pressure and thereby a reduction in left atrial pressure and the pressure at the filtration sites in the lungs. Among the potent loop diuretics administered by slow intravenous injection, furosemide is highly effective and generally regarded as the agent of choice. Dozens of different dosage regimens of furosemide have been and are still being used to treat symptomatic patients with pulmonary edema from acute congestive heart failure. One such regimen consists of a loading dose of 40 to 80 mg followed by a continuous infusion at 10 to 20 mg/hr; if there is no response in an hour, the loading dose is repeated and the infusion rate is doubled. Equivalent doses of other loop diuretics (bumetanide, torsemide, or ethacrynic acid) have essentially the same effects. Decades of clinical experience have documented that intravenous administration of loop diuretics nearly always result in prompt diuresis and symptomatic relief. Nevertheless, information about safety and efficacy is lacking and optimal use of diuretics in heart failure management needs much further study and improvement.140 If the patient is hypotensive or in frank shock, diuretics are seldom of benefit, because poor renal perfusion limits any effects they might have on kidney function. In this circumstance, continuous venovenous ultrafiltration can reduce intravascular volume even in patients who require vasopressors for blood pressure support. If the patient has severely diseased kidneys, diuresis may not be an option. In this circumstance, continuous arteriovenous hemofiltration with or without counter-current dialysis141 or venovenous ultrafiltration142 should be con­ sidered. These techniques represent considerable advances over traditional hemodialysis, which was often impossible in hemodynamically unstable patients (especially those with low cardiac output or hypotension), and peritoneal dialysis, which was both slow and poorly tolerated. Hemofiltration can be instituted, maintained, and managed suc-

cessfully by well-trained intensive care unit nurses and physicians, and it is not complicated by hypotension because the circuits have small volumes and pressures are low. Considerable fluid (<200 to 300 mL/hr) can be removed when needed, with the amount being titrated to the patient’s cardiovascular status. Inotropic Agents.  Patients with cardiogenic shock and other cardiac catastrophes that lower systemic blood pressure often require inotropic agents as a temporary lifesaving measure. In quite a different therapeutic category, patients with systolic heart failure and pulmonary edema with impaired cardiac contractility were at one time also thought to benefit from inotropic agents by increasing cardiac output and lowering the driving pressure for fluid filtration in the lungs. Although the use of pharmacologic agents to improve myocardial contractility seems rational, current opinion states that “inotropic therapies in the [systolic heart failure] population have universally failed to live up to their expectations.”143 Palliative inotropic treatment has its occasional indications, but other options such as combination implantable cardioverter-defibrillator and cardiac resynchronization may also provide benefits.144 When present, increased pressure pulmonary edema is an important associated feature of impaired myocardial contractility, has proven difficult to treat, and has a poor prognosis; individualized diagnostic fine-tuning may transiently improve clinical outcome, but results remain modest at best. Older methods, however, are being supplanted by advanced techniques of coronary reperfusion, mechanical support with intra-aortic balloon pumps or ventricularassist devices, expanded use of extracorporeal membrane oxygenators for severe respiratory failure, and improvements in heart and/or lung transplantation.

INCREASED PERMEABILITY EDEMA As already emphasized, the strategy for managing patients with increased permeability pulmonary edema caused by various types of severe lung injury differs from that for patients with increased pressure pulmonary edema in two crucial respects: first, the endothelial and epithelial barriers are damaged in permeability edema, whereas they are usually normal in high-pressure edema; and second, when barriers are damaged, edema develops even at low driving pressures. The goals of therapy (Table 62-8) are to treat the underlying cause of acute lung damage, to provide support while the repair phase begins, to use a lung-protective ventilator strategy that will not worsen the lung injury, and to reduce as much as possible the driving pressures for fluid movement across the injured barriers into the lungs.

General Principles The cause of lung injury may not always be apparent, and when the cause is not obvious, it should be assumed to be infection: the most common treatable underlying cause of increased permeability edema. Although the patient is often seriously ill, diagnostic studies must be performed to identify a possible source of infection, so that appropriate drainage and antimicrobial therapy can be instituted. Plain chest radiographs are seldom helpful. Abdominal sonograms and CT scans can be diagnostically useful. Sepsis

1112 PART 3  •  Clinical Respiratory Medicine Table 62-8  Acute Lung Injury: Important Principles of Management MINIMIZE EDEMA ACCUMULATION Ensure lowest possible pulmonary microvascular pressure Reduce vascular volume FIND AND TREAT INFECTION PROVIDE SUPPORTIVE THERAPY Administer oxygen Use lung-protective ventilation Optimize blood pressure and cardiac output DO MORE GOOD THAN HARM Avoid hypotension Avoid volume overload Avoid oxygen toxicity Avoid infection

from intra-abdominal infection is common and may be especially difficult to identify. Many of these infections require surgical drainage if antimicrobial drug therapy is to be effective. Because specific therapy for increased permeability is not usually available—unless the cause is a treatable infection—supportive therapy is extremely important. The initial concerns are to support ventilation and circulation. Patients with increased permeability edema may be hemodynamically unstable, and ventilatory support can be complicated by hypotension or frank shock; PEEP, which is usually required for adequate oxygenation, may compound the problem by impeding venous return to the heart and decreasing cardiac function. Patients with low pulmonary capillary wedge pressures (<10 mm Hg) or central venous pressure (<4 mm Hg) and systemic hypotension require fluid resuscitation to support blood pressure and end-organ perfusion. If the patient has active, ongoing blood loss or the hemoglobin concentration is low (<7 g/dL), packed red blood cells are effective, not only to expand intravascular volume and restore blood pressure, but also to increase the oxygen-carrying capacity of the blood. Patients who are not bleeding and who have normal hemoglobin concentrations should be resuscitated with crystalloid solutions. Because the barriers restricting colloid movement from the vascular space into the lungs are not functioning normally in injured lungs, protein osmotic pressure differences favoring fluid movement into the vascular space cannot be established in the lungs; therefore there is no advantage to fluid resuscitation with colloid solutions. Patients with hypotension that does not respond to fluid resuscitation or who have normal (>10 mm Hg) or elevated pulmonary capillary wedge pressure require vasopressors or positive inotropic agents. For patients with septic shock, generalized vasodilation resulting in low systemic vascular resistance is the major hemodynamic abnormality.145 Norepinephrine is the most commonly used vasopressor to support blood pressure, but dopamine is also widely used. Aggressive hemodynamic support of critically ill patients can be detrimental if the complications of therapies become more harmful than the underlying disease itself. Strategies to increase systemic oxygen delivery (e.g., with inotropes, intravascular fluids, and blood transfusions) or to achieve

supranormal values for the cardiac index or normal values for mixed venous oxygen saturation did not improve mortality when applied to all critically ill patients, but did improve outcome in patients with sepsis who were treated very early in their course.146,147 One ARDS Network clinical trial showed that, in patients without shock, as defined by lack of need for vasopressor treatment, a strategy to restrict fluids to reduce intravascular pressures improved clinical outcomes by reducing the duration of mechanical ventilation in patients with ALI; of note, pressure measurements from both central venous and pulmonary artery catheters provided equally useful guidelines for fluid restriction, but pulmonary artery catheterization was associated with significantly increased complications.148 A multicenter, randomized, controlled clinical trial of transfusion requirements in critically ill patients with euvolemia after initial treatment compared a restrictive strategy of transfusing red blood cells if the hemoglobin concentration dropped to less than 7 g/dL versus a more liberal strategy of maintaining hemoglobin concentration greater than 10 g/dL. Restricting transfusion requirements to maintain hemoglobin concentration at 7 to 9 g/dL was at least as effective and was possibly superior to the strategy of giving transfusions to maintain hemoglobin concentration at 10 to 12 g/dL; the sole possible exception to this policy included patients with active coronary ischemic syndromes, such as acute myocardial infarction and unstable angina.149

Lung-Protective Ventilator Strategies In experimental animals, ventilation with high tidal volumes increases vascular filtration pressures and produces stress fractures of microvascular endothelium, alveolar epithelium, and basement membrane.150,151 The resulting injury appears to be due to the combination of large tidal excursions at high lung volumes coupled with elevated airway pressures: so-called volutrauma. Because the evidence from animal studies and small clinical trials provided compelling experimental rationale,152,153 investigations were performed to test the potential benefit of lower tidal volumes and reduced airway pressures versus standard (high) tidal volumes and elevated pressures. A large NHLBI-sponsored multicenter trial was stopped prematurely after the enrollment of 861 patients with well-defined ALI-ARDS. The trial compared “traditional” ventilator management using initial tidal volumes of 12 mL/kg of ideal body weight with plateau pressures of 50 cm/H2O or less, with lower tidal volumes of 6 mL/kg with airway plateau pressures limited to 30 cm H2O or less: “Mortality was lower in the group treated with lower tidal volumes than in the group treated with traditional tidal volumes— 31% vs. 39.8%, P = .007—and the number of days without ventilator use during the first 28 days after randomization was greater in this group—(mean ± SD), 12 ± 11 vs. 10 ± 11; P = .007.”154 The results of this seminal trial and several follow-up studies have transformed the management and outcome of patients with ARDS,155-158 and are being successfully applied to critically ill patients without ARDS.159,160 The protocol for implementing the lung-protective ventilatory strategy is detailed in Table 100-3 and further discussed in Chapter 100.

62  •  Pulmonary Edema 1113

Interestingly, results from an Italian study showed that a low tidal volume strategy in ARDS patients attenuated inflammatory responses in both the lungs and bloodstream, as measured by a reduction in neutrophil and cytokine concentrations in BAL fluid and a reduction in cytokines in circulating blood.161 Other studies have confirmed that low tidal volume lung-protective ventilation is associated with reduction in inflammatory markers in the lung.162 In addition, alveolar epithelial injury is probably reduced, considering the decline in surfactant protein D levels (a type II epithelial cell marker) and the receptor for advanced end glycation product (a type I epithelial cell marker) in the plasma of patients treated with the lung-protective ventilatory strategy.115,163 In addition to use of the lung protective ventilatory strategy, several adjuncts in small numbers of patients have been proposed to supplement treatment of critically ill patients, including prone positioning,164 extrapulmonary gas exchange (oxygenation, carbon dioxide removal, or both), liquid ventilation, tracheal gas insufflation, permissive hypercapnia, and high-frequency ventilation. Additional discussion of these modalities is provided in Chapter 100.

Specific Pharmacologic Therapies Several therapeutic agents, which are listed in Table 62-9, have been studied in patients with various forms of lung injury and sepsis. None of these agents have shown any benefit on mortality in large-scale, prospective, randomized, controlled clinical trials, including a recent trial of statins in ARDS.164a Whether or not newer anti-inflammatory therapies will be useful in patients with increased permeability edema from damaged lung barriers remains an important question. Although inflammatory reactions can damage tissue, it is important to remember that inflammation plays an important beneficial role in the elimination of invading microorganisms. Because inflammatory pathways are redundant, blocking any one inflammatory mediator (or even multiple mediators) may have little or no effect on the overall inflammatory response. If the mechanisms of lung injury differ after various clinical events, the application of therapy directed at one particular mechanism would not be appropriate until most or all of the underlying mechanisms have been recognized and treated successfully. Table 62-9  Therapeutic Agents Tested in Increased Permeability Pulmonary Edema164a Corticosteroids Ibuprofen Nitric oxide Prostaglandin-E1 (PGE1) Liposomal prostaglandins E1 and E2 Surfactant Antiendotoxin and antitumor necrosis factor antibodies Platelet-activating factor receptor antagonist Interleukin-1 receptor antagonist Ketoconazole N-acetylcysteine Oxothiazolidine carboxylate Pentoxifylline Beta-adrenergic agonists Statins

Corticosteroids.  Of all the possible pharmacologic agents used to treat critical lung injuries, corticosteroids have the longest history. Despite a seemingly compelling rationale for their use in the setting of increased permeability edema from sepsis, four separate prospective, randomized, doubleblind, placebo-controlled trials of high-dose methylprednisolone therapy failed to show any benefit.165 Corticosteroid therapy did not prevent the development of wellcharacterized ALI or decrease its incidence in patients with the sepsis syndrome; neither did it hasten the reversal of ARDS, lower mortality, or improve respiratory function. Moreover, use of corticosteroids was associated with both a greater 14-day mortality rate in patients who developed critical lung injury and an increased frequency of associated infections. Corticosteroids also were ineffective in the sepsis syndrome and may have caused harm.166 One exception may be the fat embolism syndrome. A prospective, randomized, double-blind, placebo-controlled trial of corticosteroid treatment in 64 patients with long bone fractures showed that high-dose methylprednisolone effectively prevented the development of the fat embolism syndrome.167 Fat embolism is an uncommon clinical syndrome that is usually described in single case reports from orthopedists; corticosteroid treatment has not been routinely used and thus far a large-scale multicenter, randomized trial does not seem warranted. The generally disappointing results of corticosteroid trials in patients with early or impending lung injury have not discouraged further investigations; rather, ongoing studies indicate that corticosteroids might be beneficial in subsets of patients or when given at a particular time (e.g., during the proliferative phase) or for a more sustained period.168 Results of studies of corticosteroid therapy late in the course of ARDS (so-called rescue therapy) were encouraging, and a small (24-patient), randomized, double-blind, placebo-controlled clinical trial in patients with severe ARDS whose Lung Injury Score had failed to improve by the seventh day of respiratory failure showed improvement in lung injury and other organ dysfunction scores, as well as reduced mortality in treated patients.169 However, an NHLBI-sponsored trial of corticosteroid therapy (methylprednisolone 2 mg/kg) beginning on day 7 to 21 of ARDS showed no reduction in 60- or 180-day hospital mortality in patients treated with corticosteroids compared with placebo.170 In addition, there were important neuromuscular complications in patients treated with methylprednisolone. One study suggested potential benefit with steroids in severe lung injury although the design of the trial was suboptimal.171 Neuromuscular Blocking Agents.  Experience to date has been limited and randomized clinical trials inconclusive concerning both the efficacy of neuromuscular blocking agents in the treatment of ARDS and their association with intensive care unit–acquired muscular weakness. In 2010, a randomized-controlled trial showed improvement in the adjusted 90-day survival and time off the ventilator in patients receiving cisatracurium compared with placebo.172 A new systematic review and meta-analysis, which included three trials, all from the same French research group, reevaluated the potential benefit of 48-hour intravenous infusions of cisatracurium besylate; short-term infusion of

1114 PART 3  •  Clinical Respiratory Medicine

this neuromuscular blocking agent significantly improved mortality rate and lowered the risk for barotrauma, but had no effect on the length of mechanical ventilation among survivors.173 Data on risk of intensive care unit–acquired weakness were inconclusive. Further studies are needed to settle this pending issue.

OUTCOME The outcome of patients with pulmonary edema depends on which of the two major pathophysiologic categories of edema formation is involved. Until recently, much less was known about the resolution of pulmonary edema than about its formation. Water—and any extravasated proteins and cellular debris—must be removed from the alveoli and the interstitial spaces to restore the lungs to their pristine, healthy condition. Edema fluid clears from the lungs via five routes: lymphatics, airways, blood vessels, pleural space, and mediastinum; in contrast, cellular debris and particulate matter must be removed from the alveoli by macrophage uptake or through the airways.

RESOLUTION OF PULMONARY EDEMA The considerable advances in our understanding of the clearance of fluid and solute from the alveoli have been the subject of several reviews.2,174,175,175a Active sodium and chloride transport across the alveolar epithelial barrier into the interstitium drives edema fluid removal from the air spaces (see Chapter 9). The uninjured alveolar epithelium has a remarkable ability to clear fluid from the air spaces rapidly: for example, serial edema fluid protein concentration measurements relative to simultaneous plasma protein concentrations in a saltwater near-drowning patient showed that 50% to 60% of the excess alveolar fluid in the lungs was removed over the course of just 4 hours.38 In experimental studies in rabbits, instillation of 4 mL/kg of seawater into the lungs resulted in a 300% increase in alveolar fluid volume in less than 5 minutes—owing to the movement of mainly pure water driven by osmotic forces from the plasma into the hyperosmolar (881 ± 29 mOsm) instillate—80% of which was cleared from the alveoli in 6 hours.39 Equivalent volumes of iso-osmotic (292 ± 6 mOsm) saline instilled into rabbit lungs were cleared at a similar rate. In neither circumstance was there evidence of injury to the alveolar epithelial barrier, which is more resistant than the endothelial barrier to a wide range of injuries, including ischemia, alveolar and intravenous endotoxin and bacteria, intravenous oleic acid, acid aspiration, saltwater aspiration, hyperoxia, intratracheal bleomycin, septic and hypovolemic shock, and rewarming after severe hypothermia. Even after mild to moderate alveolar injury, the capacity to transport salt and water is often preserved. In severe injury, however, when the barrier is disrupted, the capacity to clear edema is lost, and the vascular endothelium becomes the limiting barrier between the vascular system and the air spaces. Clinically, the capacity to remove some alveolar edema fluid—as indicated by increase in the edema fluid to plasma protein concentration ratio in the first 4 to 12 hours after the development of increased permeability edema—is a favorable prognostic finding associ-

ated with a mortality of only 20%. In contrast, the inability to resorb alveolar edema fluid early in the course of severe lung injuries was associated with a mortality of nearly 80%.97 Of interest, a larger and more recent study confirmed these results.98 Thus the functional capability of the alveolar epithelial barrier in acute increased permeability edema may be a useful prognostic index, perhaps because it serves as a marker of the severity and extent of lung injury. The barrier function may also be manipulated in certain settings. Lung Na+, K+-ATPase activity was increased in rats recovering from experimental thiourea-induced increased permeability pulmonary edema.176 Moreover, alveolar fluid clearance can be increased by salmeterol in uninjured ex vivo human lung,177 and experimental studies have shown that alveolar fluid clearance can be increased pharmacologically (e.g., by catecholamines), even in the presence of increased permeability edema with alveolar flooding.2 These observations raise the potential of therapy to hasten the resolution of alveolar edema, although an increase in alveolar clearance requires an intact alveolar epithelial barrier. Because clearance of protein from flooded alveoli is much slower (1% to 2%/hr) than clearance of protein-free fluid (10% to 20%/hr),178,179 the protein left behind becomes concentrated. The increase in protein concentration in the alveoli as protein-free fluid is resorbed does not slow fluid clearance, because precipitated protein exerts no osmotic pressure and the concentration of soluble macromolecules is too small to counteract the differences in ion concentration resulting from transepithelial transport. Removal of edema fluid from flooded alveoli may be slowed if the fluid clots, which can be seen especially when lung vascular permeability is increased. Edema fluid may clot because, following extravasation of plasma into the air spaces, the clotting system may be activated by surfactant or macrophage-derived procoagulants. Fluid cleared from alveolar spaces into the alveolar interstitium can leave the lungs by flowing into the lymphatic capillaries or by moving down the prevailing pressure gradient into the loose peribronchovascular connective tissue spaces or directly into the pleural space. Large amounts of fluid in the air spaces may be partially cleared into peribronchovascular cuffs through the hypothesized “leaky” terminal-airway epithelium,180,181 leaving alveolar fluid and solute behind to be cleared more slowly through the more impermeable alveolar epithelium. Most of the interstitial water in pulmonary edema is in the peribronchovascular loose connective tissue spaces rather than in the alveolar walls. Because the lymphatic capillaries are arranged to drain only the alveolar wall interstitium, this route for edema removal is not available for most interstitial water. A study in goats showed that lung lymph originated mainly from alveolar wall interstitial fluid, and the contribution of the lung lymphatic system to the clearance of interstitial edema in bronchoalveolar cuffs and interlobular septa was small. The maximum possible contribution by lung lymphatics to the clearance of interstitial edema liquid was less than 10%, and airway loss of liquid by evaporation was about twice the rate of lymphatic clearance.182 In a study of in situ perfused sheep lungs with experimental low- and high-protein pulmonary edema, during recovery from pulmonary edema, interstitial liquid

62  •  Pulmonary Edema 1115

was resorbed into the circulation in inverse proportion to its protein concentration, and only a very small fraction of interstitial edema was cleared by the lung lymphatics during recovery from either type of edema.183 Some fluid from the loose peribronchovascular interstitium may drain directly into the bloodstream by crossing the walls of blood vessels in the lungs. A study of isolated sheep lungs made edematous by raising vascular pressures showed that the primary route of edema clearance was by vascular resorption; 60% of filtered water was cleared over 3 hours, 42% by reabsorption into the bloodstream and 18% by lymphatic, pleural, and mediastinal drainage.184 Edema may also drain into the pleural space. Pleural effusions are fairly common in increased pressure pulmonary edema—found in about 25% to 50% of patients and usually on the right side when unilateral—but are present in increased permeability as well—in about 35% of patients.185-187 Formation and removal of pleural effusions are discussed in detail in Chapter 79. As much as 25% to 30% of pulmonary edema fluid may leave the lungs through the pleural space.186,188 A significant portion of the interstitial edema probably follows the prevailing pressure gradient in the lungs to drain into the mediastinum to be removed by neighboring lymphatics. Short-term alveolar protein clearance appears to proceed primarily by paracellular diffusion and is size dependent.6 Most proteins are cleared intact rather than being degraded into smaller fragments. The general consensus is that trans­ cytosis (transport via vesicles) is not a major mechanism for clearing bulk quantities of albumin or other proteins from the alveolar space. Over the long term, cellular mechanisms, principally phagocytosis and catabolism by macrophages, account for most protein clearance from the alveolar space.56 Insoluble, precipitated proteins are removed in this way. Macrophages are also ultimately responsible for removing senescent and dead PMNs and other debris. The small ciliated surface area of the distal air spaces seems to indicate that the mucociliary route could account for only a minor fraction of alveolar protein clearance, although proteins might reach the mucociliary escalator along currents in the alveolar fluid layer. Even so, removal would be very slow: the half-time for mucociliary clearance of particles from the alveolar space is more than 4 weeks. Complete clearance of alveolar protein from pulmonary edema by any route is slow.6 Little is known about the mechanisms and signals that regulate endothelial barrier function or how increased endothelial permeability returns to normal189; on the other hand, the mechanisms for resolution of lung inflammation are beginning to be understood.56

Increased Pressure Edema The outcome of increased pressure pulmonary edema is determined by the underlying cause and the treatment used to correct it. Because the great majority of cases of increased pressure edema are caused by heart disease, outcome is largely determined by the patient’s underlying cardiac function. Patients with pulmonary edema uncomplicated by acute myocardial infarction do reasonably well with an annual mortality rate of less than 10%.190 However, when acute myocardial infarction supervenes, the prognosis

worsens, although coronary vascular reperfusion therapy, using thrombolytics and coronary angioplasty with stent placement, has considerably improved survival. Typically, patients who recover from increased pressure pulmonary edema caused by chronic congestive cardiac failure require long-term outpatient management aimed at preventing recurrent episodes.191 Some patients develop increased pressure pulmonary edema from noncardiac causes. Most cases are iatrogenic, being related to excessive, sometimes inadvertent, volume overload or to the use of drugs such as cocaine that impair cardiac function.192 Pulmonary edema associated with congenital or acquired heart disease is an uncommon but important problem in pregnancy, but pulmonary edema without heart disease is, perhaps, even more frequent. As discussed in Chapter 96, pregnancy complicated by pulmonary edema is occasionally caused by tocolytic treatment,193 fluid overload, or preeclampsia.

Increased Permeability Edema The outcome of increased permeability pulmonary edema from both ARDS and forms of milder lung injury is determined by its underlying cause and extent of lung injury, the presence of comorbidities, and the particular treatment strategy employed. The reported mortality rates range from 20% to 60% depending on the specific etiologic factor, but yearly ARDS mortality rates reported from a single institution in the period between 1983 and 1993 showed a significant decrease in patients younger than 60 years and in those with sepsis syndrome.194 Recent trends in ALI mortality from 1996 through 2005 in 2451 mechanically ventilated patients enrolled in the NHLBI ARDS Network showed “clear temporal improvement in survival”; in 1996-1997, mortality was 35% and steadily declined to a low of 26% in 2004-2005.195 Furthermore, 60-day mortality decreased even further to 22% in adult patients in the most recent ARDS Network clinical trials, despite an increased severity of illness.196 Using the NHLBI-consensus definition in patients from Iceland, the incidence of ARDS almost doubled from 1988 to 2010, whereas hospital mortality decreased from 50% in 1988-1992 to 33% in 2006-2010; conversely, the 10-year survival of ARDS patients was only 68% compared with 90% in the reference population.197 Patients with sepsis have significantly higher mortality rates than patients with other clinical disorders associated with the development of increased permeability edema.198 Mortality, also, is much higher in patients with chronic liver disease199 or with histories of chronic alcohol abuse than in other predisposing conditions. Besides damaging the liver, chronic ethanol ingestion may reduce alveolar type II cell glutathione content and impair surfactant synthesis and secretion.200 In general, mortality from ARDS increases with increasing age, which conforms to both data and expectations.201 Patients with uncommon self-limited causes of injury, including venous air and fat embolism, isolated lung contusion and other trauma, massive blood transfusions, postictal pulmonary edema, and heroin pulmonary edema (Fig. 62-7), and those with milder degrees of edema have a greater chance of survival and edema often clears rapidly. As pointed out, since at least during the 1980s, survival from increased permeability has steadily improved, but

1116 PART 3  •  Clinical Respiratory Medicine

A

B

Figure 62-7  Frontal chest radiographs in a 22-year-old patient with heroin-induced increased permeability edema. Frontal chest radiograph performed at the time of presentation (A) shows patchy bilateral peribronchial thickening and consolidation, somewhat more prominent and nodular on the right. The vascular pedicle is narrow, and no features suggestive of volume overload are present. Frontal chest radiograph 24 hours later (B) shows complete clearing of lung opacity. Increased permeability edema may resolve rapidly if the initiating agent of lung injury is transient and mild and the chronic phase of parenchymal lung injury (type II cell hyperplasia, fibrosis, connective tissue deposition, vascular remodeling) does not develop. (Courtesy Michael Gotway, MD.)

long-term sequelae pose increasing problems. In decades past, based on short-term follow-up studies, survivors of ARDS seemed to do well; many had normal chest radiograph results, minimal or no complaints of dyspnea on exertion, normal lung volumes and airflow measurements, and normal resting and exercise arterial blood gas findings, including shunt fraction. Although the numbers of patients and the duration of follow-up may have left some doubt about the final outcome, an apparent paradox has taken place. Mortality from ARDS has clearly decreased over the past decades, whereas long-term survivors seem to have suffered worsening physical, cognitive, and mental health. The post-ARDS return of pulmonary function test results to normal or near-normal values may, in fact, have obscured a significant deterioration of health-related quality of life. Disabling symptoms from muscle wasting, weakness, and fatigue were consistent with reduced 6-minute walk distances that changed little during the first 12 months in previously high-functioning, mostly young subjects.57 Cognitive and emotional function was significantly impaired in “all” surviving post-ARDS patients at hospital discharge, 78% of whom remained symptomatic after 12 months. Furthermore, psychological sequelae such as anxiety, depression, and posttraumatic stress disorder persisted in 20% to 40% of survivors.58 The roles played by treatment of ARDS with corticosteroids and neuromuscular blocking are not clear-cut, but neither agent appears to be a major factor. Healing involves a fibroproliferative response in a subset of ARDS survivors, which seems to begin early in the course of increased permeability edema,202-204 perhaps as a consequence of ventilator-induced lung injury before the era of lung-protective ventilation205 (Video 62-2). How much of this problem is due to disease and how much to treatment is not known, but a fibroproliferative reaction portends a poor prognosis, either increased mortality or prolonged ventilation dependence. It might also be possible to accelerate reconstitution of alveolar structure in injured lungs; for example, keratinocyte growth factor (fibroblast growth factor-7),206 has been effective in several preclinical models of ALI207 and is now being tested in a phase 2 trial. More information about repair and healing should open new pos-

sibilities for therapy, including a better understanding of how lung progenitor cells may play a role in lung repair and regeneration.208,209

OVERVIEW Among the significant advances since the late 1960s has been the acquisition of important new knowledge concerning the physiology of fluid, solute, and protein transport in healthy and diseased lungs. Pulmonary edema—the abnormal accumulation of extravascular fluid in the lung—is a pathologic state that arises when fluid is filtered into the lungs faster than it can be removed. The many causes of pulmonary edema have been grouped into two main pathophysiologic categories: (1) increased pressure edema that results from an increase in the hydrostatic or protein osmotic forces (or both) that act across the barriers that normally restrict movement of fluid and solutes in the lungs; (2) increased permeability edema that results from damage to the normal barrier properties of lung endothelium and/or epithelium. Although these two different types of pulmonary edema share many features, they can usually be distinguished clinically and they have different treatment requirements and prognoses; differentiation is made possible by careful clinical, radiologic, and physiologic evaluation, but both increased permeability and increased pressure edema often coexist. Increased pressure pulmonary edema typically has one of two cardiac origins, either from new-onset acute myocardial infarction or from inadequately treated and/or refractory heart failure, most commonly caused by coronary heart disease, but many other sources are possible. Revascularization techniques for acute and chronic manifestations of coronary heart disease have dramatically improved prognosis and long-term outcome. Acute cardiogenic pulmonary edema is becoming scarce and should become even scarcer. However, more attention is currently being paid to the acute complications of cardiovascular disease than to its chronic therapeutic requirements. Nevertheless, therapy of chronic heart disease has huge clinical payoffs and much more should be done to ensure satisfactory blood pressure control, regulate dyslipidemias, and manage anticoagulants, starting with aspirin.

62  •  Pulmonary Edema 1117

There have been major advances in the treatment of increased permeability edema, largely due to the successful early application of lung-protective ventilatory strategies in patients with clinical lung injury. A low tidal volume (6 mL/ kg ideal body weight) coupled with a plateau pressure limit (<30 cm H2O) is still the only therapy proven to reduce mortality in patients with well-defined ALI and ARDS. Shortduration infusion of neuromuscular blockade agents show promise in severe ARDS, but further studies are needed for confirmation. New insights into the pathogenesis of various causes of increased permeability pulmonary edema suggest that other therapies may also prove to lower mortality in this common syndrome of severe acute respiratory failure.

Key Points The two main categories of pulmonary edema are (1) increased pressure edema from an increase in the hydrostatic or protein osmotic forces (or both) that act across the barriers that normally restrict movement of fluid and solutes in the lungs and (2) increased permeability edema from damage to the normal barrier properties of lung endothelium and/or epithelium. ■ An evaluation of the cause of pulmonary edema should include a detailed history, thorough physical examination, and selected laboratory data, including a chest radiograph, electrocardiogram, cardiac enzymes, and, in some patients, microbiologic cultures, an echocardiogram, and, occasionally, pulmonary artery catheterization. ■ Increased permeability pulmonary edema may be complicated by the presence of elevated pulmonary intravascular pressures, especially from coexisting cardiac failure or volume overload. ■ Treatment for cardiogenic pulmonary edema should include the immediate administration of supplemental oxygen, morphine (usually), and pharmacologic measures to reduce preload. ■

The diagnosis of increased permeability pulmonary edema should always include a thorough search for a treatable infection. ■ Patients with increased permeability edema and, increasingly, other forms of ventilatory failure, require intubation and lung-protective ventilation, using 6 mL/kg tidal volume and less than 30 cm H2O plateau pressure. ■

Complete reference list available at ExpertConsult.

Key Readings Alhazzani W, Alshahrani M, Jaesdhke R, et al: Neuromuscular blocking agents in acute respiratory distress syndrome: a systematic review and meta-analysis of randomized controlled trials. Crit Care 17(2):R43, 2013. Burnham EL, Janssen WJ, Riches DW, et al: The fibroproliferative response in ARDS: mechanisms and clinical significance. Eur Respir J 43(1):276– 285, 2014. Chatterjee K: The Swan-Ganz catheters: past, present, and future. A viewpoint. Circulation 119:147–152, 2009. Gray A, Goodacre S, Newby DE, et al: Noninvasive ventilation in acute cardiogenic pulmonary edema. N Engl J Med 359:142–151, 2008. Guerin C, Reignier J, Richard JC, et al: Prone positioning in severe acute respiratory distress syndrome. N Engl J Med 368:2159–2168, 2013. Hastings RH, Folkesson HG, Matthay MA: Mechanisms of alveolar protein clearance in the intact lung. Am J Physiol Lung Cell Mol Physiol 286:L679–L689, 2004. Herridge MS, Tansey CM, Matte A, et al: Functional disability 5 years after acute respiratory distress syndrome. N Engl J Med 364:1293–1304, 2011. Matthay MA, Folkesson HG, Clerici C: Lung epithelial fluid transport and the resolution of pulmonary edema. Physiol Rev 82:569–600, 2002. Matthay MA, Ware L, Zimmerman GA: The acute respiratory distress syndrome. J Clin Invest 122:2731–2740, 2012. Staub NC: Pulmonary edema. Physiol Rev 54:678–811, 1974. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. The Acute Respiratory Distress Syndrome Network. N Engl J Med 342:1301–1308, 2000. Ware LB, Matthay MA: Alveolar fluid clearance is impaired in the majority of patients with acute lung injury and the acute respiratory distress syndrome. Am J Respir Crit Care Med 163:1376–1383, 2001.

62  •  Pulmonary Edema 1117.e1

References 1. Staub NC: Pulmonary edema. Physiol Rev 54:678–811, 1974. 2. Matthay MA, Folkesson HG, Clerici C: Lung epithelial fluid transport and the resolution of pulmonary edema. Physiol Rev 82:569–600, 2002. 3. Taylor AE: Capillary fluid filtration. Starling forces and lymph flow. Circ Res 49:557–575, 1981. 4. Lai-Fook SJ: Mechanical factors in lung liquid distribution. Annu Rev Physiol 55:155–179, 1993. 5. Staub NC: The pathogenesis of pulmonary edema. Prog Cardiovasc Dis 23:53–80, 1980. 6. Hastings RH, Folkesson HG, Matthay MA: Mechanisms of alveolar protein clearance in the intact lung. Am J Physiol Lung Cell Mol Physiol 286:L679–L689, 2004. 7. Erdmann AJ 3rd, Vaughan TR Jr, Brigham KL, et al: Effect of increased vascular pressure on lung fluid balance in unanesthetized sheep. Circ Res 37:271–284, 1975. 8. Guyton AC, Lindsey AW: Effect of elevated left atrial pressure and decreased plasma protein concentration on the development of pulmonary edema. Circ Res 7:649–657, 1959. 9. Lemen R, Jones JG, Graf PD, et al: “Closing volume” changes in alloxan-induced pulmonary edema in anesthetized dogs. J Appl Physiol 39:235–241, 1975. 10. Noble WH, Kay JC, Obdrzalek J: Lung mechanics in hypervolemic pulmonary edema. J Appl Physiol 38:681–687, 1975. 11. Hughes JM, Rosenzweig DY: Factors affecting trapped gas volume in perfused dog lungs. J Appl Physiol 29:332–339, 1970. 12. Bernard GR, Pou NA, Coggeshall JW, et al: Comparison of the pulmonary dysfunction caused by cardiogenic and noncardiogenic pulmonary edema. Chest 108:798–803, 1995. 13. DeFouw DO, Berendsen PB: Morphological changes in isolated perfused dog lungs after acute hydrostatic edema. Circ Res 43:72–82, 1978. 14. Bachofen H, Bachofen M, Weibel ER: Ultrastructural aspects of pulmonary edema. J Thorac Imaging 3:1–7, 1988. 15. Felson B: Thoracic calcifications. Dis Chest 56:330–343, 1969. 16. Charan NB, Turk GM, Hey DH: Effect of increased bronchial venous pressure on lung lymph flow. J Appl Physiol 59:1249–1253, 1985. 17. Bland RD, Demling RH, Selinger SL, et al: Effects of alveolar hypoxia on lung fluid and protein transport in unanesthetized sheep. Circ Res 40:269–274, 1977. 18. Landolt CC, Matthay MA, Albertine KH, et al: Overperfusion, hypoxia, and increased pressure cause only hydrostatic pulmonary edema in anesthetized sheep. Circ Res 52:335–341, 1983. 19. van der Werff YD, van der Houwen HK, Heijmans PJ, et al: Postpneumonectomy pulmonary edema. A retrospective analysis of incidence and possible risk factors. Chest 111:1278–1284, 1997. 20. Hultgren H: High-altitude pulmonary edema. In Matthay MA, Ingbar D, editors: Pulmonary edema, New York, 1998, Marcel Dekker, pp 355–378. 21. West JB, Mathieu-Costello O: Structure, strength, failure, and remodeling of the pulmonary blood-gas barrier. Annu Rev Physiol 61:543– 572, 1999. 22. Huchon GJ, Hopewell PC, Murray JF: Interactions between permeability and hydrostatic pressure in perfused dogs’ lungs. J Appl Physiol 50:905–911, 1981. 23. Jerome EH, Severinghaus JW: High-altitude pulmonary edema. N Engl J Med 334:662–663, 1996. 24. Kleger GR, Bartsch P, Vock P, et al: Evidence against an increase in capillary permeability in subjects exposed to high altitude. J Appl Physiol 81:1917–1923, 1996. 25. Maron MB, Pilati CF: Neurogenic pulmonary edema. In Matthay MA, Ingbar D, editors: Pulmonary edema, New York, 1998, Marcel Dekker, pp 319–354. 26. Kallet RH, Daniel BM, Gropper M: Acute pulmonary edema following upper airway obstruction: case reports and brief review. Respir Care 43:476–480, 1998. 27. Pavlin J, Cheney FW Jr: Unilateral pulmonary edema in rabbits after reexpansion of collapsed lung. J Appl Physiol 46:31–35, 1979. 28. Shaw TJ, Caterine JM: Recurrent re-expansion pulmonary edema. Chest 86:784–786, 1984. 29. Suzuki S, Tanita T, Koike K, et al: Evidence of acute inflammatory response in reexpansion pulmonary edema. Chest 101:275–276, 1992.

30. Sue RD, Matthay MA, Ware LB: Hydrostatic mechanisms may contribute to the pathogenesis of human re-expansion pulmonary edema. Intensive Care Med 30:1921–1926, 2004. 31. Jackson RM, Veal CF, Alexander CB, et al: Re-expansion pulmonary edema. A potential role for free radicals in its pathogenesis. Am Rev Respir Dis 137:1165–1171, 1988. 32. Ware LB, Golden JA, Finkbeiner WE, et al: Alveolar epithelial fluid transport capacity in reperfusion lung injury after lung transplantation. Am J Respir Crit Care Med 159:980–988, 1999. 33. Albert RK, Lakshminarayan S, Hildebrandt J, et al: Increased surface tension favors pulmonary edema formation in anesthetized dogs’ lungs. J Clin Invest 63:1015–1018, 1979. 34. Dodek PM, Rice TW, Bonsignore MR, et al: Effects of plasmapheresis and of hypoproteinemia on lung liquid conductance in awake sheep. Circ Res 58:269–280, 1986. 35. Bhattacharya J, Gropper MA, Staub NC: Interstitial fluid pressure gradient measured by micropuncture in excised dog lung. J Appl Physiol 56:271–277, 1984. 36. Lai-Fook SJ, Beck KC: Alveolar liquid pressure measured by micropipettes in isolated dog lung. J Appl Physiol 53:737–743, 1982. 37. Bredenberg CE, Nieman GF, Paskanik AM, et al: Microvascular membrane permeability in high surface tension pulmonary edema. J Appl Physiol 60:253–259, 1986. 38. Cohen DS, Matthay MA, Cogan MG, et al: Pulmonary edema associated with salt water near-drowning: new insights. Am Rev Respir Dis 146:794–796, 1992. 39. Folkesson HG, Kheradmand F, Matthay MA: The effect of salt water on alveolar epithelial barrier function. Am J Respir Crit Care Med 150:1555–1563, 1994. 40. Ashbaugh DG, Bigelow DB, Petty TL, et al: Acute respiratory distress in adults. Lancet 2:319–323, 1967. 41. Robbins I, Newman JH, Brigham KL: Increased-permeability pulmonary edema from sepsis/endotoxin. In Matthay MA, Ingbar D, editors: Pulmonary edema, New York, 1998, Marcel Dekker, pp 203–245. 42. Gorin AB, Stewart PA: Differential permeability of endothelial and epithelial barriers to albumin flux. J Appl Physiol 47:1315–1324, 1979. 43. Pelosi P, Cereda M, Foti G, et al: Alterations of lung and chest wall mechanics in patients with acute lung injury: effects of positive endexpiratory pressure. Am J Respir Crit Care Med 152:531–537, 1995. 44. Grossman RF, Jones JG, Murray JF: Effects of oleic acid-induced pulmonary edema on lung mechanics. J Appl Physiol 48:1045–1051, 1980. 45. Pelosi P, D’Andrea L, Vitale G, et al: Vertical gradient of regional lung inflation in adult respiratory distress syndrome. Am J Respir Crit Care Med 149:8–13, 1994. 46. Wright PE, Carmichael LC, Bernard GR: Effect of bronchodilators on lung mechanics in the acute respiratory distress syndrome (ARDS). Chest 106:1517–1523, 1994. 47. Gattinoni L, Pelosi P, Suter PM, et al: Acute respiratory distress syndrome caused by pulmonary and extrapulmonary disease. Different syndromes? Am J Respir Crit Care Med 158:3–11, 1998. 48. Tierney DF, Johnson RP: Altered surface tension of lung extracts and mechanics. J Appl Physiol 20:1253–1260, 1965. 49. Ryan SF, Ghassibi Y, Liau DF: Effects of activated polymorphonuclear leukocytes upon pulmonary surfactant in vitro. Am J Respir Cell Mol Biol 4:33–41, 1991. 50. Gregory TJ, Longmore WJ, Moxley MA, et al: Surfactant chemical composition and biophysical activity in acute respiratory distress syndrome. J Clin Invest 88:1976–1981, 1991. 51. Kobayashi T, Nitta K, Ganzuka M, et al: Inactivation of exogenous surfactant by pulmonary edema fluid. Pediatr Res 29:353–356, 1991. 52. Dantzker DR, Brook CJ, Dehart P, et al: Ventilation-perfusion distributions in the adult respiratory distress syndrome. Am Rev Respir Dis 120:1039–1052, 1979. 53. Nuckton TJ, Alonso JA, Kallet RH, et al: Pulmonary dead-space fraction as a risk factor for death in the acute respiratory distress syndrome. N Engl J Med 346:1281–1286, 2002. 54. Coss-Bu JA, Walding DL, David YB, et al: Dead space ventilation in critically ill children with lung injury. Chest 123:2050–2056, 2003. 55. Bachofen M, Weibel ER: Structural alterations of lung parenchyma in the adult respiratory distress syndrome. Clin Chest Med 3:35–56, 1982.

1117.e2 PART 3  •  Clinical Respiratory Medicine 56. Matthay MA, Ware L, Zimmerman GA: The acute respiratory distress syndrome. J Clin Invest 122:2731–2740, 2012. 57. Herridge MS, Cheung AM, Tansey CM, et al: One-year outcomes in survivors of the acute respiratory distress syndrome. N Engl J Med 348:683–693, 2003. 58. Herridge MS, Tansey CM, Matte A, et al: Functional disability 5 years after acute respiratory distress syndrome. N Engl J Med 364:1293– 1304, 2011. 59. Headley AS, Tolley E, Meduri GU: Infections and the inflammatory response in acute respiratory distress syndrome. Chest 111:1306– 1321, 1997. 60. Donnelly SC, Strieter RM, Kunkel SL, et al: Interleukin-8 and development of adult respiratory distress syndrome in at-risk patient groups. Lancet 341:643–647, 1993. 61. Park WY, Goodman RB, Steinberg KP, et al: Cytokine balance in the lungs of patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 164:1896–1903, 2001. 62. Pugin J, Ricou B, Steinberg KP, et al: Proinflammatory activity in bronchoalveolar lavage fluids from patients with ARDS, a prominent role for interleukin-1. Am J Respir Crit Care Med 153:1850–1856, 1996. 63. Ricou B, Nicod L, Lacraz S, et al: Matrix metalloproteinases and TIMP in acute respiratory distress syndrome. Am J Respir Crit Care Med 154:346–352, 1996. 64. Verghese GM, McCormick-Shannon K, Mason RJ, et al: Hepatocyte growth factor and keratinocyte growth factor in the pulmonary edema fluid of patients with acute lung injury. Biologic and clinical significance. Am J Respir Crit Care Med 158:386–394, 1998. 65. Chesnutt AN, Matthay MA, Tibayan FA, et al: Early detection of type III procollagen peptide in acute lung injury. Pathogenetic and prognostic significance. Am J Respir Crit Care Med 156:840–845, 1997. 66. Zhu S, Ware LB, Geiser T, et al: Increased levels of nitrate and surfactant protein A nitration in the pulmonary edema fluid of patients with acute lung injury. Am J Respir Crit Care Med 163:166–172, 2001. 67. Pittet JF, Mackersie RC, Martin TR, et al: Biological markers of acute lung injury: prognostic and pathogenetic significance. Am J Respir Crit Care Med 155:1187–1205, 1997. 68. Ware LB, Koyama T, Billheimer DD, et al: Prognostic and pathogenetic value of combining clinical and biochemical indices in patients with acute lung injury. Chest 137:288–296, 2010. 69. Matute-Bello G, Downey G, Moore BB, et al: An official American Thoracic Society workshop report: features and measurements of experimental acute lung injury in animals. Am J Respir Cell Mol Biol 44:725–738, 2011. 70. Murray JF, Matthay MA, Luce JM, et al: An expanded definition of the adult respiratory distress syndrome. Am Rev Respir Dis 138:720– 723, 1988. 71. Bernard GR, Artigas A, Brigham KL, et al: The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 149:818–824, 1994. 72. The ARDS Definition Task Force, Ranieri VM, Rubenfeld GD, et al: Acute respiratory distress syndrome: the Berlin definition. JAMA 307:2526–2533, 2012. 73. Trillo-Alvarez C, Cartin-Ceba R, Kor DJ, et al: Acute lung injury prediction score: derivation and validation in a population-based sample. Eur Respir J 37:604–609, 2011. 74. Gajic O, Dabbagh O, Park PK, et al: Early identification of patients at risk of acute lung injury. Evaluation of lung injury prediction score in a multicenter cohort study. Am J Crit Care Med 183:462–470, 2011. 75. Agrawal A, Matthay MA, Kangelaris KN, et al: Plasma angiopoietin-2 predicts the onset of acute lung injury in critically ill patients. Am J Respir Crit Care Med 187:736–742, 2013. 76. Thille AW, Esteban A, Fernandez-Segoviano P, et al: Comparison of the Berlin definition for acute respiratory distress syndrome with autopsy. Am J Respir Crit Care Med 187:761–767, 2013. 77. Snashall PD, Chung KF: Airway obstruction and bronchial hyperresponsiveness in left ventricular failure and mitral stenosis. Am Rev Respir Dis 144:945–956, 1991. 78. Graham SP, Vetrovec GW: Comparison of angiographic findings and demographic variables in patients with coronary artery disease presenting with acute pulmonary edema versus those presenting with chest pain. Am J Cardiol 68:1614–1618, 1991.

79. Meduri GU, Wunderink RG, Leeper KV, et al: Management of bacterial pneumonia in ventilated patients. Protected bronchoalveolar lavage as a diagnostic tool. Chest 101:500–508, 1992. 80. Sutherland KR, Steinberg KP, Maunder RJ, et al: Pulmonary infection during the acute respiratory distress syndrome. Am J Respir Crit Care Med 152:550–556, 1995. 81. Delclaux C, Roupie E, Blot F, et al: Lower respiratory tract colonization and infection during severe acute respiratory distress syndrome: incidence and diagnosis. Am J Respir Crit Care Med 156:1092–1098, 1997. 82. Chastre J, Trouillet JL, Vuagnat A, et al: Nosocomial pneumonia in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 157:1165–1172, 1998. 83. Patel SR, Karmpaliotis D, Ayas NT, et al: The role of open-lung biopsy in ARDS. Chest 125:197–202, 2004. 84. Oren S, Rapoport J, Zlotuik M, et al: Extreme hypermagnesemia due to ingestion of Dead Sea water. Nephron 47:198–201, 1984. 85. Staub NC: Clinical use of lung water measurements. Report of a workshop. Chest 90:588–594, 1986. 86. Bombino M, Gattinoni L, Pesenti A, et al: The value of portable chest roentgenography in adult respiratory distress syndrome. Comparison with computed tomography. Chest 100:762–769, 1991. 87. Aberle DR, Wiener-Kronish JP, Webb WR, et al: Hydrostatic versus increased permeability pulmonary edema: diagnosis based on radiographic criteria in critically ill patients. Radiology 168:73–79, 1988. 88. Rubenfeld GD, Caldwell E, Granton J, et al: Interobserver variability in applying a radiographic definition for ARDS. Chest 116:1347– 1353, 1999. 89. Ware LB, Neyrinck A, O’Neal HR, et al: Comparison of chest radiograph scoring to lung weight as a quantitative index of pulmonary edema in organ donors. Clin Transplant 26:665–671, 2012. 90. Brigham KL, Kariman K, Harris TR, et al: Correlation of oxygenation with vascular permeability-surface area but not with lung water in humans with acute respiratory failure and pulmonary edema. J Clin Invest 72:339–349, 1983. 91. Harris TR, Bernard GR, Brigham KL, et al: Lung microvascular transport properties measured by multiple indicator dilution methods in patients with adult respiratory distress syndrome. A comparison between patients reversing respiratory failure and those failing to reverse. Am Rev Respir Dis 141:272–280, 1990. 92. Estenne M, Yernault JC: The mechanism of CO2 retention in cardiac pulmonary edema. Chest 86:936–938, 1984. 93. De Backer D, Creteur J, Zhang H, et al: Lactate production by the lungs in acute lung injury. Am J Respir Crit Care Med 156:1099– 1104, 1997. 94. Albertine KH, Soulier MF, Wang Z, et al: Fas and fas ligand are up-regulated in pulmonary edema fluid and lung tissue of patients with acute lung injury and the acute respiratory distress syndrome. Am J Pathol 161:1783–1796, 2002. 95. Sprung CL, Rackow EC, Fein IA, et al: The spectrum of pulmonary edema: differentiation of cardiogenic, intermediate, and noncardiogenic forms of pulmonary edema. Am Rev Respir Dis 124:718–722, 1981. 96. Carlson RW, Schaeffer RC Jr, Michaels SG, et al: Pulmonary edema fluid. Spectrum of features in 37 patients. Circulation 60:1161– 1169, 1979. 97. Matthay MA, Wiener-Kronish JP: Intact epithelial barrier function is critical for the resolution of alveolar edema in humans. Am Rev Respir Dis 142:1250–1257, 1990. 98. Ware LB, Matthay MA: Alveolar fluid clearance is impaired in the majority of patients with acute lung injury and the acute respiratory distress syndrome. Am J Respir Crit Care Med 163:1376–1383, 2001. 99. Ware LB, Fremont RD, Bastarache JA, et al: Determining the aetiology of pulmonary oedema by the oedema fluid-to-plasma protein ratio. Eur Respir J 35:331–337, 2010. 100. Steinberg KP, Mitchell DR, Maunder RJ, et al: Safety of bronchoalveolar lavage in patients with adult respiratory distress syndrome. Am Rev Respir Dis 148:556–561, 1993. 101. Staub NC: The hemodynamics of pulmonary edema. Bull Eur Physiopathol Respir 22:319–322, 1986. 102. Matthay MA: Measurement of extravascular lung water in patients with pulmonary edema. Am J Physiol Lung Cell Mol Physiol 294:L1021–L1022, 2008. 103. Aberle DR, Wiener-Kronish JP, Webb WR, et al: Hydrostatic versus increased permeability pulmonary edema: diagnosis based on

62  •  Pulmonary Edema 1117.e3 radiographic criteria in critically ill patients. Radiology 168:73–79, 1988. 104. Sibbald WJ, Cunningham DR, Chin DN: Non-cardiac or cardiac pulmonary edema? A practical approach to clinical differentiation in critically ill patients. Chest 84:452–461, 1983. 105. Connelly KG, Repine JE: Markers for predicting the development of acute respiratory distress syndrome. Annu Rev Med 48:429–445, 1997. 106. Levitt JE, Gould MK, Ware LB, et al: Analytic review: the pathogenetic and prognostic value of biologic markers in acute lung injury. J Intensive Care Med 24:151–167, 2009. 107. Cheng IW, Ware LB, Greene KE, et al: Prognostic value of surfactant proteins A and D in patients with acute lung injury. Crit Care Med 31:20–27, 2003. 108. Parsons PE, Eisner MD, Thompson BT, et al: Lower tidal volume ventilation and plasma cytokine markers of inflammation in patients with acute lung injury. Crit Care Med 33:1–6, discussion 230–232, 2005. 109. Ware LB, Eisner MD, Thompson BT, et al: Significance of von Willebrand factor in septic and nonseptic patients with acute lung injury. Am J Respir Crit Care Med 170:766–772, 2004. 110. Parsons PE, Matthay MA, Ware LB, et al: National Heart LBIARDSCTN. Elevated plasma levels of soluble TNF receptors are associated with morbidity and mortality in patients with acute lung injury. Am J Physiol Lung Cell Mol Physiol 288:L426–L431, 2005. 111. Calfee CS, Eisner MD, Parsons PE, et al: Soluble intercellular adhesion molecule-1 and clinical outcomes in patients with acute lung injury. Intensive Care Med 35:248–257, 2009. 112. Ware LB, Fang X, Matthay MA: Protein C and thrombomodulin in human acute lung injury. Am J Physiol Lung Cell Mol Physiol 285:L514–L521, 2003. 113. Prabhakaran P, Ware LB, White KE, et al: Elevated levels of plasminogen activator inhibitor-1 in pulmonary edema fluid are associated with mortality in acute lung injury. Am J Physiol Lung Cell Mol Physiol 285:L20–L28, 2003. 114. Uchida T, Shirasawa M, Ware LB, et al: Receptor for advanced glycation end-products is a marker of type I cell injury in acute lung injury. Am J Respir Crit Care Med 173:1008–1015, 2006. 115. Calfee CS, Budev MM, Matthay MA, et al: Plasma receptor for advanced glycation end-products predicts duration of ICU stay and mechanical ventilation in patients after lung transplantation. J Heart Lung Transplant 26:675–680, 2007. 116. Ricciuto DR, dos Santos CC, Hawkes M, et al: Angiopoietin-1 and angiopoietin-2 as clinically informative prognostic biomarkers of morbidity and mortality in severe sepsis. Crit Care Med 39:702–710, 2011. 117. Staub NC: Pulmonary edema: physiologic approaches to management. Chest 74:559–564, 1978. 118. Calfee CS, Matthay MA: Nonventilatory treatments for acute lung injury and ARDS. Chest 131:913–920, 2007. 119. Gheorghaide M, Cody RJ, Francis GS, et al: Current medical therapy for advanced heart failure. Am Heart J 135:S231–S248, 1998. 120. Verghese GM, Ware LB, Matthay BA, et al: Alveolar epithelial fluid transport and the resolution of clinically severe hydrostatic pulmonary edema. J Appl Physiol 87:1301–1312, 1999. 121. Klein HO, Brodsky E, Ninio R, et al: The effect of venous occlusion with tourniquets on peripheral blood pooling and ventricular function. Chest 103:521–527, 1993. 122. Eiser AR, Lieber JJ, Neff MS: Phlebotomy for pulmonary edema in dialysis patients. Clin Nephrol 47:47–49, 1997. 123. Jasmer RM, Luce JM, Matthay MA: Noninvasive positive pressure ventilation for acute respiratory failure: underutilized or overrated? Chest 111:1672–1678, 1997. 124. Gray A, Goodacre S, Newby DE, et al: Noninvasive ventilation in acute cardiogenic pulmonary edema. N Engl J Med 359:142–151, 2008. 125. Antonelli M, Conti G, Rocco M, et al: A comparison of noninvasive positive-pressure ventilation and conventional mechanical ventilation in patients with acute respiratory failure. N Engl J Med 339:429– 435, 1998. 126. Weng CL, Zhao YT, Liu QH, et al: Meta-analysis: noninvasive ventilation in acute cardiogenic pulmonary edema. Ann Intern Med 152:590–600, 2010. 127. Swan HJ, Ganz W, Forrester J, et al: Catheterization of the heart in man with the use of a flow-directed catheter for cardiac monitoring. N Eng J Med 283:447, 1970.

128. Chatterjee K: The Swan-Ganz catheters: past, present, and future. A viewpoint. Circulation 119:147–152, 2009. 129. Matthay MA, Chatterjee K: Bedside catheterization of the pulmonary artery: risks compared with benefits. Ann Intern Med 109:826– 834, 1988. 130. Connors AF Jr, Speroff T, Dawson NV, et al: The effectiveness of right heart catheterization in the initial care of critically ill patients. SUPPORT Investigators. JAMA 276:889–897, 1996. 131. Connors AF Jr, Castele RJ, Farhat NZ, et al: Complications of right heart catheterization. A prospective autopsy study. Chest 88:567– 572, 1985. 132. Richard C, Warszawski J, Anguel N, et al: Early use of the pulmonary artery catheter and outcomes in patients with shock and acute respiratory distress syndrome: a randomized controlled trial. JAMA 290:2713–2720, 2003. 133. O’Quin R, Marini JJ: Pulmonary artery occlusion pressure: clinical physiology, measurement, and interpretation. Am Rev Respir Dis 128:319–326, 1983. 134. Robin ED: The cult of the Swan-Ganz catheter. Overuse and abuse of pulmonary flow catheters. Ann Intern Med 103:445–449, 1985. 135. The NHLBI Acute Respiratory Disease Syndrome (ARDS) Clinical Trial Network: Pulmonary artery versus central venous catheter to guide treatment of acute lung injury. N Engl J Med 354:2564–2575, 2006. 136. Rajaram SS, Desai NK, Kaira A, et al: Pulmonary artery catheters for adult patients in intensive care. Cochrane Database Syst Rev (2):CD003408, 2013. 137. Litton E, Morgan M: The PiCCO monitor: a review. Anaesth Intensive Care 40:393–409, 2012. 138. Brater DC: Diuretic therapy. N Engl J Med 339:387–395, 1998. 139. Faris RF, Flather M, Purcell H, et al: Diuretics for heart failure. Cochrane Database Syst Rev (2):CD003838, 2012. doi: 10.1002/ 14651858, CD003838.pub3. 140. Felker GM, O’Connor CM, Braunwald E, et al: Loop diuretics in acute decompensated heart failure: necessary? evil? a necessary evil? Circ Heart Fail 2:56–62, 2009. 141. Garzia F, Todor R, Scalea T: Continuous arteriovenous hemofiltration countercurrent dialysis (CAVH-D) in acute respiratory failure (ARDS). J Trauma 31:1277–1284, discussion 84–85, 1991. 142. Susini G, Zucchetti M, Bortone F, et al: Isolated ultrafiltration in cardiogenic pulmonary edema. Crit Care Med 18:14–17, 1990. 143. Goldhaber JI, Hamilton MA: Role of inotropic agents in the treatment of heart failure. Circulation 121:1655–1660, 2010. 144. Curtis AB, Worley SJ, Adamson PB, et al: Biventricular pacing for atrioventricular block and systolic dysfunction. N Engl J Med 368:1585–1593, 2013. 145. Wheeler AP, Bernard GR: Treating patients with severe sepsis. N Engl J Med 340:207–214, 1999. 146. Gattinoni L, Brazzi L, Pelosi P, et al: A trial of goal-oriented hemodynamic therapy in critically ill patients. SvO2 Collaborative Group. N Engl J Med 333:1025–1032, 1995. 147. Rivers E, Nguyen B, Havstad S, et al: Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 345:1368–1377, 2001. 148. Wiedemann HP, Wheeler AP, Bernard GR, et al: Comparison of two fluid-management strategies in acute lung injury. N Engl J Med 354:2564–2575, 2006. 149. Hebert PC, Wells G, Blajchman MA, et al: A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion requirements in critical care investigators, Canadian Critical Care Trials Group. N Engl J Med 340:409–417, 1999. 150. Webb HH, Tierney DF: Experimental pulmonary edema due to intermittent positive pressure ventilation with high inflation pressures. Protection by positive end-expiratory pressure. Am Rev Respir Dis 110:556–565, 1974. 151. Parker JC, Townsley MI, Rippe B, et al: Increased microvascular permeability in dog lungs due to high peak airway pressures. J Appl Physiol 57:1809–1816, 1984. 152. Slutsky AS, Tremblay LN: Multiple system organ failure. Is mechanical ventilation a contributing factor? Am J Respir Crit Care Med 157:1721–1725, 1998. 153. Marini JJ: Evolving concepts in the ventilatory management of acute respiratory distress syndrome. Clin Chest Med 17:555–575, 1996. 154. [No authors listed]: Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the

1117.e4 PART 3  •  Clinical Respiratory Medicine acute respiratory distress syndrome. The Acute Respiratory Distress Syndrome Network. N Engl J Med 342:1301–1308, 2000. 155. Brower RG, Lanken PN, MacIntyre N, et al: National Heart, Lung, and Blood Institute Clinical Trials Network. Higher versus lower positive end-expiratory pressures in patients with the acute respiratory distress syndrome. N Engl J Med 351:327–336, 2004. 156. Steinberg KP, Hudson LD, Goodman RB, et al: National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome (ARDS) Clinical Trials Network. N Engl J Med 354:167–184, 2006. 157. Thompson BT, Bernard GR: ARDS Network (NHLBI) studies: successes and challenges in ARDS clinical research. Crit Care Clin 27:459–468, 2011. 158. Petrucci N, De Feo C: Lung protective ventilation strategy for the acute respiratory distress syndrome. Cochrane Database Syst Rev (2):CD003844, 2013. 159. Serpa Neto A, Cardoso SO, Manette JA, et al: Association between use of lung-protective ventilation with lower tidal volumes and clinical outcomes among patients without acute respiratory distress syndrome: a meta-analysis. JAMA 308:1651–1659, 2012. 160. Hill NS: Review: lower rather than higher tidal volume benefits ventilated patients without ARDS. Ann Intern Med 158(6):JC4, 2013. 161. Ranieri VM, Suter PM, Tortorella C, et al: Effect of mechanical ventilation on inflammatory mediators in patients with acute respiratory distress syndrome: a randomized controlled trial. JAMA 282:54–61, 1999. 162. Puneet P, Moochhala S, Bhatia M: Chemokines in acute respiratory distress syndrome. Am J Physiol Lung Cell Mol Physiol 288:L3–L15, 2005. 163. Eisner MD, Parsons P, Matthay MA, et al, Acute respiratory distress syndrome N. Plasma surfactant protein levels and clinical outcomes in patients with acute lung injury. Thorax 58:983–988, 2003. 164. Guerin C, Reignier J, Richard JC, et al: Prone positioning in severe acute respiratory distress syndrome. N Engl J Med 368:2159–2168, 2013. 164a.  National Heart, Lung, and Blood Institute ARDS Clinical Trials Network, Truwit JD, Bernard GR, et al: Rosuvastatin for sepsisassociated acute respiratory distress syndrome. N Engl J Med 370(23):2191–2200, 2014. 165. Bernard GR, Luce JM, Sprung CL, et al: High-dose corticosteroids in patients with the adult respiratory distress syndrome. N Engl J Med 317:1565–1570, 1987. 166. Slotman GJ, Fisher CJ Jr, Bone RC, et al: Detrimental effects of highdose methylprednisolone sodium succinate on serum concentrations of hepatic and renal function indicators in severe sepsis and septic shock. The Methylprednisolone Severe Sepsis Study Group. Crit Care Med 21:191–195, 1993. 167. Schonfeld SA, Ploysongsang Y, DiLisio R, et al: Fat embolism prophylaxis with corticosteroids. A prospective study in high-risk patients. Ann Intern Med 99:438–443, 1983. 168. Hooper RG: ARDS: inflammation, infections, and corticosteroids. Chest 100:889–890, 1991. 169. Meduri GU, Headley AS, Golden E, et al: Effect of prolonged methylprednisolone therapy in unresolving acute respiratory distress syndrome: a randomized controlled trial. JAMA 280:159–165, 1998. 170. Steinberg KP, Hudson LD, Goodman RB, et al: Efficacy and safety of corticosteroids for persistent acute respiratory distress syndrome. N Engl J Med 354:1671–1684, 2006. 171. Meduri GU, Golden E, Freire AX, et al: Methylprednisolone infusion in early severe ARDS: results of a randomized controlled trial. Chest 131:954–963, 2007. 172. Papazian L, Forel JM, Gacouin A, et al: Neuromuscular blockers in early acute respiratory distress syndrome. N Engl J Med 363:1107– 1118, 2010. 173. Alhazzani W, Alshahrani M, Jaesdhke R, et al: Neuromuscular blocking agents in acute respiratory distress syndrome: a systematic review and meta-analysis of randomized controlled trials. Crit Care 17(2):R43, 2013. 174. Matalon S, O’Brodovich H: Sodium channels in alveolar epithelial cells: molecular characterization, biophysical properties, and physiological significance. Annu Rev Physiol 61:627–661, 1999. 175. Saumon G, Basset G: Electrolyte and fluid transport across the mature alveolar epithelium. J Appl Physiol 74:1–15, 1993. 175a.  Matthay MA: Resolution of pulmonary edema. Thirty years of progress. Am J Respir Crit Care Med 189:1301–1308, 2014.

176. Zuege D, Suzuki S, Berthiaume Y: Increase of lung sodium-potassiumATPase activity during recovery from high-permeability pulmonary edema. Am J Physiol 271:L896–L909, 1996. 177. Sakuma T, Folkesson HG, Suzuki S, et al: Beta-adrenergic agonist stimulated alveolar fluid clearance in ex vivo human and rat lungs. Am J Respir Crit Care Med 155:506–512, 1997. 178. Hastings RH, Grady M, Sakuma T, et al: Clearance of different-sized proteins from the alveolar space in humans and rabbits. J Appl Physiol 73:1310–1316, 1992. 179. Berthiaume Y, Albertine KH, Grady M, et al: Protein clearance from the air spaces and lungs of unanesthetized sheep over 144 h. J Appl Physiol 67:1887–1897, 1989. 180. Staub NC: Alveolar flooding and clearance. Am Rev Respir Dis 127:S44–S51, 1983. 181. Conhaim RL: Airway level at which edema liquid enters the air space of isolated dog lungs. J Appl Physiol 67:2234–2242, 1989. 182. Kambara K, Longworth KE, Serikov VB, et al: Effect of interstitial edema on lung lymph flow in goats in the absence of filtration. J Appl Physiol 72:1142–1148, 1992. 183. Fukue M, Serikov VB, Jerome EH: Recovery from increased pressure or increased leakiness edema in perfused sheep lungs. J Appl Physiol 77:184–189, 1994. 184. Pearse DB, Wagner EM, Sylvester JT: Edema clearance in isolated sheep lungs. J Appl Physiol 74:126–132, 1993. 185. Wiener-Kronish JP, Broaddus VC: Interrelationship of pleural and pulmonary interstitial liquid. Annu Rev Physiol 55:209–226, 1993. 186. Wiener-Kronish JP, Matthay MA: Pleural effusions associated with hydrostatic and increased permeability pulmonary edema. Chest 93:852–858, 1988. 187. Wiener-Kronish JP, Matthay MA, Callen PW, et al: Relationship of pleural effusions to pulmonary hemodynamics in patients with congestive heart failure. Am Rev Respir Dis 132:1253–1256, 1985. 188. Broaddus VC, Wiener-Kronish JP, Staub NC: Clearance of lung edema into the pleural space of volume-loaded anesthetized sheep. J Appl Physiol 68:2623–2630, 1990. 189. Lum H, Malik AB: Regulation of vascular endothelial barrier function. Am J Physiol 267:L223–L241, 1994. 190. Wiener RS, Moses HW, Richeson JF, et al: Hospital and long-term survival of patients with acute pulmonary edema associated with coronary artery disease. Am J Cardiol 60:33–35, 1987. 191. DiFederico EM, Harrison M, Matthay MA: Pulmonary edema in a woman following fetal surgery. Chest 109:1114–1117, 1996. 192. Thomsen GE, Morris AH: Incidence of the adult respiratory distress syndrome in the state of Utah. Am J Respir Crit Care Med 152:965– 971, 1995. 193. Abbas OM, Nassar AH, Kani NA, et al: Acute pulmonary edema during tocolytic therapy with nifedipine. Am J Obstet Gynecol 195(4):e3–e4, 2006. [Epub 2006 Jul 17]. 194. Milberg JA, Davis DR, Steinberg KP, et al: Improved survival of patients with acute respiratory distress syndrome (ARDS): 1983– 1993. JAMA 273:306–309, 1995. 195. Erickson SE, Martin GS, Davis JL, et al: NIH NHLBI Network. Recent trends in acute injury mortality: 1995–2005. Crit Care Med 37:1574–1579, 2009. 196. Spragg RG, Bernard GR, Checkley W, et al: Beyond mortality: future clinical research in acute lung injury. Am J Respir Crit Care Med 181:1121–1127, 2010. 197. Sigurdsson MI, Sigvaldason K, Gunnarsson TS, et al: Acute respiratory distress syndrome: nationwide changes in incidence, treatment and mortlaity over 23 years. Acta Anaesthesiol Scand 57:37–45, 2013. 198. Severansky JE, Martin GS, Shanholtz C, et al: Mortality in sepsis versus non-sepsis induced acute lung injury. Crit Care 13(5):R150, 2009. 199. Doyle RL, Szaflarski N, Modin GW, et al: Identification of patients with acute lung injury. Predictors of mortality. Am J Respir Crit Care Med 152:1818–1824, 1995. 200. Holguin F, Moss I, Brown LA, et al: Chronic ethanol ingestion impairs alveolar type II cell glutathione homeostasis and function and predisposes to endotoxin-mediated acute edematous lung injury in rats. J Clin Invest 101:761–768, 1998. 201. Villar J, Perez-Mendez L, Basaldua S, et al: A risk tertiles model for predicting mortality in patients with acute respiratory distress syndrome: age, plateau pressure, and P(aO2))/F(IO(2)) at ARDS onset can predict mortality. Respir Care 56:420–428, 2011.

62  •  Pulmonary Edema 1117.e5 202. Chapman HA, Li X, Alexander JP, et al: Integrin alpha6beta4 identifies an adult distal lung epithelial population with regenerative potential in mice. J Clin Invest 121:2855–2862, 2011. 203. Pugin J, Verghese G, Widmer MC, et al: The alveolar space is the site of intense inflammatory and profibrotic reactions in the early phase of acute respiratory distress syndrome. Crit Care Med 27:304–312, 1999. 204. Olman MA, White KE, Ware LB, et al: Pulmonary edema fluid from patients with early lung injury stimulates fibroblast proliferation through IL-1 beta-induced IL-6 expression. J Immunol 172:2668– 2677, 2004. 205. Martin C, Papazian L, Payan MJ, et al: Pulmonary fibrosis correlates with outcome in adult respiratory distress syndrome. A study in mechanically ventilated patients. Chest 107:196–200, 1995.

206. Panos RJ, Bak PM, Simonet WS, et al: Intratracheal instillation of keratinocyte growth factor decreases hyperoxia-induced mortality in rats. J Clin Invest 96:2026–2033, 1995. 207. Ware LB, Matthay MA: Keratinocyte and hepatocyte growth factors in the lung: roles in lung development, inflammation, and repair. Am J Physiol Lung Cell Mol Physiol 282:L924–L940, 2002. 208. Chapman HA: Disorders of lung matrix remodeling. J Clin Invest 113:148–157, 2004. 209. Gotts JE, Matthay MA: Mesenchymal stem cells and the stem cell niche: a new chapter. Am J Physiol Lung Cell Mol Physiol 302:L1147– L1149, 2012.