Acute Lung Injury, Pulmonary Edema, and Multiple System Organ Failure

Published on 01/06/2015 by admin

Filed under Pulmolory and Respiratory

Last modified 22/04/2025

Print this page

This article have been viewed 3116 times

Chapter 27

Acute Lung Injury, Pulmonary Edema, and Multiple System Organ Failure

Elliott D. Crouser, Matthew C. Exline and Ruairi J. Fahy

Chapter Objectives

After reading this chapter you will be able to:

Identify the approximate incidence rate of acute respiratory distress syndrome (ARDS) and how the mortality rate has changed over the past several decades.

State the risk factors associated with the onset of ARDS.

Describe how the normal lung prevents fluid from collecting in the parenchyma and how these mechanisms can fail and cause pulmonary edema.

Describe the effect pulmonary edema has on lung function, including gas exchange and lung compliance.

Describe the relationship between multiple organ dysfunction syndrome (MODS) and ARDS.

Identify the histopathology associated with the exudative phase and the fibroproliferative phase of ARDS.

Differentiate hydrostatic and nonhydrostatic pulmonary edema based on clinical setting.

Describe the principles of supportive care followed for patients with ARDS.

Describe how ventilator settings (e.g., tidal volume, positive end expiratory pressure, respiratory rate) are adjusted for patients with ARDS and MODS.

Describe how mechanical ventilation can cause lung injury and how ventilator-induced lung injury can be avoided.

State the approaches to the management of ARDS and MODS.

Describe the use of innovative mechanical ventilation strategies in the support of patients with ARDS.

State the effect of prone positioning on oxygenation and mortality in a patient with ARDS.

Describe the value of pharmacologic therapies such as nitric oxide and corticosteroids in the treatment of patients with ARDS.

Acute hypoxemic respiratory failure may develop in many clinical settings and is a common reason for admission to the intensive care unit (ICU). Most cases of acute hypoxemic respiratory failure develop as a result of abnormal accumulations of fluid within the lung parenchyma and alveoli. These accumulations are collectively referred to as pulmonary edema. Pulmonary edema may arise from acute illnesses associated with increased pulmonary venous pressure (hydrostatic pulmonary edema or congestive heart failure [CHF]) or may result from conditions associated with acute lung injury (ALI), in which the normal barriers to fluid movement within the lungs are disrupted (nonhydrostatic pulmonary edema). ALI of sufficient severity to cause acute hypoxemic respiratory failure is commonly referred to as acute respiratory distress syndrome (ARDS). ALI and ARDS represent a spectrum of lung injury with many patients initially presenting with ALI and progressing to ARDS with more severe gas exchange abnormalities (Table 27-1).^1,2

TABLE 27-1

Criteria Pressure	Timing	Oxygenation	Chest Radiograph	Pulmonary Artery Wedge
ALI	Acute onset	PaO₂/FiO₂ ≤ 300 mm Hg (regardless of PEEP level)	Bilateral infiltrates seen on frontal chest radiograph	≤18 mm Hg when measured or no clinical evidence of left atrial hypertension
ARDS	Acute onset	PaO₂/FiO₂ ≤ 200 mm Hg (regardless of PEEP level)	Bilateral infiltrates seen on frontal chest radiograph	≤18 mm Hg when measured or no clinical evidence of left atrial hypertension

Modified from 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 49(3 Pt 1):818–824, 1994.

The arbitrary dividing line between ALI and ARDS may have clinical relevance because patients with more severe lung injury often have simultaneous injury to other systemic organs. Acute illnesses associated with widespread systemic organ injury are referred to as multiple organ dysfunction syndrome (MODS), which is the most common cause of death in ICUs. ARDS is the pulmonary manifestation of MODS.

CHF and ARDS are distinct disease processes that require different management strategies. Failure to differentiate these two forms of pulmonary edema may delay appropriate therapy for the underlying disease, resulting in prolonged hospitalization and worse outcome. Because of similarities in clinical presentations, differentiating hydrostatic and nonhydrostatic pulmonary edema is often difficult for even the most skilled clinician.

This chapter initially focuses on the unique mechanisms responsible for acute respiratory failure caused by hydrostatic or nonhydrostatic (sometimes called cardiogenic and noncardiogenic) pulmonary edema. Emphasis is on identifying key clinical features that differentiate these two forms of respiratory failure. This discussion sets the stage for a better understanding of the established principles of supportive care and of the rationale behind innovative treatments of patients with ARDS and MODS.

Epidemiology

ARDS is a common cause of respiratory failure. It can occur as a consequence of critical illnesses of diverse causes. The exact incidence of ARDS varies depending on the population being studied, but in the United States it is thought to range from 13.5 to 64 cases per 100,000 person-years.3,4 ARDS may be present in 16% of mechanically ventilated patients on admission to the ICU.² Despite uncertainties regarding the incidence of ARDS, the mortality rate associated with ARDS has seemed to decline over the past 3 decades from more than 90% to the present level of 30% to 40%.^5,6

The explanation for this favorable trend is likely multifactorial and includes advances in supportive care, early detection, effective management of comorbid diseases such as nosocomial infection, and the broad application of innovative mechanical ventilation techniques. However, the cumulative cost of ARDS and MODS in terms of both human lives and medical resource use remains unacceptably high. The medical community awaits the results of ongoing investigations that may provide insight into the pathogenesis and management of ARDS.

Risk Factors For Acute Respiratory Distress Syndrome

It has been proposed that ARDS can develop via different mechanisms and that the risk factors for ARDS should be categorized as either direct injury via damage directly to the alveolar space or indirect injury initiated by systemic disease (Box 27-1).¹ However, all of these risk factors share the common ability to initiate a systemic inflammatory reaction, which, if sufficiently vigorous, may lead to diffuse lung injury (ARDS). In this regard, the probability of developing ARDS may depend in part on the severity and characteristics of the initial injury. Gastric aspiration and septic shock (sepsis with refractory hypotension) are associated with a greater than 25% risk of ARDS, whereas the administration of multiple blood transfusions is associated with an ARDS risk of less than 5%.⁷ The risk of ARDS seems to be additive when multiple risk factors are present.⁸

Pathophysiology

When fluid filtration exceeds the capacity for drainage through pulmonary lymphatic vessels, several “backup” systems exist for storing additional fluid and for protection against alveolar flooding. Loose connective tissue located along the peribronchovascular space and extending to the level of the bronchiole is capable of storing twice the normal fluid content of the lungs.⁹ Filling of these spaces or cuffs manifests radiographically as increased interstitial infiltrates (Figure 27-3), or Kerley’s lines, which are caused by increased fluid in interlobular septal spaces. The peribronchovascular spaces drain into the local blood vessels or follow the intrinsic pressure gradient in the lung and empty into pulmonary lymphatic vessels. As total lung fluid accumulation increases further, the gel-like matrix of the lung is capable of absorbing additional fluid without affecting interstitial pressure. The latter property is important because fluid is allowed to accumulate in the lungs without transmitting additional hydrostatic pressure to the alveolar barrier. Additional fluid movement into the lungs is avoided. The dense connective tissue making up the alveolar matrix resists edema formation in response to elevated hydrostatic or oncotic pressure.

FIGURE 27-3 Chest radiographs show typical radiographic features of CHF and ARDS. A, CHF is characterized by cardiomegaly, interstitial infiltrates, bilateral perihilar and basilar alveolar infiltrates, and bilateral pleural effusions, which cause blunting of the costophrenic angles. B, ARDS is commonly associated with normal cardiac size, diffuse peripheral alveolar infiltrates, and minimal or absent pleural effusions.

Pulmonary Edema

Despite these protective mechanisms, in certain disease states, fluid flux into the lungs exceeds the capacity of the lung to remove or store the fluid. Under such circumstances, small increases in lung fluid content produce large increases in interstitial hydrostatic pressure. The result is alveolar flooding (pulmonary edema). This pulmonary edema can be due to increased vascular hydrostatic pressure—hydrostatic pulmonary edema—or loss of the endothelial membrane integrity—nonhydrostatic pulmonary edema. Impaired gas exchange in the setting of pulmonary edema is complicated further by an increase in the work of breathing caused by dramatic reductions in lung compliance resulting from alveolar collapse and interstitial fluid accumulation. This scenario commonly results in acute respiratory failure.

Hydrostatic Pulmonary Edema

Hydrostatic pulmonary edema, also called cardiogenic pulmonary edema, is the result of increased venous hydrostatic pressure leading to fluid accumulation and alveolar flooding. The alveolar barrier is composed of the dense alveolar matrix (described earlier), the epithelial basement membrane, and the lining epithelial cells. The alveolar basement membrane is selectively permeable to only very small solutes such that osmotic forces favor the retention of fluid in the intravascular space. Alveolar pressure generally is slightly higher than interstitial pressure, and this difference provides further protection against alveolar fluid accumulation (see Figure 27-2). Under normal circumstances, there is little fluid movement into the alveoli across the alveolar barrier. Fluid that does form is composed of low-molecular-weight proteins and is actively transported back into the interstitium by type II pneumocytes.¹⁰ In contrast, when the vascular hydrostatic pressure increases and overwhelms the defense mechanisms against fluid accumulation in the lung, hydrostatic pressure increases sharply within the interstitium of the lung, and alveolar flooding ensues. The precise mechanism of alveolar flooding in hydrostatic pulmonary edema is unknown. It has been shown, however, that alveolar flooding occurs in an “all-or-none” manner. The alveolar fluid formed in the setting of increased hydrostatic pressure has characteristics identical to interstitial fluid even though the alveolar epithelium is normally impermeable to large proteins and molecules.⁹ This observation lends support to the findings of Conhaim,¹¹ who showed experimentally that high interstitial fluid pressure leads to alveolar flooding through leakage of fluid from the epithelium of respiratory bronchioles, alveolar ducts, and their associated alveoli. The epithelium of the respiratory bronchioles and alveolar ducts may be particularly prone to hydrostatic injury because these locations represent the transition zone between respiratory and alveolar epithelium.

Nonhydrostatic Pulmonary Edema

Nonhydrostatic pulmonary edema, also called noncardiogenic pulmonary edema, is the result of the loss of microvascular membrane integrity. In contrast to hydrostatic pulmonary edema, nonhydrostatic pulmonary edema is associated with increased total lung water despite normal microvascular hydrostatic pressure. The mechanisms of nonhydrostatic pulmonary edema that ultimately lead to ARDS are more complex than the mechanisms responsible for hydrostatic pulmonary edema. Although many seemingly unrelated risk factors for ARDS have been identified, all causes of ARDS evoke disruption of endothelial and epithelial barriers and typically occur under conditions associated with widespread microvascular injury to the lungs. Vascular endothelial injury in the lungs results in increased microvascular permeability and fluid filtration, such that there is uninhibited entry of protein-rich fluid into the pulmonary interstitium. Alveolar flooding develops when the osmotic gradient between the capillary and the lung becomes essentially zero and no longer opposes the hydrostatic forces favoring fluid movement from the capillary into the lung: K_fc(P_mv − P_i) >> (s_d)(TT_mv − T_ti) (see earlier equation). This process is likely facilitated both by damage to the normally impermeable alveolar epithelial barrier, which is a key feature of ARDS,¹² and by impaired alveolar fluid clearance in ALI and ARDS.¹³ Investigations support a role for both necrotic epithelial cell death and programmed cell death (apoptosis) in the pathogenesis of alveolar wall damage.^14,15 In addition, therapies aimed at improving the vascular barrier to fluid movement have shown protection against ALI animal models.

Nonhydrostatic pulmonary edema is caused by the loss of the normal osmotic gradient that normally opposes fluid movement into the lungs. Fluid accumulation in the lung is further amplified by impaired pulmonary fluid clearance and in the presence of higher hydrostatic pressures (i.e., pulmonary capillary pressure).

A common mechanism to explain how different acute illnesses (e.g., sepsis, gastric acid aspiration, pancreatitis) can lead to the development of ARDS was proposed by Weiland and colleagues.¹⁶ These investigators showed that ARDS, regardless of the cause, is associated with an influx of polymorphonuclear neutrophils (PMNs) and PMN-derived inflammatory by-products, such as neutrophil elastase and myeloperoxidase, into the lung. The PMN-activating cytokine interleukin (IL)-8 has been shown to be increased in the lungs of patients with ARDS, whereas a reduction in IL-8 and PMNs has been shown to correlate with recovery from ARDS.¹⁷ Taken together, the results of these studies suggest that the common mechanism for the development of ARDS is induction of lung inflammation leading to loss of membrane integrity, as a result of either local lung injury (direct injury) or systemic inflammation (indirect injury).

Although PMNs play a central role in the development of ARDS, other chemical (e.g., gastric aspiration) and immunologic pathways participate in the initiation and development of the systemic inflammatory response to critical illnesses associated with ARDS. Sepsis is associated with intense activation of systemic inflammatory pathways such that cytokines (e.g., tumor necrosis factor [TNF]-alpha, IL-1beta, IL-6, and IL-8), arachidonic acid metabolites (e.g., platelet-activating factor, leukotrienes), and nitric oxide (NO) all contribute to the hemodynamic and inflammatory events characteristic of this syndrome.¹⁸ The relative contributions and exact roles of these proinflammatory mediators in the pathogenesis of ARDS and MODS remain a topic of intense investigation and controversy.^19,20 It is likely that there is no single dominant pathway in the development of ARDS because attempts to control the inflammatory response in sepsis by blocking specific mediators, such as anti-TNF antibodies, have not proved beneficial and may worsen outcome.

Gas Exchange and Lung Mechanics in Acute Respiratory Distress Syndrome

ARDS is associated with restrictive physiology and refractory hypoxemia, which are largely a result of pulmonary microvascular injury. Specifically, increased pulmonary capillary permeability facilitates the influx of inflammatory fluid into the lung interstitium and alveolar spaces and causes decreased lung compliance and alveolar consolidation. The presence of intraalveolar inflammatory fluid impairs surfactant synthesis and function so that pulmonary gas exchange (i.e., related to atelectasis) and compliance are further impaired. The negative effects of alveolar consolidation and atelectasis on pulmonary gas exchange are exacerbated by a loss of the normal vascular response to alveolar hypoxemia. The body is unable to shunt blood away from the diseased alveoli, and these unaerated alveoli receive excessive blood flow, which contributes to severe ventilation/perfusion mismatching and an effective intrapulmonary right-to-left shunting of blood flow in ARDS. The pulmonary manifestations of ARDS and CHF are summarized in Box 27-2.

Box 27-2

Clinical Features of Congestive Heart Failure and Acute Respiratory Distress Syndrome

Features Common to Chf and Ards

• Symptoms of anxiety, dyspnea, tachypnea

• Reduced lung volumes and decreased compliance

• Arterial blood gases initially show respiratory alkalosis and arterial hypoxemia

• Chest radiograph shows diffuse alveolar and interstitial infiltrates

Features Favoring Chf

• Clinical history suggestive of CHF (see Box 27-1)

• Cardiomegaly or pleural effusions on chest radiograph (see Figure 27-3)

• PCWP > 18 mm Hg

• BALF nonproteinaceous and noninflammatory

Features Favoring Ards

• Clinical history of risk factors for ARDS (see Table 27-1)

• Peripheral infiltrates on chest radiograph (see Figure 27-3)

• PCWP > 18 mm Hg

• BALF proteinaceous and inflammatory

• Pathologic examination shows diffuse alveolar damage, type II pneumocyte hyperplasia with or without fibrosis

• Ratio of PaO₂/FiO₂ < 200 *

^*Impaired oxygenation is characteristic of ARDS. By definition, ARDS and ALI are distinguished by the ratio of PaO₂ to FiO₂. In ARDS, this ratio is <200, and in ALI, the ratio is <300.

BALF, Bronchoalveolar lavage fluid.

Role of Organ-Organ Interactions in the Pathogenesis of Acute Respiratory Distress Syndrome and Multiple Organ Dysfunction Syndrome

The notion that factors operating outside the lungs may participate in the initiation and progression of ARDS has generated interest in the role of organ system interactions in the pathogenesis of ARDS and MODS. For example, injury to remote systemic organs is known to occur after ALI,²¹ apparently related to PMN-mediated mechanisms.²² In this way, ALI may perpetuate the systemic inflammatory response and lead to further lung and systemic organ injury. The gut-liver-lung axis may be most influential in causing the systemic inflammatory response associated with ARDS and MODS. The gastrointestinal (GI) tract contains large quantities of potentially pathogenic bacteria against which the host is normally protected by intact mucosal barriers and the reticuloendothelial system. However, the function of the GI tract and the liver is frequently compromised in critical illness. The widespread use of broad-spectrum antibiotics in the care of critically ill patients often leads to overgrowth of antibiotic-resistant bacteria within the GI tract. These bacteria and their toxic by-products (e.g., endotoxin) escape from the GI tract and are taken up by the reticuloendothelial cells of the liver, spleen, and regional lymph nodes. The resultant activation of the reticuloendothelial system may initiate and perpetuate a systemic inflammatory response that leads to systemic organ injury (i.e., ARDS and MODS).²³ This sequence of events forms the basis for strategies designed to reduce the release of proinflammatory mediators from the GI tract, including selective decontamination, early enteral feeding, and other approaches designed to moderate the systemic inflammatory response in ARDS and MODS (see later section on Therapeutic Approach to Acute Respiratory Distress Syndrome).

Why ARDS and MODS develop in some patients with ALI and not in others is unknown. The determinants of ARDS may relate to the balance between proinflammatory and antiinflammatory factors within the body. In this regard, the liver plays a major role in both induction and modulation of the systemic inflammatory response to all kinds of initiating events and is primarily responsible for the breakdown of endogenous proinflammatory mediators, including TNF-alpha, leukotrienes, and others.²⁴ Patients with liver disease have higher levels of circulating proinflammatory mediators, are more prone to bacteremia, and have a high incidence of ARDS compared with patients without liver disease.²⁵

The liver is not the sole determinant of ARDS and MODS in critically ill patients. Other factors, such as the severity of the primary illness and comorbid diseases (e.g., cardiac disease, advanced age, renal failure, malignant disease) and perhaps the genetic profile of the patient, also seem to predispose patients to ARDS and MODS.23,26

Histopathology And Clinical Correlates Of Acute Respiratory Distress Syndrome

Exudative Phase (1 to 3 Days)

The exudative phase is characterized by diffuse damage to alveoli and blood vessels and the influx of inflammatory cells into the interstitium. Many of the alveolar spaces become filled with a proteinaceous, eosinophilic (on hematoxylin and eosin stain) material called hyaline membranes, which are composed of cellular debris and condensed plasma proteins. Pathologically, there is destruction of type I pneumocytes, which are normally the predominant cells lining the alveoli; type II pneumocytes are relatively resistant to injury.27,28 Patients with ARDS have profound dyspnea, tachypnea, and refractory hypoxemia. This phase of ARDS often is difficult to differentiate from respiratory failure related to hydrostatic pulmonary edema (CHF). The clinical presentations of these two forms of acute respiratory failure are discussed later (see the section on Differentiating Hydrostatic from Nonhydrostatic Pulmonary Edema in the Clinical Setting). The exudative phase may be self-limited or may progress to a fibroproliferative phase.

Rule of Thumb

A careful history and physical examination often are the most useful means by which CHF and ARDS can be initially differentiated in a patient who has refractory hypoxemia and bilateral infiltrates on chest radiographs. A history consistent with one of the common causes of CHF (Box 27-3) combined with physical examination findings of jugular venous distention, cardiac murmurs or gallops, bibasilar crackles, or peripheral edema suggests a diagnosis of CHF. ARDS is more likely when the history is positive for one of the established risk factors (see Box 27-1) and there is no clinical evidence in support of CHF.

Fibroproliferative Phase (3 to 7 Days)

After inflammatory injury to the lung is established and the initiating events are controlled, a process of lung repair begins. Pathologically, there is hyperplasia of alveolar type II pneumocytes and proliferation of fibroblasts within the alveolar basement membrane and intraalveolar spaces. Fibroblasts mediate the formation of intraalveolar and interstitial fibrosis.²⁷ The extent of fibrosis determines the degree of pulmonary disability in patients who survive ARDS.

The exact mechanisms controlling lung remodeling in ALI are not well established but very likely involve by-products of inflammatory cells (e.g., proteases, antiproteases, IL-6) and various growth factors (transforming growth factor [TGF]-alpha, TGF-beta).29,30 However, the remodeling process after ARDS is quite variable. In some cases, patients have nearly complete normalization of lung compliance and oxygenation for 6 to 12 months after the illness. In other cases, the architecture of the lung never returns to normal, and patients experience severe respiratory disability related to extensive pulmonary fibrosis and obliteration of the pulmonary vasculature. The pattern of fibrosis after ALI suggests that, as in repair of the skin, an intact basement membrane is necessary for normal repair of the epithelium of the alveoli. It follows that disruption of the alveolar basement membrane is a prerequisite to the development of fibrosis after ALI. This line of reasoning is supported by the observation that the extent of recovery depends on the severity of the initial lung injury and on the influence of secondary forms of injury. Secondary forms of lung injury include nosocomial infection, O₂ toxicity, and barotrauma (see later section on Therapeutic Approach to Acute Respiratory Distress Syndrome).

Differentiating Hydrostatic from Nonhydrostatic Pulmonary Edema in the Clinical Setting

The diagnostic criteria for ARDS are shown in Table 27-1. However, despite the existence of distinct pathophysiologic mechanisms of hydrostatic and nonhydrostatic pulmonary edema, differentiating these two forms of acute respiratory failure may be difficult because of similarities in their early clinical presentations. CHF is more common than ARDS and should be considered when any patient has pulmonary edema, in particular, when the history and physical examination findings suggest one of the causes of CHF listed in Box 27-3. Alveolar flooding by either hydrostatic or nonhydrostatic mechanisms results in diffuse radiographic infiltrates, altered gas exchange, and abnormal mechanical properties of the lung (see Box 27-2). A clinical history of infection, recent trauma, or risk factors for aspiration may be present in either patient group. Likewise, many patients with ARDS are older and have preexisting illnesses that place them at risk of CHF. In patients with acute hypoxemic respiratory failure, CHF must always be considered, even when the patient has obvious risk factors for ARDS.

The ability to discern CHF and ARDS solely on the basis of radiographic findings is difficult and is complicated by technical limitations associated with obtaining chest radiographs in the ICU. Both CHF and ARDS are characterized by diffuse alveolar infiltrates prevalent in dependent lung zones. CHF is more often associated with cardiomegaly and the presence of perihilar infiltrates and pleural effusions. ARDS is more often associated with the presence of peripheral alveolar infiltrates, air bronchograms, sparing of the costophrenic angles, and normal cardiac size. However, cardiac size may be difficult to interpret on an anteroposterior chest x-ray, and when images are obtained with the patient in the supine position, pleural effusions may be obscured. Consequently, the hallmark radiographic distinctions between CHF and ARDS are often invisible to the clinician, a situation further complicated by the occasional coexistence of both hydrostatic and nonhydrostatic pulmonary edema.

Alveolar flooding of any cause (i.e., hydrostatic or nonhydrostatic) is associated with impaired gas exchange and abnormal lung mechanics. Both CHF and early ARDS (exudative phase) are associated with interstitial and alveolar accumulation of fluid. As a result of ventilation/perfusion mismatching and shunt accompanying this fluid accumulation, arterial hypoxemia develops. Patients with interstitial and alveolar edema of any cause use a higher fraction (25% to 50%) of their total metabolic output to support the increased work of breathing attendant to reduced lung compliance and higher ventilatory rates. However, on the basis of gas exchange and lung mechanics alone, alveolar flooding related to hydrostatic causes may be indistinguishable from alveolar flooding related to nonhydrostatic causes.

In many cases, measurement of hemodynamic variables is necessary to determine whether hydrostatic or nonhydrostatic forces underlie the development of acute hypoxemic respiratory failure. To this end, a pulmonary artery (Swan-Ganz) catheter is a useful clinical tool because it allows estimation of clinically relevant data, such as cardiac output and pulmonary capillary wedge pressure (PCWP), also called pulmonary artery occlusion pressure. Under ideal circumstances, PCWP closely corresponds to left ventricular end-diastolic pressure and is a reflection of the hydrostatic forces applied to the pulmonary capillary system. In this regard, PCWP greater than 18 mm Hg is necessary for the development of hydrostatic pulmonary edema. Conversely, alveolar flooding resulting from nonhydrostatic pulmonary edema can occur at any PCWP and is the exclusive cause of pulmonary edema at PCWP less than 18 mm Hg.

Invasive hemodynamic monitoring may be unreliable in patients with high airway pressure. In particular, high levels of positive end expiratory pressure (PEEP) lead to expansion of non–zone 3 lung areas in which alveolar pressure exceeds venous pressure. PCWP measured in non–zone 3 lung reflects alveolar pressure and not left ventricular pressure.³¹ Hydrostatic forces may be overestimated or misdiagnosed in ARDS as hydrostatic pulmonary edema. These confounding variables may lead to misinterpretation of the hemodynamic status of the patient, resulting in mismanagement of his or her condition, and could explain why the routine use of pulmonary artery catheters in patients with ARDS is not beneficial.³² A similar estimate of pulmonary capillary hydrostatic pressure may be obtained via central venous pressure (CVP) obtained from a central line placed in the superior vena cava. CVP is less technically difficult to obtain and interpret compared with PCWP; in a large clinical trial, therapy guided by CVP estimation of the patient’s volume status was equivalent to therapy guided by PCWP.³³

Another useful method of separating nonhydrostatic from hydrostatic pulmonary edema is based on differences in the characteristics of the edema fluid. As previously discussed (see the section on Pathophysiology), ARDS is associated with inflammatory injury to the pulmonary microvasculature, which allows the influx of inflammatory cells and proteinaceous fluid into the interstitium and alveolar spaces. The inflammatory nature of this exudative fluid is reflected by the presence of large quantities of inflammatory cells (predominantly neutrophils) in bronchoalveolar lavage fluid (BALF). The BALF findings provide diagnostic insights if infectious agents or signs of aspiration (e.g., food particles) are present. In contrast, although hydrostatic pulmonary edema is associated with alveolar flooding, the edema fluid typically is noninflammatory, and the protein content is much lower than the protein content of BALF.³⁴ BALF often provides useful insight into the underlying cause of hypoxemic respiratory failure, but the role of BALF in the clinical management of patients with acute respiratory failure caused by pulmonary edema is not well established. The clinical characteristics that differentiate acute hypoxemic failure caused by hydrostatic pulmonary edema from acute hypoxemic failure caused by nonhydrostatic pulmonary edema are summarized in Box 27-2.

Mini Clini

Determination of “Optimal” PEEP

Problem

Although PEEP is capable of recruiting collapsed alveolar units in patients with ARDS, it adversely affects cardiac output and increases ventilatory pressure. The clinician must approach PEEP as a therapy that has risks and benefits.

Discussion

In patients with ARDS, increasing levels of PEEP are associated with improved oxygenation and a reduction in the measured shunt fraction (PaO₂/FiO₂ ratio) (see Figure 27-5). Routine use of high levels of PEEP has not been shown to be beneficial for all patients. However, patients with severe ARDS may benefit from higher PEEP levels to maintain oxygenation.³⁵ Despite initial increases in oxygenation above the level of PEEP designated the “optimal level” in Figure 27-5, there is a net decrease in systemic O₂ delivery. Lung compliance may begin to decline at higher levels of PEEP, and the risk of volutrauma increases. Also, healthy alveoli may become overdistended, resulting in decreased perfusion and stretch-induced injury of the “good lung.” The pressure-volume relationships of the lungs may be used to adjust PEEP such that end expiratory alveolar collapse and end inspiratory overinflation are minimized (see Figure 27-6).

Although the optimal PEEP must be determined for each patient, optimal levels of PEEP usually are in the range of 8 to 20 cm H₂O. The tidal volume may have to be decreased whenever PEEP is increased to avoid alveolar overdistention.

Therapeutic Approach To Acute Respiratory Distress Syndrome

The management of critical illness associated with ARDS previously was confined to supportive therapy (Box 27-4) designed to preserve systemic organ function and allow recovery from the underlying illness. Strategies have evolved for controlling the systemic inflammatory response that leads to lung and other organ injury. This section presents an overview of the current approach to supportive care and new potential therapies for ARDS and MODS (Figure 27-4).

Box 27-4 Supportive Care of Patients With Acute Respiratory Distress Syndrome

1. Identify and manage underlying cause of ARDS.

2. Avoid secondary (iatrogenic or nosocomial) lung injury:

a. Avoid O₂ toxicity.

b. Avoid aspiration.

c. Avoid barotrauma and volutrauma.

d. Identify and manage nosocomial infection and pneumonia.

e. Extubate as soon as is feasible.

3. Maintain adequate DO₂ to systemic organs.

a. Minimize demand by reducing metabolic rate:

Control fever.

Control anxiety and pain.

b. Support cardiovascular system with intravenous fluids and vasopressor agents, as necessary, to

Prevent hypotension (systolic blood pressure >90 mm Hg and mean arterial pressure >60-70 mm Hg).

Reverse lactic acidosis.

Maintain adequate urine output.

4. Provide nutritional support.

5. Optimize gas exchange by avoiding excessive fluid administration.

FIGURE 27-4 As the severity of lung injury worsens, clinical teams should provide the best evidence-based supportive care including low tidal volume ventilation, conservative fluid management, and daily attempts at ventilator liberation. In patients with significant ARDS and refractory hypoxemia (arbitrarily defined as a PaO₂/FiO₂ ratio <100), transfer to a center with one or more rescue therapies should be considered.

Supportive Care of Patients With Acute Respiratory Distress Syndrome and Multiple Organ Dysfunction Syndrome

The fundamental principles of supportive care of patients with ARDS and MODS are outlined in Box 27-4.

Hemodynamics and Fluid Management in Acute Respiratory Distress Syndrome

Preservation of vital organ integrity through the optimization of systemic O₂ delivery is a principal goal of supportive management in all causes of respiratory failure. Ventilator strategies designed to improve arterial oxygenation must be weighed against any potential changes in hemodynamic values incurred as a result of these strategies. In view of the deleterious cardiopulmonary effects of high-pressure mechanical ventilation (e.g., PEEP) in ARDS, it is logical to imagine that invasive monitoring of both cardiac output and systemic O₂ delivery (DO₂) would be beneficial. As previously mentioned, the role of invasive hemodynamic monitoring (e.g., pulmonary artery catheters) in the management of ARDS was historically controversial. However, clarity was provided by the National Institutes of Health National Heart, Lung and Blood Institute–sponsored ARDS Clinical Trials Network, which evaluated the benefits and risks of the use of pulmonary artery catheters to guide treatment in the setting of ALI. The study involved 1000 participants and showed no benefit in terms of mortality, ventilator-free days, and duration of ICU admission and no difference in terms of overall fluid balance. The authors of the study concluded that pulmonary artery catheters “should not be routinely used for the management of ALI.”³³

Gas exchange is highly dependent on total lung fluid during the exudative phase of ARDS, and small increases in hydrostatic forces (PCWP) lead to significant decreases in oxygenation consequent to alveolar flooding and associated right-to-left shunting. Measures that restrict intravascular volume are associated with improved oxygenation. However, as with increasing PEEP, improvements in arterial oxygenation attendant to reducing PCWP must be weighed against reduced cardiac output, as reflected by the measured DO₂ or other measures of systemic tissue oxygenation. In this context, a randomized, multicenter study conducted by the ARDS Clinical Trials Network compared conservative and liberal fluid management strategies for the first 7 days of treatment of 1000 patients with ALI. In keeping with the concept of “leaky” pulmonary capillaries in the setting of ALI resulting in increased susceptibility to lung fluid accumulation in response to increased hydrostatic forces, conservative fluid management was associated with improved arterial oxygenation, increased ventilator-free days, and shorter stays in the ICU. There was no statistical improvement in 60-day mortality, and there was no difference between the groups in terms of vital organ failures. Nonetheless, the results of this study strongly favor the routine use of a conservative fluid management strategy in patients with ARDS.³⁶

Rule of Thumb

The beneficial effects of PEEP are optimal at a pressure of 20 cm H₂O or less in most patients with ARDS. Levels of PEEP greater than 20 cm H₂O should not be routinely used unless the benefits of higher levels of PEEP are supported by objective end points, such as improved lung compliance (Figure 27-5) or optimal alveolar recruitment (see Figure 27-6).

FIGURE 27-5 Determination of “optimal PEEP” from simultaneous measurements of hemodynamic (DO₂), gas exchange (shunt fraction and arterial oxygenation [PaO₂]), and physiologic values. Optimal PEEP does not correspond to PEEP associated with optimal pulmonary gas exchange. When adjusting PEEP, the clinician should consider its effects on systemic DO₂ and lung compliance such that systemic organ injury and lung injury are minimized.

With regard to tissue oxygenation in critical illness, various investigations have shown that an abnormal dependence of O₂ consumption () on DO₂ exists throughout the physiologic range of DO₂ in many critically ill patients.³⁷ This abnormal dependence of on DO₂ is associated with impaired O₂ extraction. These observations have been interpreted to imply that tissue hypoxia may exist and is contributing to organ failure (MODS) in these patients. This notion formed the basis for studies in which DO₂ was increased to “supranormal” levels in patients at risk of sepsis and ARDS.⁷ The encouraging results of early trials using this strategy^38,39 were not supported by larger, well-designed clinical trials in cohorts of patients with established sepsis.^40,41 One study showed a higher mortality associated with the use of this strategy.⁴⁰ Because of these findings, efforts to increase DO₂ beyond normal values (3.5 L/min/kg) are not currently recommended. The “optimal DO₂” for critically ill patients may never be established because the needs of individual patients must be factored into the decision to augment DO₂. New techniques for detecting tissue hypoxia are needed for guiding the hemodynamic management of these patients. Until such tools become available, it seems prudent to prevent hypotension (systolic arterial blood pressure >90 mm Hg, mean arterial blood pressure >60 mm Hg), consider augmentation of DO₂ in the setting of hyperlactatemia, optimize hemoglobin saturation (>90%), and ensure adequate organ function (e.g., urine output).

Mechanical Ventilation in Acute Respiratory Distress Syndrome

Despite the presence of widespread pulmonary injury and altered gas exchange in ARDS, more recent investigation has shown that aerated portions of the lung have near-normal mechanical characteristics. Gattinoni and Pesenti ⁴² suggested that three distinct zones exist in the lungs of patients with ARDS. The most dependent lung zones are characterized by dense pulmonary infiltrates corresponding to nonventilated lung units. A second lung zone also has dense pulmonary infiltrates and nonventilated alveolar units but is distinct from the more dependent lung zones in that these areas may be made available for gas exchange through changes in the mode of mechanical ventilation. Finally, nondependent lung zones are fully inflated and receive most of the ventilation. The aerated lung zones have been shown to retain normal mechanical characteristics as reflected by the measured specific compliance (static compliance adjusted for the volume of ventilated lung). In ARDS, the lungs are effectively diminished in size to 20% to 30% of normal, but the aerated portions of the lungs retain near-normal physiologic properties.

Setting Tidal Volume

Because of the heterogeneous properties of the lungs in ARDS, mechanical ventilation with large tidal volumes is inappropriate for these patients. Tidal volumes of 10 to 15 ml/kg, when distributed primarily to the relatively small aerated lung zones, lead to hyperinflation and overdistention of the alveoli in these areas. In animal models, alveolar hyperinflation has been shown to result in altered alveolar capillary permeability identical to that associated with ARDS. The excessive volume, not high pressure, is responsible for lung injury.⁴³ Lung tissue injury induced by alveolar hyperinflation has been termed volume trauma or volutrauma and can be avoided with the use of lower tidal volumes. In the original ARDSNet trial, patients with ARDS ventilated with an initial tidal volume of 6 ml/kg and a targeted plateau pressure of 30 cm H₂O had a significantly reduced mortality compared with patients ventilated with 12 ml/kg.⁴⁴

Mini Clini

Managing Hydrostatic Pressure in Patients With ARDS

Problem

Critically ill patients requiring mechanical ventilation for acute respiratory failure often receive large volumes of fluid in the form of intravenous medications, maintenance fluids, and enteral or parenteral feedings. These patients consequently are at very high risk for the development of iatrogenic pulmonary edema secondary to increased pulmonary vascular hydrostatic pressure (increased intravascular volume). Excessive fluid administration is of particular concern in patients with ARDS who have “leaky capillaries,” particularly in the early phases of the disease, and are remarkably sensitive to changes in pulmonary vascular hydrostatic pressure. How can iatrogenic pulmonary edema be avoided in these patients so that liberation from the ventilator can be expedited?

Discussion

Patients with respiratory insufficiency of any cause may not tolerate additional demands on the cardiorespiratory system, including the demands associated with pulmonary edema owing to intravascular volume overload. The patient’s volume status must be monitored closely by noninvasive and, in the case of more severely ill patients, invasive means. Noninvasive estimates of total body water, such as daily weights and total fluid intake and total fluid output measurements, are prone to error related to confounding variables. For example, limited patient mobility interferes with the ability to obtain exact measurements of the patient’s weight. Likewise, total fluid intake and output measurements are limited by the loss of information related to insensible fluid losses (e.g., perspiration and respiration-related fluid loss).

Tidal volume ideally would be selected on the basis of the patient’s individual pressure-volume relationships. Pressure-volume relationships can be established for each patient by measuring airway pressure changes over a wide range of tidal volumes. These measurements are used to describe the lower inflection point (LIP) and upper inflection point (UIP) (Figure 27-6). The UIP corresponds to the development of regional lung overdistention. Ventilatory pressure exceeding the pressure associated with the UIP is likely to cause lung injury. In contrast, the LIP represents the point in the pressure-volume curve at which dynamic collapse of alveolar units begins to occur. At ventilator pressures associated with lung volumes below the LIP, alveoli begin to collapse, and oxygenation is impaired. As airway pressure decreases to less than the LIP at end-expiration and increases to more than the LIP during the ventilatory cycle, the alveoli undergo repeated collapse and reexpansion. The shear stress induced by this cyclic opening and closing of the alveoli, particularly relating to forces generated when an alveolus opens adjacent to closed alveolar units creating tremendous distortion of the alveolar walls, represents another possible mechanism of ventilator-associated lung injury.⁴⁵

FIGURE 27-6 Typical pressure-volume relationships during normal conditions and during ARDS. At low lung volume, inspiratory pressure increases faster than lung volume *(line A)* owing to high alveolar surface tension. As alveoli open, surface tension decreases, and the pressure required to increase lung volume further decreases *(line B).* The lower inflection point *(LIP)* occurs between *lines A* and B and represents the volume above which most alveolar units are open. The upper inflection point *(UIP)* occurs at near-maximal lung volume and corresponds to the point at which further increases in pressure result in minimal increases in lung volume *(line C).* Pressure applied above the *UIP* is associated with alveolar distention. During the expiratory phase of the respiratory cycle *(dotted line),* the lower deflection point *(LDEF)* is the point below which lung volumes slowly decrease (alveolar units collapse). During ARDS *(gray dashed curve)*, pressure-volume relationships change such that higher pressure is needed to maintain alveolar patency, and alveolar distention occurs at lower tidal volumes. Near-maximal lung volumes are typically achieved at an inspiratory pressure of approximately 35 cm H₂O in normal conditions and during *ARDS*. Attempts to increase inspiratory pressure to more than 35 cm H₂O provide little additional ventilation and substantially increase the risk of volume trauma to the lungs. The *LIP* corresponds to pressure below which alveolar collapse occurs *(line A).* Under normal conditions, the *LIP* may not be evident (i.e., minimal alveolar collapse occurs at end-inspiration under normal conditions). In contrast, during *ARDS*, the *LIP* often is evident during measurements of pressure-volume relationships and frequently occurs at a pressure of 5 to 15 cm H₂O. PEEP at levels greater than *LIP* may prevent end expiratory alveolar collapse and reduce lung injury secondary to alveolar shear stress.

Ventilator strategies designed to optimize pressure-volume relationships and minimize ventilator-associated lung injury were initially described by Amato and colleagues ⁴⁶ and were subsequently evaluated by the ARDS Clinical Trials Network. These strategies, termed volume-controlled ventilation (see later section on Innovative Ventilation Strategies for Acute Respiratory Distress Syndrome), have been an encouraging finding in an otherwise disappointing quest to discover effective treatments for patients with ARDS.

Adjusting Positive End Expiratory Pressure

The rationale behind the use of PEEP in the care of patients with ARDS is not directly related to its effect on lung injury. PEEP may contribute to further pulmonary injury (volutrauma) in these patients. The benefits of PEEP relate to recruitment of additional alveoli, which results in an increase in functional residual capacity (FRC) and improved oxygenation. By improving arterial oxygenation, PEEP may enable reduction of the fraction of inspired O₂ (FiO₂) and diminish the risk of O₂ toxicity to the lungs. When patency of alveolar units is maintained throughout the ventilatory cycle, the damaging effects of opening and closing alveoli with each ventilatory cycle can be avoided. The form of lung injury occurring at lower lung volumes has been termed airway shear trauma, and the point at which it occurs is reflected by the LIP of the pressure-volume curve (see Figure 27-6).

The beneficial effects of PEEP must be balanced against the negative effects. Because the primary goal of mechanical ventilation is to provide adequate oxygenation at safe levels of FiO₂ while maintaining adequate DO₂ to the body, the inverse relationship between PEEP and cardiac output must be considered. One common clinical scenario involves increasing PEEP to improve arterial oxygenation at the expense of a reduction in the overall DO₂. For this reason, patients who present with ARDS necessitating the use of PEEP may benefit from invasive hemodynamic monitoring. The more recent ARDS Clinical Trials Network study that evaluated low tidal volume with either high or low PEEP failed to show a survival advantage or shorter ventilator time with either measure.⁴⁷ It is reasonable to use the lowest level of PEEP that maintains adequate oxygenation.

As with tidal volume adjustments, the optimal level of PEEP is different for each patient. On the basis of the previous discussion, the goals of PEEP therapy are as follows:

• Provide adequate oxygenation (PaO₂ > 55 mm Hg) at a safe FiO₂ (<0.6).

• Ensure adequate tissue oxygenation.

• Maintain the patency of alveolar units throughout the ventilatory cycle (see Figure 27-6).

• Avoid barotrauma by maintaining mean airway pressure less than 35 cm H₂O or less than the pressure that corresponds to the UIP of the pressure-volume curve (see Figure 27-6).

Rule of Thumb

Administration of high levels of supplemental O₂ (FiO₂ > 0.6) for longer than 24 hours can cause lung injury as a result of O₂ toxicity (oxidative stress), which advances ARDS and lung fibrosis. As the FiO₂ becomes greater than 0.6, the time required to cause lung injury decreases. The FiO₂ should be decreased to 0.6 as soon as possible via supportive therapy, such as positive pressure ventilation, PEEP, manipulations of pulmonary vascular pressure, or another recommended therapy (Tables 27-2 and 27-3).

TABLE 27-2

Criteria for Evidence-Based Recommendations for Management of Acute Respiratory Distress Syndrome (ARDS)

Quality of Evidence
Level 1	Randomized, prospective, controlled investigations of ARDS
Level 2	Nonrandomized concurrent cohort investigations, historical cohort investigations, and case series of patients with ARDS
Level 3	Randomized, prospective, controlled investigations of sepsis or other relevant conditions that have potential application to ARDS
Level 4	Case reports of ARDS
Grading of Recommendations
A	Supported by at least two Level 1 investigations
B	Supported by only one Level 1 investigation
C	Supported by Level 2 investigations only
D	Supported by at least one Level 3 investigation
Ungraded	No available clinical investigations

Modified from Kollef MH, Schuster D: Acute respiratory distress syndrome. Dis Mon 42:270–326, 1996.

TABLE 27-3

Recommendations for Nonpharmacologic Management of Acute Respiratory Distress Syndrome

Treatment	Recommendation	Grade
Mechanical ventilation		Ungraded
Initial settings: assist control mode; FiO₂ 1.0; PEEP ≤5 cm H₂O; inspiratory flow, 60 L/min	Yes
Tidal volume, 6-10 ml/kg	Yes	C
Prophylactic PEEP (≤5 cm H₂O)	No	B
Least PEEP with SaO₂ ≥ 0.9 and FiO₂ ≤ 0.6	Yes	Ungraded
Permissive hypercapnia to maintain peak airway pressure <40-45 cm H₂O and plateau pressure <35 cm H₂O	Yes	C
Routine use of IRV	No	C
IRV for persistent hypoxemia or elevated airway pressure	Yes	C
APRV	No*
HFV	No	B
Tracheal gas insufflation	*
Partial liquid ventilation	*
ECMO	No	B
ECCO₂R	No	B
Patient repositioning (including prone position)	Yes	C
Early fluid restriction or diuresis	Yes	B
Supranormal DO₂ goals	No	D

IRV, Inverse ratio ventilation; APRV; airway pressure release ventilation; HSV, high frequency ventilation; ECMO, extracorporeal membrane oxygenation; ECCO₂R, extracorporeal carbon dioxide removal.

^*Pending results of ongoing clinical trials.

Modified from Kollef MH, Schuster D: Acute respiratory distress syndrome. Dis Mon 42:270–326, 1996.

Adjusting the Ventilatory Rate

ARDS is associated with alveolar consolidation and ventilation/perfusion mismatching, which results in a decrease in the number of normally functioning alveoli. Critically ill patients often have elevated metabolic rates so that CO₂ production is increased. Compared with individuals with normal lungs, patients with ARDS require much higher minute ventilation to maintain PaCO₂ in the normal range. In patients with ARDS, it is desirable to maintain lower tidal volumes and avoid volutrauma. The goal of reducing tidal volume and controlling ventilatory rate is achieved at the expense of considerable CO₂ retention in patients with ARDS. In most cases, the PaCO₂ increases to 60 to 80 mm Hg, and the arterial pH decreases to approximately 7.25. Subsequent metabolic compensation tends to correct the acidosis over several days.⁴⁸ In some cases, the acidosis is more severe but appears to be well tolerated as long as tissue oxygenation is maintained. This ventilatory strategy has been designated permissive hypercapnia or controlled hypoventilation and often requires increased levels of sedation and, in some cases, paralysis to avoid patient discomfort owing to air hunger and a high respiratory rate.⁴⁸

Animal models and human studies have confirmed the safety of controlled hypoventilation.⁴⁹ Several investigations have shown a survival benefit of low-volume ventilation and permissive hypercapnia in patients with ARDS. This strategy is contraindicated in the care of patients with elevated intracranial pressure because this condition may be negatively affected by elevated PaCO₂. Respiratory acidosis secondary to permissive hypercapnia may result in a further decrease in systolic blood pressure in patients with shock, especially patients receiving vasoactive medications; this generally occurs at an arterial pH less than 7.20. In these patients, parenteral replacement of bicarbonate may be considered to maintain an arterial pH of greater than 7.20 while still achieving the goal of low tidal volume ventilation.

Innovative Ventilation Strategies for Acute Respiratory Distress Syndrome

The mainstay of therapy for patients with ALI and ARDS is generally supportive care aimed at avoiding complications associated with mechanical ventilation, volutrauma, and nosocomial pneumonia. No specific ventilator modality has been shown to be consistently superior in improving outcomes in patients with ARDS. However, some ventilator strategies and pharmacotherapies have been shown to improve patients’ oxygenation acutely. Patients with ARDS requiring high levels of inspired O₂ (FiO₂ > 0.6) despite conventional supportive therapy may benefit from one or more of these ventilatory strategies. The choice of which strategy to use depends on the patient’s clinical condition; patient comfort; and the familiarity, comfort, and availability of these modalities to the clinical team.

Volume-Controlled Mechanical Ventilation

Volume-controlled mechanical ventilation represents an exciting new area of ARDS research. A large, well-designed clinical trial sponsored by the National Institutes of Health ARDS Clinical Trials Network showed a significant reduction in mortality (approximately 20%) when lower tidal volumes were used in patients with ALI and ARDS.⁴⁴ Volume-controlled ventilation is now considered a preferred ventilation strategy for patients with ARDS. This ventilation technique is typically adjusted to the specific pressure-volume relationships of each patient; however, a range of optimal initial ventilator settings can be derived from more recent experimental observations. A tidal volume of 10 ml/kg exceeds the volume at which the UIP is reached in more than 80% of patients with ARDS, and most patients need only 5 to 7 ml/kg tidal volume to reach this inflection point.⁵⁰

On the basis of these observations, it is now recommended that the tidal volume for patients with ARDS be initiated at 5 to 7 ml/kg. Subsequent tidal volume adjustments should be made on the basis of each patient’s pressure-volume relationships (see Figure 27-6). Ideal PEEP should be determined as previously described (see Figure 27-5 and Mini Clini, “Determination of ‘Optimal’ PEEP”). Despite the convincing evidence that low-volume ventilation decreases mortality, institutional practices vary considerably, and overall compliance with this recommended approach to the management of patients with ALI is currently quite poor.⁵¹

High-Frequency Ventilation

High-frequency ventilation (HFV) was designed to maintain adequate ventilation and reduce alveolar collapse simultaneously through ventilation with high expiratory lung volumes and rapid (up to 300 breaths/min), small tidal volumes (3 to 5 ml/kg). This technique has been successfully applied to the ventilation of neonates with respiratory distress related to insufficient surfactant production.⁵² However, despite anecdotal evidence suggesting that HFV may be beneficial to adults with ARDS,^53,54 larger clinical trials do not support the routine use of HFV in this patient population.⁵⁵

Experience with the H1N1 influenza epidemic of 2009 has resulted in renewed interest in HFV. In many centers, high-frequency oscillatory ventilation (HFOV) is a rescue therapy applied to patients with inadequate response to conventional ARDS ventilator strategies.⁵⁶ Although smaller trials seem to show equivalence with conventional ventilation,⁵⁷ a more recent meta-analysis suggested that HFOV may provide a survival advantage relative to conventional modes of ventilation.⁵⁸ Additionally, the combination of HFOV with pumpless interventional lung assist devices to remove CO₂ has proven useful in small trials.⁵⁹ Patients on HFOV usually have high mean airway pressures (≥30 cm H₂O), and as with the use of high levels of PEEP, care must be taken that these elevated airway pressures do not overly reduce the cardiac output and the overall O₂ delivery despite elevated arterial oxygenation.

Inverse-Ratio Ventilation

Inverse-ratio ventilation (IRV) is designed to recruit alveolar units through prolongation of the inspiratory phase of the ventilatory cycle and improve oxygenation. In conventional modes of mechanical ventilation, the respiratory cycle is characterized by inspiratory-to-expiratory (I : E) ratios exceeding 1 : 2. During IRV, the inspiratory time on the ventilator is prolonged so that the I : E ratio is reversed and may exceed 4 : 1. Initial reports of the use of this strategy showed significant improvement in oxygenation in patients with ARDS.⁶⁰ However, these studies did not take into account other variables, such as the level of PEEP. After controlling for the level of PEEP, no change in oxygenation was observed in patients with ARDS who received IRV.⁶¹ No study has shown a significant survival benefit with this ventilator mode. Because of the discomfort associated with this mode of ventilation and the risk of asynchronous spontaneous ventilatory efforts, patients often demand heavy sedation or paralysis during IRV. The routine use of IRV in ARDS cannot be advocated at this time.

Mini Clini

Avoiding Ventilator-Induced Lung Injury

Problem

How can ventilator-induced lung injury be minimized in patients with ARDS through adjustments of PEEP and tidal volume?

Discussion

Alveolar shear stress occurs when alveoli collapse during expiration and are reopened as the next tidal volume is delivered. Patients with ARDS are prone to development of this form of lung injury because surface tension increases as a result of impaired surfactant synthesis and function and decreased lung compliance. Alveolar collapse at end-expiration (the LIP of the pressure-volume curve) can be avoided with PEEP, which reduces alveolar shear stress by preventing the cyclic opening and closing of the alveoli during each ventilatory cycle. Because partially patent alveoli require less pressure to inflate than collapsed alveoli, PEEP also improves lung compliance. Improved compliance is reflected by an increase in the slope of the pressure-volume curve (see Figure 27-6).

Many newer ventilators can provide pressure-volume relationships. This information allows the clinician to estimate the PEEP needed to maintain alveolar patency (see Figure 27-6). Alternatively, the clinician can gauge the effects of PEEP on the basis of calculated changes in lung compliance (see Figure 27-5). In the clinical setting, changes in lung compliance are estimated from the compliance of the entire respiratory system, which includes the combined influences of lung compliance, chest wall compliance, and abdominal pressure:

Compliance of the respiratory system=Tidal volume÷(Plateau pressure−Total PEEP)

As PEEP increases, so do airway pressure and the risk of lung injury owing to volutrauma. Volutrauma occurs when alveoli are exposed to pressure greater than that necessary to provide optimal ventilation. Volutrauma is particularly problematic in the care of patients with ARDS because the tidal volume necessary for complete inflation of the lungs is reduced. It becomes critical during ARDS to avoid lung injury related to alveolar overinflation by reducing plateau airway pressures to less than approximately 30 cm H₂O.

Just as alveolar collapse during end-expiration is associated with decreased lung compliance, alveolar overinflation during end-inspiration is reflected by decreasing compliance and corresponds to the UIP of the pressure-volume curve (see Figure 27-6). Tidal volumes should be adjusted so that maximal inspiratory pressure does not exceed the pressure corresponding to the UIP of the pressure-volume curve. In most patients, the UIP is attained at tidal volumes less than 10 ml/kg and plateau pressure less than 35 cm H₂O.⁶¹

Pressure Control Ventilation

Pressure control ventilation (PCV) is designed to prevent ventilator-associated lung injury. The clinician sets the maximal inspiratory airway pressure. A maximal inspiratory pressure (30 to 35 cm H₂O) is chosen that is likely to avert alveolar overdistention and prevent volume-associated lung injury. Minute ventilation is maintained by setting the respiratory rate. Tidal volume becomes a dynamic variable during PCV. That is, tidal volume primarily depends on the driving pressure (maximum inspiratory pressure—PEEP), airway resistance, inspiratory time, and lung compliance. Changes in intrathoracic pressure (e.g., owing to spontaneous inspiratory efforts) or airway resistance or changes in lung compliance influence tidal volume. Large swings in ventilation (increases or decreases in PaCO₂) may be observed during PCV. Close monitoring of the ventilatory status of the patient must be maintained while the PCV mode is in use. Despite its theoretic benefits, PCV has not been shown to be superior to volume-controlled ventilation in clinical trials.62,63

Many modern ventilators now have hybrid modes that combine the benefit of pressure control ventilation, a set maximal inspiratory pressure, with a targeted tidal volume that allows the ventilator to reduce the drive pressure as the lung compliance improves and avoid overdistention. These modes rely on algorithms to adjust the drive pressure that are often proprietary to the ventilator manufacturer. To date, data showing that these hybrid modes improve outcomes in ARDS are insufficient, but they may improve patient comfort.

Airway Pressure Release Ventilation

Airway pressure release ventilation (APRV) is designed to optimize ventilation by recruiting collapsed alveolar units while minimizing ventilator-induced barotrauma in patients with ARDS. APRV generally features two levels of PEEP: a high PEEP often set to approximately 25 to 30 cm H₂O and a low PEEP that is usually set to zero. Similar to IRV, APRV prolongs the inspiratory phase of the ventilatory cycle with inspiratory times often of 4 to 6 seconds at the high PEEP level. During APRV, patients may spontaneously breathe while on this high PEEP, making it more comfortable than traditional IRV. After a period at the high PEEP level, there are brief, approximately 0.5 second, decreases in airway pressure to the low PEEP level, which may be triggered by the patient, allowing volume release and generating a “tidal volume” for this mode. Owing to the prolonged inspiratory times, APRV is associated with an increase in mean airway pressure and a tendency to stack breaths (increased intrinsic PEEP). Increased intrinsic PEEP is the likely mechanism by which APRV improves arterial oxygenation.

Comparisons of APRV with other forms of mechanical ventilation, including pressure support and synchronized intermittent mandatory ventilation, have shown APRV to be effective and well tolerated. In a study by Sydow and associates ⁶⁴ in which APRV was compared with IRV in patients with ARDS, APRV was better tolerated and was associated with lower peak airway pressures. Alveolar recruitment improved over time with APRV but not with IRV. Despite these potential advantages, APRV has not proved to be superior to conventional mechanical ventilation in large clinical trials in patients with ARDS.⁶⁵

Adjunctive Strategies for Acute Respiratory Distress Syndrome

Patient Positioning

In view of the heterogeneous distribution of lung injury in patients with ARDS, it has been proposed that changing the position of the patient could result in improved ventilation/perfusion matching within the lungs. In this regard, it is known that alveolar consolidation tends to be most pronounced in the dependent lung zones in patients with ARDS where blood flow is greatest.⁶⁶ These observations led investigators to experiment with positioning the patient so that aerated lung fields (nondependent lung zones) become dependent (by positioning of the patient in the prone position). Douglas and colleagues⁶⁷ were the first to describe improved oxygenation with prone positioning of patients with ARDS. For unclear reasons, this ventilation strategy was not popularized until more recently, when other investigators showed similar beneficial effects in patients with ARDS.⁶⁸

The mechanisms by which prone positioning improves oxygenation is the subject of ongoing debate. Some proposed mechanisms include improved matching of ventilation with perfusion, increased FRC, increased cardiac output, more effective drainage of upper and lower airway secretions, and improved diaphragmatic excursion. Investigations using animal models of ALI have shown that ventilation in the supine position causes compressive forces on the dorsal airspaces resulting in derecruitment of lung units. This phenomenon is reversed by ventilation in the prone position.69,70

A large multicenter, randomized clinical trial failed to show a survival benefit when ventilation in the prone position was used for at least 6 hours a day in the care of patients with ALI or ARDS, even though oxygenation was significantly improved in the prone position.⁶⁹ A more recent trial that evaluated the implementation of prone ventilation very early in the course of ARDS and for prolonged periods appeared to show a survival benefit.⁷¹ It is also encouraging that a more recent meta-analysis suggested a survival advantage for prone ventilation, which was evident in patients with more severe hypoxemia, defined as a PaO₂/FiO₂ ratio less than 100 mm Hg.⁷²

The advantages of prone ventilation are the improvement in oxygenation and potentially a survival advantage over supine ventilation without the need for expensive pharmacologic interventions or new ventilator modalities. Despite these advantages, there is a significant downside to this intervention. Some patients do not tolerate prone positioning because of hemodynamic instability or worsening gas exchange, and all patients treated in this way require specialized nursing care. Prone positioning of a patient with ARDS requires experienced and committed nursing staff and may require special equipment to facilitate turning, especially obese patients. The improvements in gas exchange related to repositioning the patient tend to be transient necessitating subsequent repositioning. At this point, it is difficult to offer definitive guidance for the use of prone ventilation, but centers experienced with prone ventilation may use it in patients who fail to respond to usual interventions.

Extracorporeal Membrane Oxygenation and Extracorporeal Carbon Dioxide Removal

Extracorporeal membrane oxygenation (ECMO) was first introduced in 1972 as a form of respiratory support for patients with severe, acute hypoxemic respiratory failure. This modality involves the establishment of an arteriovenous circuit for diverting a large proportion of the cardiac output through an artificial gas exchange device, or “artificial lung,” to facilitate the exchange of CO₂ and O₂. Following an initial flurry of interest during the 1970s, a trial comparing ECMO plus conventional mechanical ventilation with conventional mechanical ventilation alone in ARDS showed no survival benefit with ECMO.⁷³

Several more recent studies reported a significant improvement in survival 74,75 and less severe disability at 6 months in patients who were managed with ECMO compared with patients who received conventional therapy.⁷⁵ These promising results, along with others not mentioned here, have led many institutions to use ECMO as rescue therapy for patients with H1N1-induced ARDS who cannot be managed with conventional ventilatory modes or HFOV.⁷⁴ The merits of ECMO for routine management of all patients with ARDS is an area of intense debate and conflicting opinions, but ECMO is currently a reasonable alternative to conventional ventilation strategies in the setting of refractory hypoxemia in patients with ARDS.⁷⁶

Clinical trials comparing ECCO₂R with conventional mechanical ventilation in the care of adults with ARDS have shown improved gas exchange, lower peak airway pressure, lower ventilatory rate, and reduced thoracic volume (less overinflation of the lungs) in patients treated with ECCO₂R ventilation.⁷⁷ This trial failed to show a survival benefit with ECCO₂R at 30 days in 40 patients.

The H1N1 influenza outbreak of 2009 generated many reports supporting the use of ECCO₂R employing technology, such as the Novalung (Heilbronn, Germany), in patients with ARDS failing to respond to conventional therapy.⁷⁸ As experience grows in its use in specialized centers around the world, ECCO₂R has become a therapy of last resort especially for viral-induced ARDS. Most hospitals have put in place contingency plans to move patients to these centers when deemed appropriate by critical care clinicians.

Pharmacologic Therapies for Acute Respiratory Distress Syndrome

Exogenous Surfactant Administration

The assumed role of surfactant deficiency in the pathogenesis of infant respiratory distress is well established. Exogenous surfactant administration is a cornerstone of therapy for infant respiratory distress syndrome. The pathogenesis of ARDS is more complex, however (see earlier section on Pathophysiology). Although surfactant is known to be qualitatively and quantitatively altered during ARDS, other factors, such as the severity of pulmonary microvascular injury, contribute to the gas exchange abnormalities associated with this condition. Nonetheless, surfactant has been shown to have immunomodulating (i.e., affecting immune function) properties that may reduce microvascular injury in the lungs. Surfactant dysfunction may contribute to the development of ARDS by promoting instability of the alveolar units (airway shear trauma, atelectasis, and right-to-left shunt) and by allowing inflammatory injury to alveoli to continue unchecked. This rationale explains the observed correlation between the degree of surfactant dysfunction and the severity of gas exchange abnormalities during adult ARDS and forms the basis for clinical trials of exogenous surfactant administration during ARDS.

Exogenous surfactant replacement is of greatest benefit in models of pure surfactant deficiency, such as infant respiratory distress or the aftermath of saline lavage. However, in adult ARDS, in which deactivation of surfactant relates to the influx of inflammatory cells and mediators into the alveolar space, the effects of exogenous surfactant are less apparent. Although initial studies showed a benefit in oxygenation with surfactant administration, either synthetic (Alveofact [Lyomark Pharma, Oberhaching, Germany])⁷⁹ or a modified bovine lung surfactant (beractant [Survanta, Abbott Nutrition, Chicago, IL]),⁸⁰ subsequent trials failed to show a survival benefit, and one trial showed a trend toward increasing mortality.^81,82 Surfactant replacement therapy cannot be recommended for the routine management of patients with ARDS.

Neuromuscular Blockade

Neuromuscular blocking agents have been used for decades to enhance compliance with mechanical ventilation in patients with ARDS, especially when the ventilatory mode has involved low-volume ventilation, or inverse ratio ventilation. Although earlier studies showed improved oxygenation with neuromuscular blockade for the first 48 hours, they did not address any survival advantage gained from this intervention. A multicenter randomized trial considered 90-day in-hospital mortality as the outcome measure after therapeutic intervention with a bolus of the paralytic drug cisatracurium followed by a 48-hour infusion. The cisatracurium group showed an improvement in the adjusted 90-day survival, days off the ventilator, and incidence of barotrauma complications. In post hoc analysis of the study, the benefit seemed to be most evident in more severe disease as defined by a PaO₂/FiO₂ ratio less than 120. Although problems with muscle weakness related to neuromuscular blocker use were thought to be more likely in the intervention arm, this was not seen in practice in the study and may reflect the short duration of paralysis.⁸³ At this time, it is difficult to make firm conclusions about the routine use of neuromuscular blockade, and further trials are needed to clarify this area.

Inhaled Nitric Oxide

NO is a potent vasodilator that plays a critical role in the regulation of blood flow within the lungs. NO is soluble in liquids and gases and diffuses readily through various tissues. In vitro, the vasodilatory effects of NO are short-lived because NO is quickly bound and neutralized by hemoglobin and competes with O₂ for mitochondrial oxidization. Consequently, the effects of NO are localized and temporally self-limited.

The potential role of inhaled NO in ARDS is based on the idea that NO is preferentially distributed to well-ventilated portions of the lung, where it causes local vasodilation. In this way, well-ventilated areas of the lung receive a greater portion of the total pulmonary blood flow, which results in improved oxygenation by reducing ventilation/perfusion mismatch (Figure 27-7). In practice, the effects of inhaled NO in patients with ARDS have been variable. Generally, patients with high pulmonary vascular resistance, who presumably have more severe alterations in pulmonary vascular regulation, seem to benefit the most. The beneficial effects of NO can be achieved at low concentrations of inhaled NO, which explains the low incidence of systemic hypotension associated with its use. However, sudden discontinuation of inhaled NO may be associated with severe pulmonary vasoconstriction; slow weaning is necessary.

FIGURE 27-7 The damage of ALI and ARDS is heterogeneous. A, Despite autoregulatory changes that favor vasodilation of well-oxygenated alveoli, perfusion goes to both the healthy, unaffected alveoli and the diseased alveoli that are unable to oxygenate blood. This results in ventilation/perfusion mismatch or a shuntlike state resulting in deoxygenated blood returning from the lung and hypoxemia. B, The addition of inhaled NO to the inspired gas results in vasodilation of the capillaries adjacent to healthy alveoli promoting increased blood flow to the “good” alveoli and diminishing the percentage of blood going to diseased alveoli, resulting in improved ventilation/perfusion match and improved oxygenation.

Despite encouraging results of animal studies showing tolerance of low doses of inhaled NO for 6 months, the safety and effectiveness of inhaled NO therapy for ARDS have not been established. The breakdown products of NO are potentially toxic and include highly reactive free radicals (e.g., peroxynitrite) and methemoglobin, potentially contributing to further lung injury in patients with ARDS or in individuals with normal lung function. The use of inhaled NO during mechanical ventilation involves close monitoring and special exhaust systems to prevent exposure of health care personnel to NO and its potentially toxic by-products. Many phase II and phase III studies have been completed with NO. No study has shown that NO improves outcome or mortality. Most recently, a large multicenter randomized trial involving more than 300 patients reported that NO at a dose of 5 ppm was ineffective in altering mortality, or duration of ventilator support, although there was a short-term benefit on oxygenation.⁸⁴ Because of these uncertainties, inhaled NO, although promising, remains an experimental therapy for ARDS.

Inhaled Epoprostenol

Corticosteroids for Late, Uncomplicated Acute Respiratory Distress Syndrome

Although most patients who survive ARDS have minimal residual pulmonary impairment, a small but significant number of patients cannot be weaned from mechanical ventilatory support because of abundant fibroproliferation during recovery from ARDS. The high mortality among these patients seems to relate to the extent and severity of pulmonary fibrosis.⁸⁷ High-dose corticosteroids have been used to manage uncomplicated pulmonary fibrosis following ARDS. In a well-publicized but uncontrolled study by Meduri and coworkers,⁸⁸ patients treated with corticosteroids for ARDS-related pulmonary fibrosis had improved gas exchange and low mortality (24%). Subsequently, the same investigators performed a randomized double-blind, placebo-controlled trial to determine the effect of prolonged methylprednisolone therapy for unresolving ARDS. The results were encouraging, showing improvement in lung injury scores and reduced mortality.⁸⁹ The ARDS Clinical Trials Network conducted a large multicenter trial of corticosteroids for the treatment of patients with late, fibroproliferative ARDS (>7 days’ duration). The results showed no survival benefit with steroid use at 60 days and 180 days, and patients given steroids after 14 days from ARDS onset experienced a higher mortality. At this time, the routine use of corticosteroids for the treatment of established ARDS cannot be advocated and should be strictly avoided after 14 days from onset.⁹⁰