The Merck Manual of Diagnosis and Therapy
Section 6. Pulmonary Disorders
Chapter 67. Adult Respiratory Distress Syndrome
Topics
[General]

[General]

Adult respiratory distress syndrome: Respiratory failure caused by various acute pulmonary injuries and characterized by noncardiogenic pulmonary edema, respiratory distress, and hypoxemia.

Etiology

Adult respiratory distress syndrome (ARDS), a common medical emergency, is precipitated by various acute processes that directly or indirectly injure the lung, eg, sepsis, primary bacterial or viral pneumonias, aspiration of gastric contents, direct chest trauma, prolonged or profound shock, burns, fat embolism, near drowning, massive blood transfusion, cardiopulmonary bypass, O₂ toxicity, acute hemorrhagic pancreatitis, inhalation of smoke or other toxic gas, and ingestion of certain drugs. The incidence of ARDS is estimated to be > 30% with sepsis (see Ch. 156). Although termed "adult," this syndrome also occurs in children.

Pathophysiology

The initial lung injury is poorly understood. Animal studies suggest that activated WBCs and platelets accumulate in capillaries, the interstitium, and airspaces; they may release prostaglandins, toxic O₂ radicals, proteolytic enzymes, and other mediators (such as tumor necrosis factor and interleukins), which injure cells, promote inflammation and fibrosis, and alter bronchomotor tone and vasoreactivity.

When the pulmonary capillary and alveolar epithelia are injured, plasma and blood leak into the interstitial and intra-alveolar spaces. Alveolar flooding and atelectasis result; atelectasis is due in part to reduced surfactant activity. The injury is not homogeneous and affects mainly the dependent lung zones. Within 2 to 3 days, interstitial and bronchoalveolar inflammation develops, and epithelial and interstitial cells proliferate. Then, collagen may accumulate rapidly, resulting in severe interstitial fibrosis within 2 to 3 wk. These pathologic changes lead to low lung compliance, decreased functional residual capacity, ventilation/perfusion imbalances, increased physiologic dead space, severe hypoxemia, and pulmonary hypertension.

Symptoms, Signs, and Diagnosis

ARDS usually develops within 24 to 48 h after the initial injury or illness. Dyspnea occurs first, usually accompanied by rapid, shallow respiration. Intercostal and suprasternal retraction may be present on inspiration. The skin may appear cyanotic or mottled and may not improve with O₂ administration. Auscultation may detect crackles, rhonchi, or wheezes, or findings may be normal.

Early diagnosis requires a high index of suspicion aroused by the onset of dyspnea in settings that predispose to ARDS. A presumptive diagnosis can be made with arterial blood gas analysis and chest x-rays. This analysis initially shows acute respiratory alkalosis: a very low Pa_O2, a normal or low Pa_CO2, and an elevated pH. Chest x-rays usually show diffuse bilateral alveolar infiltrates similar to acute pulmonary edema of cardiac origin, but the cardiac silhouette is usually normal. However, changes seen on x-ray often lag many hours behind functional changes, so hypoxemia may seem disproportionately severe compared with the edema observed on chest x-ray. The extremely low Pa_O2 often persists despite high concentrations of inspired O₂ (FI_O2), indicating pulmonary right-to-left shunting through atelectatic and consolidated lung units that are not ventilated.

After immediate treatment of hypoxemia, further diagnostic steps are indicated. When there is doubt about whether a patient has heart failure, a Swan-Ganz catheter may help. Characteristically, pulmonary arterial wedge pressure (PAWP) is low (< 18 mm Hg) in ARDS and high (> 20 mm Hg) in heart failure. If pulmonary embolism, which may mimic ARDS, is considered likely (see Ch. 72), appropriate diagnostic procedures (eg, pulmonary angiography) should be performed after the patient is stabilized. Pneumocystis carinii pneumonia and occasionally other primary lung infections may mimic ARDS and should be considered, especially in immunocompromised hosts; lung biopsy or bronchoscopy-guided bronchoalveolar lavage may be helpful.

The American-European Consensus Conference defines ARDS as Pa_O2/FI_O2< 200 (regardless of positive end-expiratory pressure), bilateral infiltration on frontal chest x-rays, and PAWP <= 18 mm Hg when measured or no clinical evidence of left atrial hypertension.

Complications and Prognosis

Secondary bacterial superinfection of the lungs, particularly by aerobic gram-negative bacteria (such as Klebsiella, Pseudomonas, and Proteus spp) and by gram-positive Staphylococcus aureus, especially methicillin-resistant strains; multiple organ system failure, especially renal failure (see Table 67-1); and complications of invasive life support may occur and are associated with high morbidity and mortality. Tension pneumothorax associated with the placement of central venous catheters and with the use of positive pressure ventilation (PPV) and positive end-expiratory pressure (PEEP) may occur suddenly. Prompt recognition and treatment are necessary to prevent death. Tachycardia, hypotension, and a sudden increase in the peak inspiratory pressures required for mechanical ventilation suggest possible pneumothorax. Pneumothorax occurring late in ARDS is an ominous sign because it is usually associated with severe lung damage and a need for high ventilatory pressures. Without adequate replacement of intravascular volume, PPV and PEEP may decrease venous return, resulting in depression of cardiac output and of overall O₂ transport to the tissues, which contributes to secondary multiple organ system failure.

The survival rate for patients with severe ARDS who receive appropriate treatment is about 60%; if the severe hypoxemia of ARDS is not recognized and treated, cardiopulmonary arrest occurs in 90% of patients. Those who respond promptly to treatment usually have little or no residual pulmonary dysfunction or disability. Patients needing prolonged ventilatory support with FI_O2 > 50% are more likely to develop lung fibrosis. In most patients who survive acute illness, lung fibrosis resolves after several months, but the mechanism of resolution is unknown.

Treatment

Management principles are similar despite different etiologies. Oxygenation must be maintained, and the underlying cause of acute lung injury corrected. Meticulous attention is necessary to prevent nutritional depletion, O₂ toxicity, superinfection, barotrauma, and renal failure, which may be worsened by intravascular volume depletion. While the diagnosis is being considered, life-threatening hypoxemia must be treated with a high FI_O2 and monitored with repeated arterial blood gases or noninvasive oximetry. Prompt endotracheal intubation with mechanical ventilation and PEEP may be needed to deliver O₂ because hypoxemia is frequently refractory to O₂ inhalation by face mask.

Intravascular volume is often depleted with the onset of ARDS, because sepsis is the underlying cause, because diuretic therapy was given before ARDS was suspected, or because initiation of PPV decreases venous return. Despite the presence of alveolar edema, IV fluids should be given if needed to restore peripheral perfusion, urine output, and BP. Monitoring vascular volume is crucial because both hypovolemia and overhydration are deleterious. Physical findings and central venous pressure may be misleading in critically ill patients undergoing mechanical ventilation, and if severe hypoxemia persists, if skin perfusion is poor, if mentation is impaired, or if urinary output decreases (< 0.5 mL/kg/h), a reliable index of intravascular volume is needed immediately. A Swan-Ganz catheter generally is used to guide volume infusions, particularly if PEEP is needed. However, the use of Swan-Ganz catheters is not without risk. Close daily monitoring of the patient's weight and total intake and output of fluids is also essential for fluid management. As a rule, patients with ARDS do better if kept on the "dry" side--by restricting fluids and judiciously using diuretics--as long as cardiac output and tissue perfusion are not impaired.

If sepsis is or could be the cause of ARDS, empirical antibiotic therapy should be initiated pending culture results. Surveillance cultures and Gram stains of sputum or tracheal aspirates can help detect lung superinfection early and guide antibiotic therapy. Closed-space infections should be drained. Alimentation should be begun within 48 to 72 h; the enteral route is preferred because it protects the gut mucosal lining.

Corticosteroids are of no proven benefit in acute ARDS, although anecdotal reports suggest benefit in some patients with ARDS in the late fibroproliferative phase, which may develop after 7 to 10 days of mechanical ventilation. Concurrent lung infection must be excluded in these patients, who often are febrile and have leukocytosis, with or without infection.

Many approaches to the prevention and management of ARDS have been unsuccessful or inconclusive. Treatments that have not improved outcome or prevented ARDS include monoclonal antibody to endotoxin, monoclonal antibody to tumor necrosis factor, interleukin-1 receptor antagonist, prophylactic (early) PEEP, extracorporeal membrane oxygenation and extracorporeal CO₂ removal, IV albumin, volume expansion and cardiotonic drugs to increase systemic O₂ delivery, corticosteroids in early ARDS, parenteral ibuprofen to inhibit cyclooxygenase, prostaglandin E₁, and pentoxifylline. Several approaches show promise but need further study.

The prone position can substantially improve oxygenation in some patients, probably because this position shifts perfusion and gas exchange to more normal, previously nondependent lung zones. Whether this technique improves gas exchange in severe ARDS and whether it can reduce the duration of mechanical ventilation and improve overall survival is unclear. Patient positioning is difficult to perform.

Inhaled nitric oxide can significantly improve pulmonary hypertension and arterial oxygenation in patients who have severe ARDS without causing systemic hypotension. Whether nitric oxide improves survival and whether prolonged use promotes further lung injury from toxic nitric oxide by-products, such as the peroxynitrite anion, remains to be shown.

Ketoconazole may help prevent ARDS by suppressing the formation and release of tumor necrosis factor from macrophages. Its clinical benefit in small preliminary studies needs to be confirmed in larger well-controlled studies. Initial studies of aerosolized synthetic surfactant in adult ARDS patients have been disappointing. Improved aerosol delivery devices and natural mammalian surfactant preparations may improve alveolar stability, reduce atelectasis and intrapulmonary blood shunting, and enhance the antibacterial and anti-inflammatory properties of the alveolar lining fluid; additional studies using these approaches are in progress.

Mechanical ventilation: Most patients require endotracheal intubation and assisted ventilation with a volume-limited mechanical ventilator. Endotracheal intubation and PPV should be considered if the respiratory rate is > 30 breaths/min or if an FI_O2 > 60% by face mask is required to maintain arterial PO₂ at 70 mm Hg for more than a few hours. As an alternative to intubation, a continuous positive airway pressure mask may effectively deliver PEEP in patients with mild or moderate ARDS. Such masks are not recommended for patients with depressed consciousness because of the risk of aspiration and should be replaced by a ventilator if patients progress to severe ARDS or show signs of respiratory muscle fatigue with increasing respiratory rate and arterial PCO₂.

Conventional settings on a volume-limited ventilator in ARDS are a tidal volume of 10 to 15 mL/kg, a PEEP of 5 to 10 cm H₂O, an FI_O2 of <= 60%, and a patient-triggered assist-control mode. Intermittent mandatory ventilation with an initial rate of 10 to 12 breaths/min with PEEP may be used instead.

There is concern that high ventilator pressures and volumes in ARDS can worsen lung injury, but this effect has not been proved. A PEEP that is too low can also damage the lung by allowing unstable terminal lung units to open and close repeatedly. This problem may be circumvented with small tidal volumes (6 to 8 mL/kg) and a higher PEEP (between 10 and 18 cm H₂O).

The goal of small tidal volumes is to prevent ventilator-generated breaths from exceeding the upper inflection (deflection) point of the patient's pressure-volume curve and from causing lung overdistention (see Fig. 67-1). Beyond this point, the lung becomes quite stiff, and small increments in tidal volume cause large increases in ventilator plateau pressure (the pressure needed to maintain lung and chest wall inflation after inspiratory flow has ended). For technical reasons, the upper inflection point is often not measured directly. Instead, ventilator plateau pressure is measured, which in most patients should not exceed 25 to 30 cm H₂O (or 20 to 25 cm H₂O according to some investigators). With a reduced tidal volume, the respiratory rate of the ventilator can be increased to maintain an adequate arterial pH and PCO₂. Some patients still develop hypercapnia and respiratory acidosis, which is usually well tolerated. If arterial pH falls below 7.20, a slow bicarbonate infusion can be started.

Theoretically, the PEEP selected should be several centimeters of water above the lower inflection point on the patient's pressure-volume curve (see Fig. 67-1) to promote generous alveolar recruitment and inflation. If the lower inflection point is not directly determined, a PEEP of 10 to 15 cm H₂O often suffices. At a satisfactory PEEP setting, the ventilator FI_O2 can usually be decreased to a safe level of < 50 to 60%, so that the patient has a satisfactory Pa_O2 of >= 60 mm Hg or arterial O₂ saturation (Sa_O2) >= 90%. For adequate tissue O₂ transport, the cardiac index should be >= 3 L/min/m²; occasionally, volume infusion or parenteral cardiotonic drugs are needed.

Alternatively, pressure-controlled mechanical ventilation may be used, mainly for patients with severe ARDS. Inspiratory pressure and duration are selected, and tidal volume varies with inspiratory impedance; high inspiratory ventilator pressures are thus avoided, but permissive hypercapnia often results. This approach is often combined with inverse ratio ventilation, in which the duration of inspiration is set to equal or exceed that of exhalation. This technique may recruit and reexpand more lung units than PEEP can alone (partly by producing intrinsic or auto-PEEP), so that injuriously high FI_O2 can be reduced further. This technique is uncomfortable and requires patient sedation, often together with a muscle-paralyzing drug.

Weaning readiness is based on continued evidence of improved lung function (ie, a decreasing need for O₂ and PEEP), improvement seen on x-rays, and resolution of tachypnea. Patients without previous underlying pulmonary disease can usually be weaned smoothly; difficulty in weaning may indicate an untreated or a new site of infection, overhydration, bronchospasm, anemia, electrolyte disturbance, cardiac dysfunction, or poor nutritional status causing respiratory muscle weakness. If these conditions are treated, weaning can usually be accomplished by the use of intermittent mandatory ventilation to decrease the mechanical rate, often with some pressure support ventilation (see Ch. 66), or by trials of spontaneous breathing for progressively longer periods through a T-piece attached to the endotracheal tube. A low PEEP ( 5 cm H₂O) is usually maintained throughout the weaning process.