CLINICAL REVIEW

Respiratory Aspiration of Stomach Contents JAMES W. WYNNE, M.D.; and JEROME H. MODELL, M.D.; Gainesville, Florida

The aspiration of stomach contents is a common clinical problem of concern to all physicians. Its consequences are varied, depending on the amount and distribution of the aspirate, its pH, and the presence or absence of food, particulate matter, and bacteria. Because multiple factors are involved, aspiration of stomach contents can lead to several distinct syndromes of pulmonary injury, all of which unfortunately have been labeled "aspiration pneumonitis." We review the pathophysiology of each of these syndromes and discuss important diagnostic and therapeutic consequences.

As pH decreases below 1.5, however, little additional damage to the lung occurs (2). The volume of an acid aspirate (8) and its distribution also are important: large aspirates that are poorly localized cause the highest mortality rate (4, 8-10). Of limited importance is the presence of digestive enzymes or bile in the stomach at the time of aspiration (2). Finally, aspirates grossly contaminated with bacteria, such as those occurring in patients with bowel obstruction, are uniformly fatal (11, 12). PATHOLOGY

Experimentally, as the pH of the aspirate decreases below the critical level of 2.5 (2, 4, 7), lung injury increases.

Most of our knowledge of the pathologic changes of acid aspiration is based on experiments done on animals. After aspiration, acid is rapidly distributed throughout the lungs, and damage occurs immediately. Acid gastric juice stained with methylene blue and aspirated into an isolated dog lung can be seen on the surface of the lung within 12 to 18 seconds. Isolated areas of atelectasis become visible when the dye appears and become extensive within 3 minutes (11). Pathologic examination within the first few hours of acid aspiration reveals epithelial degeneration of the bronchi, pulmonary edema, and hemorrhage. With electron microscopy, necrosis of type I alveolar cells and the presence of free lamellated inclusion bodies in the pulmonary transudate are noted. Within 4 h, there is an acute infiltration of polymorphonuclear cells, and fibrin can be seen in the alveolar space. Degeneration of alveolar type II cells and further necrosis of type I cells with detachment from the basement membrane can also be noted. During the next 24 to 36 h, marked polymorphonuclear infiltration results in alveolar consolidation, and damage to the airways may result in mucosal sloughing. After 48 h, hyaline membranes can be seen (8). Gross examination shows lungs that are boggy, edematous, and hemorrhagic. At 72 h, resolution has already begun. There is regeneration of bronchial epithelium, proliferation of fibroblasts, and a decrease in acute inflammation (13). Lungs obtained from experimental animals 2 to 3 weeks after acid aspiration usually are normal or slightly increased in weight and show parenchymal scarring with pleural retraction. These scars vary in size and contain macrophages, lymphocytes, and hemosiderin granules. Often they are associated with bronchiolitis obliterans (13) and atypical bronchial regeneration (14).

• From the Division of Pulmonary Medicine, Department of Medicine, and the Department of Anesthesiology, University of Florida College of Medicine; Gainesville, Florida.

The severe chemical burn induced by acid aspiration

ASPIRATIONS often are classified according to their pH. Mendelson (1) was the first to stress the importance of acidity in determining the extent of pulmonary injury when he described the clinical course of 66 women who aspirated stomach contents during labor and delivery. He also showed that unneutralized liquid gastric contents introduced into the lungs of rabbits caused severe pulmonary injury indistinguishable from that caused by an equal amount of 0.1 N hydrochloric acid. If the pH of the vomitus were neutralized before aspiration, lung injury was minimal. He concluded that acid was the main determinant of the pulmonary injury seen after aspiration. As a result of this work, the clinical syndrome of severe aspiration injury has been termed "Mendelson's syndrome" and is generally thought to be synonymous with acid aspiration. Subsequent researchers corroborated Mendelson's experimental findings (2-4) and described a critical pH below which severe lung damage occurs. This varies from species to species, for example, 1.7 for rats (2) and 2.1 to 2.4 for rabbits (2, 3). A critical pH of 2.5 has been suggested for humans (5, 6) but has not been proved. Nevertheless, clinically and for purposes of discussion, aspirations in humans are termed "acid" if the pH is less than 2.5 and "nonacid" or "neutral" if the pH is greater than 2.5. Acid Aspiration

466

PHYSIOLOGY

Annals of Internal Medicine 8 7 : 4 6 6 - 4 7 4 . 1 9 7 7

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19526/ by a University of California San Diego User on 03/19/2017

causes loss of alveolar capillary integrity and exudation of fluid and protein into the alveoli and bronchi (8). Serum protein electrophoresis done on material aspirated from a small bronchus after acid aspiration reveals a pattern identical to that of serum (4). This exudation causes increased lung weight, decreased pulmonary compliance, and pulmonary edema (8, 15, 16). The accompanying loss of intravascular volume may cause severe hypotension (7, 8, 15). Hypoxia can occur within minutes of acid aspiration (4, 7, 8, 11, 15) and has multiple causes. First, reflex airway closure occurs in response to the aspiration of fluid (16-18). Second, surfactant activity decreases when surfactant is destroyed, diluted, or altered by acid; this reduction leads to alveolar instability and atelectasis (4, 18, 19). Third, the outpouring of fluid and protein into damaged tissues causes interstitial and alveolar edema resulting in further airway closure. Finally, alveolar hemorrhage and consolidation occur, followed by hyaline membrane formation. All these conditions contribute to the large alveolar-arterial oxygen differences and the significant increase in venous admixture. Acid aspiration also causes changes in the pulmonary vasculature. Initially, pulmonary artery pressure may rise rapidly (4, 11); however, it falls quickly in association with decreased cardiac output resulting from loss of intravascular volume (4, 8). As a consequence, pulmonary artery pressure is usually low or normal (15, 20). On the other hand, pulmonary vascular resistance is elevated (8, 21, 22). This may be due to hypoxic vasoconstriction or anatomic obstruction. Marked constriction of pulmonary arterioles has been seen arteriographically (23) and histologically (11) after aspiration, and in-situ thrombus formation also has been reported (23, 24). CLINICAL CORRELATION

Clinical correlation of experimental studies on acid aspiration has been limited, chiefly because of the difficulty in measuring the pH of the aspirate. Using gastric contents obtained immediately after aspiration to approximate the pH of the aspirate, Lewis, Burgess, and Hampson (7) studied 18 patients with documented aspiration. Reduction of plasma volume, arterial blood pressure, and dynamic compliance were noted. Blood gas tensions showed arterial hypoxemia with increased venous admixture and large alveolar-arterial oxygen differences. Although the volume of the aspirate was not known, there was a direct relation between mortality and gastric acidity. Particularly alarming was the 100% mortality rate among patients with a gastric pH of less than 1.8. This rate prevailed despite vigorous supportive measures such as intubation, positive-pressure ventilation, and administration of steroids, antibiotics, and intravenous fluids. When the gastric pH was between 1.8 and 2.5, however, the mortality rate was only 25%. These rates correlate with results from several experimental studies (8, 11, 13, 20, 21, 24) and with other clinical studies in which the pH was unknown but suspected of being low (10, 25, 26). Most patients who survive aspiration recover completely (27). However, roentgenographic evidence of fi-

brosis and continued physiologic derangements has been reported after long-term follow-up (28). Carefully analyzed, much of the experimental and clinical information available suggests that the term "Mendelson's syndrome," used to describe severe acid aspiration, may be a misnomer. Mendelson's original work described clinical observations of 66 patients. The character of the aspirated material was known in 45. Five aspirates contained large food particles. These patients presented with symptoms of large airway obstruction. The remaining 40 aspirates were liquid. After aspiration these patients developed an asthmalike syndrome that included tachycardia, cyanosis, dyspnea, wheezing, rales, rhonchi, and, on occasion, gross pulmonary edema. On the basis of his experimental data, Mendelson assumed this clinical response was caused by acid aspiration. However, the mortality rate in his patients was zero, in marked contrast to experimental studies of acid aspiration and other clinical studies in which mortality rates were as high as 35% to 60% (7, 10, 25, 26). Furthermore, in about 75% of his patients the course was uncomplicated. Finally, although none of his patients received ventilatory support, all recovered in 36 h. The clinical syndrome described in Mendelson's patients resembled only slightly the findings of severe life-threatening pulmonary injury that Mendelson demonstrated experimentally and that we now associate with acid aspiration. Most of Mendelson's patients may have aspirated only small amounts of acid or may have experienced some form of "nonacid" aspiration. Nonacid Aspiration

Nonacid or neutral aspirates (pH > 2.5) can cause either transient or sustained damage to the lung. The nature and extent of this damage depend not only on the volume of the aspirate but also on its composition, especially its tonicity and the presence of large particles or irritating food material. Classification of neutral aspirates is not clear-cut. We have arbitrarily divided them into neutral clear liquid aspirates (saline, water), neutral large particulate aspirates (those containing particles large enough to obstruct airways down to the segmental level), and foodstuff aspirates (those containing small food particles, dairy products, or hypertonic solutions). CLEAR LIQUIDS

Pathology: Many of the early pathologic changes seen after acid aspiration can also be seen when clear liquids such as saline or water are aspirated. These changes include pulmonary edema, diapedesis of erythrocytes, separation of endothelial cells from basement membranes, and peribronchial neutrophilic infiltration (16, 29, 30). Compared with acid aspiration, aspiration of nonacid solutions causes little necrosis of alveolar cells and minimal neutrophilic infiltration (30). Physiology: Many of the early physiologic changes of aspiration, whether acid or neutral, are also nonspecific and reflect only the response of the lung to fluid. Mendelson himself (1) remarked on the brief period of "labored respiration and cyanosis" seen in experimental animals Wynne and Model I • Aspiration of Stomach Contents

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19526/ by a University of California San Diego User on 03/19/2017

467

after saline aspiration. Teabeaut (2) was the first to point out that the pulmonary edema and congestion occurring after aspiration could be produced by any type of injected liquid. Other investigators also have made this observation when comparing acid and neutral clear liquid aspiration (15, 16,23,29, 30). Studies on drowning indicate that the instillation of as little as 1 ml/kg of water into the lungs of experimental animals causes marked changes in arterial Po 2 and oxygen saturation and decreases pulmonary compliance (16, 17, 19, 31). Reflex airway closure causes some of these variations, since these changes are improved temporarily (19) by vagotomy and administration of atropine or isoproterenol (17, 18). Other mechanisms, such as alveolar collapse and atelectasis resulting from surfactant changes or filling of alveoli with fluid, have also been suggested (19, 32). The main physiologic distinction between the aspiration of acid and neutral clear liquid is that the acute respiratory decompensation caused by the latter is frequently shortlived and more easily reversible (2). On the other hand, not all neutral aspirates are clear liquids. During the course of a day, material of varying amounts, composition, and particle size may be found in a patient's stomach. Aspiration of certain types of even nonacid materials may have dramatic effects on the lung. L A R G E PARTICLES

The aspiration of large particulate matter causes obstruction of major airways. In Mendelson's original series, five patients suffered from this type of aspiration: three had complete airway obstruction, and two of these died from suffocation. Large particle aspiration, though often easily recognized, can be rapidly fatal. FOODSTUFF

More intriguing, and certainly more perplexing, is the lung injury caused by aspiration of neutral stomach contents containing small, nonobstructing particles of foodstuff. Teabeaut (2) administered such an aspirate to experimental animals and demonstrated a prolonged inflammatory response clinically similar to that caused by acid. The severity of the reaction varied; particle size played a role, but equally important was the chemical composition of the particles and their ability to be phagocytized, disintegrated, or removed from the lung. Other authors have reported similar observations (14, 33). Hypertonic fluids can cause the same reaction, as can milk and formula, an especially bothersome problem in the pediatric age group (34, 35). Pathology: In contrast to aspiration of neutral clear liquids that resolve rapidly, a continued pathologic response occurs after aspiration of partially digested meat, vegetable, or dairy products in which small food particles may be present (2, 33). Within 6 h, there is extensive hemorrhagic pneumonia, with erythrocytes, granulocytes, and macrophages in the alveoli and bronchi. Unlike the neutrophilic response seen in acid aspiration, a widespread granulomatous reaction with numerous macrophages and giant cells is present within 48 h. Alveolar walls are thickened by infiltrating mononuclear cells. 468

Within another 24 h most of the reaction is mononuclear, and numerous granulomas are present. Obstruction of airways by food particles is not the cause of this reaction, although obstructive bronchiolitis caused by inflammatory exudate often is noted. Within 5 days, the macrophage exudate is decreased, although focal areas resembling hard tubercles are present in large numbers (2, 33). Food particles can be identified at the center of these granulomas, and the exact nature of the aspirate may be ascertained by their characteristic microscopic appearance (36, 37). Roentgenograms may show granuloma formation similar to that of miliary tuberculosis when repeated small aspirations occur over a long period of time (38). Physiology: The physiologic changes that occur in animals receiving a neutral foodstuff aspirate differ in several respects from those of animals receiving an acid aspirate. First, the shift of fluid from the intravascular space into the lungs occurs much later, about 3 to 4 h after aspiration, and is not as great as occurs when acid is aspirated. Second, heavy bronchial transudation usually is not seen. Finally, although arterial Po 2 levels are as low as those seen after acid aspiration, arterial Pco 2 levels are much higher after food aspiration, probably indicating a greater degree of hypoventilation. In general, most changes are dose-related, and when large volumes of neutral food aspirates are given, mortality rates are similar to those seen in acid aspiration (4). The frequency of this type of nonacid aspiration is unknown because few clinical studies have addressed themselves to the issue. However, many aspirations undoubtedly occur during or shortly after a meal, when food present in the stomach is only partially digested and when the gastric pH may be above 2.5. Studies done on obstetric patients receiving emergency anesthesia have shown that most of these patients have a gastric pH greater than 2.5 (39-41), but these studies have not discussed the presence or absence of foodstuff. In summary, the aspiration of significant amounts of fluid into the lung causes acute respiratory decompensation. If the aspirate has a pH greater than 2.5, is free of particulate matter, and has normal tonicity, recovery may be rapid. If the fluid has a pH less than 2.5, is hypertonic, or contains food particles or irritating but nonparticulate food substances, a continued inflammatory reaction will result. The reaction will be primarily hemorrhagic, granulocytic, and necrotizing when the aspirate is acid, and mononuclear and granulomatous when it is foodstuff. The degree of impairment will depend on the nature and volume of the aspirate and its distribution. Acid aspirates cause the greatest degree of damage, both anatomic and physiologic; however, determining whether the aspirate was acid or foodstuff may be difficult clinically. Finally, there is reason to believe that the clinical syndrome described by Mendelson (1) was not caused by acid aspiration, although the terms "acid aspiration" and "Mendelson's syndrome" have now become synonymous. Infection

The exact role of infection in the pathogenesis of lung

October 1977 • Annals of Internal Medicine • Volume 87 • Number 4

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19526/ by a University of California San Diego User on 03/19/2017

injury caused by aspiration is unclear, although the frequency of infection complicating aspiration is quite high. Of 18 patients with documented aspirations studied by Lewis, Burgess, and Hampson (7), 15 developed infection. In 88 cases of aspiration reported by Arms, Dines, and Tinstman (26), the infection rate was nearly 50%. Two distinct patterns of infection are seen in patients who aspirate. A small, heavily infected inoculum is probably the initial focus of infection in patients who develop nonspecific lung abscess, empyema, or necrotizing bacterial pneumonia after aspiration (42). Cultures taken from these lesions will reflect closely the oropharyngeal flora. In nonhospitalized patients, anaerobic organisms predominate. Bartlett, Gorbach, and Finegold (43), for instance, found anaerobes present in 93% of isolates obtained directly from infected lesions in patients who had aspirated. In 46% of these isolates, anaerobes were the only organisms found. In hospitalized patients, facultative anaerobes and aerobic organisms are more common (44), since they frequently colonize the oropharynx (45). The aspirate in these patients is usually small, and aspiration may go unrecognized until signs of infection develop, the onset of symptoms taking perhaps days or weeks to occur. On the other hand, Bartlett, Gorbach, and Finegold (43) found that in about one fourth of their patients, signs of infection developed and cultures were positive within 24 h. The major determinants governing the occurrence and progression of this type of infection are the patient's dental hygiene, general state of health, integrity of cough reflex and other pulmonary defense mechanisms, and, possibly, the degree of damage induced by the aspirate. The second pattern of infection is that which occurs after large aspiration, usually of the acid type. The initial lung damage is caused by the aspirate; whether infection plays a role in the early clinical findings is not clear. Most physicians believe it does not (26, 46-49). This conclusion often is substantiated by reference to experiments in which specimens obtained from animals after acid aspiration have been sterile (2, 11, 13, 50). Experimental aspirates, however, always are introduced directly into the tracheobronchial tree by catheter or endotracheal tube, bypassing the oropharynx and thus failing to duplicate normal mechanisms of aspiration, thereby eliminating a well-recognized source of bacterial contamination. Whatever the case, in patients with this pattern of infection, the predominant organisms are aerobic rather than anaerobic. Gram-negative organisms, such as Pseudomonas aeruginosa, and Gram-positive organisms, such as staphylococci, are frequent isolates (26, 51). The lung damaged by aspiration probably provides an excellent setting for infection. Progression of infection in these cases depends not only on whether there is contamination by oropharyngeal flora but also on the need for ventilatory support, the quality of airway care, and the use of other therapeutic modalities, such as steroids or antibiotics, which might favor bacterial overgrowth. Predisposing Factors

The clinical setting in which aspiration of stomach

contents occurs has been reviewed extensively (5, 47, 52). Patients particularly at risk are those with altered states of consciousness, those with diseases that might affect normal swallowing or protective mechanisms, or those undergoing anesthesia. This is especially true during emergency surgical procedures on patients with full stomachs, or when pain, trauma, or analgesics have slowed intestinal motility. Aspiration is often the result of regurgitation, a passive process that contrasts with vomiting, a complicated and coordinated series of reflex maneuvers. Regurgitation, usually clinically silent, is common in patients with depressed mental status and is difficult to recognize (5). A high frequency of aspiration has been reported in patients who have had nasogastric tubes inserted or tracheostomies (52, 53). Theoretically, nasogastric tubes keep the stomach free of significant volumes of acid; however, unless the tubes are expertly placed and maintained in position, large amounts of gastric secretions can accumulate. In addition, the presence of the tube passing through both upper and lower esophageal sphincters renders them incompetent and increases the likelihood of aspiration. The incidence of aspiration in patients with tracheostomies in whom either a standard uncuffed metal tube or a small-volume, high-pressure cuff" is used has been recorded as high as 87%, whether or not the cuff was inflated (9, 53). The use of large-volume, low-pressure cuffs has reduced this to 15% (9). Tracheostomies, theoretically, contribute to aspiration by interfering with normal mechanisms of glottic closure. Interestingly, the use of nasotracheal and orotracheal tubes does not appear to increase the frequency of aspiration (52). Prevention

There are many factors that predispose a patient to aspirate, and virtually any patient may be at risk. The frequency of aspiration in the unconscious patient may be reduced by careful observation and by positioning the patient in a head-down position. The risk of aspiration associated with general anesthesia in the patient requiring emergency surgery may be reduced by the use of regional anesthetic techniques when appropriate, or awake endotracheal intubation preceding induction of general anesthesia. If the last-mentioned method is not possible, rapid induction and intubation of the trachea with simultaneous application of cricoid pressure to protect the patient's airway would be appropriate. The preoperative use of oral antacids to reduce gastric acidity also has been recommended (1, 39-41, 54). Although giving antacids has never been shown to reduce morbidity and mortality rates, a standard dose of an effective antacid given 30 minutes to 1 h before surgery usually will raise gastric pH above 2.5 (40, 41, 54, 55). For obstetric patients whose time of receiving anesthesia cannot be predicted, routine administration of antacids during labor at 2- to 3-h intervals has been recommended (39). Unfortunately, despite all preventive measures, aspiration may still occur; in that event, proper therapy is of the greatest importance. Wynne and Model/

• Aspiration of Stomach Contents

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19526/ by a University of California San Diego User on 03/19/2017

469

Therapy E N D O T R A C H E A L SUCTIONING

Most authors agree that when aspiration is observed, endotracheal suctioning should always be attempted (42, 46). It stimulates coughing, removes aspirated material, and may aid in confirming the diagnosis. However, dispersion of an aspirate is rapid, and in an acid aspiration, damage to the lung is almost instantaneous. Under the best conditions, suctioning removes only part of the aspirate (46). Therefore, it cannot be depended on to remove the aspirate entirely or to prevent lung injury and should be supplemented by other forms of therapy. BRONCHOSCOPY A N D L A V A G E

Bronchoscopy has been used to remove aspirated material and is indicated for any patient who has aspirated large particulate matter, especially if clinical or radiographic signs of loss of lung volume are present. Some physicians have lavaged the lung with neutral or alkaline solutions to neutralize aspirated acid (11). However, as previously mentioned, acid damages the lung almost instantaneously. In addition, bronchial secretions are buffered within minutes after an aspiration (4). These facts leave little rationale for lavaging the lung with neutral or alkaline solutions. In addition, most experimental studies of lavage have shown either no improvement or increased lung damage after lavage, especially when large volumes of lavage fluid were used (3, 4, 56, 57). An exception to this general experience was the study by Simenstad, Galway, and MacLean (58) that reported beneficial effects from large-volume lavage. However, their results were not statistically significant, and animals receiving lavage were also given 100% oxygen. Therefore, it is recommended that bronchoscopy be done only when particulate aspiration has been observed or is highly suspected. During bronchoscopy, small amounts of saline may be used to clear airways of secretions or aspirated material, if necessary. The use of large volumes of fluid should be avoided. CORTICOSTEROIDS

Corticosteroids often are recommended in the treatment of aspiration pneumonia, although their use is controversial. Clinical impressions, though optimistic, have been almost entirely anecdotal and based on uncontrolled factors (59-64). The theoretical basis for the use of steroids rests, among other things, on their ability to decrease inflammation, stabilize lysosomal membranes (65, 66), prevent platelet and leukocyte agglutination (67), and improve the peripheral release of oxygen from erythrocytes by shifting the oxyhemoglobin dissociation curve (68). Experimentally, steroids have been administered both intratracheally and parenterally. In 1965, Lewinski (69) administered intratracheal steroids to cats minutes after they had received an acid aspiration. He reported dramatic improvement in the gross and microscopic appearance of their lungs, as well as a reduction in the total area of injury. However, in 1968 Taylor and Pryse-Davies (3) repeated Lewiriski's experiment. They found no patho470

logic evidence of benefit from administering intratracheal steroids. A third study by Wamberg and Zeskov (57) reported a beneficial effect from the use of intratracheal steroids; however, the response of the animals to acid aspiration was variable, and the results were expressed in a fashion too vague to allow any useful conclusion. The use of intratracheal steroids, therefore, remains a matter of preference and intuition. A number of experimental studies using parenteral steroids tp treat aspiration also have been done, but their results are inconsistent. In 1961, Bannister, Sattilaro, and Otis (56) administered large intramuscular doses of hydrocortisone to eight rabbits after acid aspiration. Four of these rabbits were also given antibiotics. The steroid-treated animals had less extensive areas of damage than did control animals. However, survival rate was the same in both groups; the exact methods by which lung damage was quantitated were not given, and the results were expressed only in general descriptive terms. In 1964, Hamelberg and Bosomworth (11) reported reduction in the extent of pneumonitis seen on roentgenograms of five dogs treated with intramuscular hydrocortisone after an acid aspiration. Once again, survival rates were similar in control and treated animals. In addition, positive-pressure ventilation was administered to both groups in an uncontrolled fashion. Although physiologic features were measured and pathologic examinations carried out, the results were neither reported nor correlated with the roentgenographic observations. Exarhos and colleagues (70) later evaluated the effects of steroids on two dogs that had received acid aspiration followed by pulmonary lavage. The type, dosage, and rate of steroid administration and the nature and volume of the lavage fluid used were not stated. In addition, positive-pressure ventilation was used but was not controlled. Finally, the results were compared with those from other small treatment groups rather than with those of control animals. In 1966, Lawson and co-workers (71) compared the effects of acid aspiration on experimental animals that were either [a] untreated; [b] treated with conservative medical management, including oxygen, positive-pressure ventilation as needed, bronchodilators, antibiotics, and suctioning; or [c] treated with conservative therapy and steroids. Animals in both groups that were treated had a lower mortality rate and a more stable clinical course than did untreated animals. Comparison of the steroid-treated and the conservatively treated animals revealed no differences in mortality rate or extent of pathologic involvement. Physiologic differences did occur but were limited to the first 24 h. Arterial Po 2 , measured when the animals were breathing room air, was significantly higher in the steroid-treated group, despite lower respiratory rates and minute volumes. However, after 24 h, the clinical courses of both treatment groups were indistinguishable. Unfortunately, once again the use of positive-pressure ventilation was not controlled. Other studies showing beneficial effects from steroid administration are subject to similar criticisms (50, 57,

October 1977 • Annals of Internal Medicine • Volume 87 • Number 4

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19526/ by a University of California San Diego User on 03/19/2017

72). In general, all have used treatment groups too small to allow valid statistical conclusions, have used poor control of experimental conditions, or have expressed results in vague, general terms difficult to quantitate. In addition, little attention has been paid to the large variability among animals in response to the experimental conditions, to sampling errors caused by patchy and nonuniform involvement of the lung in this condition, or to the severity of the experimental models, which could easily mask subtle changes resulting from therapy. Finally, the dosages of steroids used have varied from study to study. Often they have been administered in single doses or concomitantly with an aspirate, practices that contrast with the usual clinical situation. Similar criticisms have been directed at studies showing no benefits from steroid administration (4, 17). However, a series of recent studies (13, 20, 73) strongly suggests that steroids are of little benefit in experimental acid aspiration. In the original study by Downs and co-workers (13), a small number of dogs were given methylprednisolone, either in a single high dose (30 mg/kg), multiple high doses (30 mg/kg every 8 h for 3 days), or multiple low doses (0.3 mg/kg every 8 h for 3 days) after receiving an acid aspirate (pH 1.0 to 1.1). They were then compared with untreated control animals. Evaluation of various physiologic features as well as survival rates and morphologic changes revealed no differences between the groups. However, the small number of animals studied and the severity of the experimental model made evaluating the clinical applicability of the results difficult. In a following study by Chapman and colleagues (20) involving a larger number of animals, all dogs were supported with controlled mechanical ventilation and with 10 cm of water positive end-expiratory pressure after acid aspiration (pH 1.0). One half of the animals received methylprednisolone (30 mg/kg) shortly after aspiration. N o statistically significant differences could be detected between control animals and animals treated with steroids regarding blood gas tensions, intrapulmonary shunt, cardiac output, pulmonary artery and capillary wedge pressures, and a variety of other laboratory and clinical features. Four of the 10 dogs in each group survived, and morphologic appearance of the lungs at autopsy was similar in both groups. In a third study (73), animals received either [a] no treatment; [b] high-dose methylprednisolone (30 mg/kg every 8 h for 3 days); [c] continuous positive-pressure ventilation; or [d] continuous positive-pressure ventilation plus high-dose methylprednisolone after an acid aspiration (pH 1.8). Using this less toxic model (95% survival rate), no beneficial effect of methylprednisolone could be demonstrated, although arterial Po 2 increased with continuous positive-pressure ventilation. In summary, there are no conclusive clinical or experimental data on which to base the use of steroids in aspiration pneumonia. Experimental evidence indicates that there are minimal, if any, benefits from using steroids in severe acid aspiration. The effects of steroids on nonacid aspiration are not known, although recent studies in dogs (74) and humans (75) have shown that they are of no

benefit in treating near-drowning. Justification for using steroids in aspiration rests on theoretical considerations which in themselves are controversial. One who uses steroids in the treatment of this disorder must weigh their unproven benefits against possible complications. ANTIBIOTICS

The prevention and the treatment of infection associated with aspiration are complicated matters. The presence or absence of infection in a patient who has aspirated is difficult to document. The development of fever, leukocytosis, pulmonary infiltrates, and thick, tenacious sputum are nonspecific responses that can result from uncomplicated chemical pneumonitis. Furthermore, samples for culture not obtained directly from the lung may be misleading because of contamination by oropharyngeal flora. Finally, even if well-collected culture samples have positive results, there remains the problem of determining whether the respiratory tract is truly infected or merely colonized. As previously discussed, the organisms responsible for infection complicating aspiration are variable. In general, patients presenting with lung abscess, empyema, and pneumonia after aspiration are more likely to grow organisms that reflect the oropharyngeal flora, especially anaerobes. Patients who present with severe respiratory failure requiring ventilatory support are more likely to harbor aerobic bacteria such as Staphylococcus and Pseudomonas. Despite difficulties in predicting whether infection will occur with aspiration and what organisms will be involved, many authors recommend giving prophylactic antibiotics after an observed aspiration (25, 26, 46, 47, 49). The effectiveness of this practice is doubtful. Although no well-controlled clinical studies have been done, retrospective review of the results has been discouraging. Of 18 patients with aspiration reported by Lewis, Burgess, and Hampson (7), 15 received prophylactic antibiotics; 13 of these subsequently developed infection, four with resistant organisms. In reviewing 47 patients with documented aspiration, Cameron, Mitchell, and Zuidema (10) were unable to detect any beneficial effect from the routine use of antibiotics, either alone or in combination with steroids or ventilation. Bynum and Pierce (51) also reported that initial treatment with antibiotics had no apparent effect on the clinical outcome or subsequent development of infection in 54 cases of aspiration presenting with severe respiratory failure. In their series, when infection occurred, it was usually due to Gram-negative bacteria and was thought to be nosocomial in origin. If prophylactic antibiotics are to be used in aspiration, allowances must be made for coverage of all possible infecting organisms. Completeness would require protection against all anaerobes, including Bacteroides fragilis, as well as similar broad-spectrum coverage for aerobes. A more logical approach, in our opinion, would be to withhold antibiotics initially, monitor the patient clinically for evidence of infection, and treat on the basis of results from well-collected smear and culture specimens. Wynne and Mode// • Aspiration of Stomach Contents

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19526/ by a University of California San Diego User on 03/19/2017

471

VENTILATORY SUPPORT

Positive-pressure ventilation with and without positive end-expiratory pressure has been shown to be successful in reversing blood-gas abnormalities and in improving survival rate after aspiration (15, 20, 23, 24, 73, 76). In addition to its obvious supportive function, positive-pressure ventilation also may have therapeutic benefits in acid aspiration (23, 24). In 1968, Cameron and co-workers (24) studied the effects of positive-pressure ventilation on three groups of dogs that received a 2-ml/kg dose of 0.1 N hydrochloric acid in the right lung through a Carlen's endotracheal catheter. The animals were either left untreated, given immediate positive-pressure ventilation for 6 h followed by no further treatment, or given positivepressure ventilation for 6 h after a 24-h delay, during which time no therapy was given. In the untreated and delayed-treatment groups, the mortality rate was about 80%; in contrast, all animals that were immediately ventilated survived. Radiographic evidence of lung damage was the same in all groups. Lung scans done on selected animals from each of the groups showed an initial loss of perfusion to the affected lungs. A gradual return of perfusion was seen in surviving animals, although this did not occur in nonsurvivors. In a second experiment using a similar model (23), pulmonary arteriograms were taken before and 6 h after an acid aspirate was given to control animals and animals treated with positive-pressure ventilation. All of the untreated group developed vascular abnormalities detected by pulmonary arteriograms, including pulmonary arterial vasospasm and premortem thrombus formation in extensively involved sections of lung. None of these changes were seen in the treated animals. These studies and subsequent work done by the same group (15) suggest that a poor survival rate in aspiration is associated with irreversible changes in the pulmonary vasculature that can be prevented by the early initiation of positive-pressure ventilation. Although other investigators have reported vasospasm after aspiration (11), none has noted thrombus formation, even when it has been sought specifically (21). In reviewing their clinical experience, Cameron, Mitchell, and Zuidema (10) have reported continued high mortality rates from aspiration despite the use of positive-pressure ventilation. This certainly should be an area for further research, since clinical application of these studies would require, to prevent irreversible changes, that positivepressure ventilation be initiated early in all patients known or suspected of aspirating, even when not specifically indicated by blood-gas determinations. Besides the work of Cameron and associates (10), there have been few experimental studies that have directly assessed the role of positive-pressure ventilation in aspiration pneumonia. Chapman and associates (20) were able to increase the survival rate from 20% to 40% in animals receiving a strongly acid aspiration (pH 1.0) by applying positive-pressure ventilation with 10 cm of water positive end-expiratory pressure for 4 h after aspiration. Toussaint, Chiu, and Hampson (21) also reported an improvement in survival rate in small groups of animals receiving 4 7 2

positive-pressure ventilation with room air and 100% oxygen, despite the fact that arterial Po 2 was unchanged in animals ventilated with room air and inconsistently improved when 100% oxygen was used. Other investigators (4, 50) have reported beneficial effects with positivepressure ventilation in experimental acid aspiration, but in general these results were noted only in passing and were not obtained under controlled conditions. Because of the lack of solid experimental data dealing specifically with the use of positive-pressure ventilation in aspiration, most clinicians treat patients with respiratory failure due to aspiration as they would patients with "adult respiratory distress syndrome" due to other causes (77). Since the earliest recognition of adult respiratory distress syndrome, positive end-expiratory pressure has been the cornerstone of therapy (78). Although the exact role of positive end-expiratory pressure is not clear, it does elevate functional residual capacity and prevent airway closure and atelectasis. As a result, ventilation-to-perfusion matching is improved and venous admixture reduced (76). At the very least, this allows the use of less toxic levels of oxygen and gives the lung a chance to recover. Whether positive end-expiratory pressure also plays a therapeutic role and directly aids in the healing process is still a matter of debate. Positive end-expiratory pressure is indicated when adequate oxygenation cannot be maintained without the use of toxic levels of oxygen (77). Positive end-expiratory pressure can depress cardiac output by interfering with venous return, and the ideal level of positive end-expiratory pressure is one that allows maximum reduction in venous admixture with minimum reduction in cardiac output (79, 80). The amount of positive end-expiratory pressure necessary to achieve a therapeutic response will vary from patient to patient, and extremely high levels have been used with success in selected patients (81). When higher levels of positive end-expiratory pressure are used, it is necessary to monitor the patient's cardiac and pulmonary status carefully. This generally requires inserting a catheter into the pulmonary artery and making multiple blood-gas measurements of both arterial and mixed venous samples. Recently it has been shown that spontaneous breathing with positive end-expiratory pressure is less likely to depress cardiac output than controlled ventilation with positive end-expiratory pressure (82). This finding provides a rationale for further investigation of alternate methods of applying positive end-expiratory pressure, such as the positive end-expiratory pressure mask (83) or intermittent mandatory ventilation (84). The latter allows patients to breathe spontaneously between breaths supplied by a ventilator, thereby reducing the mean intrathoracic pressure and enhancing cardiac output. In some cases where hypoventilation is not a problem, the use of a positive end-expiratory pressure mask may preclude entirely the need for intubation and mechanical ventilation. Besides reducing cardiac output, positive end-expiratory pressure can cause barotrauma. Subcutaneous and mediastinal emphysema and pneumothorax have been re-

October 1977 • Annals of Internal Medicine • Volume 87 • Number 4

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19526/ by a University of California San Diego User on 03/19/2017

ported. In general, the last is the only problem of major clinical significance. Undoubtedly, many factors contribute to the development of pneumothorax in patients being ventilated with positive end-expiratory pressure: underlying lung disease, regional difference in lung compliance, high inspiratory pressures, and the level of positive end-expiratory pressure used. Not all patients with aspiration will require ventilatory assistance. Ventilatory support should be initiated when indicated by clinical assessment and blood-gas measurements. OTHER SUPPORTIVE MEASURES

Proper use of fluids is critically important in patients with severe aspiration. Loss of intravascular volume may occur (4, 7, 8) and should be corrected quickly. If patients require prolonged ventilatory support, they may tend to accumulate fluid (85) and should be monitored carefully with accurate intake and output measurements and daily weighings, if possible. The pulmonary edema associated with aspiration is not caused by elevated left ventricular pressures. Commonly used measures of treating pulmonary edema secondary to cardiac failure (for example, digitalis, diuretics, rotating tourniquets, and phlebotomy) are of little help and may aggravate intravascular volume loss. If cardiac failure is suspected, placement of a pulmonary arterial catheter will allow measurement of pulmonary capillary wedge pressures. We have not found therapy with albumin and diuretics, as advocated by Skillman, Parikh, and Tanenbaum (86), to be of benefit in patients with aspiration, despite encouraging experimental studies (87, 88). In summary, optimum management of patients with severe aspiration pneumonia should entail immediate endotracheal suctioning if aspiration is observed, bronchoscopy if large-particle aspiration is seen or suspected, aggressive ventilatory support with supplemental oxygen and positive end-expiratory pressure if indicated by clinical assessment and blood-gas measurements, and adequate fluid replacement. Lavaging with large volumes of neutral or alkaline solutions and administering prophylactic antibiotics are not indicated. Corticosteroids, once thought to be a cornerstone of therapy, are of unproven benefit. ACKNOWLEDGMENTS: The authors thank Pauline Snider for editorial assistance. Grant support: in part by the Florida Lung Association, and by Public Health Service Pulmonary Academic Award K07 HL00122 from the National Heart and Lung Institute. Received 9 September 1976; revision accepted 7 January 1977. • Requests for reprints should be addressed to James W. Wynne, M.D.; Division of Pulmonary Medicine, Department of Medicine, J. Hillis Miller Health Center, Box J-225; Gainesville, F L 32610.

References 1. MENDELSON CL: The aspiration of stomach contents into the lungs during obstetric anesthesia. Am J Obstet Gynecol 52:191-205, 1946 2. TEABEAUT JR II: Aspiration of gastric contents. An experimental study. Am J Pathol 28:51-67, 1952 3. TAYLOR G, PRYSE-DAVIES J: Evaluation of endotracheal steroid therapy in acid pulmonary aspiration syndrome (Mendelson's syndrome). Anesthesiology 29:17-21, 1968 4. A W E WC, FLETCHER WS, JACOB SW: The pathophysiology of aspiration pneumonitis. Surgery 60:232-239, 1966 5. BANNISTER WK, SATTILARO AJ: Vomiting and aspiration during anesthesia. Anesthesiology 23:251-264, 1962

6. VANDAM LD: Aspiration of gastric contents in the operative period. N Engl J Med 273:1206-1208, 1965 7. L E W I S RT, BURGESS JH, H A M P S O N LG: Cardiorespiratory studies in

critical illness. Arch Surg 103:335-340, 1971 8. G R E E N F I E L D LJ, S I N G L E T O N R P , M C C A F F R E E DR, COALSON JJ: Pul-

monary effects of experimental graded aspiration of hydrochloric acid. Ann Surg 170:74-86, 1969 9. B O N E DK, D A V I S JL, Z U I D E M A G D , C A M E R O N JL: Aspiration pneu-

monia. Prevention of aspiration in patients with tracheostomies. Ann ThoracSurg 18:30-37, 1974 10. C A M E R O N JL, M I T C H E L L WH, Z U I D E M A G D : Aspiration pneumonia.

Clinical outcome following documented aspiration. Arch Surg 106:4952, 1973 11. HAMELBERG W, BOSOMWORTH PP: Aspiration pneumonitis: experimental studies and clinical observations. Anesth Analg (Cleve) 43:669677, 1964 12. VILINSKAS J, S C H W E I Z E R RT, F O S T E R J H : Experimental studies on

aspiration of contents of obstructed intestine. Surg Gynecol 135:568-570, 1972

Obstet

13. D O W N S JB, C H A P M A N R L J R , M O D E L L JH, H O O D CI: An evaluation

of steroid therapy in aspiration pneumonitis. Anesthesiology 40:129-135, 1974 14. M O R A N TJ: Experimental aspiration pneumonia. IV. Inflammatory and reparative changes produced by intratracheal injections of autologous gastric juice and hydrochloric acid. Arch Pathol 60:122-129, 1955 15. C A M E R O N JL, C A L D I N I P, T O U N G J-K, Z U I D E M A G D : Aspiration

pneumonia: physiologic data following experimental aspiration. Surgery 72:238-245, 1972 16. D A V I D S O N JT, R U B I N S, E Y A L Z, POLLIACK A: A comparison of the

pulmonary response to the endotracheal instillation of 0.1 N hydrochloric acid and Hartmann's solution in the rabbit. Br J Anaesth 46:127132, 1974 17. H A L M A G Y I DFJ, C O L E B A T C H H J H , STARZECKI B: Inhalation of blood,

saliva, and alcohol: consequences, mechanism, and treatment. Thorax 17:244-250, 1962 18. COLEBATCH HJH, HALMAGYI DFJ: Reflex airway reaction to fluid aspiration. J Appl Physiol 17:787-794, 1962 19. M O D E L L JH, M O Y A F, N E W B Y EJ, R U I Z BC, S H O W E R S AV: The ef-

fects of fluid volume in seawater drowning. Ann Intern Med 67:68-80, 1967 20. C H A P M A N R L J R , D O W N S JB, M O D E L L JH, H O O D CI: The ineffective-

ness of steroid therapy in treating aspiration of hydrochloric acid. Arch Surg 108:858-861, 1974 21. TOUSSAINT G P M , C H I U CJ, H A M P S O N LG: Experimental aspiration

pneumonia: hemodynamics, ventilator and membrane oxygenator support. / Surg Res 16:324-329, 1974 22. F I S K RL, SYMES J F , A L D R I D G E LL, C O U V E S CM: The pathophysiolo-

gy and experimental therapy of acid pneumonitis in ex vivo lungs. Chest 57:364-370, 1970 23. BOOTH DJ, Z U I D E M A G D , CAMERON JL: Aspiration pneumonia: pulmonary arteriography after experimental aspiration. J Surg Res 12:4852, 1972 24. C A M E R O N JL, SEBOR J, A N D E R S O N R P , Z U I D E M A G D : Aspiration

pneumonia. Results of treatment by positive-pressure ventilation in dogs. J Surg Res 8:447-457, 1968 25. TINSTMAN TC, D I N E S DE, A R M S RA: Postoperative aspiration pneumonia. Surg Clin North Am 53:859-862, 1973 26. A R M S RA, D I N E S DE, TINSTMAN TC: Aspiration pneumonia. Chest 65:136-139, 1974 27. STEINER J, BACHOFEN M, BACHOFEN H: Recovery from aspiration pneumonitis. Pneumonologie 151:127-134, 1974 28. S L A D E N A, Z A N C A P, H A D N O T T W H : Aspiration

pneumonitis—the

sequelae. Chest 59:448-450, 1971 29. HARBORD RP, H O M I J: Microscopic changes in living lung after fluid aspiration. Anesth Analg (Cleve) 49:835-843, 1970 30. A L E X A N D E R IGS: The ultrastructure of the pulmonary alveolar vessels in Mendelson's (acid pulmonary aspiration) syndrome. Br J Anaesth 40:408-414, 1968 31. M O D E L L J H , M O Y A F, W I L L I A M S H D , W E I B L E Y TC: Changes in blood

32.

33. 34. 35. 36. 37.

gases and A-aD0 2 during near-drowning. Anesthesiology 29:456-465, 1968 G I A M M O N A ST, M O D E L L JH: Drowning by total immersion. Effects on pulmonary surfactant of distilled water, isotonic saline, and sea water. AmJDis Child 114:612-616, 1967 M O R A N TJ: Experimental food-aspiration pneumonia. Arch Pathol 52:350-354, 1951 M O R A N TJ: Milk-aspiration pneumonia in human and animal subjects. Arch Pathol 55:286-301, 1953 M O R A N TJ: Pulmonary edema produced by intratracheal injection of milk, feeding mixtures, and sugars. Am J Dis Child 86:45-50, 1953 SCHWARTZ P: Asylum pneumonitis. Gerontol Clin (Basel) 15:281-351, 1973 KNOBLICH R: Pulmonary granulomatosis caused by vegetable particles. Wynne and Model!

• Aspiration of Stomach Contents

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19526/ by a University of California San Diego User on 03/19/2017

473

So-called lentil pulse pneumonia. Am Rev Respir Dis 99:380-389, 1969 38. VIDYARTHI SC: Diffuse miliary granulomatosis of the lungs due to aspirated vegetable cells. Arch Pathol 83:215-218, 1967 39. ROBERTS RB, SHIRLEY MA: Reducing the risk of acid aspiration during cesarean section. Anesth Analg (Cleve) 53:859-868, 1974 40. TAYLOR G, PRYSE-DAVIES J: The prophylactic use of antacids in the prevention of the acid-pulmonary-aspiration syndrome (Mendelson's syndrome). Lancet 1:288-291, 1966 41. LAHIRI SK, THOMAS TA, HODGSON R M H : Single-dose antacid therapy for the prevention of Mendelson's syndrome. Br J Anaesth 45:11431146, 1973 42. BARTLETT JG, GORBACH SL: The triple threat of aspiration pneumonia. Chest 68:560-566, 1975

116:451-466, 1962 66. STARLING JR, R U D O L F LE, F E R G U S O N W, W A N G E N S T E E N SL: Bene-

fits of methylprednisolone in the isolated perfused organ. Ann Surg 177:566-571, 1973 67. WILSON JW: Treatment or prevention of pulmonary cellular damage with pharmacologic doses of corticosteroid. Surg Gynecol Obstet 134:675-681, 1972 68. B R Y A N - B R O W N CW, B A E K S, M A K A B A L I G, S H O E M A K E R WC: Con-

sumable oxygen: availability of oxygen in relation to oxyhemoglobin dissociation. Grit Care Med 1:17-21, 1973 69. LEWINSKI A: Evaluation of methods employed in the treatment of the chemical pneumonitis of aspiration. Anesthesiology 26:37-44, 1965 70. E X A R H O S N D , L O G A N W D J R , A B B O T T OA, H A T C H E R CR J R : The

importance of pH and volume in tracheobronchial aspiration. Dis Chest 47:167-169, 1965

43. B A R T L E T T JG, G O R B A C H SL, F I N E G O L D SM: The bacteriology of aspi-

44.

45.

46.

47. 48. 49. 50.

51. 52. 53. 54.

55.

ration pneumonia. Am J Med 56:202-207, 1974 LORBER B, SWENSON RM: Bacteriology of aspiration pneumonia. A prospective study of community- and hospital-acquired cases. Ann Intern Med 81:329-331, 1974 JOHANSON WG, PIERCE AK, SANFORD JP: Changing pharyngeal bacterial flora of hospitalized patients. Emergence of gram-negative bacilli. N Engl J Med 281:1137-1140, 1969 CAMERON JL, Z U I D E M A G D : Aspiration pneumonia, in Tice's Practice of Medicine, vol. 5, chapter 11. Hagerstown, Maryland, Harper & Row, 1972, pp. 1-20 D I N E S DE, TITUS JL, SESSLER AD: Aspiration pneumonitis. Mayo Clin Proc 45:347-360, 1970 OLSEN AM: The spectrum of aspiration pneumonitis. Ann Otol Rhinol Laryngol79:%75-S&%, 1970 R I B A U D O CA, G R A C E WJ: Pulmonary aspiration. Am J Med 50:510520, 1971 BOSOMWORTH PP, HAMELBERG W: Etiologic and therapeutic aspects of aspiration pneumonitis. Experimental study. Surg Forum 13:158-159, 1962 BYNUM LJ, PIERCE AK: Pulmonary aspiration of gastric contents. Am RevRespirDis 114:1129-1136, 1976 CAMERON JL, Z U I D E M A G D : Aspiration pneumonia. Magnitude and frequency of the problem. JAMA 219:1194-1196, 1972 CAMERON JL, REYNOLDS J, Z U I D E M A G D : Aspiration in patients with tracheostomies. Surg Gynecol Obstet 136:68-70, 1973 PESKETT WGH: Antacids before obstetric anaesthesia. A clinical evaluation of the effectiveness of mist, magnesium trisilicate BPC. Anaesthesia 28:509-513, 1973 LiTTMAN A, P I N E BH: Antacids and anticholinergic drugs. Ann Intern Med 82:544-551, 1975

71. L A W S O N DW, D E F A L C O AJ, P H E L P S J A, B R A D L E Y BE, M C C L E N A -

THAN JE: Corticosteroids as treatment for aspiration of gastric contents: an experimental study. Surgery 59:845-852, 1966 72. D U D L E Y WR, MARSHALL BE: Steroid treatment for acid-aspiration pneumonitis. Anesthesiology 40:136-141, 1974 73. C H A P M A N RL J R , M O D E L L JH, R U I Z BC, C A L D E R W O O D HW, H O O D

74.

75. 76. 77.

CI, G R A V E S SA: Effect of continuous positive-pressure ventilation and steroids on aspiration of hydrochloric acid (pH 1.8) in dogs. Anesth Analg (Cleve) 53:556-562, 1974 C A L D E R W O O D HW, M O D E L L JH, R u i z BC: The ineffectiveness of steroid therapy for treatment of fresh-water near-drowning. Anesthesiology 43:642-650, 1975 M O D E L L JH, G R A V E S SA, K E T O V E R A: Clinical course of 91 consecutive near-drowning victims. Chest 70:231-238, 1976 PONTOPPIDAN H, G E F F I N B, LOWENSTEIN E: Acute respiratory failure in the adult. I. N Engl J Med 287:690-698, 1972 BREDENBERG CE: Acute respiratory distress. Surg Clin North Am 54:1043-1066, 1974

78. A S H B A U G H D G , P E T T Y TL, B I G E L O W DB, H A R R I S TM: Continuous

positive-pressure breathing (CPPB) in adult respiratory distress syndrome. / Thorac Cardiovasc Surg 57:31-41, 1969 79. S U T E R PM, F A I R L E Y HB, ISENBERG M D : Optimum

end-expiratory

airway pressure in patients with acute pulmonary failure. N Engl J Med 292:284-289, 1975 80. LUTCH JS, M U R R A Y J F : Continuous positive-pressure ventilation: effects on systemic oxygen transport and tissue oxygenation. Ann Intern Med 76:193-202, 1972 81. K I R B Y RR, D O W N S JB, C I V E T T A JM, M O D E L L J H , D A N N E M I L L E R

FJ, K L E I N EF, HODGES M: High level positive end expiratory pressure (PEEP) in acute respiratory insufficiency. Chest 67:156-163, 1975

56. B A N N I S T E R WK, SATTILARO AJ, O T I S R D : Therapeutic aspects of as-

piration pneumonitis in experimental animals. Anesthesiology 22:440443, 1961 57. W A M B E R G K, ZESKOV B: Experimental studies on the course and treatment of aspiration pneumonia. Anesth Analg (Cleve) 45:230-236, 1966 58. SIMENSTAD JO, G A L WAY CF, M A C L E A N LD: Tracheobronchial lavage for treatment of aspiration and atelectasis. Surg Forum 13:155-157, 1962

82. K I R B Y RR, P E R R Y JC, C A L D E R W O O D HW, R u i z BC, L E D E R M A N DS:

59. N I C H O L L RM, H O L L A N D EL, B R O W N SS: Mendelson's syndrome: its

84. D O W N S JB, K L E I N E F J R , D E S A U T E L S D, M O D E L L JH, K I R B Y RR:

treatment by tracheostomy and hydrocortisone. Br Med J 2:745-746, 1967 60. G R A H A M EC, C H O Y D: Corticosteroids in aspiration pneumonia. JAMA 184:976-977, 1963

Intermittent mandatory ventilation: a new approach to weaning patients from mechanical ventilators. Chest 64:331-335, 1973

Cardiorespiratory effects of high positive end-expiratory pressure. Anesthesiology 43:533-539, 1975 83. TAYLOR GJ, BRENNER W, SUMMER WR: Severe viral pneumonia in young adults. Therapy with continuous positive airway pressure. Chest 69:722-728, 1976

85. S L A D E N A, L A V E R MB, P O N T O P P I D A N H: Pulmonary complications

and water retention in prolonged mechanical ventilation. N Engl J Med 279:448-453, 1968

61. D I N E S DE, B A K E R WG, S C A N T L A N D WA: Aspiration pneumonitis—

Mendelson's syndrome. JAMA 176:229-231, 1961 62. M A R X G F : Aspiration pneumonitis. JAMA 201:129-130, 1967 63. HAUSMANN W, L U N T RL: Problem of treatment of peptic aspiration pneumonia following obstetric anaesthesia (Mendelson's syndrome). / Obstet Gynaecol Br Commonw 62:509-512, 1955 64. MARSHALL BM, G O R D O N RA: Vomiting, regurgitation and aspiration in anaesthesia. I. Can Anaesth Soc 7 5:274-281, 1958

86. SKILLMAN JJ, P A R I K H BM, T A N E N B A U M BJ: Pulmonary arteriovenous

65. J A N O F F A, W E I S S M A N N G, Z W E I F A C H BW, T H O M A S L: Pathogenesis

Effects of albumin and/or furosemide therapy on pulmonary edema induced by hydrochloric acid aspiration in rabbits. / Trauma 16:788-791, 1976

admixture. Improvement with albumin and diuresis. Am J Surg 119:440-447, 1970 87. T O U N G TJK, BORDOS D, BENSON DW, C A R T E R D, Z U I D E M A G D ,

P E R M U T T S, CAMERON JL: Aspiration pneumonia: experimental evaluation of albumin and steroid therapy. Ann Surg 183:179-184, 1976 88. G E E R RT, SOMA LR, B A R N E S C, L E A T H E R M A N JL, M A R S H A L L BE:

of experimental shock. IV. Studies on lysosomes in normal and tolerant animals subjected to lethal trauma and endotoxemia. / Exp Med

4 7 4

October 1977 • Annals of Internal Medicine • Volume 87 • Number 4

Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19526/ by a University of California San Diego User on 03/19/2017