Barotrauma: Detection, Recognition, and Management

Barotrauma: Detection, Recognition, and Management* Theodore W Marcy, M.D. , F.C.C.P.t

(Chest 1993; 104:578-84) ARDS =adult respiratory distress syndrome; Cf =computed tomography; Flo1 =fractional inspired oxygen concentration; MAP=mean airway pressure; PA=mean alveolar pressure; PCV =pressure-controlled ventilation; PEEP= positive endexpiratory pressure; PIP= peak inspiratory pressure

B

arotrauma is an important and potentially lethal complication of mechanical ventilation. All physicians working with patients receiving mechanical ventilation must be familiar with the different clinical manifestations of barotrauma, the recognition of these manifestations, and their management. Perhaps the most important aspect of management is the use of ventilatory strategies that may prevent or limit the development of barotrauma. This portion of the symposium will first review different types of lung injury related to barotrauma, including one type that has only recently been recognized (systemic gas embolism), and another type that has only recently been attributed to barotrauma (diffuse lung injury). Next, the pathogenesis of these manifestations of barotrauma will be reviewed as this underlies the rationale of newer ventilatory strategies. CLINICAL MANIFESTATIONS

Extra-alveolar Gas

The most widely recognized form of barotrauma is the presence of extra-alveolar air. The strict definition of this phenomena is that gas presents in different tissue spaces after rupturing from alveoli .' Based on work originally performed by Macklin and Macklin in the 1940s! it is thought that gas in the alveoli first ruptures across the boundary between the alveoli and the bronchovascular sheath . The gas then dissects within this sheath toward the mediastinum from which it decompresses through other fascial planes producing the radiographic and clinical manifestations of pneumothorax, subcutaneous emphysema, subpleural air cysts, pneumomediastinum, and pneumoperitoneum . The incidence of pneumothorax in patients receiving mechanical ventilation is cited to be between 4 and 15 percent,'-7 although in patients with the adult respiratory distress syndrome (ARDS), status asthmaticus, and aspiration pneumonia, the incidence may be significantly higher.' -" Gammon and colleagues' recently published a study in which they retrospectively reviewed the films and medical records of 139 patients receiving ventilation in their ICU over a 10-month time period . While the incidence of pneumothoraces in the entire group was 14 percent, similar to earlier reports, the incidence in patients with ARDS was 60 percent, while none was observed in patients with *From the Universitv of Minnesota and St. Paul Ramsey Medical Center, St. Paul , Minn. tPresentlv at the Universitv of Ve rmont. BurlinJ-,
578

congestive heart failure or patients intubated for neurologic disease .7 The consequences of a pneumothorax in a patient receiving positive pressure ventilation can include impaired gas exchange, but, perhaps more importantly, adverse hemodynamic effects. Intrathoracic extra-alveolar air may increase intrathoracic pressure, decreasing vef!OUS return to the right side of the heart. Compression of the lung and pulmonary vessels may increase afterload to the right side of the heart, and geometric changes in the heart itself may decrease myocardial performance."-'" Studies in animals on positive pressure ventilation demonstrate that there is an almost linear decrease in cardiac output as the volume of an experimentally induced pneumothorax is increased .•• A significant fall in cardiac output may occur without a corresponding decrease in oxygenation. 10 Therefore, changes in arterial blood gases may not be sensitive indicators of a hemodynamically important pneumothorax. These data also support the clinical maxim that even a small pneumothorax must be evacuated in a patient receiving positive pressure ventilation as it may be causing hemodynamic effects or it could rapidly progress to one that would . Unfortunately, it is difficult to make a radiographic diagnosis of a pneumothorax on portable x-ray films taken in the ICU setting. ._., _, The traditional radiographic hallmark of a pneumothorax-an apicolateral collection of air with the visceral pleura visible as a thin white line in the upper hemithorax- may not be present on portable supine films of patients with pneumothoraces. Free air will often be located in the anterior costophrenic sulcus as this is the most superior portion of the pleural space in the supine patient.•• In addition, concurrent lung disease may lead to different distributions of free air in the pleural space than in patients with relatively normal lungs, 15 or to locula~ed gas collections if there is associated pleural disease." "Less common" radiographic patterns for pneumothorax may, in fact, be the most common findings in ICU patients. Tocino and colleagues" retrospectively determined the location of gas collections on the portable chest radiograp~s of patients in the ICU. The majority of pneumothoraces were located either in the anteromedial (38 percent) or subpulmonic (26 percent) areas. Only 22 percent of the pneumothoraces were located in the traditional apicolateral location. For this reason, the clinician must be alert to subtle radiographic signs of pneumothoraces in these other locations: (I) relative hyperlucency over the upper abdomina) quadrants;••-" (2) deep costophrenic angle (the deep sulcus sign);" -'" and (3) the visualization of the anterior costophrenic sulcus that results in a curvilinear change in density over the upper quadrant. ' 2 Despite careful review of the supine chest film, it is possible for pneumothoraces to be missed without the use of other imaging modalities or techniques. Several studies have reported that CT scans of the chest may have a greater Barotrauma (Theodore W. Marcy)

sensitivity to detect pneumothoraces in ICU patients than portable chest radiographs. 17''"1n one study, the investigators performed cr scan cuts through the chest in patients with multiple trauma who were having a head Cf scan performed .•• Twenty one pneumothoraces were visualized in 15 of the 25 patients by either chest radiograph or Cf scan . However, 11 of the pneumothoraces (52 percent) were detected by Cf scan alone. Of note, three pneumothoraces were missed by Cf scans because the cuts were not extended down into the subpulmonic area in these patients. Unfortunately, it is not always practical or safe to transport critically ill patients for a cr scan to exclude an occult pneumothorax, particularly when the patient is hemodynamically unstable. Are there methods of improving the sensitivity of portable chest radiographs taken in the ICU? Recently, Carr and colleagues"" investigated the sensitivity of the lateral decubitus chest radiograph for experimentally induced pneumothoraces in human cadavers. Lateral decubitus films for pneumothoraces are technically easier than those taken to detect pleural fluid as the attention is on the superior hemithorax and not the dependent hemithorax which may be obscured by the bed or other objects. Pneumothoraces were induced in the cadavers with a catheter in the right pleural space. Supine, erect, and left lateral decubitus films were taken at baseline and at each increment in injected volume. Computed radiography was also performed simultaneously. All films were interpreted randomly and in a blinded fashion by five radiologists who graded the films on a five-point scale for the presence or absence of a pneumothorax. The lateral decubitus film was superior to that of both the erect and supine films in sensitivity and could detect a pneumothorax as small as 50 mi. Conventional screen-film radiography provided similar sensitivity as computed radiography in all positions."' It is uncertain if lateral decubitus films would reliably detect loculated as opposed to free air collections. Therefore, a study in patients is required to confirm the results of these findings. Nonetheless, these results suggest that lateral decubitus films may be the next radiographic exam to obtain after a supine or erect chest radiograph when there is a high index of suspicion for a pneumothorax. Other forms of extra-alveolar air may themselves not be life threatening but may herald the development of pneumothoraces. Extra-alveolar air in the pulmonary parenchyma (pulmonary interstitial emphysema, subpleural air cysts)21 · 23 or in the mediastinum"·7 - 13 ·,.. may rupture through fascial planes into the pleural space and lead to hemodynamically significant pneumothoraces. In a retrospective review by Gammon et al7 of patients receiving mechanical ventilation, mediastinal emphysema was the initial manifestation of extra-alveolar air in 21 percent of the patients. Once again, the subgroup of ARDS patients had the highest incidence (62 percent). Subsequent pneumothoraces occurred in 42 percent of these patients within 3 days (50 percent of the patients with ARDS mediastinal emphysema). Prophylactic chest tubes in this series of patients would have been unnecessary in a large proportion of patients. Moreover, one cannot be certain that the chest tube will be located in the appropriate position. If the pneumothorax is loculated it may remain unevacuated by a previously placed but inadequately positioned chest tube." It may be more appropriate

to carefully monitor patients with mediastinal emphysema and be prepared to emergently place directed chest tubes if a pneumothorax develops. Other complications of extra-alveolar air in the pulmonary parenchyma can occur. Albeda and colleagues"' reported that in their series of five patients with subpleural air cysts, two patients developed secondary infections in the cysts. One of these patients subsequently died of Pseudomonas sepsis; Pseudomonas was cultured from purulent material present in the cysts at autopsy. Systemic Gas Embolism It is possible for extra-alveolar air to enter the systemic circulation if there is a bronchovenous communication and an adequate pressure gradient. Gas that has ruptured from alveolar compartments into the bronchovascular sheath may preferentially enter the pulmonary venous system if parenchymal infiltrates block the decompression of this gas into other tissue planes, and if the vascular structures have been disrupted by a necrotizing process or shear forces. The clinical manifestations occur because of entry of gas into the systemic circulation and subsequent occlusion of the vascular supply to multiple organs. Marini and Culver' described the simultaneous onset of livido reticularis over the right anterior hemithorax, focal neurologic changes with seizures, and myocardial infarction in two young patients with ARDS, high peak airway pressures, and tension air cyst formation . The distribution of the tissue injury could be explained by microembolization of gas into large vessels that would preferentially receive buoyant gas in the supine position: right internal thoracic artery, the right carotid artery, and the right coronary artery. In one patient, gas in the left atrium was visualized by a transthoracic echocardiogram without evidence of an atrial septal defect at autopsy. Systemic gas embolization is a well recognized complication in neonates with the neonatal respiratory distress syndrome,"" as well as in diving accidents 27 or when there is accidental injection of air into a central vein."" These cases in adult patients with ARDS suggest that this diagnosis should be considered when patients with other evidence of extra-alveolar air develop unexplained myocardial or cerebral injury. It is possible , though unproven, that more subtle abnormalities in acutely ill patients receiving mechanical ventilation are related to systemic gas embolization.

Diffuse Lung Injury

Forms of barotrauma that occur from the rupture of alveoli and the extravasation of alveolar gas may not be the only manifestation of tissue injury. In 1976, Teplitz"' reported his experience as a pathologist doing autopsies on burn victims at the United States Army Surgical Research Unit during the time when routine clinical blood gas monitoring became available in the early 1960s. Prior to 1963, he rarely saw evidence of hyaline membrane formation in patients dying with pulmonary edema. After 1963, when the aggressive ventilatory management of gas exchange abnormalities was instituted, he frequently observed hyaline membrane formation in these patients. It was his hypothesis that ARDS was nothing more than the iatrogenic complication of ventilatory management for noncardiogenic pulmonary edema."' He could not distinguish , however, if these histoCHEST I 104 I 2 I AUGUST, 1993

579

pathologic changes were related to the ventilatory management, the increased inspired oxygen concentrations, or to the natural history of this complication in patients who received aggressive life support. Subsequently, numerous investigators have demonstrated that diffuse lung edema and tissue histopathologic changes indistinguishable from that of ARDS can occur in normal animals ventilated with moderately high peak airway pressures (30 to 50 em H20).30-"" Webb and Tierney'" demonstrated that pulmonary edema and hypoxemia would occur in normal rats ventilated at peak inspiratory pressures (PIP) of 30 em H,O. Interestingly, applied PEEP at 10 em H,O decreased the level of injury. Kolobow and colleagues"' demonstrated at autopsy that normal sheep ventilated at a PIP of 50 em H 20 developed, within 35 h, severely reduced static lung compliance, decreased FRC, hypoxemia, and grossly abnormal lungs. In a related study,"" sheep ventilated with high tidal volumes developed not only lung injury but hypotension and renal failure similar to that seen in patients with multiple organ system failure. Although not yet supported by data from patients, the implication of these studies is that lung edema may be perpetuated or extended by the very technique designed to compensate for it. These studies have provoked a reexamination of the underlying pathogenesis of barotrauma, and the ventilatory strategies applied to certain patient groups predisposed to this complication. PATHOGENESIS OF BAROTRAUMA Peak inspiratory pressures over 40 to 50 em H,O are associated with an increased risk of alveolar rupture during mechanical ventilation, and the incidence of barotrauma exceeds 40 percent in patients exposed to a PIP above 70 em H20. • Other data suggest an association of barotrauma with the requirement for high levels of PEEP.• As reviewed above, certain types oflung injury-notably ARDS-appear to increase the risk of barotrauma. However, how these factors interact, and their relative contributions to the risk of barotrauma are not completely understood . Early investigations• into the pathophysiology of pneumothoraces demonstrated that alveolar disruption appeared to occur in alveoli bordering the bronchovascular sheath when there was an excessive pressure gradient across this boundary. Subsequent studies suggested that alveolar volume was also an important factor. 40 In fact , high airway pressures can be applied to the respiratory system without injury if excessive lung inflation is prevented. For example, Dreyfuss and colleagues"' reported that rats had impressive alveolar flooding and microvascular injury when exposed to 20 min of peak airway pressures of 45 em H 20 . This degree of injury, however, could be prevented if the chests of the animals were strapped. This relationship of lung injury to hyperinflation has led some investigators to suggest that "barotrauma" might be more appropriately termed "volutrauma;· and that the relevant pressure is not peak airway pressure or PEEP level, but the transalveolar pressure .41 · 42 In fact , Dreyfuss et al"' could reproduce lung injury in his rat model by using negative pressure ventilators that created similar transalveolar pressures. Because the normal lung is inflated to total lung capacity with transalveolar pressures of 35 em H 20 , 580

there is an emerging consensus that peak alveolar pressures (or transthoracic pressure) exceeding 35 to 40 em H 20 may increase the risk of tissue rupture . Above this pressure, normal alveoli are almost certainly overdistended , and fragile alveoli may rupture. High transalveolar pressures and alveolar overdistention are unlikely to be the only causative factors in barotrauma. In neonates, significant barotrauma can occur during high frequency ventilation even with modest peak and mean alveolar pressures. In animal models of ARDS, large tidal volumes contributed to increased lung edema even if peak airway pressures, end-inspiratory lung volume, and pulmonary capillary wedge pressures were held constant.43 These and other data suggest that a variety of factors influence the risk of barotrauma: maximal distending pressures, mean airway pressure, the "mechanical stress of hyperventilation," tissue fragility, secretion retention, surfactant depletion, shear forces, and the duration of ventilation. The role of underlying lung injury as a risk factor for barotrauma can be illustrated by examining the underlying characteristics of the lung in patients with early ARDS. Diffuse lung injury is one of the defining characteristics of ARDS . However, despite the widespread tissue injury and the diffuse infiltrates present on the frontal chest radiograph, there does not appear to be a homogeneous distribution of radiographic abnormalities, gas exchange abnormalities, or mechanical alterations. When performed within the first 10 days of disease onset, CT scans demonstrate marked heterogeneity in the radiographic density of the lung........ Analyses of gas exchange using inert gas techniques demonstrate that, despite the marked increases in shunt and dead space, a significant proportion oflung units retain normal ventilation to perfusion r:\TIQ) ratios... Another indication that many lung units are uninvolved, or at least are mechanically normal, is that the compliance of the lung corrected for aerated lung volume (the specific compliance) is comparable to that of anesthetized normal subjects.47 •48 Thus, the functioning lung in ARDS is not so much stiff as it is small. The ventilator must apply high pressures to the airway to deliver a tidal volume of"standard size" (10 to 12 ml/kg) into these small lungs. These high applied pressures can lead to hyperinflation of regions of the lung that have relatively normal lung units, putting them at risk for overdistention, rupture, or injury. VENTILATORY STRATEGIES TO DECREASE BAROTRAUMA The ideal ventilatory strategy would relieve the patient of an excessive breathing workload , assure adequate gas exchange to support organ function, and recruit collapsed or flooded alveoli without injuring normally compliant lung units. Unfortunately, the goal of achieving adequate gas exchange to support organ function may conflict with the goal of limiting regional hyperinflation. For example, the conventional ventilatory strategy in ARDS patients is to use tidal volumes of 10 to 15 mllkg and respiratory rates adjusted to normalize pH and PaCO,. Positive end-expiratory pressure is applied as necessary to achieve acceptable 0 2 saturations at nontoxic inspired oxygen concentrations. In some ARDS patients, this strategy will be associated with peak transthoracic pressures exceeding 50 em H.O-well within the range that may cause alveolar overdistention and Barotrauma (Theodore W. Marcy)

tissue injury. Pressure-Targeted \fmtilation (Permissive Hypercapnia)

Newer ventilatory strategies in ARDS patients use a "pressure-targeted" approach in that they limit peak distending pressures applied to the respiratory system rather than targeting a nonnal PaC02 or pH. Common objectives of these strategies are to maintain some minimum effective level of positive alveolar pressure throughout the respiratory cycle, thereby preventing collapse of unstable lung units; and to prevent peak alveolar (plateau) pressures from exceeding 35 em H 20 ....... Once the diagnosis of ARDS is established, we initiate conventional volume-cycled ventilation with the assist-control mode using tidal volumes of7 to 10 mVkg. Next, PEEP, at a minimum level of 5 to 7 em H2 0, is applied in an effort to recruit nonaerated alveoli and to prevent airway closure and alveolar collapse at end-expiration. Increments of PEEP are added as needed to reduce fractional inspired oxygen concentration (Flo.) requirements, but a PEEP level of 15 em H20 is rarely exceeded as this will further increase peak distending pressures. We aggressively control respiratory infections, work to clear secretions, and reposition the patient frequently (if tolerated) to maximize volume recruitment and minimize alveolar collapse. Tidal volume is adjusted to maintain peak plateau pressures at, or less than, 35 em H20. Occasionally, in a patient with a very stiff chest wall, higher maximal alveolar pressures may be acceptable because the corresponding transalveolar pressure will be less. To limit cycling pressures and lung injury, it may be desirable and necessary to use relatively small tidal volumes (5 to 8 mVkg). In some patients, it will not be possible to maintain a nonnal PaC02 and still stay within acceptable pressure limits. In these patients, the PaCO, is allowed to rise gradually despite the development of a respiratory acidosis, a strategy tenned "pennissive hypercapnia.''"·'" While acute hypercapnia has potential deleterious effects, including intracellular acidosis, increased pulmonary vascular resistance, increased cerebral blood flow, and increased sympathetic activity, the gradual development of hypercapnia allows for compensatory changes to occur. Hypercapnia has been relatively well tolerated in animal experiments,"" in humans during anesthesia,,. and in some patient groups where this strategy has been employed either by clinicians (status asthmaticus)"" or by the patients themselves (COPD with chronic hypercapnia). Hickling and colleagues'" used a pressure-targeted stategy in a series of 50 patients with severe ARDS (defined as having a lung injury score 2:2.5-'4). Patients were ventilated with tidal volumes as low as 350 ml (5 mVkg) to maintain peak inspiratory pressure less than 30 to 40 em H,O. The maximum PaC02 in these patients averaged 59 mm Hg, but exceeded 90 mm Hg in four patients, three of whom survived. The survival rate was higher than that predicted by APACHE II scores (corrected for hypercapnia), and eight surviving patients had a "ventilator score" (detennined by age, oxygenation, and peak inspiratory pressures) that had previously been associated with 100 percent mortality."' The study, however, did not include a concurrent control group. A separate investigation""·"" compared extracorporeal CO,

removal to a pressure-targeted ventilation strategy in ARDS patients who met the extracorporeal membrane oxygenation (ECMO) study criteria. There were no differences in survival between the two treatment groups, but survival in both groups was significantly better than that previously reported in patients meeting these same criteria who were managed with conventional ventilatory strategies. Once again, there was not a concurrent group treated with conventional ventilation. Controlled studies of pressure-targeted ventilation versus conventional ventilation are needed to confirm these preliminary findings. Pressure-Controlled Ventilation

Some investigators advocate the use of pressure-controlled ventilation (PCV) in patients with ARDS as a means of improving gas exchange while limiting barotrauma.""·"' During PCV, a rectangular ("square") wave of set pressure is applied at the airway opening for a defined duration (expressed usually as a fraction of the total respiratory cycle, or as a ratio in relation to expiratory time). Inspiratory flow is initially high, reflecting the large gradient between alveolar and airway pressure, then decelerates as this pressure gradient falls. There are several potential advantages to this mode . First, the set pressure represents a maximum pressure that alveolar pressure cannot exceed under conditions of passive ventilation, thereby preventing uncontrolled and dangerous increases in alveolar pressure. Second, gas exchange may improve under conditions of decelerating flow; rapid expansion of alveoli may facilitate collateral ventilation or improve gas exchange by other mechanisms. As with any pressure-limited mode of ventilation, the volume actually delivered varies with respiratory system compliance and resistance . Furthennore, any level of inadvertent end-expiratory pressure generated by dynamic hyperinflation will exert a direct (and possibly undetected) opposing pressure. Therefore, careful monitoring of exhaled volume and minute ventilation is mandatory. Increments of applied PEEP will also decrease delivered tidal volume unless the set pressure is also increased. If minute ventilation is inadequate, it may seem logical to raise respiratory frequency. However, minute ventilation does not increase linearly with frequency during PCV and actually reaches a plateau ... Alveolar ventilation reaches a peak value and then actually falls with further increments of frequencies. 63 Finally, not all ventilators designed for adult patients are able to provide PCV, and not all clinicians are familiar with its use and complexities. Extended Inspiratory 1ime Ventilatory Modes (Inverse Ratio Ventilation)

During positive pressure ventilation of a passive subject, mean airway pressure (MAP) is the pressure measured at the airway opening averaged over the entire respiratory cycle, and reflects the average pressure applied by the ventilator. Under these same conditions, mean alveolar pressure (PA) is the average pressure acting to distend the alveoli against the combined recoil of lung and chest wall. In the setting of lung edema, PA corresponds directly to alveolar recruitment, shunt reduction, and blood oxygenation. Although P A cannot be measured directly during the CHEST I 104 I 2 I AUGUST, 1993

581

tidal cycle, it is closely related to changes in the MAP of the patient receiving passive ventilation ... As a result, MAP is a major determinant of oxygenation as well as a key determinant of hemodynamic compromise ...... The MAP, in tum, is a function of the minute ventilation, the amount of applied PEEP, and the pattern of applied airway pressure ... One approach to refractory hypoxemia in ARDS patients may be to augment MAP by extending the inspiratory time rather than by increasing applied PEEP. The inspiratory time can be extended, and MAP increased in small increments during volume-controlled ventilation by decreasing the inspiratory flow rate, by adding an endinspiratory pause, or by changing to a decelerating flow waveform .rn Alternatively, PCV can be used with protracted inspiratory times.'"' In concept, prolongation of the inspiratory time should maintain tidal volume and increase MAP at lower levels of PEEP and peak distending pressures, provided that excessive gas trapping does not occur. Sustained elevations in airway pressure may recruit lung units more effectively than transient increases.""·70 Sustained alveolar inflation also appears to decrease dead space, perhaps by enhancing the efficacy of collateral ventilation or by facilitating the mixing of gas in well and poorly perfused regions.""·70 Eventually, as expiratory time is shortened, autoPEEP (dynamic hyperinflation) will develop. Unlike the situation with patients who have dynamic airway collapse, PEEP and auto PEEP are generally additive in ARDS patients. 71 As end-expiratory alveolar pressure increases, either peak alveolar pressure will rise during volume-controlled ventilation, or tidal volume will fall during PCV. The total endexpiratory pressure (determined by end-expiratory port occlusion) must be monitored and maintained between 10 and 15 em H10 by altering the level of applied PEEP or the inspiratory time . There are several real or theoretic problems that can occur with this ventilatory strategy. As expiratory time falls, there may be air-trapping in lung units with high expiratory resistance, and these disadvantaged lung units may become overdistended. Thus, increases in MAP and inspiratory time may exacerbate or extend barotrauma. The second potential problem with extended inspiratory time ventilation is that cardiac output may fall as MAP increases. A third problem is that many patients do not tolerate extended inspiratory times without sedation, and in some cases, neuromuscular paralysis. This is particularly important in volume-controlled ventilation as alveolar pressures will increase dangerously if the respiratory frequency is excessive. For these reasons, the ventilator should be adjusted to provide the shortest inspiratory time and the lowest MAP associated with adequate oxygenation. Many patients do not require actual inversion of their I:E ratio. Unfortunately, many patients do not achieve sufficient benefit from this strategy to justify its hazards. Although there are many anecdotal reports, '"'·71-7• there are no controlled studies comparing survival in ARDS patients treated with extended inspiratory time ventilation to those treated with conventional ventilation. CoNCLUSIONS

The hazards of excessive applied pressures and, perhaps 582

more directly, regional hyperinflation extend beyond pneumothoraces to other forms of lung and systemic injury. Newer ventilatory strategies such as pressure targeted ventilation with permissive hypercapnia attempt to limit overdistension of relatively normal lung units. Preliminary studies suggest, but as yet do not confirm, that this approach may reduce the incidence of barotrauma. Until there is more investigational and clinical experience, these methods must be applied cautiously and with full awareness of their possible hazards. REFERENCES

1 Rohlfing BM, Webb WR, Schlobohm RM . Ventilator-related extra-alveolar air in adults. Radiology 1976; 121:25-31 2 Macklin MT, Macklin CC. Malignant interstitial emphysema of the lungs and mediastinum as an important occult complication in many respiratory diseases and other conditions: an interpretation of the clinical literature in the light of laboratory experiment. Medicine 1944; 23:281-352 3 Zwillich CW, Pierson DJ, Creagh CE, Sutton FD, Schatz E, Petty TL. Complications of assisted ventilation: a prospective study of354 consecutive episodes. Am J Med 1974; 57:161-70 4 de Latorre FJ, Tomasa A, Klamburg J, Leon C, Soler M, Rius J. Incidence of pneumothorax and pneumomediastinum in patients with aspiration pneumonia requiring ventilatory support. Chest 1977; 72:141-44 5 Petersen GW, Baier H. Incidence of pulmonary barotrauma in a medical ICU . Crit Care Med 1983; 11:67-9 6 Haake R, Schlichtig R, Ulstad DR, Henschen RR. Barotrauma: pathophysiology, risk factors, and prevention. Chest 1987; 91:608-13 7 Gammon RB, Shin MS, Buchalter SE. Pulmonary barotrauma in mechanical ventilation. Chest 1992; 102:568-72 8 Scoggin CH, Sahn SA, Petty TL. Status asthmaticus: anine year experience. JAMA 1977; 238:1158-62 9 Pinsky MR. The effects of mechanical ventilation on the cardiovascular system. Crit Care Clin 1990; 6:663-78 10 Bitto T, Mannion JD, Stephenson LW, Hammond R, Lanken PN , Miller W, et al. Pneumothorax during positive-pressure mechanical ventilation. JThorac Cardiovasc Surg 1985; 89:58591 11 Kurlander GJ, Helmen CH. Subpulmonary pneumothorax. AJR 1966; 96:1019-21 12 Rhea JT, vanSonnenberg E, McLoud TC. Basilar pneumothorax in the supine adult. Radiology 1979; 133:593-95 13 Gobien RP, Reines HD, Schabel Sl. Localized tension pneumothorax: unrecognized fonn of barotrauma in adult respiratory distress syndrome. Radiology 1982; 142:15-19 14 Tocino IM, Miller MH, Fairfax WR. Distribution of pneumothorax in the supine and semirecumbent critically ill adult. AJR 1985; 144:901-05 15 Chiles C, Ravin CE. Radiographic recognition of pneumothorax in the intensive care unit. Crit Care Med 1986; 14:677-80 16 Gordon R. The deep sulcus sign. Radiology 1980; 136:25-7 17 Wall SD, Federle MP, Jeffrey RB, Brett CM. CT diagnosis of unsuspected pneumothorax after blunt abdominal trauma. AJR 1983; 141:919-21 18 Tocino IM, Miller MH, Frederick PR, Bahr AL, Thomas F. CT detection of occult pneumothorax in head trauma. AJR 1984; 143:987-90 19 Mirvis SE , Tobin KD, Kostrubiak 1, Belzberg H. Thoracic CT in detecting occult disease in critically ill patients. AJR 1987; 148:685-89 20 Carr JJ, Reed JC, Choplin RH, Pope TL, Case LD. Plain and computed radiography for detecting experimentally induced pneumothorax in cadavers: implications for detection in patients. Barotrauma (Theodcxe W. Maroy)

Radiology 1992; 183:193-99 21 Johnson TH, Altman AR. Pulmonary interstitial gas: first sign of barotrauma due to PEEP therapy. Crit Care Med 1979; 7:53235 22 Albelda SM, Gefter WB, Kelley MA, Epstein DM , Miller WT. Ventilator-induced subpleural air cysts: clinical, radiologic, and pathologic significance. Am Rev Respir Dis 1983; 127:360-65 23 Woodring JH . Pulmonary interstitial emphysema in the adult respiratory distress syndrome. Crit Care Med 1985; 13:786-91 24 Maunder RJ, Pierson DJ, Hudson LD. Subcutaneous and mediastinal emphysema: pathophysiology, diagnosis, and management. Arch Intern Med 1984; 144:1447-53 25 Marini JJ, Culver BH . Systemic gas embolism complicating mechanical ventilation in the adult respiratory distress syndrome . Ann Intern Med 1989; 110:699-703 26 Gregory GA, Tooley WH. Gas embolization in hyaline membrane disease. N Eng! J Med 1970; 282:1141-42 27 Leitch DR, Green RD. Pulmonary barotrauma in divers and the treatment of cerebral arterial embolism. Aviat Space Environ Med 1986; 57: 931-38 28 Murphy BP, Harford FJ, Cramer FS. Cerebral air embolism resulting from invasive medical procedures. Ann Surg 1985; 201:242-45 29 Teplitz C . The core pathobiology and integrated medical science of adult acute respiratory insuliiciency. Surg Clin North Am 1976; 56:1091-1113 30 Barsch J, Birbara C, Eggers GWN , Krumlofsky F, Sanit VW, Smith W, et al. Positive pressure as a cause of respirator-induced lung disease [abstract]. Ann Intern Med 1970; 72:810 31 Webb HH, Tierney DF. Experimental pulmonary edema due to intennittent positive pressure ventilation with high inflation pressures: protection by positive end-expiratory pressure. Am Rev Respir Dis 1974; 110:556-65 32 Bone RC, Francis PB, Pierce AK. Pulmonary barotrauma complicating positive end-expiratory pressure [abstract]. Am Rev Respir Dis 1975; 111:921 33 Dreyfuss D, Basset G, Soler P, Saumon G . Intennittent positivepressure hyperventilation with high inflation pressures produces microvascular injury in rats. Am Rev Respir Dis 1985; 132:88084 34 Dreyfuss D, Soler P, Basset G, Saumon G. High inflation pressure pulmonary edema: respective effects of high airway pressure, high tidal volume, and positive end-expiratory pressure. Am Rev Respir Dis 1988; 137:1159-64 35 Kolobow T, Moretti MP, Fumagalli R, Mascheroni D, Prato P, Chen V. et al. Severe impairment in lung function induced by high peak airway pressure during mechanical ventilation: an experimental study. Am Rev Respir Dis 1987; 135:312-15 36 Borelli M, KolobowT, Spatola R, Prato P, Tsuno K. Severe acute respiratory failure managed with continuous positive airway pressure and partial extracorporeal carbon dioxide removal by an artificial membrane lung. Am Rev Respir Dis 1988; 138:148087 37 Parker JC, Hernandez LA, Longenecker GL, Peevy K, Johnson W. Lung edema caused by high peak inspiratory pressures in dogs: role of increased microvascular filtration pressure and penneability. Am Rev Respir Dis 1990; 142:321-28 38 Tsuno K, Prato P, Kolobow T. Acute lung injury from mechanical ventilation at moderately high airway pressures. J Appl Physiol 1990; 69:91XH>1 39 Tsuno K, Miura K, Takeya M, Kolobow T, Morioka T. Histopathologic pulmonary changes from mechanical ventilation at high peak airway pressures. Am Rev Respir Dis 1991; 143:111520 40 Schaefer KE, McNulty WP, Carey C , Liebow AA. Mechanisms in development of interstitial emphysema and air embolism in decompression from depth. J Appl Physiol1958; 13:15-29

41 Pierson DJ. Alveolar rupture during mechanical ventilation: role of PEEP, peak airway pressure, and d istending volume. Respir Care 1988; 33:472-86 42 Dreyfuss D, Saumon G . Barotrauma is volutrauma, but which volume is the one responsible? Intensive Care Med 1992; 18:139-41 43 Corbridge TC, Wood LDH, Crawford GP, Chudoba MJ, Yanos J, Sznajder JI. Adverse effects of large tidal volume and low PEEP in canine acid aspiration. Am Rev Respir Dis 1990; 142:311-15 44 Gattinoni L , Pesenti A, Bombino M, Baglioni S, Rivolta M, Rossi F. Relationships between lung computed tomographic density, gas exchange, and PEEP in acute respiratory failure . Anesthesiology 1988; 69:824-32 45 Maunder RJ, Shuman WP, McHugh Jw, Marglin Sl , Butler J. Preservation of nonnal lung regions in the adult respiratory distress syndrome. JAMA 1986; 255:2463-65 46 Dantzker DR. Gas exchange in the adult respiratory distress syndrome. Clin Chest Med 1982; 3:57-67 47 Benito S, Lemaire F. Pulmonary pressure-volume relationship in acute respiratory distress syndrome in adults: role of positive and expiratory pressure. J Crit Care 1990; 5:27-34 48 Gattinoni L, Pesenti A, Avalli L, Rossi F, Bombino M. Pressurevolume curve of total respiratory system in acute respiratory failure: computed tomographic scan study. Am Rev Respir Dis 1987; 136:730-36 49 Marini JJ, Kelsen SG. Re-targeting ventilatory objectives in adult respiratory distress syndrome. Am Rev Respir Dis 1992; 146:2-3 50 Pesenti A. Target blood gases during ARDS ventilatory management. Intensive Care Med 1990; 16:349-51 51 Hickling KG . Low volume ventilation with pennissive hypercapnia in the adult respiratory distress syndrome. Clin Intensive Care 1992; 3:67-78 52 Hickling DG, Henderson SJ, Jackson R. Low mortality associated with low volume, pressure limited ventilation with pennissive hypercapnia in severe adult respiratory distress syndrome. Intensive Care Med 1990; 16:372-77 53 Stinson JM, Mattsson JL. Tolerance of rhesus monkeys to graded increase in environmental C01 : serial changes in heart rate and cardiac rhythm . Aerospace Med 1970; 41:415-18 54 Ellison RG, Ellison LT, Hamilton WF. Analysis of respiratory acidosis during anesthesia. Ann Surg 1955; 141:375-82 55 Darioli R, Perret C . Mechanical controlled hypoventilation in status asthmaticus. Am Rev Respir Dis 1984; 129:385-87 56 Murray JF, Matthay MA, Luce JM, Flick MR. An expanded definition of the adult respiratory distress syndrome. Am Rev Respir Dis 1988; 138:720-23 57 Smith PEM, Gordon IJ. An index to predict outcome in adult respiratory distress syndrome. Intensive Care Med 1986; 12:869 58 East TD, Bohm SH, Wallace CJ, Clemmer TP, Weaver LK, Onne JF, et al. A successful computerized protocol for clinical management of pressure control inverse ratio ventilation in ARDS patients. Chest 1992; 101:697-710 59 Morris AH , Wallace CJ, Clemmer TP, Onne JF, Weaver LK, Thomas F, et al. Final report: computerized protocol controlled clinical trial of new therapy which includes EC co. R for ARDS [abstract] . Am Rev Respir Dis 1992; 145:184 60 Abraham E, Yosihara G. Cardiorespiratory effects of pressure controlled ventilation in severe respiratory failure. Chest 1990; 98:1445-49 61 Lachman B. Open up the lung and keep the lung open. Intensive Care Med 1992; 18:319-21 62 Marini JJ, Crooke PS, Truwit JD . Detenninants and limits of pressure preset ventilation: a mathematical model of pressure control. J Appl Physiol1989; 67:1081-92 CHEST I 104 I 2 I AUGUST. 1993

583

63 Nahum A, Burke WC, Ravenscraft SA, Marcy Tw, Adams AB, Crooke PS, et al. Lung mechanics and gas exchange during pressure-rontrol ventilation in dogs: augmentation of C01 elimination by an Intratracheal catheter. Am Rev Respir Dis 1992; 146:965-73 64 Marini JJ, Ravenscraft SA. Mean airway pressure: physiologic determinants and clinical im~rtance-Part 1: physiologic determinants and measurements. Crit Care Med 1992; l1A146172 65 Boros SJ. Variations in lnspiratory:expiratory ratio and airway pressure wave rorm during mechanical ventilation: the sigilificance of mean airway pressure. J Pediatr 1979; 94:114-17 66 Ciszek TA, Modanlou HD, Owings D , Nelson P. Mean airway pressure: significance during mechanical ventilation in neonates. J Pediatr 1981; 99:121-26 67 Marcy Tw, Marini JJ. Inverse ratio ventilation: rationale and implementation. Chest 1991; 100:494-504 68 Lachman B, Danzmann E, Haendley B, Jonson B. Ventilator settings and gas exchange In respiratory distress syndrome. In: Pralcash 0 , ed. Applied physiology In clinical respiratory care.

The Hague: Martinus NijholfPublishers, 1_982; 141-76 69 AI-Saady N, Bennett ED. Decelerating Inspiratory Bow waveform improves lung mechanics and gas exchange in patients on Intermittent positive-pressure ventilation. Intensive Care Med 1985; 11:68-75 70 Knelson JH, Howatt WF, DeMuth GR. Effect of respiratory pattern on alveolar gas exchange. J Appl Physiol 1970; 29:32831 71 Marini JJ. Should PEEP be used In airftow obstruction? Am Rev Respir Dis 1989; 140:1-3 72 Abraham E, Yoshihara G. Cardiorespiratory effects of pressure controlled Inverse ratio ventilation In severe respiratory failure. Chest 1989; 96:1356-59 73 Cole AGH, \Yeller SF, Sykes MK. Inverse ratio ventilation compared with PEEP In adult respiratory failure. Intensive Care Med 1984; 10:227-32 74 Gurevitch MJ, Van Dyke J, Young ES, Jackson K. Improved oxygenation ·and lower peak airway pressure In severe adult respiratory distress syndrome: treatment with inverse ratio ventilation. Chest 1986; 89:211-13