Beneficial effects of chest tube drainage of pleural effusion in acute respiratory failure refractory to positive end-expiratory pressure ventilation

Beneficial effects of chest tube drainage of pleural effusion in acute respiratory failure refractory to positive endexpiratory pressure ventilation Mia Talmor, MD, Lynn Hydo, BS, Jeffrey G. Gershenwald, MD,a and Philip S. Barie, MD, FCCM, FACS, New York, N.Y.

Background. As part of an ongoing prospective evaluation of the response of acute respiratory failure (ARF) to ventilation with titrated amounts of positive end-expiratory pressure (PEEP), a subset of patients with a poor response to the initial application of PEEP and radiographic evidence of pleural effusion was identified. The effusion(s) were treated by tube thoracostomy (TT) to test the hypothesis that drainage would have a favorable effect on oxygenation and compliance in critically ill patients with substantial pulmonary dysfunction. Methods. Consecutive patients with ARF underwent a titrated progressive application of PEEP if arterial oxygen saturation was less than 90% on fraction of inspired oxygen less than 0.5. One or two thoracostomy tubes (TT) were placed afterward in patients with radiologic evidence of effusion who had a poor response to PEEP therapy. The lung injury score (LIS), PaO2:FiO2 (P:F), peak airway pressure, dynamic compliance, and TT output were recorded. Changes over time were analyzed by one-way analysis of variance with repeated measures. Results. Nineteen of 199 patients needed TT. LIS was 3.0 ± 0.1. Maximum PEEP was 16.6 ± 1.0 cm H2O. TT drainage was 863 ± 164 ml in the first 8 hours. Mortality was 63% (12 of 19) but only 41% (74 of 180) in the patients who did not require TT (p = 0.11). TT improved oxygenation and compliance immediately after insertion in 17 of 19 patients, and P:F remained statistically higher (245 ± 29 versus 151 ± 13, p < 0.01) 24 hours after TT drainage. There was no correlation between the volume of fluid removed and P:F either immediately (R2, 0.16) or 24 hours after TT (R2, 0.07). Conclusions. Drainage of pleural fluid resulted in a significant improvement in oxygenation in ARF patients with pleural effusions who were refractory to treatment with mechanical ventilation and PEEP. TT represents a simple and safe alternative for aggressive management of selected patients, obviating the inherent risk of pneumothorax with thoracentesis and possibly avoiding the need for more complex forms of support in this critically ill patient population. (Surgery 1998;123:137-43.) From the Department of Surgery, The New York Hospital-Cornell Medical Center, New York, N.Y.

VENTILATORY MANAGEMENT OF SEVERE acute respiratory failure (ARF) requires support of oxygen delivery while minimizing both the risk of barotrauma from high airway pressures and pulmonary fibrosis from oxygen toxicity, goals that may be mutually exclusive. The mainstay of support for patients with ARF has been mechanical ventilation with positive end-expiratory pressure (PEEP).1-3 Accepted for publication May 20, 1997. Reprint requests: Philip S. Barie, MD, Anne & Max A. Cohen Surgical Intensive Care Unit, The New York Hospital-Cornell Medical Center, 525 East 68 St., L-384, New York, NY 10021. aCurrent address: Department of Surgery, University of Texas M.D. Anderson Cancer Center, Houston, TX. Copyright © 1998 by Mosby, Inc. 0039-6060/98/$5.00 + 0 11/56/85945

Unfortunately, PEEP ventilation is not a panacea for patients with ARF. Prophylactic administration of PEEP does not prevent adult respiratory distress syndrome (ARDS),4,5 and PEEP itself has not unequivocally decreased mortality since its introduction despite extensive evaluations of how best to use it.3 Some patients fail entirely to respond to PEEP therapy, despite application at high levels (greater than 15 cm H2O). Alternatives for therapy are limited, portending the high mortality reported in this subgroup of patients with ARF.6 Pressure-controlled inverse-ratio ventilation with or without permissive hypercapnia,7-9 high frequency ventilation,10 and even extracorporeal gas exchange11 have been attempted, with results that have been equivocal. SURGERY 137

138 Talmor et al.

Surgery February 1998

Table 1. ARF patients with pleural effusion(s): descriptive statistics (n = 19) Characteristic Age (yr) APACHE II score APACHE III score MOD score LIS Maximum PEEP (cm H2O) History of pulmonary disease Smoking history Cause of ARF Sepsis Shock/massive transfusion Multifactorial Pulmonary contusion Congestive heart failure Mortality

Mean ± standard error 68 ± 4 21 ± 2 82 ± 7 13 ± 1 3.0 ± 0.1 16.6 ± 1.0 6 6 9 4 4 1 1 12

PEEP enhances oxygenation in patients with ARF by alleviating the ventilation/perfusion (VA/Q) mismatch that is characteristic of the disease.12 Restoration of normal VA/Q is accomplished principally by increased lung volume as measured by the functional residual capacity (FRC). PEEP may increase the FRC by direct increases in alveolar volume13 when PEEP up to 10 cm H2O is applied to normal alveoli and may also recruit reexpanded alveoli that collapsed previously (e.g., atelectasis) or filled partially with edema fluid.14 This mechanism may account for the sometimes dramatic improvement in lung compliance seen with PEEP therapy, although PEEP levels greater than 10 cm H2O are probably required to accomplish it.3 The advantages of PEEP—adequate oxygenation at lower (and therefore nontoxic or less toxic) inspired oxygen concentrations, reexpansion of collapsed alveoli, and improved lung compliance— decrease the work of breathing and appear in some studies to promote pulmonary parenchymal repair.15 However, physiologically important decreases of cardiac output should be avoided because there is no clinical advantage to therapy if overall oxygen delivery is reduced. The therapeutic end point should be achievement of adequate oxygenation—PaO2 greater than 60 mm Hg (90% oxyhemoglobin saturation assuming a normal saturation curve) on a fraction of inspired oxygen (FiO2) of 0.5 or less—without a significant reduction in cardiac output. Most large series suggest that 95% of patients can be supported with less than 20 cm H2O PEEP.9 As part of an ongoing prospective evaluation of

the response of ARF to ventilation with titrated amounts of PEEP, a subset of patients with a poor response to the initial application of PEEP and radiographic evidence of pleural effusion were identified. The effusion(s) were treated by tube thoracostomy (TT) to test the hypothesis that drainage would have a favorable effect on oxygenation and compliance in critically ill patients with substantial pulmonary dysfunction. PATIENTS AND METHODS This study was conducted in the general surgical intensive care unit (SICU) of an urban, tertiary care medical center that provides surgical care to its local community and to tertiary care referrals and also functions as a level I trauma center. Although the institution serves in addition as a regional center for burn care and cardiovascular surgery, those patients are cared for in separate dedicated units and are thus excluded from this analysis. All care was indicated clinically and thus exempted from Institutional Review Board approval. We reviewed prospectively the data from 3197 admissions to the SICU between August 1990 and October 1996. Of these, 199 patients with ARF who required mechanical ventilation with PEEP were identified. Demographic data, including pertinent medical history, pulmonary history, and risk factors for the development of ARF, were collected for each patient. Acute physiology and chronic health evaluation (APACHE) II and APACHE III scores were measured after 24 hours of ICU care,16 and the lung injury score (LIS)17 was calculated immediately after a “PEEP trial”. For the PEEP trial the patients underwent a titrated progressive application of PEEP if arterial oxygen saturation was less than 90% on FiO2 less than 0.5 with PEEP 5 cm H2O. FiO2 was increased to 1.0 to maintain a safe PaO2 during application, and all patients were sedated with a stable continuous infusion of a benzodiazepine (lorazepam or midazolam) or propofol. Neuromuscular blockade was not used except in rare cases of patient-ventilator dyssynchrony. Hemodynamic stability was assured with the infusion of crystalloid fluids or inotropic agents (e.g., dobutamine or dopamine). PEEP was then increased by 2.5 cm H2O increments to 20 cm H2O or until cardiac output decreased by 20% (thermal dilution, performed in all patients). FiO2 was then decreased to 0.5 only if PaO2:FiO2 (P:F) increased to 200 or greater after PEEP. If FiO2 could not be decreased and there was plain radiographic evidence of pleural effusion,

Surgery Volume 123, Number 2

Talmor et al. 139

A

B

Fig. 1. A, Radiographic evidence of right pleural effusion was present in this patient. At the time, the patient was on FiO2 1.0, PEEP 17.5 cm H2O with P:F 104 and LIS of 3.0, consistent with acute respiratory distress syndrome. Drainage of 800 ml through TT (B) increased P:F to 188 and improved compliance from 29 to 41 ml/cm H2O. Despite improvement, the patient died later of multiple organ dysfunction syndrome.

one or two TTs were placed at the bedside in a standard manner by using local infiltration of 1% lidocaine to supplement the stable sedative infusion. No TT was placed prophylactically; TTs placed for barotrauma were excluded, as were those procedures where a “rush” of air during insertion identified a previously unrecognized pneumothorax. Data collected included fluid output on initial placement, daily total output, and cumulative total output. Chest radiographs were obtained daily. Peak airway pressure (PAW), FiO2, PEEP, P:F, and dynamic compliance (Cdyn) were recorded daily or more often if changes were noted. The multiple organ dysfunction score of Marshall et al.18 was cal-

culated daily and cumulatively. Mortality was recorded for each patient. Changes over time were analyzed by one-way analysis of variance (ANOVA) with repeated measures, with Tukey’s post hoc test. Relationships between coordinate variables at discrete time points were assessed by linear regression analysis. Differences in mortality were determined by Fisher’s exact test. Statistical analysis was performed by microcomputer (Macintosh Performa 6115CD; Apple Computer Inc., Cupertino, Calif.), with commercial software (STATVIEW 4.0.1 with Super ANOVA 1.1.1; Abacus Concepts Inc., Berkeley, Calif.). Statistical significance was deter-

140 Talmor et al.

Surgery February 1998

Fig. 2. P:F increased immediately after TT in 17 of 19 patients and was unchanged in one additional patient. The effect persisted 24 hours after TT in all 17 responders. The patient who initially had no response had a higher P:F after 24 hours.

mined at an alpha of less than 0.05. Data are expressed as mean values ± standard error of the mean. RESULTS Of 199 patients with ARF who underwent a titrated application of PEEP, 19 (9.5%) were identified who had a poor response to the initial application of PEEP and who had evidence of pulmonary effusion on portable plain chest radiographs. Six patients had a history of cigarette smoking immediately before the current hospital admission. Six patients had a known history of primary pulmonary disease, which was not associated with tobacco use in three cases. The average LIS of these patients was 3.0 ± 0.1. The average maximum PEEP applied was 16.6 ± 1.0 cm H2O. Sixty-eight percent (13 of 19) of patients had multiple organ dysfunction (MOD). Demographic data are summarized in Table I. Fourteen patients had a unilateral effusion and therefore had only one TT placed, whereas five patients had bilateral effusions, necessitating two TTs. Mortality was 63% (12 of 19) but only 41% (74 of 180) (p = 0.11) in the patients who did not require TT. TT drainage on average was 863 ± 164 ml during the first 8 hours. A representative radiographic effusion and response to TT are shown in Fig. 1. Drainage of pleural effusion improved oxygenation (Fig. 2) and compliance (data not shown) immediately after insertion in 17 of 19 patients,

with no change in one of the two nonresponders. Although the PaO2 increased significantly (p < 0.01) right after TT placement, it was not significantly different from the post-PEEP PaO2 after 24 hours (Table II). The PaO2 value does not reflect the decrease in FiO2 that was permissible after TT insertion in all patients. The P:F value was statistically higher (p < 0.01) both immediately after TT drainage and 24 hours after placement, indicating improved oxygenation. An effect was observed after drainage of as little as 80 ml fluid in one patient. However, there was no correlation between the volume of fluid removed and P:F either immediately (R2, 0.16) or 24 hours after TT (R2, 0.07). In contrast, Cdyn improved only transiently after TT placement (Table II). There was no correlation between the volume of fluid removed and Cdyn either immediately (R2, 0.23) or 24 hours after TT (R2, 0.10). DISCUSSION The incidence of pleural effusion and its impact on patients with ARF are somewhat controversial. In the past, pleural effusion in ARF was considered an infrequent occurrence,19,20 and its absence has even been considered necessary for a radiographic diagnosis of the disorder. More recently, with the improved availability and use of computed tomographic (CT) scanning, smaller effusions are recognized, and the reported incidence has increased. In evaluating 74 patients with ARDS who had both

Talmor et al. 141

Surgery Volume 123, Number 2 Table II. Changes in oxygenation, airway pressure, and compliance after TT PaO2 P:F PAW Cdyn

Pre-PEEP

Post-PEEP

TT

24 Hours after TT

p Value

75.8 ± 4.3 121.9 ± 9.0 40.8 ± 3.6 23.4 ± 2.6

124.7 ± 12.2 151.0 ± 15.3 44.3 ± 3.2 27.1 ± 3.5

199.1 ± 26.8* 253.6 ± 26.5* 42.9 ± 4.3 35.7 ± 4.7

132.3 ± 14.0 244.5 ± 29.1* 38.2 ± 3.4 32.9 ± 3.5

<0.0001 <0.0001 0.6865 0.0818

*Compared with pre-PEEP and post-PEEP values.

routine chest radiographs and chest CT scans, Tagliabue et al.21 found that pleural effusion was a frequent finding (50%) on chest CT scan. Whereas pleural effusions were noted on CT scan in 37 cases, only 15 of these cases were identified by plain chest radiography. In seven of their cases the plain radiographic findings were unilateral, but the CT scan findings were bilateral. The cause of pleural effusion in patients with ARF is unknown, but it is likely multifactorial. Nine patients (47%) in the TT group had a history of tobacco use or primary pulmonary disease before their admission to the SICU. A plurality of patients (also 47%, but not the same subset of patients) had sepsis as their major risk factor for the development of ARF. The impact of pleural effusion on the overall course of the patient with ARF is unknown. Some studies have suggested that pleural effusion is a sign of an inflammatory or embolic complication and as such is an ominous prognostic sign.22 In contrast, more recent studies have suggested that the finding of pleural effusion (on CT scan) does not worsen prognosis.21 In our study the mortality among patients with pleural effusion(s) and a poor response to PEEP was 63%, as compared with 41% in ARF patients without an effusion. Although there was no statistical difference in mortality, that may have been an artifact of the discrepancy in sample sizes. It is difficult to draw conclusions as to the impact of the effusion on mortality, because we only identified patients who were poor responders to therapy and who may therefore have been more likely to have an adverse outcome. It is known that pleural effusion is associated with a restrictive ventilatory impairment and hypoxemia,23 which is usually manifest as dyspnea in the conscious, nonintubated patient. In one study in which patients with inversion of a hemidiaphragm caused by pleural effusion underwent thoracentesis,24 there was a small but significant increase in the forced expiratory volume in 1 second and forced vital capacity within 24 hours after removal of 600 to 2700 ml fluid. In addition, the alveolar-arterial oxygen gradient (D[A-a] O2) and PaO2 showed a significant improvement.24 When fluid is aspirated from the thoracic cavity, the vol-

ume removed may be associated with a decrease in size of the thoracic cavity25 or alternately in an increase of pulmonary volume.26 Perpina et al.27 performed thoracentesis on 33 patients and found an increased PaO2, decreased D[A-a]O2, and a decrease in both anatomic and physiologic shunt.27 Thoracentesis may improve oxygenation by improving VA/Q inequality, particularly by increasing ventilation to low VA/Q areas. This may explain why TT drainage improves oxygenation markedly in patients with high degrees of physiologic shunt, as manifested by their failure to improve with even high levels of PEEP. The population of patients ventilated with PEEP may gain greater benefit from pleural drainage than the general population of patients with ARF. Light et al.26 argued that improvements in spirometric pulmonary function (lung volumes) after thoracentesis may be related to the absolute magnitude of pleural pressure or to changes in pressure in response to intervention. In their study the mean vital capacity improved 410 ml in response to drainage of a mean volume of 1740 ml pleural fluid by thoracentesis. The improvement in vital capacity correlated best with the volume of fluid removed in patients who had at least 800 ml removed, but patients with higher pleural pressures (or smaller decreases in pleural pressure) after thoracentesis of at least 800 ml had greater improvement in pulmonary function. This is directly analogous to patients ventilated with PEEP, in whom airway and transpleural pressures are always positive. The association of barotrauma with PEEP is well known.28 The reported incidence is as high as 50%,29 but only 5% of the 199 patients titrated with PEEP overall in the present series had a pneumothorax at any time during their course. Pneumothorax may be difficult to detect on a plain film, given low lung compliance and a low rate of overt lung collapse.30 In the supine position, lateral and apical pneumothoraces are particularly hard to detect, given the propensity of air to collect in the least dependent (anteromedial and infrapulmonary) regions of the pleural cavity.31,32 In one study that compared CT scans with plain films in patients with ARDS, 24 pneumothoraces were

142 Talmor et al.

detected on CT scan, whereas only 15 were detected on plain film.21 Although no patients who had a TT placed for the purpose of relieving a pneumothorax were included in this study, there is a small possibility that a few patients had small pneumothoraces that were missed on plain films. If so, insertion of a TT might have improved gas exchange or hemodynamic abnormalities on that basis, but TT insertions where a rush of air was noted during insertion were also excluded from this analysis. The possibility of occult pneumothorax is remote in the present study because TT, when indicated, was inserted immediately after the PEEP trial, leaving virtually no time for one to develop. In draining pleural fluid from patients with ARF refractory to PEEP, TT has several distinct advantages over thoracentesis. Thoracentesis is cumbersome because of difficulty with patient positioning and is associated with a significant risk of pneumothorax, especially in mechanically ventilated patients with high airway pressures. In contrast, TT is the treatment of choice for evacuating a pneumothorax. An indwelling thoracostomy tube allows for continuous drainage of pleural fluid, which may reaccumulate after thoracentesis. The mortality of patients who have ARF and pneumothorax is significantly higher compared with patients without pneumothorax,29 and all pneumothoraces detected under the circumstances of positive-pressure ventilation with PEEP must be evacuated, no matter how small.33 Pleural effusion, which is associated with a restrictive ventilatory impairment, contributed to pulmonary dysfunction when present in severe ARF. Drainage of even a small amount of pleural fluid resulted in a significant improvement in oxygenation in 89% of patients who had an effusion and were refractory to treatment with mechanical ventilation and PEEP. Whether the improvement was due primarily to restoration of VA/Q matching or to improvements in vital capacity and FRC, TT represents a simple and safe alternative for aggressive management that obviates the inherent risk of pneumothorax from thoracentesis in this critically ill patient population. If patients with refractory ARF and pleural effusion can be stabilized by TT, additional maneuvers that often require neuromuscular blockade (which itself can reduce FRC),34 such as pressure-controlled inverse-ratio ventilation with permissive hypercapnia10 or prone patient positioning,35 may be unneccessary. For the patient who has a poor response to PEEP therapy but who does not have radiographic evidence of effusion, the dilemma is whether to trans-

Surgery February 1998

port the patient for a CT scan or to place one or (likely two) TT empirically. For patients who are very hypoxemic on a high level of support, transport to CT scan is potentially hazardous because levels of PEEP greater than 15 cm H2O cannot be maintained by adult manual resuscitator bag. However, our safety record in transporting patients on PEEP greater than 5 cm H2O has been excellent,36 because such patients are accompanied by a physician and respiratory therapist and remain on their ICU ventilator for the duration of the transport. In some centers it may be safer and simpler to just insert the chest tubes empirically, but the data do not support the practice.21 Faced with this dilemma, we would prefer a CT scan to empiric TT. REFERENCES 1. Weigelt JA, Norcross JF, Borman KR, et al. Early steroid therapy for respiratory failure. Arch Surg 1985;119:536-40. 2. Stevens PM. General assessment and support of the adult respiratory distress syndrome. Ann N Y Acad Sci 1982;384:477-88. 3. Shapiro BA, Cand RD, Harrison RA. Positive end-expiratory pressure therapy in adults with special reference to acute lung injury: a review of the literature and suggested clinical correlations. Crit Care Med 1984;12:127-41. 4. Pepe PE, Hudson LD, Carrico CJ. Early application of positive end-expiratory pressure in patients at risk for the adult respiratory distress syndrome. N Engl J Med 1984;311:2816. 5. Springer RR, Stevens PM. The influence of positive endexpiratory pressure on survival of patients in respiratory failure. Am J Med 1979;66:196-200. 6. Fowler AA, Hamman RF, Good JT. Adult respiratory distress syndrome: risk with common predispositions. Ann Intern Med 1983;98:593-7. 7. MacIntyre NR. New forms of mechanical ventilation in the adult. Clin Chest Med 1988;9:47-54. 8. Morris AH, Wallace CJ, Menlove RL. Randomized clinical trial of pressure-controlled inverse ratio ventilation and extracorporeal CO2 removal for adult respiratory distress syndrome. Am J Respir Crit Care Med 1994;149:295-305. 9. Barie PS. Organ-specific support in multiple organ failure: pulmonary support. World J Surg 1995;19:581-91. 10. Hickling KG, Henderson HJ, Jackson R. Low mortality associated with low volume pressure limited ventilation with permissive hypercapnia in severe adult respiratory distress syndrome. Intensive Care Med 1990;16:372-7. 11. Egan MT, Duffin J, Glynn MF, et al. Ten year experience with extracorporeal membrane oxygenation for severe respiratory failure. Chest 1988;94:681-7. 12. Ralph DD, Robertson HT, Weaver LJ, et al. Distribution of ventilation and perfusion during positive end-expiratory pressure in the adult respiratory distress syndrome. Am Rev Respir Dis 1985;131:54-60. 13. Daly BDT, Edmonds CH, Norman JC. In vivo alveolar morphometrics with positive end-expiratory pressure. Surg Forum 1973;24:217-9. 14. Tyler DC, Cheney FW. Comparison of positive end-expiratory pressures and inspiratory positive plateau in ventilation of rabbits with experimental pulmonary edema. Anesth Analg 1979;58:288-92.

Talmor et al. 143

Surgery Volume 123, Number 2 15. Block EB. Recovery from hyperoxic depression of pulmonary 5-hydroxytryptamine clearance: effect of inspired PO2 tension. Lung 1978;155:131-8. 16. Barie PS, Hydo L, Fischer E. Comparison of APACHE II and III scoring systems for mortality prediction in critical surgical illness. Arch Surg 1995;130:77-82. 17. Murray JF, Matthay MA, Luce JM, et al. An expanded definition of the adult respiratory distress syndrome. Am Rev Respir Dis 1988;138:720-3. 18. Marshall JC, Sibbald WJ, Cook DA, et al. The multiple organ dysfunction (MOD) score: a reliable indicator of a complex clinical outcome. Crit Care Med 1995;23:1638-52. 19. Milne ENC, Pistolesi M, Miniati M, et al. The radiologic distinction of cardiogenic and noncardiogenic edema. AJR 1985;144:879-94. 20. Swensen SJ, Peters SG, LeRoy AJ, et al. Radiology in the intensive-care unit. Mayo Clin Proc 1991;66:396-410. 21. Tagliabue M, Casella TC, Zincone GE, et al. CT and chest radiography in the evaluation of adult respiratory distress syndrome. Acta Radiol 1994;35:230-4. 22. Greene R. Adult respiratory distress syndrome: acute alveolar damage. Radiology 1987;163:57-66. 23. Mulvey RB. The effect of pleural fluid on the diaphragm. Radiology 1965;84:1080-6. 24. Wang JS, Tseng CH. Changes in pulmonary mechanics and gas exchange after thoracentesis on patients with inversion of a hemidiaphragm secondary to large pleural effusion. Chest 1995;107:1610-4. 25. Estenne M, Yernault JC, De Troyer A. Mechanism of relief of dyspnea after thoracentesis in patients with large pleural effusions. Am J Med 1983;74:813-9. 26. Light RW, Stansbury DW, Brown SE. The relationship between pleural pressures and changes in pulmonary func-

27. 28.

29.

30.

31.

32. 33. 34.

35.

36.

tion after therapeutic thoracentesis. Am Rev Respir Dis 1986;133:658-61. Perpina M, Benlloch E, Marco V, et al. Effect of thoracentesis on pulmonary gas exchange. Thorax 1983;38:747-50. De La Torre FJ, Tomasa A, Klamburg J, et al. Incidence of pneumothorax and pneumomediastinum in patients with aspiration pneumonia requiring ventilatory support. Chest 1977;72:141-4. Gattinoni L, Bombino M, Pelosi P, et al. Lung structure and function in different stages of severe adult respiratory distress syndrome. JAMA 1994;271:1772-9. Gobien RP, Reines HD, Schabel SI. Localized tension pneumothorax: unrecognized form of barotrauma in adult respiratory distress syndrome. Radiology 1982;142:15-9. Chiles C, Ravin CE. Radiographic recognition of pneumothorax in the intensive care unit. Crit Care Med 1986;14:677-80. Petersen GW, Baier H. Incidence of pulmonary barotrauma in a medical ICU. Crit Care Med 1983;11:67-9. Marcy TW. Barotrauma: detection, recognition, and management. Chest 1993;104:578-84. Hedenstierua G, Stranberg A, Brismar B, et al. Functional residual capacity, thoracoabdominal dimensions and central blood volume during general anesthesia with muscle paralysis and mechanical ventilation. Anesthesiology 1985;62:247-54. Gattinoni L, Pelos P, Vitale G, et al. Body position changes redistribute lung computed tomographic density in patients with respiratory failure. Anesthesiology 1991;74:15-23. Szem JW, Hydo LJ, Fischer E, et al. High risk intrahospital transport of critically ill patients: safety and outcome of the necessary “road trip”. Crit Care Med 1995;23:1660-6.

ANNOUNCEMENT Effective October 1, 1997, please send all manuscripts and other submissions for Surgery to the following address: Kim Fons Managing Editor Surgery 11830 Westline Industrial Drive St. Louis, MO 63146-3318