Resection of large bullae has been performed as an accepted method of emphysema surgery with good results [1,2]. In contrast, Brantigan et al. [3,4] introduced a concept of multiple wedge resections of diffuse emphysematous lung tissue to reduce lung volume, aiming at improving radial traction of the airways. However, a high perioperative mortality of 16% appeared to be prohibitive for this operation. This concept was resumed recently by Cooper et al. , encouraged by experiences in lung transplantation of patients with severe chronic obstructive pulmonary disease (COPD). They used a median sternotomy for multiple wedge resections of emphysematous lung tissue (20%-30%) in patients with severe hyperinflation due to diffuse pulmonary emphysema. They reported good functional improvement, and no case of worsening or even death. Similar results with a documented increase in elastic recoil of the lung were reported in patients after unilateral lung resection using thoracoscopy or bilateral resection using sternotomy . Currently, the number of investigations being published that focus predominantly on surgical or pneumological aspects of this new type of surgery is increasing [7-11].
Our group used video-assisted thoracoscopy as a surgical approach to bilateral volume reduction surgery (VRS). We also found good functional improvement and a low morbidity rate without any perioperative mortality , comparable to the results of Cooper's group after sternotomy. Thoracoscopic VRS in these severely impaired patients represents a particular anesthesiological challenge, because one-lung ventilation (OLV) is inherently required. This is usually associated with compromised pulmonary gas exchange due to a considerable ventilation-perfusion mismatch, especially if combined with lateral positioning of the patient. Furthermore, reexpansion pulmonary edema after prolonged OLV and/or postinjury pulmonary edema formation after laser ablation of bullous emphysema have been described [13,14]. Although the underlying pathophysiology in diffuse pulmonary emphysema is different, and multiple wedge resections represent a different surgical approach, reexpansion pulmonary edema must be considered after multiple wedge resections. Also, Barker et al.  described that prolonged postoperative ventilation was necessary after laser ablation of bullous emphysema. However, the immediate postoperative tracheal extubation is highly desirable after VRS to prevent both an increased risk for air leaks with consecutive bronchopleural fistulas and nosocomial bronchopulmonary infections in these patients. In an attempt to blunt postoperative morbidity we thus designed an anesthetic technique aiming at the earliest possible extubation.
Pulmonary gas exchange during OLV has been investigated in patients with bullous emphysema . However, there are no data on COPD patients with diffuse pulmonary emphysema and on the specific anesthesiological requirements for thoracoscopic VRS. We therefore describe the anesthesiological management in detail, reporting specifically our prospective data of continuously recorded intraarterial blood gases during VRS as well as the experience with immediate postoperative extubation in a group of severely impaired patients with advanced diffuse pulmonary emphysema.
Between August 1994 and October 1995, 24 consecutive patients (17 men, 7 women), aged 64 +/- 10 yr (mean +/- SD, minimum 41/maximum 78 yr), were selected to be enrolled in the VRS program according to the criteria previously described in detail . Diffuse pulmonary emphysema was caused by smoking in 24 patients, 2 of whom also had alpha1-antitrypsin deficiency. All patients received regular inhaled beta adrenergics and inhaled corticosteroids. Eight patients had received long-term oxygen therapy. Radiological signs of bilateral hyperinflation with no isolated bullae (more than 20% of the volume of either hemithorax), and dyspnea at rest or at minimal physical activity was present in all patients. Lung functional characteristics are described in Table 1. Relevant coronary artery disease was excluded and preserved ventricular function was documented by cardiac catheterization with coronary angiography in 23 patients. As an exception, one patient was accepted for VRS after implantation of a coronary stent.
Wedge resections of about 20%-30% of total lung volume were performed by video-assisted thoracoscopic surgery according to a method previously described in more detail . The most destroyed lung tissue on both sides-identified by preoperative computed tomography scans and perfusion scintigraphy-was resected using endoscopic staplers. Surgery was performed by the same surgeon (WW) in the lateral decubitus position-with an intraoperative change to the other side-in 18 patients and in the supine position in 6 patients.
No premedication was given, but regular inhaled antiobstructive treatment was continued including on the morning of operation. A thoracic epidural catheter was inserted in the awake patient using local anesthesia in the lateral decubitus position via the paramedian approach between T-5 and T-8. A continuous infusion of bupivacaine 0.5% 4-6 mL/h was given during surgery. General anesthesia was induced with propofol (1.5-2 mg/kg), fentanyl (0.1 mg), and vecuronium (0.1 mg/kg) and maintained using a continuous infusion of propofol (4-8 mg [centered dot] kg-1 [centered dot] h-1), small doses of ketamine (20-40 mg per bolus, maximal 200 mg), fentanyl (limited to a total of 0.4 mg), and vecuronium (1-2 mg per bolus). In two patients with soya allergy, thiopental and isoflurane were used. A left-sided double-lumen tube (Sheri-I-Bronch[R]; Sheridan Catheter Corp., Argyle, NY) was placed under fiberoptic control. Usual monitoring was used. A 20-gauge radial artery cannula was inserted prior to induction of general anesthesia for continuous intraarterial blood gas measurement using a previously calibrated Paratrend 7[R] sensor system (Biomedical Sensors Ltd., Pfizer Hospital Products Group, High Wycombe, England), as well as for continuous blood pressure monitoring (VICOM-SM SMU 612[R]; PPG Hellige, Freiburg, Germany). After anesthetic induction, a pulmonary artery catheter for continuous cardiac output and mixed venous oxygen saturation (SvO2) monitoring was inserted via the right internal jugular vein and connected to the corresponding cardiac output computer. In vivo SvO2 calibration was performed using in vitro values obtained by the CO-oximeter (IL 482[R]; Instrumentation Laboratory, Milano, Italy) as a reference. A Siemens Servo 900 C[R] ventilator (Siemens Life Support Systems, Solna, Sweden) was used for controlled ventilation of the lungs. The inspired oxygen fraction (FIO (2), oxygen in air) was generally >or=to0.4 and was set to 1.0 prior to OLV. The ventilator settings (including the decision for a volume or pressure controlled mode) were tailored and adjusted during surgery to the individual requirements of the patient according to the following guidelines: respiratory rate (usually 12-14 breaths/min), inspiration/expiration ratio (25%-33%), and extrinsic positive end-expiratory pressure (PEEP) (0-5 cm H2 O) as low as possible to counteract air-trapping with hypercapnia, and peak inspiratory pressure <or=to 30 cm H2 O to avoid barotrauma with pneumothorax and/or air leak. If necessary during OLV (PaO2 < 6 kPa), apneic oxygen insufflation (0.5-2 L/min), and brief periods of two-lung ventilation were intermittently used. Continuous positive airway pressure of the nonventilated lung was avoided, and the corresponding part of the double-lumen tube was opened to the atmosphere to facilitate lung collapse as a prerequisite for thoracoscopic surgery. Prior to reinflation, the collapsed lungs were suctioned through a fiberoptic bronchoscope. An infusion of lactated Ringer's solution (2 mL [centered dot] kg-1 [centered dot] h-1) was given. If necessary, vasopressors and/or catecholamines were used for maintenance of the perfusion pressure or cardiac output.
The epidural infusion of bupivacaine 0.5% was continued during extubation and the early postoperative course. A suction of 10 cm H2 O was applied to the two thoracic drainage tubes of each hemithorax. A pressure-supported mode of ventilation (trigger -2 cm H2 O, peak inspiratory pressure <or=to 20 cm H2 O, PEEP 0-5 cm H2 O, FIO2 1.0), which was gradually reduced, was applied. Higher peak inspiratory pressure or PEEP were avoided because of the risk of increasing air leakage. If the patient was conscious and ventilation with minimal pressure support (6 cm H2 O) was clinically adequate with acceptable blood gases (PaO2 > 53 mm Hg (7 kPa) on FIO2 <or=to 0.5, PaCO2 < 67 mm Hg (9 kPa), and pHa > 7.15), the trachea was extubated and oxygen was given by mask. The spontaneously breathing patients were transferred to the intensive care unit (ICU) without interruption of chest tube suction. Blood gas monitoring at the ICU was performed by repeated measurements. Chest physiotherapy was initiated at the day of operation. Epidural analgesia was continued using an infusion of bupivacaine 0.25% 4-6 mL/h. Oral paracetamol 3 x 500 mg daily was given routinely, and intravenous opioids (nicomorphine) were additionally administered if necessary.
The time spent in the operating room was divided into different time periods: total case (patient entering-patient leaving the operating room); anesthesia (insertion of epidural catheter-stop of anesthetics); weaning (stop of anesthetics-endotracheal extubation); postextubation care (endotracheal extubation-patient leaving the operating room); surgery (skin incision-bandage).
PaO2, PaCO2, and pHa values as well as cardiac index (CI) and SvO2 were recorded continuously. Discrete data were chosen from these continuous recordings as indicated below, and the PaO2/FIO2 ratio was calculated for each patient at the given time point: (a) T1 = PaCO2, pHa, PaO2, corresponding FIO2: awake, before induction of anesthesia; (b) T2 = PaCO2, pHa, PaO2, corresponding FIO2: anesthetized patient, two-lung ventilation, before operation; (c) T3 = Maximum PaCO2 (PaCO2max), minimum pHa (pHamin), minimum PaO2 (PaO2min), corresponding FIO2: intraoperatively, OLV; (d) T4 = PaCO2, pHa, PaO2, corresponding FIO2: extubated patient, before transfer to the ICU. Additionally, the extreme values (PaCO2max, pH2max, PaO2min, corresponding FIO2, minimum CI, minimum SvO2) observed during the whole period in the operating room were recorded.
The postoperative period was subdivided (0-2 h, 2-4 h, 4-8 h, 8-16 h, 16-24 h, 24-36 h, 36-48 h), and the values of PaO2min, PaCO2max, pHamin, minimum base excess, minimum oxygen saturation obtained by pulse oximetry (SpO (2min)) and the corresponding FIO2 were extracted retrospectively from the ICU charts for each subperiod.
Statistical analysis was performed using StatView 4.1 and SuperANOVA for Macintosh (Abacus Concepts, Inc., Berkeley, CA). Data are given as mean +/- SD with minimum/maximum values. Analysis of variance for repeated measures with Greenhouse-Geisser correction was performed, and P values for overall comparison of groups were calculated. P < 0.05 was considered significant. Paired t-test was used for the post hoc comparison of means, and the significance level was adjusted according to the Bonferroni method (number of corrections: n = 4 for intraoperative data; n = 8 for postoperative data).
Bilateral VRS was performed in 21 of 24 patients. Unilateral surgery only was performed in three patients due to massive air leakage after the operation on the first side in two patients and due to tension pneumothorax on the dependent lung during operation on the first side in a third case. Because the risk of operating on the other side was considered too high, only one side was resected. In two patients with initial unilateral operation, VRS of the contralateral side was performed subsequently a few weeks later, resulting in a total of 26 operations. Time spent on the different procedures is shown in Table 2. The duration of OLV was 135 +/- 48 min (minimum 60 min/maximum 240 min).
The mean minimum values observed for SvO2 and CI during the whole procedure were 65% +/- 16% and 2.6 +/- 0.9 L [centered dot] min-1 [centered dot] m-2, respectively. In the case with tension pneumothorax a CI < 1.0 L [centered dot] min-1 [centered dot] m-2 with SvO2 < 30% occurred. Mean PaO2min was within normal ranges for the whole procedure, but mean PaCO2max was increased during OLV, and hypercapnia was present in all patients (Table 3). Mean pHamin values were considerably lowered during OLV and thereafter (Table 3). Mean peak inspiratory pressure observed was 26.4 +/- 7.0 cm H2 O (range 24-42 cm H2 O). A continuous infusion of salbutamol was used in 11 patients, dopamine (200-800 micro g/min) was given in 12 patients and dobutamine (400-800 micro g/min) in two patients. Intravenous bolus doses of 2 mg methoxamine (nine patients), 10 micro g epinephrine (nine patients) or 10 micro g norepinephrine (12 patients) were given repeatedly according to the clinical situations.
Immediate endotracheal extubation in the operating room was successfully performed in 23 of the 26 operations, including the cases with severe air leakage. The patient with the intraoperative tension pneumothorax was extubated 6 h later in the ICU. In one patient reintubation was necessary in the operating room, and another patient was reintubated 30 min after ICU admission due to severe hypoventilation. Both patients could be tracheally extubated with success 8 h later.
The postoperative period in the ICU was characterized in general by an acceptable oxygenation of the spontaneously breathing patients, but several minimum PaO2 values < 45 mm Hg (6 kPa) and SaO2 values < 80% were observed (Table 4). Hypercapnia with acidosis was present at the beginning, and some markedly increased PaCO2 values were observed during the first 48 h (Table 4). One patient with known coronary artery disease, whose intraoperative and early postoperative course was uneventful, developed cardiopulmonary decompensation 10 h after surgery. Endotracheal intubation with controlled ventilation and catecholamine therapy was necessary. However, the patient developed a multiple organ failure with adult respiratory distress syndrome and died 36 h after surgery.
In terms of functional improvement, as well as morbidity and mortality, VRS can be performed by video-assisted thoracoscopic surgery in selected patients with results similar to those obtained by median sternotomy [6-9,12]. Recovery of pulmonary function as early as two weeks after video-assisted thoracoscopic surgery has been documented . The purpose of this study was to investigate the perioperative pulmonary gas exchange in a group of selected patients with severe, advanced diffuse pulmonary emphysema (forced expiratory volume in 1 s 25% +/- 7%, diffusing capacity for carbon monoxide 42% +/- 20% of predicted) who underwent VRS by video-assisted thoracoscopic surgery. Our results indicate that adequate oxygenation can be preserved during and after VRS and OLV, but CO2 elimination is impaired. However, using standardized anesthetic and monitoring techniques, intraoperative hypercapnia is well tolerated, and immediate postoperative extubation appears to be adequate.
OLV is mandatory for thoracoscopy, and the lateral decubitus positioning of the patient (18 of 24 patients) is usually necessary for video-assisted thoracoscopic surgery, resulting in a significant ventilation-perfusion mismatch. However, hypoxemia was not a limiting factor during OLV (Table 3) in our patients with severe diffuse pulmonary emphysema. The early postoperative course was also characterized in general by an acceptable oxygenation, although some episodes of low PaO2 values were observed during the 48 hours studied (Table 4). After thoracoscopic laser resection of bullous emphysema, the postoperative course of pulmonary gas exchange has often been reported to be impaired [13,14]. Although the underlying mechanisms remain undetermined [15-17], several factors have been suggested to play a major role in the formation of noncardiogenic pulmonary edema after lung resection. An increased permeability of the alveolo-capillary membrane or an increase in filtration pressure across the membrane with impaired lymphatic drainage have been observed [18,19]. Surgical compression and trauma (collapsed lung during OLV, resections by laser or staplers) , a release of mediators with concomitant increase in membrane permeability  or pulmonary artery pressure , but also hyperinflation, right ventricular failure due to a decreased size of the pulmonary vascular bed after lung resection, or simply fluid overload  may contribute to lung edema formation and preservation. However, there was no evidence for pulmonary edema during or after VRS in our patients. One of our patients with known coronary artery disease died 36 hours after surgery. Coronary artery disease thus may be a contraindication for VRS.
CO2 elimination was impaired during OLV in this group of patients with severe COPD, since alveolar ventilation-if excessive peak inspiratory pressures were avoided-had to be limited. Although the ventilatory patterns were tailored and frequently readjusted according to demand, air trapping with CO2 retention was a common problem, and significant dead-space ventilation must be assumed [22,23]. A maximum PaCO2 value of 56 +/- 12 mm Hg (7.5 +/- 1.6 kPa) with a minimum pH (a) value of 7.29 +/- 0.08 were measured, and some extreme values were observed during OLV (Table 3). The "permissive intraoperative hypercapnia" during VRS, however, was well tolerated, and no hemodynamic instability related to it was observed.
Although peak inspiratory airway pressures were moderate (26.4 +/- 7.0 cm H2 O), a tension pneumothorax of the ventilated (dependent) lung with concomitant severe hemodynamic instability occurred in one case. This complication must be regarded as life-threatening, even if detected immediately. Furthermore, air leakage is a common problem after wedge resections of emphysematous lung [5,8,9] and may also contribute to hypercapnia. The anesthesiologist's means to improve CO2 elimination are very limited. If more than 20%-30% of the tidal volume is lost by air leakage, it may be advisable to refrain from surgery on the contralateral lung. We performed unilateral VRS only in two cases due to severe air leakage after operation of the first side. One of these patients was subsequently operated on the other side a few weeks later.
For this study we used continuous intraarterial blood gas measurement (Paratrend 7[R]), a method which was previously evaluated in patients in ICUs  and in cardiac surgery during cardiopulmonary bypass , as well as in patients during thoracoscopic surgery . In these studies, a satisfying accuracy between this system and in vitro blood gas determinations was found [PaO2, r = 0.95/bias +/- 2SD = 0.38 +/- 4.86 kPa; PaCO2, 0.90/0.31 +/- 0.39 kPa; pHa, 0.83/-0.017 +/- 0.033 ]. We consider continuous blood gas monitoring very helpful during VRS. Alternatively, multiple in vitro blood gas analyses are to be performed, the results of which arrive in the operating room only with a certain time delay.
Barker et al.  described that prolonged postoperative ventilation was necessary after laser ablation of bullous emphysema. Therefore, it was anticipated that many patients suffering from severe diffuse pulmonary emphysema, who underwent bilateral lung resection, might require prolonged postoperative mechanical respiratory support. The results of this study show that this concern was unfounded. Using the anesthetic and monitoring techniques described, immediate endotracheal extubation in the operating room was performed after 25 of the 26 operations, and only two patients needed reintubation, but were successfully extubated a few hours later. Cooper et al.  reported similar results in their group of patients after median sternotomy. However, prolonged periods of postoperative ventilation (5.3 +/- 5.4 days, maximum 45 days) and severe air leakage ("often 50% of inspired tidal volume") were observed by Barker et al.  after thoracoscopic laser ablation of bullous emphysema. We assume that the difference in surgical technique (i.e., the use of laser versus stapler) and anesthesiological management may explain this different postoperative course. In fact, laser ablation of emphysematous lung tissue was abandoned by Keenan et al.  because poor outcome (death, prolonged hospital stay) was more likely to occur as compared with the stapler technique. Also, McKenna et al.  reported a higher frequency of delayed pneumothorax if laser resection was performed. Furthermore, we suggest that pain relief by thoracic epidural analgesia with local anesthetics may be important in these patients. In contrast to Barker et al.  who used halothane or isoflurane and additional fentanyl, we used a combination of thoracic epidural with total intravenous anesthesia. In two patients with known soya allergy, propofol was replaced by inhalation anesthetics. We did not compare different methods of anesthesia in this study. However, to obtain optimal postoperative pain relief with minimal respiratory and central nervous depression, a combination of thoracic epidural with some form of general anesthesia may be advantageous. Furthermore, in a study by Kellow et al.  propofol was not associated with a significant increase in shunt fraction during OLV, which increased three-fold in patients who received isoflurane. On the other hand, propofol was associated with a greater reduction in CI and right ventricular ejection fraction in this study . We did not specifically investigate the perioperative hemodynamic profile, and used vasopressors and/or catecholamines according to the clinical situations. However, we observed acceptable mean minimal values for CI (2.6 +/- 0.9 L [centered dot] min-1 [centered dot] m-2) and SVO2 (65% +/- 16%) during OLV, and there were no hemodynamic complications before or after endotracheal extubation. Rapid weaning from mechanical ventilation has been associated with left ventricular dysfunction in patients with severe COPD due to increasing left ventricular preload and afterload . Thoracic epidural analgesia with local anesthetics, however, may counteract both increased preload and afterload by pooling blood to peripheral venous vessels and by reducing arterial vasoconstriction induced by increased sympathetic activity. Furthermore, a rather descreet infusion rate as used in this study (lactated Ringer's solution 2 mL [centered dot] kg-1 [centered dot] h-1) may contribute to adequate cardiac filling after endotracheal extubation.
In conclusion, oxygenation was preserved during and after thoracoscopic VRS in a selected group of patients with advanced diffuse pulmonary emphysema. CO (2) elimination, however, was impaired, but temporary intraoperative hypercapnia was well tolerated. An immediate postoperative extubation seems to be adequate for preventing prolonged air leakage as well as prolonged postoperative mechanical respiratory support in these patients.
The authors are indebted to Mrs. Eliana Lucchinetti, physicist at the Federal Institute of Technology, Zurich, and to Burkhardt Seifert, PhD, Department of Biostatistics, Institute of Social and Preventive Medicine, University of Zurich, for the helpful discussions and the assistance in the statistical analysis of the data.
1. Gaensler EA, Cugell DW, Knudson RJ, FitzGerald MX. Surgical management of emphysema. Clin Chest Med 1983;4:443-63.
2. Wakabayashi A. Thoracoscopic technique for management of giant bullous lung disease. Ann Thorac Surg 1993;56:708-12.
3. Brantigan OC, Mueller EA, Kress MB. Surgical treatment of pulmonary emphysema. Am Surg 1957;23:789-804.
4. Brantigan OC, Mueller EA, Kress MB. A surgical approach to pulmonary emphysema. Am Rev Respir Dis 1959;80:194-202.
5. Cooper JD, Trulock EP, Triantafillou AN, et al. Bilateral pneumonectomy (volume reduction) for chronic obstructive pulmonary disease. J Thorac Cardiovasc Surg 1995;109:106-19.
6. Sciurba FC, Rogers RM, Keenan RJ, et al. Improvement in pulmonary function and elastic recoil after lung-reduction surgery for diffuse emphysema. N Engl J Med 1996;334:1095-9.
7. Gaissert HA, Trulock EP, Cooper JD, et al. Comparison of early functional results after volume reduction or lung transplantation for chronic obstructive pulmonary disease. J Thorac Cardiovasc Surg 1996;111:296-307.
8. Keenan RJ, Landreneau RJ, Sciurba FC, et al. Unilateral thoracoscopic surgical approach for diffuse emphysema. J Thorac Cardiovasc Surg 1996;111:308-16.
9. McKenna RJ, Brenner M, Gelb AF, et al. A randomized, prospective trial of stapled lung reduction versus laser bullectomy for diffuse emphysema. J Thorac Cardiovasc Surg 1996;111:317-22.
10. Roue C, Mal H, Sleiman C, et al. Lung volume reduction in patients with severe diffuse emphysema. A retrospective study. Chest 1996;110:28-34.
11. O'Donnell DE, Webb KA, Bertley JC, et al. Mechanisms of relief of exertional breathlessness following unilateral bullectomy and lung volume reduction surgery in emphysema. Chest 1996;110:18-27.
12. Bingisser R, Zollinger A, Hauser M, et al. Bilateral volume reduction surgery for diffuse pulmonary emphysema by video-assisted thoracoscopy. J Thorac Cardiovasc Surg 1996;112:875-82.
13. Wakabayashi A, Brenner M, Kayaleh RA, et al. Thoracoscopic carbon dioxide laser treatment of bullous emphysema. Lancet 1991;337:881-3.
14. Barker SJ, Clarke C, Trivedi N, et al. Anesthesia for thoracoscopic laser ablation of bullous emphysema. Anesthesiology 1993;78:44-50.
15. Waller DA, Gebitekin C, Saunders NR, Walker DR. Noncardiogenic pulmonary edema complicating lung resection. Ann Thorac Surg 1993;55:140-3.
16. Turnage WS, Lunn JL. Postpneumonectomy pulmonary edema. A retrospective analysis of associated variables. Chest 1993;103:1646-50.
17. Slinger PD. Perioperative fluid management for thoracic surgery: the puzzle of postpneumonectomy pulmonary edema. J Cardiothorac Vasc Anesth 1995;9:442-51.
18. Zeldin RA, Normandin D, Landtwing D, Peters RM. Postpneumonectomy pulmonary edema. J Thorac Cardiovasc Surg 1984;87:359-65.
19. Mathru M, Blakeman B, Dries DJ, et al. Permeability pulmonary edema following lung resection. Chest 1990;98:1216-8.
20. Van Mieghem W, Demedts M. Cardiopulmonary function after lobectomy or pneumonectomy for pulmonary neoplasm. Respir Med 1989;83:199-206.
21. Cope DK, Grimbert F, Downey JM, Taylor AE. Pulmonary capillary pressure: a review. Crit Care Med 1992;20:1043-56.
22. Baker RW, Burki NK. Alterations in ventilatory pattern and ratio of dead-space to tidal volume. Chest 1987;92:1013-7.
23. Klingstedt C, Hedenstierna G, Baehrendtz S, et al. Ventilation-perfusion relationship and atelectasis formation in the supine and lateral positions during conventional mechanical and differential ventilation. Acta Anaesthesiol Scand 1990;34:421-9.
24. Venkatesh B, Clutton-Brock TH, Hendry SP. A multiparameter sensor for continuous intra-arterial blood gas monitoring: A prospective evaluation. Crit Care Med 1994;22:588-94.
25. Venkatesh B, Clutton-Brock TH, Hendry SP. Evaluation of the Paratrend 7 intravascular blood gas monitor during cardiac surgery: comparison with the C4000 in-line blood gas monitor during cardiopulmonary bypass. J Cardiothorac Vasc Anesth 1995;9:412-9.
26. Zollinger A, Spahn DR, Ganter M, et al. Accuracy of continuous intra-arterial blood gas monitoring during thoracoscopic surgery [abstract]. Anesthesiology 1995;83:423A.
27. Kellow NH, Scott AD, White SA, Feneck RO. Comparison of the effects of propofol and isoflurane anaesthesia on right ventricular function and shunt fraction during thoracic surgery. Br J Anaesth 1995;75:578-82.
28. Lemaire F, Teboul JL, Cinotti L, et al. Acute left ventricular dysfunction during unsuccessful weaning from mechanical ventilation. Anesthesiology 1988;69:171-9.