Department of Anesthesiology and Critical Care Medicine, Tokyo Medical and Dental University, School of Medicine, 1-5-45, Yushima, Bunkyo-ku, Tokyo 113-8519, Japan E-mail: ishikawa.mane@tmd.ac.jp
Accepted for publication: September 1, 2002
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods. We studied patients during anaesthesia and one-lung ventilation, with an inspiratory oxygen fraction of 0.8. Arterial blood gas values were recorded every 10 s with a continuous intra-arterial sensor. The non-dependent lung was compressed several times during the surgical procedure, using a retractor. The change in PaO2 during and after compression of the non-dependent lung was measured.
Results. PaO2 increased significantly when the non-dependent lung was compressed, and decreased when the compression was released. The first compression of the non-dependent lung transiently increased PaO2, but the effect of the second compression on oxygenation was more marked and persistent. PaO2 increased by more than 13 kPa at 10 min after the second compression in four patients (responder group). Arterial oxygenation improved markedly in patients in this group during the surgical procedure.
Conclusion. Oxygenation can improve during one-lung ventilation in some patients. This improvement is partly related to a marked increase in PaO2 during compression of the non-dependent lung.
Br J Anaesth 2003; 90: 216
Keywords: arteries, oxygenation; monitoring, continuous intra-arterial blood gas; surgery, thoracotomy; ventilation, one-lung
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
We have found that continuous monitoring of arterial blood gas values during oesophagectomy using the continuous intra-arterial blood gas monitoring system (Paratrend 7TM, Diametrics Medical Limited, High Wycombe, UK) is useful to follow oxygenation during OLV.12 This monitor allows hypoxaemia to be detected quickly and treated promptly, and is often used in our institution for patients who need OLV.
Review of these data showed that oxygenation can improve with time during OLV.13 Changes in oxygenation with time during OLV do not seem to have been examined, although the time course of hypoxic pulmonary vasoconstriction (HPV) has been described in patients14 and animals.1517 Our finding led to a prospective assessment of the time course of oxygenation in patients undergoing thoracotomy.
We also studied the response to lung retraction. We noted that retraction of the non-dependent lung by the surgeons increased arterial oxygen partial pressure (PaO2) during continuous measurement. Conversely, the release of the retractor at the end of the procedure coincided with a decrease in PaO2. Thus, compression of the non-dependent lung appeared to affect oxygenation, so we studied the effects of manipulation of the non-dependent lung on oxygenation.
The effects of surgical manipulation on arterial oxygenation may depend on the degree of physical stimulus.18 Compression and retraction of the non-dependent lung may impede its blood flow, potentially increasing arterial oxygenation. Surgical manipulation could also cause local release of prostaglandins19 or endothelium-derived relaxing factor,18 which would have opposite effects. We assume the former effect is predominant as this would explain improved oxygenation. The effects of compression and decompression of the non-dependent lung on arterial oxygenation have not been measured continuously before.
![]() |
Methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Hydroxyzine 2550 mg and atropine 0.250.5 mg were given i.m. 30 min before the patient was taken to the operating room. Before induction of general anaesthesia, an epidural catheter was inserted at the 67th or 78th thoracic interspace. General anaesthesia was induced with propofol 2 mg kg1 i.v. and tracheal intubation was facilitated with vecuronium 0.2 mg kg1 i.v. A left-sided double-lumen tracheal tube (Broncho-Cath®, Mallinckrodt Inc., Argyle, NY, USA) was placed for OLV and the correct position was confirmed by auscultation and fibreoptic bronchoscopy. Anaesthesia was maintained with isoflurane 0.51.5%. Patients were ventilated mechanically at a constant tidal volume (10 ml kg1), and the respiratory rate was adjusted to maintain end-tidal carbon dioxide pressure at approximately 35 mm Hg. The fraction of inspiratory oxygen (FIO2) was set at 0.8. Rectal temperature was measured, and kept constant using a warm-water blanket. A dose of bupivacaine 0.25%, 812 ml was injected epidurally, followed by a continuous infusion of 48 ml h1. Systolic blood pressure was maintained within 20% of the preoperative value by controlling the concentration of isoflurane, rate of continuous epidural administration, rate of i.v. infusion, or by i.v. administration of vasopressor (ephedrine or dopamine) or vasodilator (nicardipine), as needed.
A 20-gauge intravascular catheter was inserted into the radial artery. The intravascular sensor, which had been calibrated with precision gases bubbled in sequence through the tonometer, was advanced through the arterial catheter into the radial artery to a length of 15 cm. The arterial catheter was kept patent with a continuous flush of heparinized saline 0.4 units ml1 through the sideport. The sideport allowed simultaneous arterial pressure monitoring and access for blood sampling. After turning the patient to the left lateral position, the position of the double-lumen tube was checked. Before surgery started, the monitoring system was calibrated using conventional blood gas analysis according to the manufacturers recommendations. The data collected by the monitoring system were corrected to the corresponding values at 37°C.
OLV was started just before the pleura were opened. The tracheal lumen of the double-lumen tube was simply opened to the atmosphere at the beginning of OLV. The dependent lung was ventilated with a tidal volume of 8 ml kg1 and an FIO2 of 0.8. Patients who require OLV are routinely ventilated at 10 ml kg1 during two-lung ventilation, and a smaller tidal volume during OLV20 to avoid damage from increased airway pressure (barotrauma). An FIO2 of 0.8 was chosen to reduce the risk of oxygen toxicity and absorption atelectasis and to maintain adequate oxygenation. All patients were ventilated with an Acoma AF-900 ventilator and an AF-1200 anaesthesia machine (Acoma, Tokyo, Japan). The inspiratory:expiratory ratio was 1:2. The respiratory frequency was adjusted to maintain PaCO2 at approximately 40 mm Hg. If arterial oxygen saturation (SpO2) monitored by pulse oximetry decreased below 90% during OLV, possible reasons for hypoxaemia such as tube malposition, circulatory problems, leaks or disconnection were excluded. If required, FIO2 was increased to 1.0 and continuous positive airway pressure and/or intermittent ventilation21 were applied to the non-dependent lung. Data from these patients were excluded from analysis. The Paratrend 7TM was connected to a personal computer (PC-9801NS/T, NEC, Tokyo) by an RS232C serial communication port. PaCO2, PaO2 and pH values were recorded on floppy disk every 10 s.
The non-dependent lung was compressed by the surgeons using a retractor to improve exposure of the surgical field several times during the surgical procedure. The time (s) from the start of OLV to each application and release of the retractor was recorded. The pressure applied to the lung by the retractor was constant because once the lung was compressed, the retractor was fixed rigidly to the surgical table. The surgeons were not informed of the objectives of the present study, and application and release of the retractor was done at the surgeons discretion independently of the result of the continuous intra-arterial blood gas monitoring. Pulmonary artery catheters were not used.
After anaesthesia, the PaO2 values were analysed. First, the change in PaO2at t s after the compression/decompression of the non-dependent lung [PaO2(t)] was calculated as the difference between PaO2(t) and PaO2(0).
PaO2(t) was compared between the pre-compression/decompression value (t=10) and the post-compression/decompression value. Second, PaO2 was compared at different times during OLV. The time course of arterial oxygenation was compared in patients who showed changes during the second compression of the non-dependent lung and those who did not. The second compression improved oxygenation more than the first compression, suggesting a relationship between the response to the second compression and the overall time course of oxygenation during OLV. Patients whose
PaO2(600) value of the second compression was 13.3 kPa or greater were assigned to the responder group, and those whose
PaO2(600) value was less than 13.3 kPa were assigned to the non-responder group.
PaCO2, PaO2, pH and cardiovascular data at different times were compared using repeated measures analysis of variance followed by the Tukey multiple comparison test. PaO2 values were analysed using the Friedman test followed by the ShirleyWilliams multiple comparison test.22 Because the data were not normally distributed, values for each subgroup were expressed as the median (range). P<0.05 indicated statistical significance.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Cardiovascular measurements during OLV are shown in Table 1. Arterial pressure and heart rate did not change significantly, although systolic pressure tended to decrease 20 min after starting OLV.
|
|
After the third compression of the non-dependent lung, PaO2 increased and peak values were recorded at 250 s (3.6 kPa) in one patient and at 500 s (8.9 kPa) in the other patient.
Among the eight patients in whom PaO2 was recorded after removal of compression, OLV was terminated at 100200 s in two patients, at 200300 s in one patient and at 400500 s in one patient after the decompression of the non-dependent lung. PaO2 decreased immediately after the retractor was released in all but one patient, although the PaO2 of two patients transiently returned to the baseline at approximately 300 s, and positive PaO2 values were persistently recorded in one patient after 380 s. When PaO2 values before and 200 s after removal of the retractor are considered in the six patients in whom OLV continued, the median
PaO2 was less after stopping retraction than before (P<0.05; Table 2).
Arterial blood gas data during OLV are shown in Table 3. Mean PaO2 at 1080 min after starting OLV was less than at 0 min (P<0.01). In the latter half of the OLV period (4080 min after starting OLV), PaO2 gradually increased, by an average of approximately 11 kPa. Neither PaCO2 nor pH changed throughout OLV.
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Oxygenation is affected by several factors during OLV. First, after starting OLV, blood flow through the non-dependent lung is not oxygenated. Pulmonary blood flow is then redistributed predominantly to the dependent lung, in part by hypoxic vasoconstriction in the non-dependent lung. Redistribution of blood flow to the dependent lung can also be aided by the lateral position of the patient. During thoracic surgery, oxygenation is better in the lateral position than the supine position in patients with chronic obstructive pulmonary disease.23 In the present study, surgery started with the patient in the lateral position, and the table was rolled or tilted (head-up or head-down position) as required. We assume that tilting or rotating the surgical table may affect oxygenation, although this was not shown by Fiser and colleagues.24 Rolling the surgical table during OLV in the lateral position reduces the difference in the height between the lungs, thus theoretically reducing the effects of the vertical gradient on the distribution of pulmonary blood flow. Thus, changes in position can affect the oxygenation of the patients during OLV, but this does not explain the rapid changes in PaO2 when the lung is retracted and when retraction is stopped.
The extent of HPV can be affected by factors such as anaesthetic agents, vasodilators, PaCO2,25 pH,26 27 lung manipulation18 and epidural anaesthesia.28 Isoflurane, which was used in the present study, has not been shown to inhibit HPV at a concentration of 11.5%,29 and only slightly impairs arterial oxygenation during OLV.30 In the present study, neither PaCO2 nor arterial pH changed with time (Table 3), and thus HPV was probably not affected by these factors. Local anaesthetic was given continuously, so there was probably no relationship between epidural anaesthesia and the changes in PaO2 with retraction of the lung.
Another factor that impairs arterial oxygenation during OLV is atelectasis of the dependent lung from compression by the mediastinum or abdominal contents, or by a poor patient position. It seems unlikely that compression of the non-dependent lung would cause expansion of the atelectatic dependent lung, which could then improve oxygenation with time during OLV.
Lastly, cardiac output can affect arterial oxygenation during OLV.3133 Cardiovascular measurements in the present study were limited (arterial pressure and heart rate), but these did not change significantly during the study (Table 1). These results suggest that cardiovascular changes are not an important factor affecting the time course of oxygenation during OLV, but the lack of cardiac output and shunt data limit the present study.
The most likely explanation of changes in arterial oxygenation after lung retractor application is that blood flow to the non-dependent lung is re-directed to the dependent lung by physical compression and kinking of the lung vessels. The lung may have been compressed to a greater extent when the retractor was applied the second time to give better exposure of the surgical field, so that blood flow may have been further redistributed to the dependent lung, causing greater and more persistent improvement in oxygenation. The decrease in PaO2 after stopping retraction is explained by resumption of the distribution of pulmonary blood flow to the conditions before compression. Other mechanisms may affect oxygenation after the second compression: the compression/decompression effect might resemble a hypoxic challenge14 or the HPV may have been maximal at the second compression since the HPV response has been shown to be maximal after 15 min in dogs undergoing unilateral hypoxic ventilation.17 The time course of HPV may contribute to the greater improvement in oxygenation after the second compression, but it was impossible to prove this in the present study.
We set out to determine the time course of arterial oxygenation during the procedure. As FIO2 is usually kept high during OLV, it is difficult to assume trends in oxygenation without continuous intra-arterial blood gas monitoring. Under these conditions, pulse oximetry values do not change, even with a wide range of PaO2 values. With knowledge of the time course of oxygenation, anaesthetists may be able to anticipate the need for treatment to prevent hypoxaemia, such as after the release of the retractor. Compression of the non-dependent lung could be used to improve arterial oxygenation if hypoxaemia occurs in the early stage of OLV before the application of the retractor.
Knowledge of the time course of arterial oxygenation may help in planning relevant and prospective clinical research in which the effects of anaesthetics or drugs on oxygenation are compared during OLV. Many studies have been performed during oesophagectomy to determine which anaesthetics or drugs improve oxygenation during OLV.710 34 Some researchers selected patients undergoing oesophagectomy because minimal trauma is applied to the non-ventilated non-dependent lung.8 Other investigators have compared volatile anaesthetics and propofol in a fixed order instead of the random order or cross-over design.35 From the present study, we suggest that studies of anaesthetics or drugs would be better performed in patients undergoing surgery where effects on oxygenation are less than with oesophagectomy, to avoid confounding results from changes in oxygenation independent of the anaesthetics or drugs to be tested.
The side of thoracotomy is important in interpreting the effects of surgical manipulation on oxygenation because the distribution of pulmonary blood flow partially depends on the side of thoracotomy.35 36 All patients underwent oesophagectomy with right thoracotomy, so the results of the present study may not apply to patients who undergo oesophagectomy with left thoracotomy.
The effects of dopamine on HPV are controversial. I.V. dopamine depresses the HPV response in dogs at 25 µg kg1 min1 but not at 2.5 µg kg1 min1.37 Nomoto and colleagues33 found no significant effect of dopamine 5 µg kg1 min1 on oxygenation in patients during OLV. Since the dose of dopamine in the present study (15 µg kg1 min1) was lower than in the former study, and similar to the latter study, we believe that dopamine had a minimal effect on arterial oxygenation.
In conclusion, arterial oxygenation is improved by compression of the non-dependent lung with a retractor to improve exposure of the surgical field during OLV in patients undergoing oesophagectomy. Conversely, removal of retraction of the non-dependent lung impaired oxygenation. Oxygenation tended to improve in the latter half of the thoracic procedure of oesophagectomy in four patients, and this improvement can be related in part to the marked increase in PaO2 resulting from the compression of the non-dependent lung. Whether arterial oxygenation improves with time during OLV in patients undergoing oesophagectomy may partially depend on the response of arterial oxygenation to compression of the non-dependent lung.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Barker SJ, Clarke C, Trivedi N, Hyatt J, Fynes M, Roessler P. Anesthesia for thoracoscopic laser ablation of bullous emphysema. Anesthesiology 1993; 78: 4450[ISI][Medline]
3 Fradj K, Samain E, Delefosse D, Farah E, Marty J. Placebo-controlled study of inhaled nitric oxide to treat hypoxaemia during one-lung ventilation. Br J Anaesth 1999; 82: 20812
4 Chen TL, Lee YT, Wang MJ, Lee JM, Lee YC, Chu SH. Endothelin-1 concentrations and optimisation of arterial oxygenation and venous admixture by selective pulmonary artery infusion of prostaglandin E1 during thoracotomy. Anaesthesia 1996; 51: 4226[ISI][Medline]
5 Abe K, Mashimo T, Yoshiya I. Arterial oxygenation and shunt fraction during one-lung ventilation: a comparison of isoflurane and sevoflurane. Anesth Analg 1998; 86: 126670[Abstract]
6 Karzai W, Haberstroh J, Priebe HJ. Effects of desflurane and propofol on arterial oxygenation during one-lung ventilation in the pig. Acta Anaesthesiol Scand 1998; 42: 64852[ISI][Medline]
7 Shimizu T, Abe K, Kinouchi K, Yoshiya I. Arterial oxygenation during one lung ventilation. Can J Anaesth 1997; 44: 11626[Abstract]
8 Reid CW, Slinger PD, Lenis S. A comparison of the effects of propofol-alfentanil versus anesthesia on arterial oxygenation during one-lung ventilation. J Cardiothorac Vasc Anesth 1996; 10: 8603[ISI][Medline]
9 Slinger P, Scott WAC. Arterial oxygenation during one-lung ventilation: a comparison of enflurane and isoflurane. Anesthesiology 1995; 82: 9406[CrossRef][ISI][Medline]
10 Wang JYY, Russell GN, Page RD, Jackson M, Pennefather SH. Comparison of the effects of sevoflurane and isoflurane on arterial oxygenation during one lung ventilation. Br J Anaesth 1998; 81: 8503
11 Pagel PS, Fu JL, Damask MC, et al. Desflurane and isoflurane produce similar alterations in systemic and pulmonary hemodynamics and arterial oxygenation in patients undergoing one-lung ventilation during thoracotomy. Anesth Analg 1998; 87: 8007[Abstract]
12 Ishikawa S, Makita K, Nakazawa K, Amaha K. Continuous intra-arterial blood gas monitoring during oesophagectomy Can J Anaesth 1998; 45: 2736[Abstract]
13 Ishikawa S. Oxygenation may improve with time during one-lung ventilation. Anesth Analg 1999; 89: 2589
14 Carlsson AJ, Bindslev L, Santesson J, Gottlieb I, Hedenstierna G. Hypoxic pulmonary vasoconstriction in the human lung: the effect of prolonged unilateral hypoxic challenge during anaesthesia. Acta Anaesthesiol Scand 1985; 29: 34651[ISI][Medline]
15 Benumof JL. Intermittent hypoxia increases lobar hypoxic pulmonary vasoconstriction. Anesthesiology 1983; 58: 399404[ISI][Medline]
16 Tucker A, Reeves JT. Nonsustained pulmonary vasoconstriction during acute hypoxia in anesthetized dogs. Am J Physiol 1975; 228: 75661
17 Domino KB, Chen L, Alexander CM, Williams JJ, Marshall C, Marshall BE. Time course and responses of sustained hypoxic pulmonary vasoconstriction in the dog. Anesthesiology 1984; 60: 5626[ISI][Medline]
18 Benumof JL. Special respiratory physiology of the lateral decubitus position, the open chest, and one-lung ventilation. In: Benumof JL, ed. Anesthesia for Thoracic Surgery, 2nd Edn. Philadelphia: WB Saunders, 1995; 12351
19 Piper P, Vane J. The release of prostaglandins from lung and other tissues. Ann NY Acad Sci 1971; 180: 36385
20 Ishikawa S, Ohmi S, Nakazawa K, Makita K. Continuous intra-arterial blood gas monitoring during thoracic surgery. J Anesth 2000; 14: 11923[CrossRef][ISI]
21 Benumof JL. Conventional and differential lung management of one-lung ventilation. In: Benumof JL, ed. Anesthesia for Thoracic Surgery, 2nd Edn. Philadelphia: WB Saunders, 1995; 40631
22 Shirley E. A non-parametric equivalent of Williams test for contrasting increasing dose levels of a treatment. Biometrics 1977; 33: 3869[ISI][Medline]
23 Bardoczky GI, Szegedi LL, dHollander AA, Moures JM, de Francquen P, Yernault JC. Two-lung and one-lung ventilation in patients with chronic obstructive pulmonary disease: the effects of position and FIO2. Anesth Analg 2000; 90: 3541
24 Fiser WP, Friday CD, Read RC. Changes in arterial oxygenation and pulmonary shunt during thoracotomy with endobronchial anesthesia. J Thorac Cardiovasc Surg 1982; 83: 52331[Abstract]
25 Benumof JL, Wahrenbrock EA. Blunted hypoxic pulmonary vasoconstriction by increased lung vascular pressures. J Appl Physiol 1975; 38: 84650
26 Lejeune P, Brimioulle S, Leeman M, Hallemans R, Melot C, Naeije R. Enhancement of hypoxic pulmonary vasoconstriction by metabolic acidosis in dogs. Anesthesiology 1990; 73: 25664[ISI][Medline]
27 Brimioulle S, Lejeune P, Vachiery JL, Leeman M, Melot C, Naeije R. Effects of acidosis and alkalosis on hypoxic pulmonary vasoconstriction in dogs. Am J Physiol 1990; 258: H34753
28 Garutti I, Quintana B, Olmedilla L, Cruz A, Barranco M, Garcia de Lucas E. Arterial oxygenation during one-lung ventilation: combined versus general anesthesia. Anesth Analg 1999; 88: 4949
29 Jolin-Carlsson A, Bindslev L, Hedenstierna G. Hypoxia-induced pulmonary vasoconstriction in the human lung: the effect of isoflurane anesthesia. Anesthesiology 1987; 66: 31216[ISI][Medline]
30 Benumof JL, Augustine SD, Gibbons JA. Halothane and isoflurane only slightly impair arterial oxygenation during one-lung ventilation in patients undergoing thoracotomy. Anesthesiology 1987; 67: 91015[ISI][Medline]
31 Westbrooke JL, Sykes MK. Preoperative arterial hypoxaemia: the interaction between intrapulmonary shunt and cardiac output. Anaesthesia 1992; 47: 30710[ISI][Medline]
32 Eisenkraft LB. Effects of anaesthetics on the pulmonary circulation. Br J Anaesth 1990; 65: 6378[ISI][Medline]
33 Nomoto Y, Kawamura M. Pulmonary gas exchange effects by nitroglycerine, dopamine and dobutamine during one-lung ventilation in man. Can J Anaesth 1989; 36: 2737[ISI][Medline]
34 Abe K, Shimizu T, Takashina M, Shiozaki H, Yoshiya I. The effects of propofol, isoflurane, and sevoflurane on oxygenation and shunt fraction during one-lung ventilation. Anesth Analg 1998; 87: 11649[Abstract]
35 Wulff KE, Aulin I. The regional lung function in the lateral decubitus position during anaesthesia and operation. Acta Anaesthesiol Scand 1972; 16: 195205[ISI][Medline]
36 Rehder K, Wenthe FM, Sessler AD. Function of each lung during mechanical ventilation with ZEEP and with PEEP in man anesthetized with thiopental meperidine. Anesthesiology 1973; 39: 597606[ISI][Medline]
37 Marin JLB, Orchard C, Chakrabarti MK, Sykes MK. Depression of hypoxic pulmonary vasoconstriction in the dog by dopamine and isoprenaline. Br J Anaesth 1979; 51: 30312[Abstract]