Anesth Analg 2001;93:1434-1438
© 2001 International Anesthesia Research Society
CARDIOVASCULAR ANESTHESIA
The Effects of Increasing Concentrations of Isoflurane and Desflurane on Pulmonary Perfusion and Systemic Oxygenation During One-Lung Ventilation in Pigs
Konrad Schwarzkopf, MD,
Torsten Schreiber, MD,
Reinhard Bauer, MD*,
Harald Schubert, VD
,
Niels-Peter Preussler, MD,
Elke Gaser, MD,
Uwe Klein, MD, and
Waheedullah Karzai, MD
Department of Anesthesiology and Intensive Care Medicine, *Department of Pathophysiology, and Institute for Experimental Animals, University of Jena, Germany
Address correspondence and reprint requests to Waheedullah Karzai, MD, Department of Anesthesia and Intensive Care, Zentralklinik Bad Berka, Robert-Koch-Allee 9, 99437 Bad Berka, Germany. Address e-mail to W.Karzai.ana{at}Zentralklinik-Bad-Berka.de
 |
Abstract
|
|---|
During one-lung ventilation (OLV), hypoxic pulmonary vasoconstriction (HPV) reduces venous admixture and attenuates the decrease in arterial oxygen tension by diverting blood from the nonventilated lung to the ventilated lung. In vitro, desflurane and isoflurane depress HPV in a dose-dependent manner. Accordingly, we studied the effects of increasing concentrations of desflurane and isoflurane on pulmonary perfusion, shunt fraction, and PaO2 during OLV in vivo. Fourteen pigs (30–42 kg) were anesthetized, tracheally intubated, and mechanically ventilated. After placement of femoral arterial and thermodilution pulmonary artery catheters, a left-sided double-lumen tube (DLT) was placed via tracheotomy. After DLT placement, FIO2 was adjusted at 0.8 and anesthesia was continued in random order with 3 concentrations (0.5, 1.0, and 1.5 minimal alveolar concentrations) of either desflurane or isoflurane. Differential lung perfusion was measured with colored microspheres. All measurements were made after stabilization at each concentration. Whereas mixed venous PO2, mean arterial pressure, cardiac output, nonventilated lung perfusion, and shunt fraction decreased in a dose-dependent manner, PaO2 remained unchanged with increasing concentrations of desflurane and isoflurane during OLV. In conclusion, increasing concentration of desflurane and isoflurane did not impair oxygenation during OLV in pigs.
IMPLICATIONS: In an animal model of one-lung ventilation, increasing concentrations of desflurane and isoflurane dose-dependently decreased shunt fraction and perfusion of the nonventilated lung and did not impair oxygenation. The decreases in shunt fraction are likely the result of anesthetic-induced marked decreases in cardiac output and mixed venous saturation.
|
Introduction
|
|---|
During one-lung ventilation (OLV), hypoxic pulmonary vasoconstriction (HPV) diverts blood flow from the nonventilated to the ventilated lung, thereby reducing perfusion of the nonventilated lung, decreasing shunt fraction, and ameliorating the decrease in arterial oxygen tension (PaO2) (1). In vitro, volatile anesthetics directly depress HPV in a dose-dependent manner (2–4). An important question related to the use of volatile anesthetics during OLV is therefore whether they increase the perfusion of the nonventilated lung by inhibiting HPV, and whether this increase is dose-dependent. In vivo, however, volatile anesthetics may affect HPV not only directly, but also indirectly by their influence on cardiac output (CO) and venous oxygen tension (PvO2) (5). Because indirect effects of volatile anesthetics are also dose-dependent, it is unpredictable how increasing concentrations of volatile anesthetics will affect perfusion of the nonventilated lung, the shunt fraction, and the PaO2. In a previous study we found that increasing concentrations of desflurane do not decrease oxygenation during OLV (6). However, in that study, the animals were positioned supine, the chest was closed, and lung perfusion was not measured. Accordingly, in the present study, we evaluated the effects of increasing concentrations of desflurane and isoflurane on differential lung perfusion, on shunt fraction, and on oxygenation during OLV in a pig model of OLV, which closely simulates the clinical situation. We compared desflurane with isoflurane because of the conflicting data regarding their effects on HPV. In chronically instrumented dogs, isoflurane inhibits HPV (7) whereas desflurane does not (8).
|
Methods
|
|---|
The study protocol of this study was approved by the local Animal Protection Committee (Landesverwaltungsamt, Thüringen, Germany). Fourteen female pigs (30–42 kg) were studied. After overnight fasting, the animals were premedicated with ketamine (500 mg IM) to allow placement of an IV catheter and to initiate continuous electrocardiogram and pulse oximetry monitoring. Anesthesia was induced with propofol (2–3 mg/kg IV) and rocuronium (1.2 mg/kg IV), and the trachea was orally intubated with a 5.0–8.0 mm inner diameter endotracheal tube. Mechanical ventilation was adjusted to maintain arterial CO2 tension (PaCO2) at approximately 35–40 mm Hg. Anesthesia was maintained with a 1:1 mixture of nitrous oxide (N2O) and oxygen (O2), isoflurane 1–1.5% (end-tidal, measured by Capnomac® Ultima, Datex, Helsinki, Finland), and continuous infusion of remifentanil (10–20 µg · kg-1 · h-1) and rocuronium (1–1.5 mg · kg-1 · h-1). Using sterile technique, a femoral arterial catheter was placed to measure arterial blood pressure and a flow-directed thermodilution pulmonary artery catheter was passed through a 8.5F introducer through the right external jugular vein to measure CO and pulmonary pressure. The tip of the pulmonary artery catheter was positioned just beyond the pulmonary valve to ensure placement in the main pulmonary artery (i.e., the catheter was not advanced to the wedge-position). A central venous catheter was placed in the right internal jugular vein. Subsequently, a tracheotomy was performed, and the orotracheal tube was replaced under fiberoptic control by a left-sided, specially designed 39 Ch double-lumen tube (DLT) (Mallinckrodt, Dublin, Ireland). This specially designed DLT ensured that the right upper bronchus could also be ventilated or accessed through the tracheal limb. After DLT placement, the animals were positioned in the left lateral decubitus position, and an 8.5-mm inner diameter endotracheal tube was passed through a right-sided mini-thoracotomy into the right pleural space. Ventilation to the right lung was then discontinued, and lung collapse was verified by fiberoptic observation of the right pleural space. During the study, correct DLT placement was verified by continuous dual capnography, by fiberoptic bronchoscopy, and by thoracoscopy.
After these measures, remifentanil and N2O were discontinued, OLV (left lung) was started, and the FIO2 (in air) was adjusted at 0.8. A FIO2 of 0.8 was chosen to control for 15% reduction in FIO2 during 1.5 minimal alveolar concentration (MAC) desflurane administration. Subsequently, anesthesia was continued in random order with desflurane or isoflurane of either 0.5 MAC (5% and 1.05% resp), 1 MAC (10% and 2.1% resp) or 1.5 MAC (15% or 3.15% resp) end-tidal concentration (9). After recording stable cardiorespiratory variables (mean arterial pressure, heart rate, end-tidal carbon dioxide concentration, oxygen saturation via pulse oximetry) respiratory and hemodynamic variables were determined during each anesthetic concentration. Equilibration times were at least 30 min, and blood pressure, heart rate, and end-tidal anesthetic concentrations varied by no more than 
10% for at least 20 min before measurements were made. At the end of each time period, colored microspheres were administered through a central venous line and cardiorespiratory measurements performed and noted.
The animals were kept in the left lateral decubitus position throughout the experiment. Body temperature was maintained by covering the animals with a WarmTouch blanket and was continuously monitored by the thermistor of the thermodilution catheter. The animals received 15 mL/kg of body-warm balanced electrolyte solutions during induction and preparation that was continued at a rate of 10 mL/kg/h during the study period.
Ventilation during OLV was provided by a constant volume/pressure ventilator. We set ventilation pressure at 25–30 cm H2O, expiratory pressure at 5 cm H2O, and varied respiratory frequency to achieve end-tidal CO2 at 33–38 mm Hg.
Application and methodological consideration of microsphere measurements in pigs have been presented and discussed in detail elsewhere (10). Briefly, for measurements of regional pulmonary perfusion, 1.2 x 106 colored microspheres (Dye-Trak, Triton Technology, San Diego, CA) with a nominal diameter of 15 µm and suspended in Tween 80 (Fluka, Neu-Ulm, Germany) were mixed for 3 min by sonification (Transsonic T 310; Bender and Hobein, Singen, Germany) and injected slowly over 120 s via the central venous catheter into the superior vena cava. Injection was followed by flushing the catheter with 10 mL of saline. Microsphere injections were repeated at the end of the six experimental phases using different colored microspheres (white, blue, eosin, orange, yellow) in random sequence. One experimental period per animal was omitted because there were five colors and six periods. At the end of the experiment the lungs were removed, dissected, and digested by placing them in a 4N concentrated solution of KOH. The right and left lungs were digested separately. To obtain the microspheres, the digested samples were then filtered under vacuum suction through 8 µm pore polyester membranes filters (Costar, Bodenheim, Germany). The microspheres were washed with a 2% Tween 80 solution and subsequently with ethanol. The colored microspheres were quantified by their dye content. The dye was removed from the microspheres by adding 150 µL dimethylformamide as a solvent. The photometric absorption of each dye solution was determined using a spectrophotometer at wavelengths 190–820 mm. The number of microspheres was calculated using the specific absorbance value of the different dyes. All reference and tissue samples contained >400 microspheres. Percentage of the right lung perfusion was calculated as the proportion of the microsphere number obtained from right lung on total number of microspheres.
Shunt fraction was calculated using the standard formula: Qs/Qt = (CcO2 -CaO2)/(CcO2 - CvO2), where Qs represents shunt flow, Qt = total flow, CcO2 = capillary, CaO2 = arterial, and CvO2 = mixed venous oxygen content. The data were statistically analyzed by a multivariate analysis of variance (ANOVA) with the factors anesthetic (desflurane and isoflurane) and concentration (0.5, 1.0, and 1.5 MAC). A stepwise linear regression analysis was performed for PaO2 with the factors CO, Qs/Qt, PvO2, mean pulmonary artery pressure (MPAP), mixed venous O2 saturation (SvO2), type of anesthetic and MAC of anesthetic. Statistical tests were performed with the computing program "Statistical Packet for the Social Sciences" SPSS. A P value of <0.05 was considered statistically significant.
|
Results
|
|---|
During OLV, pH, PaCO2, tidal volume, respiratory frequency, and peak airway pressure remained unchanged throughout the study and were comparable during anesthesia with increasing concentrations of either isoflurane of desflurane (Table 1). PvO2, SvO2, MPAP and systemic arterial pressure (MAP), CO, and Qs/Qt decreased comparably during desflurane and isoflurane anesthesia in a dose-dependent fashion (all P < 0.05) (Fig. 1 and Table 2). Perfusion of the nonventilated lung as measured with colored microspheres decreased with increasing concentrations of desflurane and isoflurane (ANOVA P = 0.013) but was not significantly different between 1.0 and 1.5 MAC isoflurane (paired Student’s t-test) (Fig. 1). PaO2 did not change significantly with increasing desflurane and isoflurane concentrations. (Fig. 1). Hemodynamic variables are reported in Table 1 and respiratory variables in Table 2. Hemoglobin levels and pulse oximetric saturation (data not shown) did not change between groups. There were no significant differences between pulse oximetric O2 saturation and arterial O2 saturation (blood gas analysis). Stepwise linear regression identified CO (negative correlation) and PvO2 (positive correlation) as the two main determinants of PaO2 (P < 0.001; F = 9.5; r2 = 0.25) during OLV. Regardless of the anesthetic or its concentration, PaO2 at the beginning of the study (135 ± 24) and at the end the study (144 ± 20) did not differ (P = 0.5).
View larger version (17K):
[in this window]
[in a new window]
|
Figure 1. The effects of 0.5, 1.0, and 1.5 minimal alveolar concentrations of desflurane and isoflurane on selected hemodynamic variables during one-lung ventilation. Values are depicted as mean ± SE. * P < 0.05 compared with preceding values. Qs/Qt = calculated shunt fraction; PvO2 = mixed venous O2 tension; PaO2 = arterial O2 tension.
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. The Effects of Increasing Concentrations of Desflurane and Isoflurane on Hemodynamic Variables During One-Lung Ventilation
|
|
|
Discussion
|
|---|
In our porcine model of OLV, increasing levels of desflurane and isoflurane dose-dependently decreased CO, SvO2, Qs/Qt, and perfusion of the nonventilated lung, but failed to show consistent dose-dependent effects on oxygenation.
Many studies show that volatile anesthetics impair HPV in vitro. A smaller number of studies show that HPV is operative in vivo and may affect oxygenation during OLV. In one controlled experiment, Domino et al. (11) ventilated the right lung with 100% oxygen, whereas the left lung was ventilated with a hypoxic gas mixture. Increasing concentrations of isoflurane administered only to the hypoxic left lung increased intrapulmonary shunt and decreased PaO2 in a dose-dependent fashion. The objective of this study was to show that the in vitro determined dose-dependent effects of isoflurane on HPV are also operative in vivo. However, the direct effects of increasing concentrations of volatile anesthetics on HPV may not always translate to impairments in oxygenation, largely because they are modified by the concurrent effects of volatile anesthetics on hemodynamic variables (5). Using increasing concentrations of volatile anesthetics during OLV in our animal model provides us with an opportunity to study these effects in more detail.
One factor that may have a major effect on HPV and on oxygenation during OLV is CO. Depending on study setting, changes in CO affect HPV in different ways. In experiments in which HPV is studied in the hypoxic lung region, the magnitude of HPV is flow-dependent (7,12). In experiments simulating clinical OLV, a decrease in flow leads to a preferential increase in the perfusion of the normoxic (ventilated) lung (5). Because most volatile anesthetics decrease CO in a dose-related fashion, the net effect of increasing concentrations of volatile anesthetics on oxygenation is probably the net result of dose-dependent direct inhibiting and indirect enhancing effects on HPV. In our study, CO, Qs/Qt, and perfusion of the nonventilated lung progressively decreased with increasing concentrations of both isoflurane and desflurane. Thus the effects of decreasing CO may have attenuated the dose-dependent inhibiting effect of the two volatile anesthetics on HPV and resulted in a decrease in Qs/Qt and no decrease in oxygenation during OLV.
PvO2 may also modify the HPV response in vivo. Decreases in PvO2 may enhance HPV (1,13) because the sum of alveolar and mixed-venous O2 affect the magnitude of the HPV response. In the nonventilated lung, a low PvO2 would increase HPV, a high PvO2 would decrease HPV. In our study, isoflurane and desflurane led to dose-dependent decreases in PvO2. Thus, decreasing PvO2 during larger concentrations of isoflurane and desflurane may have, in part, counteracted the increasing direct inhibitory effects of isoflurane and desflurane on HPV, thereby maintaining oxygenation during increasing isoflurane and desflurane concentrations.
Theoretically, any decrease in MPAP may aggravate the effects of gravitation leading to preferential perfusion of the lower lung. However, the effects of gravitation on MPAP do not adequately explain the effects of increasing doses of volatile anesthetics on oxygenation in our study. First, airway pressure in the nonventilated, collapsed lung can be neglected thus precluding zone-1 effects. Second, the differences in the MPAP between the smallest and the largest dose of volatile anesthetics was rather small, approximately 3–5 mm Hg. Third, regression analysis did not identify MPAP as a major factor affecting PaO2.
When Qs/Qt and perfusion of the nonventilated lung decreases with increasing anesthetic concentration, one would expect oxygenation to increase. However, as in our previous study (6), the dose-dependent decrease in shunt volume during desflurane and isoflurane anesthesia did not lead to corresponding increases in PaO2. Regression analysis in our study showed that neither Qs/Qt nor perfusion of the nonventilated lung but rather CO and PvO2 were the two (physiologically interrelated) determinants of oxygenation during OLV with increasing levels of volatile anesthetics. Most likely, at any given intrapulmonary shunt volume, decreases in PvO2 or oxygen content of the shunted (venous) blood, by ultimately mixing with oxygenated blood, will lead to related decreases in systemic oxygenation. Thus, during OLV, decreases in PvO2 may, on the one hand, improve systemic oxygenation by enhancing HPV and reducing Qs/Qt and perfusion of the nonventilated lung, but may, on the other hand, impair systemic oxygenation by admixture of venous blood with low oxygen saturation. Thus oxygenation during OLV with increasing levels of volatile anesthetics depends on the interaction between HPV, CO, SvO2, and Qs/Qt.
In our previous study, increasing inspiratory desflurane concentrations from 0.5 to 1.5 MAC decreased Qs/Qt from 0.34 to 0.28. In this study, increasing inspiratory desflurane concentrations from 0.5 to 1.5 MAC decreased Qs/Qt levels from 0.29 to 0.20. The reason why Qs/Qt was comparably lower during this study as compared with our previous study can be attributable to a number of reasons. In our previous study, the hemithorax was closed, the hypoxic lung was expanded, and the pigs were positioned supine. In this study, the hemithorax was open, the hypoxic lung was collapsed, and the pigs were in the lateral decubitus position. Whereas perfusion to a nonventilated hypoxic lung does not change whether the lung is collapsed (hemithorax open) or expanded (hemithorax closed) (14), body position does make a difference (15). During OLV, perfusion to the ventilated (dependent) lung is better during the lateral decubitus position as compared with the supine position. Therefore, Qs/Qt can be expected to be less and oxygenation higher in the lateral decubitus position as compared with supine position. A recent clinical study has shown that oxygenation was higher in the lateral decubitus position as compared with supine position during OLV (16).
Our finding that increasing concentrations of volatile anesthetics do not impair oxygenation during OLV cannot be directly extrapolated to the clinical scenario. In clinical anesthesia we may either use a concentration of volatile anesthetics that do not lead to unsafe levels of CO and MAP or we may choose to correct adverse hemodynamic effects of volatile anesthetics by administration of fluids and catecholamines. Also, surgical manipulation leads to increases in endogenous sympathomimetic response and increases in CO and MAP. Because of all these factors, a safe clinical study using increasing concentrations of volatile anesthetics may find different results than those in our animal model.
In conclusion, although in vitrodesflurane and isoflurane inhibit HPV in a dose-dependent fashion, increasing concentrations of these anesthetics do not necessarily increase perfusion of the nonventilated lung, do not increase Qs/Qt or worsen oxygenation during one-lung ventilation in vivo.
|
References
|
|---|
-
Benumof JL. One-lung ventilation and hypoxic pulmonary vasoconstriction: implications for anesthetic management. Anesth Analg 1985; 64: 821–33.[Free Full Text]
-
Loer SA, Scheeren TW, Tarnow J. Desflurane inhibits pulmonary hypoxic vasoconstriction in isolated rabbit lungs. Anesthesiology 1995; 83: 552–6.[Web of Science][Medline]
-
Ishibe Y, Gui X, Uno H, et al. Effect of sevoflurane on hypoxic pulmonary vasoconstriction in isolated rabbit lungs. Anesthesiology 1993; 79: 1348–53.[Web of Science][Medline]
-
Marshall C, Lindgren L, Marshall BE. Effects of halothane, enflurane, and isoflurane on hypoxic pulmonary vasoconstriction in rat lungs in vitro. Anesthesiology 1984; 60: 304–8.[Web of Science][Medline]
-
Eisenkraft JB. Effects of anaesthetics on pulmonary circulation. Br J Anaesth 1990; 65: 63–78.[Free Full Text]
-
Karzai W, Haberstroh J, Priebe HJ. The effects of increasing concentrations of desflurane on systemic oxygenation during one-lung ventilation in pigs. Anesth Analg 1999; 89: 215–7.[Abstract/Free Full Text]
-
Lennon PF, Murray PA. Attenuated hypoxic pulmonary vasoconstriction during isoflurane is abolished by cyclooxygenase inhibition in chronically instrumented dogs. Anesthesiology 1996; 84: 404–14.[Web of Science][Medline]
-
Lesitsky MA, Davis S, Murray PA. Preservation of hypoxic pulmonary vasoconstriction during sevoflurane and desflurane anesthesia compared to the conscious state in chronically instrumented dogs. Anesthesiology 1998; 89: 1501–8.[Web of Science][Medline]
-
Eger II EI Johnson BH, Weiskopf RB. Minimum alveolar concentration of I-653 and isoflurane in pigs: definition of a supramaximal stimulus. Anesth Analg 1988; 67: 1174–6.[Abstract/Free Full Text]
-
Walter B, Bauer R, Gaser E, Zwiener U. Validation of the multiple colored microsphere technique for regional blood flow measurements in newborn piglets. Basic Res Cardiol 1997; 92: 191–200.[Web of Science][Medline]
-
Domino KB, Borowec L, Alexander CM, et al. Influence of isoflurane on hypoxic pulmonary vasoconstriction in dogs. Anesthesiology 1986; 64: 423–9.[Web of Science][Medline]
-
Nakayama M, Murray PA. Ketamine preserves and propofol potentiates hypoxic pulmonary vasoconstriction compared with conscious state in chronically instrumented dogs. Anesthesiology 1999; 91: 760–71.[Web of Science][Medline]
-
Domino KB, Wetstein L, Glasser SA, et al. Influence of mixed venous oxygen tension (PvO2) on blood flow to atelectatic lung. Anesthesiology 1983; 59: 428–34.[Web of Science][Medline]
-
Pirlo AF, Benumof JL, Trousdale FR. Atelectatic lobe blood flow: open vs. closed chest, positive pressure vs. spontaneous ventilation. J Appl Physiol 1981; 50: 1022–6.[Abstract/Free Full Text]
-
Hales CA, Ahluwalia B, Kazemi H. Strength of pulmonary vascular response to regional alveolar hypoxia. J Appl Physiol 1975; 38: 1083–7.[Abstract/Free Full Text]
-
Bardoczky GI, Szegedi LL, d’Hollander AA, et al. Two-lung and one-lung ventilation in patients with chronic obstructive pulmonary disease: the effects of position and FIO2. Anesth Analg 2000; 90: 35–41.[Abstract/Free Full Text]
Accepted for publication August 15, 2001.
This article has been cited by other articles:
|
|
|
|
 
T. T. V. Pechet, L. Bogar, and Z. Grunwald
Anesthetic Considerations for Thoracic Trauma
Seminars in Cardiothoracic and Vascular Anesthesia,
June 1, 2002;
6(2):
95 - 103.
[Abstract]
[PDF]
|
|
|
Top
Abstract
Introduction
