- Email: [email protected]
Effects of xenon anaesthesia on the circulatory response to hypoventilation
British Journal of Anaesthesia 95 (2): 166–71 (2005)
doi:10.1093/bja/aei153
Advance Access publication May 20, 2005
Effects of xenon anaesthesia on the circulatory response to hypoventilation J.-H. Baumert*, K. E. Hecker, M. Hein, M. Reyle-Hahn1, N. A. Horn and R. Rossaint Anaesthesiology Clinic, Universitaetsklinikum Aachen, Germany. 1Anaesthesia Department, Waldkrankenhaus Berlin-Spandau, Germany *Corresponding author: Anaesthesiology Clinic, Universitaetsklinikum Aachen, Pauwelsstrasse 30, 52074 Aachen, Germany. E-mail: [email protected] Background. Circulatory response to hypoventilation is aimed at eliminating carbon dioxide and maintaining oxygen delivery (DO2) by increasing cardiac output (CO). The hypothesis that this increase is more pronounced with xenon than with isoflurane anaesthesia was tested in pigs.
Results. CO increased by 10–20% with both anaesthetics, with an equivalent rise in HR, maintaining DO2 in spite of a 20% reduction in arterial oxygen content. Decreased left ventricular (LV) afterload during hypoventilation increased FAC, and this was more marked with xenon (0.60– 0.66, P<0.05 compared with baseline and isoflurane). This difference is attributed to negative inotropic effects of isoflurane. Increased pulmonary vascular resistance during hypoventilation was found with both anaesthetics. Conclusion. The cardiovascular effects observed in this model of moderate hypoventilation were sufficient to maintain DO2. Although the haemodynamic response appeared more pronounced with xenon, differences were not clinically relevant. An increase in FAC with xenon is attributed to its lack of negative inotropic effects. Br J Anaesth 2005; 95: 166–71 Keywords: anaesthetics gases, xenon; complications, hypercarbia; complications, hypoxaemia; monitoring, trans-oesophageal echocardiography Accepted for publication: March 24, 2005
The cardiovascular reaction to hypoxaemia is an increase in cardiac output (CO), achieved mainly by increasing heart rate (HR) and contractility, and by decreasing systemic vascular resistance (SVR). This response maintains oxygen delivery (DO2) in a situation where arterial oxygen content is reduced. Hypercarbia has similar effects on HR and contractility but may increase SVR, thereby raising mean arterial pressure (MAP). The effects of isolated hypoxaemia have been studied in numerous models but data on hypercarbia and the combined effects are limited. The present study design arose from the assumption that in acute pulmonary failure without intervention there would be hypoxaemia and hypercarbia. Thus, the real physiological response to ventilatory impairment, is most likely to be produced by combined rather than isolated failure. The respective circulatory mechanisms are inhibited by most general anaesthetics, which tend to blunt the autonomic
cardiovascular reflexes and, in addition, diminish cardiac inotropy and induce vasodilation. Consequently, the increase in CO may be absent or reduced during anaesthesia, and may not maintain DO2. A multicentre study1 suggests that xenon anaesthesia shows cardiovascular stability and appears to have less effect on autonomic reflexes than volatile anaesthetics. In vitro, it has been shown that xenon does not have the inhibitory effects on myocardial calcium channels and the contractile force of isolated muscle bundles produced by volatile anaesthetics.2 3 Experimental studies suggest that xenon had no influence on LV performance,4 5 as does the only clinical study using transoesophageal echocardiography (TOE) published so far.6 This led to the hypothesis that with xenon anaesthesia there would be larger increases in cardiac output and inotropy than with a standard anaesthetic during conditions of acute hypoventilation.
# The Board of Management and Trustees of the British Journal of Anaesthesia 2005. All rights reserved. For Permissions, please e-mail: [email protected]
Downloaded from http://bja.oxfordjournals.org/ at UCSF Library on March 22, 2015
Methods. Twenty pigs received anaesthesia with xenon 0.55 MAC/remifentanil 0.5 mg kg1 min1 (group X, n=10) or isoflurane 0.55 MAC/remifentanil 0.5 mg kg1min1 (group I, n=10). CO, heart rate (HR), mean arterial pressure (MAP) and left ventricular fractional area change (FAC) were measured at baseline, after 5 and 15 min of hypoventilation and after 5, 15 and 30 min of restored ventilation.
Xenon anaesthesia and hypoventilation
Methods The above hypothesis was tested in an experimental study in pigs. Circulatory changes induced by hypoventilation (and its reversal) with xenon anaesthesia were compared with those with isoflurane as a standard anaesthetic, in a prospective, randomized, single-blind design, in two groups of 10 animals. Blinding of the investigators during the experiment was not possible because of respirator technology, and was only achieved for TOE tape processing. The primary target criteria were CO, HR, MAP and left ventricular (LV) systolic function. Secondary criteria were systemic oxygen consumption (VO2) and delivery (DO2), mean pulmonary artery pressure (PAP), systemic and pulmonary vascular resistance (SVR, PVR), and LV diastolic function (estimated by filling pressures CVP and PAOP and trans-mitral diastolic flow).
Following approval by the local animal care authorities (Reg. Pra¨s. Ko¨ln), female German Landrace pigs (30–35 kg) were investigated. After overnight fasting, pigs were premedicated by i.m. injection of azaperone 4–5 mg kg1. A 20-gauge cannula was inserted in an ear vein, and anaesthesia was induced by injection of propofol 2–3 mg kg1, as required for orotracheal intubation, using a 7.0 mm internal diameter cuffed tube. Neuromuscular blocking drugs were not used. Ringer’s solution 6 ml kg1 h1 was infused and the urinary bladder was catheterized. Ventilation was with pure oxygen using a closed circuit (PhysioFlex; Dra¨ger, Lu¨beck, Germany), a tidal volume of 10 ml kg1 and a respiratory rate sufficient to maintain an end-tidal PCO2 of 4.9–5.7 kPa. An arterial line was inserted percutaneously into a femoral artery, and an 8.5 F introducer sheath was placed in a femoral vein. A pulmonary artery catheter was advanced, and correct position was confirmed by obtaining pulmonary artery (PAP) and occlusion pressure (PAOP) curves.
Maintenance of anaesthesia During preparation, anaesthesia was maintained with repeated bolus injections of propofol 2–3 mg kg1 and continuous infusion of remifentanil 0.5 mg kg1 min1. After preparation was completed, the Fio2 was reduced to 0.21. Animals were randomly allocated to groups X and I: In group X, anaesthesia was maintained by adding xenon 65–68% to the respirator gas. In group I, isoflurane was added at an end-tidal concentration of 0.95–1.05%. As previously reported, these are equivalent to about 0.55 MAC for xenon and isoflurane in pigs.7 The animals were not restrained and none showed any spontaneous movement throughout the experiment.
Data collection HR, MAP, PAP and PAOP were monitored using a Datex AS/3 anaesthesia monitor (Datex-Engstrom, Helsinki, Finland). Values, averaged over 30 s, were taken every
DO2 ðml min-1 Þ¼ðSao2 ·1:34·Hb+Pao2 ·0:003Þ·CO VO2 ðml min-1 Þ¼avDO2 ·CO where Sao2 is arterial oxygen saturation (ranging from 0 to 1), Hb is arterial haemoglobin concentration (g dl1), Pao2 is arterial oxygen partial pressure (mm Hg), and avDO2 is the arteriovenous oxygen difference.
Echocardiography Before starting the protocol, an Omniplane II TOE probe connected to a Sonos 5500 machine (Philips, Leiden, The Netherlands) was placed into the distal oesophagus. Thus, a long-axis left atrial/left ventricular (LA/LV) view was obtained and 15 cycles recorded on videotape. Without moving the probe, transmitral flow was visualized in the same long-axis view, using a continuous-wave Doppler signal, and another 15 cycles were recorded. Videotapes were examined retrospectively by two independent anaesthetists with special TOE training who were blinded to the anaesthetic used. From early diastolic (E) and atrial (A) filling peak flows, E/A ratio was calculated as a measure of diastolic LV function. A modified fractional area change (FAC) was calculated by dividing the difference between end-diastolic (EDA) and end-systolic (ESA) LV area by EDA, using mean values of planimetry from three consecutive cycles. The long-axis view was used to avoid moving the TOE probe between the recordings. Standard (short-axis) FAC may be underestimated by doing so but relative changes over time should not be affected. EDA was taken as a measure of LV preload. For estimating afterload, the use of LV end-systolic wall stress (LVESWS) has been proposed. Because this again would have made changes in probe position necessary and wall thickness should not be important in healthy young animals, we used the closely related end-systolic pressure–area product (ESPA), as has been suggested by Greim and colleagues.8
167
Downloaded from http://bja.oxfordjournals.org/ at UCSF Library on March 22, 2015
Animal preparation
5 min and stored on a personal computer. CO was measured by thermodilution using injection of 10 ml of Ringer’s solution (temperature of 8 C) into the right atrium. Mean values from three consecutive measurements were stored, and SVR and PVR were calculated. End-tidal concentrations of oxygen, carbon dioxide and isoflurane were monitored using infrared spectroscopy and recorded every 5 min. Xenon concentration was monitored by thermoconductive analysis in the inspired gas. The PhysioFlex respirator mixes expired and fresh gas at 70 litre min1. As xenon uptake after the initial wash-in is less than 30 ml min1, inspired and end-tidal xenon concentrations are virtually identical. At each point of data collection, arterial and mixed venous blood gas analyses were performed using a Radiometer ABL 100/ABL 500 analyser (Radiometer Copenhagen, Copenhagen, Denmark). Arterial PO2, PCO2, pH, arterial and mixed venous oxygen saturation, haemoglobin concentration and haematocrit were stored on a personal computer. DO2 and VO2 were calculated as follows:
Baumert et al.
Table 1 Arterial PO2 and PCO2 at baseline, 5 and 15 min of hypoventilation (hypo), and 5, 15, and 30 min after ventilation had been restored to baseline level (norm). Mean values (SD) of n=10 in each group. #P<0.001 compared with baseline; *P<0.05 between groups (interaction with group variable significant; ANOVA) PO2 (kPa) Group
Group X
Group I
Group X
Group I
Baseline 5 min hypo 15 min hypo 5 min norm 15 min norm 30 min norm
12 (1.2) 7.0 (1.1)# 6.7 (1.4)# 11 (1.4)# 11 (1.3)# 11 (1.3)#
12 (1.7) 7.0 (0.7)# 6.6 (0.6)# 10 (0.9)# 11 (0.9)# 11 (0.9)#
5.6 7.2 7.7 6.5 6.3 6.1
5.2 6.4 7.0 5.8 5.4 5.5
(0.5) (0.5)# (0.8)# (0.7)# (0.7)# * (0.6)#
(0.4) (0.6)# (0.7)# (0.5)# (0.4)# (0.4)#
6.0 5.5
Statistics After testing for normal distribution, data were analysed using two-way repeated measures analysis of variance (ANOVA). In an explorative approach, the effect of the intervention (time variable or within-subjects effect) was investigated first. At that stage, P<0.05 indicated that, at the respective point of time, there was a significant change from baseline for all animals together. In the second step, the effect of the anaesthetic (group variable or between-subjects effect) on these changes was tested, using a post hoc contrast analysis. Again, P<0.05 for the influence of this group variable indicated a significant interaction; that is, the effect of the intervention was significantly different between the two groups (Generalized Linear Model [GLM] procedure; SAS software, Cary, NC, USA).
PCO2 (kPa)
Isoflurane Xenon
5.0 4.5 4.0
*
3.5 3.0 2.5 2.0 Base
Hyp 5 Hyp 15 Norm 5 Norm 15 Norm 30 Data point
Fig 1 Cardiac output (CO) in both groups (n=10 each, mean and SD). *Significantly greater decrease in group X, P<0.05; base, baseline; hyp 5, hyp 15: after 5 and 15 min of hypoventilation respectively; norm 5, norm 15, norm 30: after 5, 15 and 30 min of reset normal ventilation respectively.
Results 140 120 Heart rate (beats min–1)
During hypoventilation, arterial PO2 was reduced by about 40% in both groups, with little progression after the first 5 min. In contrast, PCO2 increased progressively, by 27% after 5 min and 37% after 15 min, with a small group difference. Both variables returned towards normal but did not reach control values after reversal of hypoventilation. The level of arterial PCO2 was slightly higher with xenon (Table 1). CO was increased significantly during hypoventilation, with no significant difference between the groups [mean relative increase 10% (5 min) and 17% (15 min) with isoflurane, and 12 and 25% respectively for xenon] (Fig. 1). This was the result of a corresponding increase in HR (12 and 15% with isoflurane vs 14 and 28% with xenon), again not significantly different between groups (Fig. 2). With a parallel reduction in SVR, MAP declined during hypoventilation, but showed a short-lasting increase after return to normal ventilation and SVR had returned to baseline level. At that time, CO was still elevated, remaining higher than at baseline in group I. There was a significant increase in PVR during
Isoflurane Xenon
*
100
*
80 60 40 20 Base
Hyp 5 Hyp 15 Norm 5 Norm 15 Norm 30 Data point
Fig 2 Heart rate (HR) in both groups (n=10 each, mean and SD). *Significant increase compared with baseline in both groups, P<0.05; base, baseline; hyp 5, hyp 15: after 5 and 15 min of hypoventilation respectively; norm 5, norm 15, norm 30: after 5, 15 and 30 min of reset normal ventilation respectively.
168
Downloaded from http://bja.oxfordjournals.org/ at UCSF Library on March 22, 2015
The protocol was started no earlier than 60 min after target concentrations of xenon 65–68% (group X) or isoflurane 0.95–1.05% (group I) had been reached. This was at least 3 h after premedication, and at least 1.5 h after the last administration of propofol. To produce hypoventilation, the respiratory rate was set to about 45% of the control value, with target arterial PO2 and PCO2 values around 7.3 kPa. This setting (with Fio2 0.21, tidal volume 10 ml kg1 as before) was kept for 15 min. Our preliminary experiments with these conditions found marked hypoxaemia without progressive haemodynamic instability. After 15 min, respiratory rate was set back to baseline value. The time course of data collection points was: baseline, 10 min before hypoventilation (1), after 5 (2) and 15 (3) min of hypoventilation, and 5 (4), 15 (5) and 30 min (6) after returning to baseline respiratory rate.
Cardiac output (litre min–1)
Study protocol
Xenon anaesthesia and hypoventilation
Table 2 Mean arterial pressure (MAP), mean pulmonary artery pressure (PAP), central venous pressure (CVP) and pulmonary occlusion pressure (PAOP). Mean (SD). # P<0.05 compared with baseline; *P<0.05 between groups (group variable significant; ANOVA); hypo, hypoventilation with 45% of baseline respiratory minute volume; norm, restoration of baseline ventilator settings MAP (mm Hg)
Baseline 5 min hypo 15 min hypo 5 min norm 15 min norm 30 min norm
PAP (mm Hg)
CVP (mm Hg)
PAOP (mm Hg)
Group X
Group I
Group X
Group I
Group X
Group I
Group X
Group I
74 (10) 72# (12) 73 (12) 81# (12) 79# (12) 78 (14)
77 (11) 74# (9.8) 73 (10) 79# (11) 79# (11) 78 (12)
24 (6.6) 34# (8.9) 36# (7.8) 29# (8.1) 28 (8.1) 30#* (8.0)
21 (6.8) 29# (8.7) 32# (6.5) 24# (6.0) 23 (5.5) 22 (6.6)
9.6 (2.6) 9.1 (3.7) 10 (2.2) 10 (2.7) 10 (2.4) 10 (2.7)
7.5 8.2 8.2 7.9 7.7 8.0
11 11 11 12 11 12
10 10 11 11 11 11
(3.6) (3.3) (3.2) (3.9) (4.2) (3.7)
(3.0) (2.5) (2.2) (3.0) (2.8) (3.1)
(3.0) (3.8) (3.6) (4.6) (4.8) (4.9)
Table 3 Systemic (SVR) and pulmonary (PVR) vascular resistance, oxygen delivery (DO2) and total oxygen consumption (VO2). Mean (SD) of n=10 in each group. # P<0.05 compared with baseline; *P<0.05 between groups (group variable significant; ANOVA); hypo, hypoventilation with 45% of baseline respiratory minute volume; norm, restoration of baseline ventilator settings PVR (dyn s cm5)
DO2 (ml min1)
Group X
Group I
Group X
Group I
Group X
Group I
Group X
Group I
1630 (338) 1420# (318) 1280# (276) 1730*(337) 1730 (359) 1740 (471)
1550 (354) 1350# (291) 1240# (262) 1430 (410) 1440 (496) 1410 (538)
318 (147) 517# (192) 523# (178) 438# (181) 435#* (223) 491#* (239)
257 (186) 388# (204) 410# (142) 281# (166) 248 (151) 250 (227)
377 (48.3) 365 (44.4) 395 (70.1) 396# (65.2) 374 (59.6) 369 (57.4)
441 (106) 418 (110) 426 (104) 481# (122) 485 (135) 480 (111)
119 (12.5) 153# (33.2) 169# (39.6) 123 (13.3) 125 (16.3) 129 (11.9)
107 (21.2) 143# (28.0) 148# (19.4) 126 (24.6) 129 (19.9) 125 (13.7)
hypoventilation in both groups which normal ventilation reversed only in the isoflurane group. DO2 was virtually unchanged in spite of persisting hypoventilation and hypoxaemia (5 and 15 min), without a difference between the groups. VO2 was increased significantly during hypoventilation and returned to control level after ventilation had been reset, with no difference between the groups (Tables 2 and 3).
0.85
0.75
*
FAC
0.70 0.65 0.60 0.55
FAC was increased significantly at time points 2, 3 and 4, with a significantly greater increase in group X (Fig. 3). There were no significant changes in EDA during hypoventilation. ESPA was decreased significantly in both groups during hypoventilation and returned to baseline levels with normal ventilation. E/A ratio was decreased in both groups and remained significantly lower than at baseline throughout, with a 20% increase in atrial filling peak flow and unchanged early filling peak flow (Table 4).
0.50
The circulatory effect of hypoventilation was an increase in HR, raising CO, sufficient to keep MAP stable during a decrease in SVR. There was no difference between groups with the exception of a persistent elevation of CO in group I. Left ventricular FAC was increased, along with a
Isoflurane Xenon
0.80
Echocardiography
Discussion
VO2 (ml min1)
0.45 Base
Hyp 5 Hyp 15 Norm 5 Norm 15 Norm 30 Data point
Fig 3 Fractional area change (FAC) by TOE in both groups (n=10 each, mean and SD); increase from baseline significant for all animals together from hyp 5 to norm 5 (not indicated). *Significantly greater increase in group X, P<0.05; base, baseline; hyp 5, hyp 15: after 5 and 15 min of hypoventilation respectively; norm 5, norm 15, norm 30: after 5, 15 and 30 min of reset normal ventilation respectively.
decrease in afterload (ESPA) during hypoventilation. The early increase in FAC was significantly greater in group X. Systemic oxygen delivery was maintained and consumption was increased during hypoventilation, but returned to baseline with normal ventilation, with both xenon and isoflurane.
169
Downloaded from http://bja.oxfordjournals.org/ at UCSF Library on March 22, 2015
Baseline 5 min hypo 15 min hypo 5 min norm 15 min norm 30 min norm
SVR (dyn s cm5)
Baumert et al.
Table 4 Left ventricular end-diastolic area (EDA), end-systolic pressure–area product (ESPA), early (E) and atrial (A) diastolic filling peak flows, and respective ratio (E/A). Mean (SD) of n=10 in each group. #P<0.05 compared with baseline (ANOVA); hypo, hypoventilation with 45% of baseline respiratory minute volume; norm, restoration of baseline ventilator settings EDA (cm2)
Baseline 5 min hypo 15 min hypo 5 min norm 15 min norm 30 min norm
ESPA (cm2 mm Hg)
E peak flow (cm s1)
A peak flow (cm s1)
E/A ratio
Group X
Group I
Group X
Group I
Group X
Group I
Group X
Group I
Group X
Group I
17.3 16.2 15.6 17.9 17.9 16.9
17.5 16.4 16.8 17.2 17.4 17.6
749 (248) 574# (126) 594# (155) 700 (188) 717 (160) 679 (122)
791 (274) 707# (267) 677# (216) 800 (349) 784 (309) 830 (367)
60.3 57.6 59.2 56.9 57.0 55.0
58.3 57.6 58.7 59.5 61.5 58.6
23.6 (4.0) 29.9# (4.8) 33.9# (5.0) 28.4# (6.4) 26.7# (3.9) 27.2# (6.0)
31.1 (11.1) 37.4# (12.8) 39.4# (15.9) 35.3# (14.1) 33.5# (14.1) 36.1# (14.3)
2.58 (0.76) 2.00# (0.62) 1.77# (0.48) 2.05# (0.47) 2.17# (0.54) 2.05# (0.33)
2.04 (0.76) 1.62# (0.44) 1.49# (0.37) 1.77# (0.33) 1.85# (0.41) 1.72# (0.45)
(4.1) (4.5) (2.9) (3.8) (3.9) (3.7)
(3.4) (2.8) (2.8) (3.4) (3.5) (4.5)
(17.9) (13.0) (17.5) (15.3) (16.2) (14.2)
Xenon does not appear to affect contractility18 19 or to impair left ventricular performance, in spite of an increased SVR.4 20 Consequently, the increase in FAC is regarded as the unimpaired reaction to an acute decrease in afterload. Altogether, the changes in HR, CO and FAC induced by hypoventilation, and their reversal with normal ventilation, appear faster with xenon. This is in agreement with our hypothesis that xenon produces less inhibition of the physiological response to acute ventilatory impairment. However, the changes are minimal and probably of limited clinical relevance. A missing feature of the response to hypoventilation was the increase in LV preload, which was unexpected, as venous constriction is reported to be one of the compensating mechanisms in hypoxia.17 Together with unchanged early and increased atrial filling peak flows, this finding indicates an alteration of diastolic LV filling. There was a consistent, significant increase of 20–40% in atrial peak flow, possibly caused by the rise in HR, but which persisted after HR had returned to baseline level. Such an effect has only been described with acute myocardial ischaemia21 but not with hypoxia/hypercarbia. However, the normal value of E/A ratio in pigs is unknown, and the method is affected by various technical problems concerning the quality of the Doppler signal. Thus, it is not clear if the decrease we found is of functional relevance. The main limitation of this study may be that the degree of hypoventilation was insufficient to reduce oxygen delivery. However, DO2 was kept stable by the increase in CO despite a 30% reduction in arterial oxygen content, and VO2 was increased. This suggests that the autonomic response to the decrease of almost 50% in arterial PO2 and the 40% increase in PCO2 is sufficient to prevent hypoxia, at this level of hypoxaemia. It would have been interesting to compare different degrees of hypoventilation, but our preliminary studies demonstrated that more pronounced hypoventilation is associated with marked haemodynamic instability. We therefore decided that our target was the circulatory effect of hypoventilation without sustained hypoxic or hypercarbic damage to the heart. As the MAC of xenon exceeds 100% in pigs,22 it could not be used as the sole anaesthetic. We used equipotent
170
Downloaded from http://bja.oxfordjournals.org/ at UCSF Library on March 22, 2015
Hypoxaemia and hypercarbia appear to have conflicting effects on the circulation. While tachycardia and an increase in cardiac output are attributed to both, hypoxaemia is generally accompanied by vasodilation,9 10 which increases cerebral and myocardial blood flow.11 The early circulatory response is a result of autonomic nervous system activation by vascular oxygen receptors which respond to the concentration of dissolved oxygen. Hypoxic inhibition of pig tracheal smooth muscle contraction has been demonstrated in vitro,12 but there are no data on vascular smooth muscle and the mechanism is not clear. Nevertheless, a reduction in LV afterload was demonstrated using direct echocardiography in pigs.9 A negative correlation (r=0.56) between LV end-systolic wall stress and ejection fraction area (i.e. FAC) was reported, but there was no correlation between LVESWS and SVR. In contrast to the above findings, hypercarbia alone provoked hypertension, and thus the combined effect will be difficult to predict. It has been suggested that the circulatory effects of hypoxia are likely to be potentiated by hypercarbia,13 but there are no data available on which the relative contributions to a net effect can be based. Isoflurane has been shown not to impair the chemosensitivity of the carotid body (0.5 MAC),14 the hypoxic relaxation of the aorta (1 MAC)10 or the hypoxic increase in hepatic blood flow.15 Surprisingly, in this latter study hypoxia alone decreased cardiac index, but this is the only available report of such an effect. Isoflurane also seems to inhibit other compensatory mechanisms during hypoxia, e.g. increases in cerebral and coronary blood flow16 as well as venous contraction and—most importantly—the increase in sympathetic outflow.17 There are no reports of xenon in this setting. The haemodynamic effects of hypoventilation in our study are in agreement with most other reports and were similar with xenon and isoflurane anaesthesia, except for the more rapid decrease in CO with xenon after restoring ventilation. There was an increase in FAC with xenon but almost no increase with isoflurane. This may be a direct myocardial effect, where, in spite of a decrease in LV afterload, FAC could not be increased. Xenon does not inhibit myocardial function at the cellular2 or muscle fibre level.3
(15.9) (18.7) (14.0) (18.0) (19.6) (20.4)
Xenon anaesthesia and hypoventilation
11
Acknowledgements
16
8
9
10
12
13 14
15
The work was supported by the institutional research fund and Air Liquide Deutschland GmbH, Krefeld/Germany (xenon gas supply).
17
References 1 Rossaint R, Reyle-Hahn M, Schulte am Esch J, et al. Multicenter randomized comparison of the efficacy and safety of xenon and isoflurane in patients undergoing elective surgery. Anesthesiology 2003; 98: 6–13 2 Huneke R, Jungling E, Skasa M, Rossaint R, Luckhoff A. Effects of the anesthetic gases xenon, halothane, and isoflurane on calcium and potassium currents in human atrial cardiomyocytes. Anesthesiology 2001; 95: 999–1006 3 Schroth SC, Schotten U, Alkanoglu O, Reyle-Hahn MS, Hanrath P, Rossaint R. Xenon does not impair the responsiveness of cardiac muscle bundles to positive inotropic and chronotropic stimulation. Anesthesiology 2002; 96: 422–7 4 Hettrick DA, Pagel PS, Kersten JR, et al. Cardiovascular effects of xenon in isoflurane-anesthetized dogs with dilated cardiomyopathy. Anesthesiology 1998; 89: 1166–73 5 Stowe DF, Rehmert GC, Kwok WM, Weigt HU, Georgieff M, Bosnjak ZJ. Xenon does not alter cardiac function or major cation currents in isolated guinea pig hearts or myocytes. Anesthesiology 2000; 92: 516–22 6 Luttropp HH, Romner B, Perhag L, Eskilsson J, Fredriksen S, Werner O. Left ventricular performance and cerebral haemodynamics during xenon anaesthesia. A transoesophageal echocardiography and transcranial Doppler sonography study. Anaesthesia 1993; 48: 1045–9 7 Hecker KE, Reyle-Hahn M, Baumert JH, Horn N, Heussen N, Rossaint R. Minimum alveolar anesthetic concentration of
171
18
19
20
21
22
23
24
25
isoflurane with different xenon concentrations in swine. Anesth Analg 2003; 96: 119–24 Greim CA, Roewer N, Schulte am EJ. Assessment of changes in left ventricular wall stress from the end-systolic pressure–area product. Br J Anaesth 1995; 75: 583–7 Vedrinne JM, Curtil A, Martinot S, et al. The hemodynamic effects of hypoxemia in anesthetized pigs: a comparison between right heart catheter and echocardiography. Anesth Analg 1998; 87: 21–6 Haddad E, Lebuffe G, Boillot A, Imbenotte M, Vallet B. Does halothane or isoflurane affect hypoxic and post-hypoxic vascular response in rabbit aorta? Acta Anaesth Scand 2000; 44: 423–8 Duong TQ, Iadecola C, Kim SG. Effect of hyperoxia, hypercapnia, and hypoxia on cerebral interstitial oxygen tension and cerebral blood flow. Magn Reson Med 2001; 45: 61–70 Chen X, Yamakage M, Tsujiguchi N, Kamada Y, Namiki A. Interaction between volatile anesthetics and hypoxia in porcine tracheal smooth muscle. Anesth Analg 2000; 91: 996–1002 Nunn JF. Hypoxia. In: Nunn JF, ed. Applied Respiratory Physiology, 3rd edn. London: Butterworths, 1987: 471–7 Joensen H, Sadler CL, Ponte J, Yamamoto Y, Lindahl SG, Eriksson LI. Isoflurane does not depress the hypoxic response of rabbit carotid body chemoreceptors. Anesth Analg 2000; 91: 480–5 Cray SH, Crawford MW, Khayyam N, Carmichael FJ. Effects of hypoxia and isoflurane on liver blood flow: the role of adenosine. Br J Anaesth 2001; 86: 425–7 Durieux ME, Sperry RJ, Longnecker DE. Effects of hypoxemia on regional blood flows during anesthesia with halothane, enflurane, or isoflurane. Anesthesiology 1992; 76: 402–8 Stekiel TA, Stekiel WJ, Tominaga M, Stadnicka A, Bosnjak ZJ, Kampine JP. Isoflurane-mediated inhibition of the constriction of mesenteric capacitance veins and related circulatory responses to acute graded hypoxic hypoxia. Anesth Analg 1995; 80: 994–1001 Nakayama H, Takahashi H, Okubo N, Miyabe M, Toyooka H. Xenon and nitrous oxide do not depress cardiac function in an isolated rat heart model. Can J Anaesth 2002; 49: 375–9 Preckel B, Schlack W, Heibel T, Rutten H. Xenon produces minimal haemodynamic effects in rabbits with chronically compromised left ventricular function. Br J Anaesth 2002; 88: 264–9 Picker O, Schindler AW, Schwarte LA, et al. Xenon increases total body oxygen consumption during isoflurane anaesthesia in dogs. Br J Anaesth 2002; 88: 546–54 Labovitz AJ, Lewen MK, Kern M, Vandormael M, Deligonal U, Kennedy HL. Evaluation of left ventricular systolic and diastolic dysfunction during transient myocardial ischemia produced by angioplasty. J Am Coll Cardiol 1987; 10: 748–55 Hecker KE, Horn N, Baumert JH, Reyle-Hahn SM, Heussen N, Rossaint R. Minimum alveolar concentration (MAC) of xenon in intubated swine. Br J Anaesth 2004; 92: 421–4 Schuttler J, Albrecht S, Breivik H, et al. A comparison of remifentanil and alfentanil in patients undergoing major abdominal surgery. Anaesthesia 1997; 52: 307–17 Shinohara K, Aono H, Unruh GK, Kindscher JD, Goto H. Suppressive effects of remifentanil on hemodynamics in baro-denervated rabbits. Can J Anaesth 2000; 47: 361–6 Ishibe Y, Gui X, Uno H, Shiokawa Y, Umeda T, Suekane K. Effect of sevoflurane on hypoxic pulmonary vasoconstriction in the perfused rabbit lung. Anesthesiology 1993; 79: 1348–53
Downloaded from http://bja.oxfordjournals.org/ at UCSF Library on March 22, 2015
doses of 0.55 MAC for xenon and isoflurane, supplemented with remifentanil, which, because of its known vagomimetic effects,23 24 may have influenced our results and may have resulted in different levels of anaesthesia. As there is no gold standard for measuring anaesthetic depth in pigs, we used the absence of spontaneous movement in our nonparalysed animals and took this as an indicator of adequate anaesthesia. It is possible that 0.55 MAC may be too low a concentration to produce cardiovascular effects. However, there are reports of inhibition of hypoxic pulmonary vasoconstriction (HPV)25 and efferent sympathetic activity17 even with subMAC doses of isoflurane. As the concentrations of volatile (or gaseous) anaesthetics in combination with an opioid, as used in the present study, correspond to a clinical situation, we believe that this design is adequate. In summary, hypoventilation to moderate hypoxaemia and hypercarbia produced similar circulatory effects during xenon and isoflurane anaesthesia. Although the response appeared faster with xenon, our initial hypothesis that the increase in cardiac output would be greater with this agent is not supported.
doi:10.1093/bja/aei153
Advance Access publication May 20, 2005
Effects of xenon anaesthesia on the circulatory response to hypoventilation J.-H. Baumert*, K. E. Hecker, M. Hein, M. Reyle-Hahn1, N. A. Horn and R. Rossaint Anaesthesiology Clinic, Universitaetsklinikum Aachen, Germany. 1Anaesthesia Department, Waldkrankenhaus Berlin-Spandau, Germany *Corresponding author: Anaesthesiology Clinic, Universitaetsklinikum Aachen, Pauwelsstrasse 30, 52074 Aachen, Germany. E-mail: [email protected] Background. Circulatory response to hypoventilation is aimed at eliminating carbon dioxide and maintaining oxygen delivery (DO2) by increasing cardiac output (CO). The hypothesis that this increase is more pronounced with xenon than with isoflurane anaesthesia was tested in pigs.
Results. CO increased by 10–20% with both anaesthetics, with an equivalent rise in HR, maintaining DO2 in spite of a 20% reduction in arterial oxygen content. Decreased left ventricular (LV) afterload during hypoventilation increased FAC, and this was more marked with xenon (0.60– 0.66, P<0.05 compared with baseline and isoflurane). This difference is attributed to negative inotropic effects of isoflurane. Increased pulmonary vascular resistance during hypoventilation was found with both anaesthetics. Conclusion. The cardiovascular effects observed in this model of moderate hypoventilation were sufficient to maintain DO2. Although the haemodynamic response appeared more pronounced with xenon, differences were not clinically relevant. An increase in FAC with xenon is attributed to its lack of negative inotropic effects. Br J Anaesth 2005; 95: 166–71 Keywords: anaesthetics gases, xenon; complications, hypercarbia; complications, hypoxaemia; monitoring, trans-oesophageal echocardiography Accepted for publication: March 24, 2005
The cardiovascular reaction to hypoxaemia is an increase in cardiac output (CO), achieved mainly by increasing heart rate (HR) and contractility, and by decreasing systemic vascular resistance (SVR). This response maintains oxygen delivery (DO2) in a situation where arterial oxygen content is reduced. Hypercarbia has similar effects on HR and contractility but may increase SVR, thereby raising mean arterial pressure (MAP). The effects of isolated hypoxaemia have been studied in numerous models but data on hypercarbia and the combined effects are limited. The present study design arose from the assumption that in acute pulmonary failure without intervention there would be hypoxaemia and hypercarbia. Thus, the real physiological response to ventilatory impairment, is most likely to be produced by combined rather than isolated failure. The respective circulatory mechanisms are inhibited by most general anaesthetics, which tend to blunt the autonomic
cardiovascular reflexes and, in addition, diminish cardiac inotropy and induce vasodilation. Consequently, the increase in CO may be absent or reduced during anaesthesia, and may not maintain DO2. A multicentre study1 suggests that xenon anaesthesia shows cardiovascular stability and appears to have less effect on autonomic reflexes than volatile anaesthetics. In vitro, it has been shown that xenon does not have the inhibitory effects on myocardial calcium channels and the contractile force of isolated muscle bundles produced by volatile anaesthetics.2 3 Experimental studies suggest that xenon had no influence on LV performance,4 5 as does the only clinical study using transoesophageal echocardiography (TOE) published so far.6 This led to the hypothesis that with xenon anaesthesia there would be larger increases in cardiac output and inotropy than with a standard anaesthetic during conditions of acute hypoventilation.
# The Board of Management and Trustees of the British Journal of Anaesthesia 2005. All rights reserved. For Permissions, please e-mail: [email protected]
Downloaded from http://bja.oxfordjournals.org/ at UCSF Library on March 22, 2015
Methods. Twenty pigs received anaesthesia with xenon 0.55 MAC/remifentanil 0.5 mg kg1 min1 (group X, n=10) or isoflurane 0.55 MAC/remifentanil 0.5 mg kg1min1 (group I, n=10). CO, heart rate (HR), mean arterial pressure (MAP) and left ventricular fractional area change (FAC) were measured at baseline, after 5 and 15 min of hypoventilation and after 5, 15 and 30 min of restored ventilation.
Xenon anaesthesia and hypoventilation
Methods The above hypothesis was tested in an experimental study in pigs. Circulatory changes induced by hypoventilation (and its reversal) with xenon anaesthesia were compared with those with isoflurane as a standard anaesthetic, in a prospective, randomized, single-blind design, in two groups of 10 animals. Blinding of the investigators during the experiment was not possible because of respirator technology, and was only achieved for TOE tape processing. The primary target criteria were CO, HR, MAP and left ventricular (LV) systolic function. Secondary criteria were systemic oxygen consumption (VO2) and delivery (DO2), mean pulmonary artery pressure (PAP), systemic and pulmonary vascular resistance (SVR, PVR), and LV diastolic function (estimated by filling pressures CVP and PAOP and trans-mitral diastolic flow).
Following approval by the local animal care authorities (Reg. Pra¨s. Ko¨ln), female German Landrace pigs (30–35 kg) were investigated. After overnight fasting, pigs were premedicated by i.m. injection of azaperone 4–5 mg kg1. A 20-gauge cannula was inserted in an ear vein, and anaesthesia was induced by injection of propofol 2–3 mg kg1, as required for orotracheal intubation, using a 7.0 mm internal diameter cuffed tube. Neuromuscular blocking drugs were not used. Ringer’s solution 6 ml kg1 h1 was infused and the urinary bladder was catheterized. Ventilation was with pure oxygen using a closed circuit (PhysioFlex; Dra¨ger, Lu¨beck, Germany), a tidal volume of 10 ml kg1 and a respiratory rate sufficient to maintain an end-tidal PCO2 of 4.9–5.7 kPa. An arterial line was inserted percutaneously into a femoral artery, and an 8.5 F introducer sheath was placed in a femoral vein. A pulmonary artery catheter was advanced, and correct position was confirmed by obtaining pulmonary artery (PAP) and occlusion pressure (PAOP) curves.
Maintenance of anaesthesia During preparation, anaesthesia was maintained with repeated bolus injections of propofol 2–3 mg kg1 and continuous infusion of remifentanil 0.5 mg kg1 min1. After preparation was completed, the Fio2 was reduced to 0.21. Animals were randomly allocated to groups X and I: In group X, anaesthesia was maintained by adding xenon 65–68% to the respirator gas. In group I, isoflurane was added at an end-tidal concentration of 0.95–1.05%. As previously reported, these are equivalent to about 0.55 MAC for xenon and isoflurane in pigs.7 The animals were not restrained and none showed any spontaneous movement throughout the experiment.
Data collection HR, MAP, PAP and PAOP were monitored using a Datex AS/3 anaesthesia monitor (Datex-Engstrom, Helsinki, Finland). Values, averaged over 30 s, were taken every
DO2 ðml min-1 Þ¼ðSao2 ·1:34·Hb+Pao2 ·0:003Þ·CO VO2 ðml min-1 Þ¼avDO2 ·CO where Sao2 is arterial oxygen saturation (ranging from 0 to 1), Hb is arterial haemoglobin concentration (g dl1), Pao2 is arterial oxygen partial pressure (mm Hg), and avDO2 is the arteriovenous oxygen difference.
Echocardiography Before starting the protocol, an Omniplane II TOE probe connected to a Sonos 5500 machine (Philips, Leiden, The Netherlands) was placed into the distal oesophagus. Thus, a long-axis left atrial/left ventricular (LA/LV) view was obtained and 15 cycles recorded on videotape. Without moving the probe, transmitral flow was visualized in the same long-axis view, using a continuous-wave Doppler signal, and another 15 cycles were recorded. Videotapes were examined retrospectively by two independent anaesthetists with special TOE training who were blinded to the anaesthetic used. From early diastolic (E) and atrial (A) filling peak flows, E/A ratio was calculated as a measure of diastolic LV function. A modified fractional area change (FAC) was calculated by dividing the difference between end-diastolic (EDA) and end-systolic (ESA) LV area by EDA, using mean values of planimetry from three consecutive cycles. The long-axis view was used to avoid moving the TOE probe between the recordings. Standard (short-axis) FAC may be underestimated by doing so but relative changes over time should not be affected. EDA was taken as a measure of LV preload. For estimating afterload, the use of LV end-systolic wall stress (LVESWS) has been proposed. Because this again would have made changes in probe position necessary and wall thickness should not be important in healthy young animals, we used the closely related end-systolic pressure–area product (ESPA), as has been suggested by Greim and colleagues.8
167
Downloaded from http://bja.oxfordjournals.org/ at UCSF Library on March 22, 2015
Animal preparation
5 min and stored on a personal computer. CO was measured by thermodilution using injection of 10 ml of Ringer’s solution (temperature of 8 C) into the right atrium. Mean values from three consecutive measurements were stored, and SVR and PVR were calculated. End-tidal concentrations of oxygen, carbon dioxide and isoflurane were monitored using infrared spectroscopy and recorded every 5 min. Xenon concentration was monitored by thermoconductive analysis in the inspired gas. The PhysioFlex respirator mixes expired and fresh gas at 70 litre min1. As xenon uptake after the initial wash-in is less than 30 ml min1, inspired and end-tidal xenon concentrations are virtually identical. At each point of data collection, arterial and mixed venous blood gas analyses were performed using a Radiometer ABL 100/ABL 500 analyser (Radiometer Copenhagen, Copenhagen, Denmark). Arterial PO2, PCO2, pH, arterial and mixed venous oxygen saturation, haemoglobin concentration and haematocrit were stored on a personal computer. DO2 and VO2 were calculated as follows:
Baumert et al.
Table 1 Arterial PO2 and PCO2 at baseline, 5 and 15 min of hypoventilation (hypo), and 5, 15, and 30 min after ventilation had been restored to baseline level (norm). Mean values (SD) of n=10 in each group. #P<0.001 compared with baseline; *P<0.05 between groups (interaction with group variable significant; ANOVA) PO2 (kPa) Group
Group X
Group I
Group X
Group I
Baseline 5 min hypo 15 min hypo 5 min norm 15 min norm 30 min norm
12 (1.2) 7.0 (1.1)# 6.7 (1.4)# 11 (1.4)# 11 (1.3)# 11 (1.3)#
12 (1.7) 7.0 (0.7)# 6.6 (0.6)# 10 (0.9)# 11 (0.9)# 11 (0.9)#
5.6 7.2 7.7 6.5 6.3 6.1
5.2 6.4 7.0 5.8 5.4 5.5
(0.5) (0.5)# (0.8)# (0.7)# (0.7)# * (0.6)#
(0.4) (0.6)# (0.7)# (0.5)# (0.4)# (0.4)#
6.0 5.5
Statistics After testing for normal distribution, data were analysed using two-way repeated measures analysis of variance (ANOVA). In an explorative approach, the effect of the intervention (time variable or within-subjects effect) was investigated first. At that stage, P<0.05 indicated that, at the respective point of time, there was a significant change from baseline for all animals together. In the second step, the effect of the anaesthetic (group variable or between-subjects effect) on these changes was tested, using a post hoc contrast analysis. Again, P<0.05 for the influence of this group variable indicated a significant interaction; that is, the effect of the intervention was significantly different between the two groups (Generalized Linear Model [GLM] procedure; SAS software, Cary, NC, USA).
PCO2 (kPa)
Isoflurane Xenon
5.0 4.5 4.0
*
3.5 3.0 2.5 2.0 Base
Hyp 5 Hyp 15 Norm 5 Norm 15 Norm 30 Data point
Fig 1 Cardiac output (CO) in both groups (n=10 each, mean and SD). *Significantly greater decrease in group X, P<0.05; base, baseline; hyp 5, hyp 15: after 5 and 15 min of hypoventilation respectively; norm 5, norm 15, norm 30: after 5, 15 and 30 min of reset normal ventilation respectively.
Results 140 120 Heart rate (beats min–1)
During hypoventilation, arterial PO2 was reduced by about 40% in both groups, with little progression after the first 5 min. In contrast, PCO2 increased progressively, by 27% after 5 min and 37% after 15 min, with a small group difference. Both variables returned towards normal but did not reach control values after reversal of hypoventilation. The level of arterial PCO2 was slightly higher with xenon (Table 1). CO was increased significantly during hypoventilation, with no significant difference between the groups [mean relative increase 10% (5 min) and 17% (15 min) with isoflurane, and 12 and 25% respectively for xenon] (Fig. 1). This was the result of a corresponding increase in HR (12 and 15% with isoflurane vs 14 and 28% with xenon), again not significantly different between groups (Fig. 2). With a parallel reduction in SVR, MAP declined during hypoventilation, but showed a short-lasting increase after return to normal ventilation and SVR had returned to baseline level. At that time, CO was still elevated, remaining higher than at baseline in group I. There was a significant increase in PVR during
Isoflurane Xenon
*
100
*
80 60 40 20 Base
Hyp 5 Hyp 15 Norm 5 Norm 15 Norm 30 Data point
Fig 2 Heart rate (HR) in both groups (n=10 each, mean and SD). *Significant increase compared with baseline in both groups, P<0.05; base, baseline; hyp 5, hyp 15: after 5 and 15 min of hypoventilation respectively; norm 5, norm 15, norm 30: after 5, 15 and 30 min of reset normal ventilation respectively.
168
Downloaded from http://bja.oxfordjournals.org/ at UCSF Library on March 22, 2015
The protocol was started no earlier than 60 min after target concentrations of xenon 65–68% (group X) or isoflurane 0.95–1.05% (group I) had been reached. This was at least 3 h after premedication, and at least 1.5 h after the last administration of propofol. To produce hypoventilation, the respiratory rate was set to about 45% of the control value, with target arterial PO2 and PCO2 values around 7.3 kPa. This setting (with Fio2 0.21, tidal volume 10 ml kg1 as before) was kept for 15 min. Our preliminary experiments with these conditions found marked hypoxaemia without progressive haemodynamic instability. After 15 min, respiratory rate was set back to baseline value. The time course of data collection points was: baseline, 10 min before hypoventilation (1), after 5 (2) and 15 (3) min of hypoventilation, and 5 (4), 15 (5) and 30 min (6) after returning to baseline respiratory rate.
Cardiac output (litre min–1)
Study protocol
Xenon anaesthesia and hypoventilation
Table 2 Mean arterial pressure (MAP), mean pulmonary artery pressure (PAP), central venous pressure (CVP) and pulmonary occlusion pressure (PAOP). Mean (SD). # P<0.05 compared with baseline; *P<0.05 between groups (group variable significant; ANOVA); hypo, hypoventilation with 45% of baseline respiratory minute volume; norm, restoration of baseline ventilator settings MAP (mm Hg)
Baseline 5 min hypo 15 min hypo 5 min norm 15 min norm 30 min norm
PAP (mm Hg)
CVP (mm Hg)
PAOP (mm Hg)
Group X
Group I
Group X
Group I
Group X
Group I
Group X
Group I
74 (10) 72# (12) 73 (12) 81# (12) 79# (12) 78 (14)
77 (11) 74# (9.8) 73 (10) 79# (11) 79# (11) 78 (12)
24 (6.6) 34# (8.9) 36# (7.8) 29# (8.1) 28 (8.1) 30#* (8.0)
21 (6.8) 29# (8.7) 32# (6.5) 24# (6.0) 23 (5.5) 22 (6.6)
9.6 (2.6) 9.1 (3.7) 10 (2.2) 10 (2.7) 10 (2.4) 10 (2.7)
7.5 8.2 8.2 7.9 7.7 8.0
11 11 11 12 11 12
10 10 11 11 11 11
(3.6) (3.3) (3.2) (3.9) (4.2) (3.7)
(3.0) (2.5) (2.2) (3.0) (2.8) (3.1)
(3.0) (3.8) (3.6) (4.6) (4.8) (4.9)
Table 3 Systemic (SVR) and pulmonary (PVR) vascular resistance, oxygen delivery (DO2) and total oxygen consumption (VO2). Mean (SD) of n=10 in each group. # P<0.05 compared with baseline; *P<0.05 between groups (group variable significant; ANOVA); hypo, hypoventilation with 45% of baseline respiratory minute volume; norm, restoration of baseline ventilator settings PVR (dyn s cm5)
DO2 (ml min1)
Group X
Group I
Group X
Group I
Group X
Group I
Group X
Group I
1630 (338) 1420# (318) 1280# (276) 1730*(337) 1730 (359) 1740 (471)
1550 (354) 1350# (291) 1240# (262) 1430 (410) 1440 (496) 1410 (538)
318 (147) 517# (192) 523# (178) 438# (181) 435#* (223) 491#* (239)
257 (186) 388# (204) 410# (142) 281# (166) 248 (151) 250 (227)
377 (48.3) 365 (44.4) 395 (70.1) 396# (65.2) 374 (59.6) 369 (57.4)
441 (106) 418 (110) 426 (104) 481# (122) 485 (135) 480 (111)
119 (12.5) 153# (33.2) 169# (39.6) 123 (13.3) 125 (16.3) 129 (11.9)
107 (21.2) 143# (28.0) 148# (19.4) 126 (24.6) 129 (19.9) 125 (13.7)
hypoventilation in both groups which normal ventilation reversed only in the isoflurane group. DO2 was virtually unchanged in spite of persisting hypoventilation and hypoxaemia (5 and 15 min), without a difference between the groups. VO2 was increased significantly during hypoventilation and returned to control level after ventilation had been reset, with no difference between the groups (Tables 2 and 3).
0.85
0.75
*
FAC
0.70 0.65 0.60 0.55
FAC was increased significantly at time points 2, 3 and 4, with a significantly greater increase in group X (Fig. 3). There were no significant changes in EDA during hypoventilation. ESPA was decreased significantly in both groups during hypoventilation and returned to baseline levels with normal ventilation. E/A ratio was decreased in both groups and remained significantly lower than at baseline throughout, with a 20% increase in atrial filling peak flow and unchanged early filling peak flow (Table 4).
0.50
The circulatory effect of hypoventilation was an increase in HR, raising CO, sufficient to keep MAP stable during a decrease in SVR. There was no difference between groups with the exception of a persistent elevation of CO in group I. Left ventricular FAC was increased, along with a
Isoflurane Xenon
0.80
Echocardiography
Discussion
VO2 (ml min1)
0.45 Base
Hyp 5 Hyp 15 Norm 5 Norm 15 Norm 30 Data point
Fig 3 Fractional area change (FAC) by TOE in both groups (n=10 each, mean and SD); increase from baseline significant for all animals together from hyp 5 to norm 5 (not indicated). *Significantly greater increase in group X, P<0.05; base, baseline; hyp 5, hyp 15: after 5 and 15 min of hypoventilation respectively; norm 5, norm 15, norm 30: after 5, 15 and 30 min of reset normal ventilation respectively.
decrease in afterload (ESPA) during hypoventilation. The early increase in FAC was significantly greater in group X. Systemic oxygen delivery was maintained and consumption was increased during hypoventilation, but returned to baseline with normal ventilation, with both xenon and isoflurane.
169
Downloaded from http://bja.oxfordjournals.org/ at UCSF Library on March 22, 2015
Baseline 5 min hypo 15 min hypo 5 min norm 15 min norm 30 min norm
SVR (dyn s cm5)
Baumert et al.
Table 4 Left ventricular end-diastolic area (EDA), end-systolic pressure–area product (ESPA), early (E) and atrial (A) diastolic filling peak flows, and respective ratio (E/A). Mean (SD) of n=10 in each group. #P<0.05 compared with baseline (ANOVA); hypo, hypoventilation with 45% of baseline respiratory minute volume; norm, restoration of baseline ventilator settings EDA (cm2)
Baseline 5 min hypo 15 min hypo 5 min norm 15 min norm 30 min norm
ESPA (cm2 mm Hg)
E peak flow (cm s1)
A peak flow (cm s1)
E/A ratio
Group X
Group I
Group X
Group I
Group X
Group I
Group X
Group I
Group X
Group I
17.3 16.2 15.6 17.9 17.9 16.9
17.5 16.4 16.8 17.2 17.4 17.6
749 (248) 574# (126) 594# (155) 700 (188) 717 (160) 679 (122)
791 (274) 707# (267) 677# (216) 800 (349) 784 (309) 830 (367)
60.3 57.6 59.2 56.9 57.0 55.0
58.3 57.6 58.7 59.5 61.5 58.6
23.6 (4.0) 29.9# (4.8) 33.9# (5.0) 28.4# (6.4) 26.7# (3.9) 27.2# (6.0)
31.1 (11.1) 37.4# (12.8) 39.4# (15.9) 35.3# (14.1) 33.5# (14.1) 36.1# (14.3)
2.58 (0.76) 2.00# (0.62) 1.77# (0.48) 2.05# (0.47) 2.17# (0.54) 2.05# (0.33)
2.04 (0.76) 1.62# (0.44) 1.49# (0.37) 1.77# (0.33) 1.85# (0.41) 1.72# (0.45)
(4.1) (4.5) (2.9) (3.8) (3.9) (3.7)
(3.4) (2.8) (2.8) (3.4) (3.5) (4.5)
(17.9) (13.0) (17.5) (15.3) (16.2) (14.2)
Xenon does not appear to affect contractility18 19 or to impair left ventricular performance, in spite of an increased SVR.4 20 Consequently, the increase in FAC is regarded as the unimpaired reaction to an acute decrease in afterload. Altogether, the changes in HR, CO and FAC induced by hypoventilation, and their reversal with normal ventilation, appear faster with xenon. This is in agreement with our hypothesis that xenon produces less inhibition of the physiological response to acute ventilatory impairment. However, the changes are minimal and probably of limited clinical relevance. A missing feature of the response to hypoventilation was the increase in LV preload, which was unexpected, as venous constriction is reported to be one of the compensating mechanisms in hypoxia.17 Together with unchanged early and increased atrial filling peak flows, this finding indicates an alteration of diastolic LV filling. There was a consistent, significant increase of 20–40% in atrial peak flow, possibly caused by the rise in HR, but which persisted after HR had returned to baseline level. Such an effect has only been described with acute myocardial ischaemia21 but not with hypoxia/hypercarbia. However, the normal value of E/A ratio in pigs is unknown, and the method is affected by various technical problems concerning the quality of the Doppler signal. Thus, it is not clear if the decrease we found is of functional relevance. The main limitation of this study may be that the degree of hypoventilation was insufficient to reduce oxygen delivery. However, DO2 was kept stable by the increase in CO despite a 30% reduction in arterial oxygen content, and VO2 was increased. This suggests that the autonomic response to the decrease of almost 50% in arterial PO2 and the 40% increase in PCO2 is sufficient to prevent hypoxia, at this level of hypoxaemia. It would have been interesting to compare different degrees of hypoventilation, but our preliminary studies demonstrated that more pronounced hypoventilation is associated with marked haemodynamic instability. We therefore decided that our target was the circulatory effect of hypoventilation without sustained hypoxic or hypercarbic damage to the heart. As the MAC of xenon exceeds 100% in pigs,22 it could not be used as the sole anaesthetic. We used equipotent
170
Downloaded from http://bja.oxfordjournals.org/ at UCSF Library on March 22, 2015
Hypoxaemia and hypercarbia appear to have conflicting effects on the circulation. While tachycardia and an increase in cardiac output are attributed to both, hypoxaemia is generally accompanied by vasodilation,9 10 which increases cerebral and myocardial blood flow.11 The early circulatory response is a result of autonomic nervous system activation by vascular oxygen receptors which respond to the concentration of dissolved oxygen. Hypoxic inhibition of pig tracheal smooth muscle contraction has been demonstrated in vitro,12 but there are no data on vascular smooth muscle and the mechanism is not clear. Nevertheless, a reduction in LV afterload was demonstrated using direct echocardiography in pigs.9 A negative correlation (r=0.56) between LV end-systolic wall stress and ejection fraction area (i.e. FAC) was reported, but there was no correlation between LVESWS and SVR. In contrast to the above findings, hypercarbia alone provoked hypertension, and thus the combined effect will be difficult to predict. It has been suggested that the circulatory effects of hypoxia are likely to be potentiated by hypercarbia,13 but there are no data available on which the relative contributions to a net effect can be based. Isoflurane has been shown not to impair the chemosensitivity of the carotid body (0.5 MAC),14 the hypoxic relaxation of the aorta (1 MAC)10 or the hypoxic increase in hepatic blood flow.15 Surprisingly, in this latter study hypoxia alone decreased cardiac index, but this is the only available report of such an effect. Isoflurane also seems to inhibit other compensatory mechanisms during hypoxia, e.g. increases in cerebral and coronary blood flow16 as well as venous contraction and—most importantly—the increase in sympathetic outflow.17 There are no reports of xenon in this setting. The haemodynamic effects of hypoventilation in our study are in agreement with most other reports and were similar with xenon and isoflurane anaesthesia, except for the more rapid decrease in CO with xenon after restoring ventilation. There was an increase in FAC with xenon but almost no increase with isoflurane. This may be a direct myocardial effect, where, in spite of a decrease in LV afterload, FAC could not be increased. Xenon does not inhibit myocardial function at the cellular2 or muscle fibre level.3
(15.9) (18.7) (14.0) (18.0) (19.6) (20.4)
Xenon anaesthesia and hypoventilation
11
Acknowledgements
16
8
9
10
12
13 14
15
The work was supported by the institutional research fund and Air Liquide Deutschland GmbH, Krefeld/Germany (xenon gas supply).
17
References 1 Rossaint R, Reyle-Hahn M, Schulte am Esch J, et al. Multicenter randomized comparison of the efficacy and safety of xenon and isoflurane in patients undergoing elective surgery. Anesthesiology 2003; 98: 6–13 2 Huneke R, Jungling E, Skasa M, Rossaint R, Luckhoff A. Effects of the anesthetic gases xenon, halothane, and isoflurane on calcium and potassium currents in human atrial cardiomyocytes. Anesthesiology 2001; 95: 999–1006 3 Schroth SC, Schotten U, Alkanoglu O, Reyle-Hahn MS, Hanrath P, Rossaint R. Xenon does not impair the responsiveness of cardiac muscle bundles to positive inotropic and chronotropic stimulation. Anesthesiology 2002; 96: 422–7 4 Hettrick DA, Pagel PS, Kersten JR, et al. Cardiovascular effects of xenon in isoflurane-anesthetized dogs with dilated cardiomyopathy. Anesthesiology 1998; 89: 1166–73 5 Stowe DF, Rehmert GC, Kwok WM, Weigt HU, Georgieff M, Bosnjak ZJ. Xenon does not alter cardiac function or major cation currents in isolated guinea pig hearts or myocytes. Anesthesiology 2000; 92: 516–22 6 Luttropp HH, Romner B, Perhag L, Eskilsson J, Fredriksen S, Werner O. Left ventricular performance and cerebral haemodynamics during xenon anaesthesia. A transoesophageal echocardiography and transcranial Doppler sonography study. Anaesthesia 1993; 48: 1045–9 7 Hecker KE, Reyle-Hahn M, Baumert JH, Horn N, Heussen N, Rossaint R. Minimum alveolar anesthetic concentration of
171
18
19
20
21
22
23
24
25
isoflurane with different xenon concentrations in swine. Anesth Analg 2003; 96: 119–24 Greim CA, Roewer N, Schulte am EJ. Assessment of changes in left ventricular wall stress from the end-systolic pressure–area product. Br J Anaesth 1995; 75: 583–7 Vedrinne JM, Curtil A, Martinot S, et al. The hemodynamic effects of hypoxemia in anesthetized pigs: a comparison between right heart catheter and echocardiography. Anesth Analg 1998; 87: 21–6 Haddad E, Lebuffe G, Boillot A, Imbenotte M, Vallet B. Does halothane or isoflurane affect hypoxic and post-hypoxic vascular response in rabbit aorta? Acta Anaesth Scand 2000; 44: 423–8 Duong TQ, Iadecola C, Kim SG. Effect of hyperoxia, hypercapnia, and hypoxia on cerebral interstitial oxygen tension and cerebral blood flow. Magn Reson Med 2001; 45: 61–70 Chen X, Yamakage M, Tsujiguchi N, Kamada Y, Namiki A. Interaction between volatile anesthetics and hypoxia in porcine tracheal smooth muscle. Anesth Analg 2000; 91: 996–1002 Nunn JF. Hypoxia. In: Nunn JF, ed. Applied Respiratory Physiology, 3rd edn. London: Butterworths, 1987: 471–7 Joensen H, Sadler CL, Ponte J, Yamamoto Y, Lindahl SG, Eriksson LI. Isoflurane does not depress the hypoxic response of rabbit carotid body chemoreceptors. Anesth Analg 2000; 91: 480–5 Cray SH, Crawford MW, Khayyam N, Carmichael FJ. Effects of hypoxia and isoflurane on liver blood flow: the role of adenosine. Br J Anaesth 2001; 86: 425–7 Durieux ME, Sperry RJ, Longnecker DE. Effects of hypoxemia on regional blood flows during anesthesia with halothane, enflurane, or isoflurane. Anesthesiology 1992; 76: 402–8 Stekiel TA, Stekiel WJ, Tominaga M, Stadnicka A, Bosnjak ZJ, Kampine JP. Isoflurane-mediated inhibition of the constriction of mesenteric capacitance veins and related circulatory responses to acute graded hypoxic hypoxia. Anesth Analg 1995; 80: 994–1001 Nakayama H, Takahashi H, Okubo N, Miyabe M, Toyooka H. Xenon and nitrous oxide do not depress cardiac function in an isolated rat heart model. Can J Anaesth 2002; 49: 375–9 Preckel B, Schlack W, Heibel T, Rutten H. Xenon produces minimal haemodynamic effects in rabbits with chronically compromised left ventricular function. Br J Anaesth 2002; 88: 264–9 Picker O, Schindler AW, Schwarte LA, et al. Xenon increases total body oxygen consumption during isoflurane anaesthesia in dogs. Br J Anaesth 2002; 88: 546–54 Labovitz AJ, Lewen MK, Kern M, Vandormael M, Deligonal U, Kennedy HL. Evaluation of left ventricular systolic and diastolic dysfunction during transient myocardial ischemia produced by angioplasty. J Am Coll Cardiol 1987; 10: 748–55 Hecker KE, Horn N, Baumert JH, Reyle-Hahn SM, Heussen N, Rossaint R. Minimum alveolar concentration (MAC) of xenon in intubated swine. Br J Anaesth 2004; 92: 421–4 Schuttler J, Albrecht S, Breivik H, et al. A comparison of remifentanil and alfentanil in patients undergoing major abdominal surgery. Anaesthesia 1997; 52: 307–17 Shinohara K, Aono H, Unruh GK, Kindscher JD, Goto H. Suppressive effects of remifentanil on hemodynamics in baro-denervated rabbits. Can J Anaesth 2000; 47: 361–6 Ishibe Y, Gui X, Uno H, Shiokawa Y, Umeda T, Suekane K. Effect of sevoflurane on hypoxic pulmonary vasoconstriction in the perfused rabbit lung. Anesthesiology 1993; 79: 1348–53
Downloaded from http://bja.oxfordjournals.org/ at UCSF Library on March 22, 2015
doses of 0.55 MAC for xenon and isoflurane, supplemented with remifentanil, which, because of its known vagomimetic effects,23 24 may have influenced our results and may have resulted in different levels of anaesthesia. As there is no gold standard for measuring anaesthetic depth in pigs, we used the absence of spontaneous movement in our nonparalysed animals and took this as an indicator of adequate anaesthesia. It is possible that 0.55 MAC may be too low a concentration to produce cardiovascular effects. However, there are reports of inhibition of hypoxic pulmonary vasoconstriction (HPV)25 and efferent sympathetic activity17 even with subMAC doses of isoflurane. As the concentrations of volatile (or gaseous) anaesthetics in combination with an opioid, as used in the present study, correspond to a clinical situation, we believe that this design is adequate. In summary, hypoventilation to moderate hypoxaemia and hypercarbia produced similar circulatory effects during xenon and isoflurane anaesthesia. Although the response appeared faster with xenon, our initial hypothesis that the increase in cardiac output would be greater with this agent is not supported.