Minimum alveolar concentration (MAC) of xenon in intubated swine

K. E. Hecker*,1, N. Horn1, J. H. Baumert1, S. M. Reyle-Hahn3, N. Heussen2 and R. Rossaint1

1 Department of Anaesthesiology and 2 Department of Medical Statistics, Klinikum der RWTH Aachen, Pauwelsstr. 30, D-52074 Aachen, Germany. 3 Department of Anaesthesiology and Intensive Care, Waldkrankenhaus Berlin, Germany

*Corresponding author. E-mail: klaus.hecker@post.rwth-aachen.de

Accepted for publication: September 26, 2003


    Abstract
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 References
 
Background. The minimum alveolar concentration (MAC) is a traditional index of the hypnotic potency of an inhalational anaesthetic. To investigate the anaesthetic as well as the unwanted effects of xenon (Xe) in a swine model, it is useful to know MACXe.

Methods. The study was performed using ten swine (weight 27.8–35.4 kg) anaesthetized with halothane and Xe 0, 15, 30, 40, 50 and 65% in oxygen. With each Xe concentration, various concentrations of halothane were administered in a step-by-step design. For each combination, a supramaximal pain stimulus (claw clamp) was applied and the appearance of a withdrawal reaction was recorded. The MACXe with halothane was calculated using a logistic regression model.

Results. During stable ventilation, haemodynamics and temperature, MACXe value was determined as 119 vol. % (95% confidence limits 103–135).

Conclusion. MACXe in swine was calculated by extrapolation of a logistic regression model. Its theoretical value is 119 vol. %.

Br J Anaesth 2004; 92: 421–4

Keywords: anaesthesia, closed circuit; anaesthetics volatile; pharmacology


    Introduction
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 References
 
A common method for obtaining knowledge about the properties and possible side-effects of anaesthetic gases is to initially investigate these in an animal model and then to transfer the results to conditions found in humans. In order to achieve equipotent doses for such studies, the MAC value of the gas under test needs to be known for the species investigated. So far, data on the MAC value of Xe (MACXe) exist only for dogs and humans.1 2 Data for swine are not available.

Reduction of the MAC of one volatile anaesthetic by administration of another permits extrapolation of an unknown MAC. Thus, we are able to determine a MAC that exceeds 100% and cannot be reached under normobaric conditions. This method has been used to determine the MAC of nitrous oxide in swine.3 The aim of this study was to determine MACXe in swine from the ability of Xe to reduce the MAC of halothane (MAChalo).


    Methods and results
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 References
 
The study was approved by the local animal care committee and by the Government Animal Care Office (Bez.-Reg. Koeln AZ 23.203.2-AC 38, 27/99). The study protocol was designed according to the recommendations of the Helsinki convention for the use and care of animals. Determination of MAC values was performed in ten female German Landrace swine weighing 27.8–35.4 kg (mean 31.1 (SD 3.1) kg). The animals were included in the study after five days of appropriate feeding.

After premedication with azaperone 4 mg kg–1 i.m., an ear vein was cannulated. Anaesthesia was induced with propofol 2 mg kg–1 and orotracheal intubation was performed using a 7.5 mm woodbridge tube. Thereafter, anaesthesia was maintained with repeated boluses of propofol.

Animals were mechanically ventilated using a Draeger PhysioFlex closed system ventilator (Draeger, Luebeck, Germany). Target end-tidal carbon dioxide values were 38–45 mmHg. Body temperature was maintained between 37.5 and 39.0°C by use of an airflow warming system (Warm Touch, Mallinckrodt Medical, Ireland). Heart rate, mean arterial pressure, end-tidal carbon dioxide and temperature were monitored and recorded continuously using a Datex AS/3 anaesthesia monitor (Datex-Engstrom, Helsinki, Finland). Inspiratory and end-tidal concentrations of oxygen, carbon dioxide and halothane were monitored using Datex AS/3 infrared spectroscopy. The inhaled concentration of Xe was measured via thermoconductive analysis using the PhysioFlex ventilator. Because Xe consumption after the initial wash-in is less than 20 ml min–1, inspiratory and expiratory Xe concentrations are virtually identical.

At the end of instrumentation, anaesthesia was maintained with halothane 1.3 MAC in 100% oxygen until the effect of propofol was no longer likely to influence the study protocol. The experimental protocol started at least 3 h after premedication, and at a mean of 45 min after the last propofol administration.

In the presence of Xe concentrations of 0, 15, 30, 40, 50 and 65%, the halothane concentration was changed in 0.1 vol. % steps, with an expected MAC of 1.0 vol. % for halothane in oxygen. Ten animals were randomly assigned to one of two arms of the protocol. Arm 1 started with Xe 0% and received increasing doses of halothane; arm 2 started with Xe 65% and received decreasing doses of halothane. After any change in gas concentrations, we allowed a minimum of 20 min for equilibration before continuing the experiment.

A supramaximal pain stimulus was applied using the dew claw clamp technique as previously described.4 For each concentration of Xe, halothane was either steadily increased or decreased until a change in reaction occurred: a withdrawal reaction when the individual had been asleep and vice versa. This was then confirmed by increasing or decreasing the concentration one step further, respectively. At the end of the experiments, the animals were killed according to German laws for animal studies.

Statistics
MAC determination
To evaluate the relationship between halothane and Xe, we considered a multiple logistic regression model with an interaction term:

logit (p) = ß0 + ß1X1 + ß2X2 + ß12X1X2(1)

where p=probability of no withdrawal reaction, X1=end-tidal halothane concentration, X2=Xe concentration, ß0=regression intercept, ß1=coefficient for halothane, ß2=coefficient for Xe and ß12=coefficient for the product of halothane and Xe (interaction coefficient).

As a result of our study design we were faced with correlated data. Therefore the approach used was the method of generalized estimating equations.5 Resulting estimates of equation (1) and their corresponding P values are presented in Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1 Logistic regression model coefficient estimates. ß0=regression intercept constant; ß1=coefficient for halothane; ß2=coefficient for Xe; ß12=coefficient for the product of the end-tidal halothane and Xe concentration (interaction coefficient); significance was assumed if P<0.05. The interaction coefficient (ß12) between halothane and Xe was not significantly different from 0 (P=0.1447), indicating that Xe has a linear rather than a non-linear effect on reducing halothane requirements
 
Because a linear relationship between halothane and Xe could not be excluded – indicated by a non-significant interaction term –MAChalo and its reduction by Xe were determined using a multiple logistic regression model without an interaction term:

logit (p) = ß0 + ß1X1 + ß2X2(1b)

The hypothetical MACXe without an additional anaesthetic was determined by setting the response probability p in equation (1) to 0.5 and solving the equation for Xe with a halothane concentration of 0% as follows:


The MACXe determined by the logistic regression without interaction term was extrapolated to 119 vol. % (95% confidence limits 103–135).

All statistical analyses were performed using SAS V8.02 Software (SAS Institute Inc., Cary, NC, USA).


    Comment
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 References
 
The present study estimated MACXe in swine with a multiple logistic regression model without an interaction term at 119% (103;135) CL. This is within the range of MACXe values reported for dogs,2 but higher than for humans.1 6

Our value of 1.0% for MAChalo is within the range of MAC values from other studies involving swine.7 8 This suggests that the current protocol is comparable to those of a number of previous studies investigating MAC in swine. Our results show that the interaction between Xe and halothane in swine is likely to be linear. MAC studies are subject to several influencing factors. To exclude the influence of premedication, the study protocol did not start until at least 3 h after administration. During the instrumentation period, halothane and propofol were used to maintain anaesthesia. Propofol was stopped 45–60 min before the study protocol was started. This enabled us to virtually exclude any influence of these medications on the study. Indeed, Cockshott and colleagues9 have demonstrated that the plasma propofol concentration after i.v. boluses of 2–5 mg kg–1 is reduced to 10% of the initial concentration after 45 min. The reason for using propofol for induction of anaesthesia was mainly to reduce the required concentration of halothane and thereby minimize the amount stored in fat, which could have reduced the equilibration time during the study protocol, as described above. To reduce bias caused by diurnal variability and duration of the experiment, the animals were randomly assigned to two arms of the protocol.

Other potential influences like hyper- and hypothermia, age, application of the supramaximal pain stimulus, hypo- or hypernatraemia, hypotension, abnormal perfusion–ventilation ratio, right-to-left shunt or insufficient equilibration time could be excluded. All parameters and values were within a normal range and maintained stable during the entire study protocol.

The main limitation of our study is that we did not directly measure MACXe but estimated it from data obtained in combination with halothane. It is difficult to administer more than 70% Xe to determine MAC because it would put the animals at risk of hypoxia. Therefore, we used Xe 65% as the highest concentration to keep the study design comparable in the two groups. Thus, we determined the MAC for the combination of Xe and halothane and we regard this method of extrapolating MACXe to be appropriate.

One may be concerned about a potential error resulting from extrapolation beyond the range of data. In contrast to Nakata and colleagues,1 we did not find a small antagonistic interaction between Xe and halothane. Therefore we used the assumptions of additivity of MAC between Xe and halothane and calculated MAC from the logistic regression model and excluded an interaction coefficient.


    References
 Top
 Abstract
 Introduction
 Methods and results
 Comment
 References
 
1 Nakata Y, Goto T, Ishiguro Y, et al. Minimum alveolar concentration (MAC) of xenon with sevoflurane in humans. Anesthesiology 2001; 94: 611–14[ISI][Medline]

2 Eger EI2, Brandstater B, Saidman LJ, Regan MJ, Severinghaus JW, Munson ES. Equipotent alveolar concentrations of methoxyflurane, halothane, diethyl ether, fluroxene, cyclopropane, xenon and nitrous oxide in the dog. Anesthesiology 1965; 26: 771–7[ISI][Medline]

3 Tranquilli WJ, Thurmon JC, Benson GJ. Anesthetic potency of nitrous oxide in young swine (Sus scrofa). Am J Vet Res 1985; 46: 58–60[ISI][Medline]

4 Eger EI2, Johnson BH, Weiskopf B, et al. Minimum alveolar concentration of I-653 and isoflurane in pigs: Definition of a supramaximal stimulus. Anesth Analg 1988; 67: 1174–7[Abstract]

5 Liang KY, Zeger SL, Albert PS. Models for longitudinal data: a generalized estimation approach. Biometrics 1988; 44: 1049–60[ISI][Medline]

6 Cullen SC, Eger II EI, Cullen BF, Gregory P. Observations on the anesthetic effect of the combination of xenon and halothane. Anesthesiology 1969; 31: 305–9[ISI][Medline]

7 Tranquilli WJ, Thurmon JC, Benson GJ, Steffey EP. Halothane potency in pigs (Sus scrofa). Am J Vet Res 1983; 44: 1106–7[ISI][Medline]

8 Weiskopf RB, Bogetz MS. Minimum alveolar concentrations (MAC) of halothane and nitrous oxide in swine. Anesth Analg 1984; 63: 529–32[Abstract]

9 Cockshott ID, Douglas EJ, Plummer GF, Simons PJ. The pharmacokinetics of propofol in laboratory animals. Xenobiotica 1992; 22: 369–75[ISI][Medline]