Secondary Logo

Journal Logo

Original Article

The influence of xenon, nitrous oxide and nitrogen on gas bubble expansion during cardiopulmonary bypass

Grocott, H. P.1; Sato, Y.2; Homi, H. M.1; Smith, B. E.1

Author Information
European Journal of Anaesthesiology: May 2005 - Volume 22 - Issue 5 - p 353-358
doi: 10.1017/S0265021505000608


An increasing number of studies investigating the use of xenon anaesthesia are being reported in the anaesthetic literature [1-5]. Specifically, the use of xenon anaesthesia for cardiac surgery has recently been described experimentally [6,7] and, coupled with its safety and efficacy profile described in non-cardiac surgery [1], it is possible that it may find a clinical application in the setting of cardiopulmonary bypass (CPB) surgery. While the advantages of xenon are many (including its stable haemodynamic and environmental safety profiles [8]), it is not without theoretical disadvantages. In addition to its relative expense and requirement for alternative (closed-loop) delivery systems [9], most pertinent to the setting of cardiac surgery relates to its potential to expand enclosed gas spaces such as the microbubbles that are both entrained as well as generated within the CPB circuit. This is related to its relative insolubility as reflected by its low blood gas/partition coefficient [8]. Whereas it has long been known that nitrous oxide with a blood gas/partition coefficient of 0.41 can rapidly expand enclosed gas spaces, particularly nitrogen-containing air bubbles, the blood gas coefficient of xenon is even lower at 0.12. As a result it should theoretically rapidly expand enclosed gas spaces [10], including the microbubbles occurring during CPB.

The danger of expanding microbubbles that commonly occur during CPB relates to their ability to embolize to various vascular beds, particularly cerebral. Embolization to the cerebral microvasculature may cause both immediate ischaemia in brain regions receiving blood from the occluded vessel, as well as damage to vascular endothelium which, long after the air bubble dissipates, may lead to inflammatory processes that obstruct vessels with subsequent further cerebral ischaemia [11]. Indeed, at least in aqueous solutions, xenon has been demonstrated to cause some expansion of air bubbles [12], but not to the same degree as nitrous oxide [13]. Similarly, the effect of xenon on the expansion of the enclosed gas space of an isolated loop of bowel [10] has not been as significant as one would theorize. The ability of xenon to expand microbubbles has not been studied previously within the setting of CPB. The purpose of this study was to compare the relative abilities of xenon, nitrous oxide and nitrogen to expand gas bubbles in vivo during CPB in rats.


The following experimental protocol was approved by the Duke University Animal Care and Use Committee with procedures used herein meeting the guidelines of the National Institutes of Health guidelines for animal care (Guide for the Care and Use of Laboratory Animals, Health and Human Services, National Institute of Health Publication No. 86-23, revised 1996).

Fasted male Sprague-Dawley rats (12-14 weeks; weight, 350-400 g; Harlan, Indianapolis, IN, USA) were anaesthetized with 3% isoflurane in oxygen in a plastic box. Following induction of anaesthesia, they were intubated and mechanically ventilated. Ventilation was adjusted to maintain an arterial carbon dioxide tension (PaCO2) of 4.8-5.6 kPa. During subsequent surgical preparation, anaesthesia was maintained using 1.5-2.0% isoflurane.

Rectal temperature was monitored and servo-regulated to 37.0 ± 0.1°C using a heating blanket and convective forced-air heating system. Mean arterial pressure was monitored via the left common carotid artery, which was cannulated with polyethylene tubing (PE-10). The ventral tail artery was cannulated with a 20-G, 28-mm intravenous (i.v.) catheter, which later served as the inflow for CPB circuit. A 4.5-F multi-orifice venous cannula was inserted via the external jugular vein for CPB venous return. Rats were given heparin 150 IU i.v. after placement of the first arterial cannula with a further 100 IU added to the CPB priming solution. All cannulae were secured in situ with silk ties in order to eliminate the entrainment of extraneous air.

Upon initiation of CPB, during which the lungs were disconnected from the ventilator, the rats (n = 4 per group, with three microbubble injections per animal) were randomized to receive 70% N2O, 26% O2, 4% CO2 (nitrous oxide group); 70% Xe, 26% O2, 4% CO2 (xenon group) or 70% N2, 26% O2, 4% CO2 (nitrogen group) all with isoflurane (1.0% end-tidal concentrations) continuing throughout. Four percent CO2 was added to the oxygenator fresh gas flow due to the efficiency of the oxygenator to remove CO2. From the oxygenator fresh gas inflow, xenon gas was measured using a calibrated in-line analyser (KGIR 150 Gas Analyzer; Hitech Instruments Ltd., Luton, UK) with the other gases measured using a Capnomac Ultima monitor (Datex-Ohmeda, Helsinki, Finland). In order to avoid excessive haemodilution, the bypass circuit was primed with approximately 40 mL whole blood obtained from two heparinized (100 IU i.v. heparin per rat) donor rats. In addition, 6% hetastarch (3-5 mL) was added to the circuit, as needed, with the resulting haemoglobin concentration ranging from 80 to 120 g L−1. A membrane oxygenator (a modified Cobe Micro™ neonatal oxygenator with a surface area of 0.33 m2; Cobe Cardiovascular, Inc., Arvada, CO, USA) was used in this study allowing for an efficient equilibration of the various gas mixtures.

CPB at 160-180 mL kg−1 min−1 proceeded for 90 min. Ten minutes following the start of CPB, a pre-measured gas bubble (ranging in size from 300-500 μL) composed of room air (FiO2 = 0.21) was injected, using a calibrated glass microlitre syringe, into a blood-filled bubble chamber in the venous outflow circuit (Fig. 1). After 10 min, the bubble was removed and its volume was re-measured. Following a 10-min period of stabilization, a second gas bubble was injected, allowed to equilibrate, and then similarly removed for measurement. In each of the four animals per group, three separate gas bubbles were injected and measured serially over time resulting in 12 bubble determinations per group.

Figure 1.
Figure 1.:
Schematic diagram of the rat CPB apparatus and bubble chamber for measuring changes in gas bubbles volume during CPB.

Statistical analysis

Physiological values were compared among groups using analysis of variance (ANOVA). The difference in gas bubble size was compared among groups using ANOVA. If significant, post hoc between-group comparisons were performed using Fisher's PLSD test [14]. Statistical significance was considered when P < 0.05.


Each of the four animals in each group had a series of three gas bubbles (300, 400 and 500 μL) injected and subsequently measured during the course of CPB (n = 12 bubbles per group). Table 1 represents the arterial blood gas and haemodynamic parameters during each of the gas bubble measurement time periods. There were no differences among groups with respect to arterial blood gas measurements or other physiological parameters during the microbubble measurement time periods (Table 1).

Table 1
Table 1:
Physiological parameters during gas bubble equilibration.

The nitrous oxide group had significantly more expansion compared to the xenon and nitrogen groups (P = 0.0001; Fig. 2). The ratio of the post-equilibration bubble size to the pre-equilibration bubble size demonstrated that xenon expanded the microbubble 17 ± 6%, nitrous oxide caused 63 ± 23% expansion and nitrogen expanded the gas bubble 2 ± 2%. The nitrous oxide group has significantly more expansion compared to the xenon group (P = 0.0001; Fig. 2). The initial bubble size itself had no effect on the within-group bubble size changes (Table 2).

Figure 2.
Figure 2.:
A comparison of gas bubble volume changes in animals exposed to nitrous oxide, xenon or nitrogen during CPB.
Table 2
Table 2:
Ratio of post-to pre-equilibration bubble volume according to original bubble volume and group.


This study reconfirms that nitrous oxide can rapidly and significantly expand air bubbles during CPB, similarly reconfirming the general premise that it should be avoided during CPB. The question then becomes, ‘how much expansion is acceptable?’ That is, if 67% for nitrous oxide is too much, would the 17% by xenon be allowable? This question is particularly relevant today with respect to the fact that xenon is increasingly being considered for use in both cardiac and non-cardiac surgery.

Xenon has been studied as an anaesthetic for over 50 yr [15]. Yet, it is only in the past 5-10 yr that interest and new technologies have grown significantly in its potential clinical application to warrant further detailed study. Specific to cardiac surgery, xenon's haemodynamic profile offers considerable attraction [5]. In addition, xenon has been demonstrated to possess protective properties in the setting of experimental myocardial infarction. In a study examining the effects of xenon on the extent of myocardial injury in a rabbit model of myocardial infarction, xenon significantly decreased the infarct size [16]. This, coupled with its recent demonstration of neuroprotection, both in the setting of stroke [17], and also in experimental post-CPB cognitive decline [7], increases the likelihood that further clinical applications of this anaesthetic agent may be developed. Cardiac surgery represents a unique setting in which, despite its numerous advantages, xenon may have a significant theoretical disadvantage. Its lower solubility makes it theoretically more likely to expand gas bubbles, which clinically manifest as cerebral microemboli and are known to be present in most cardiac surgery procedures [18].

Despite its theoretical potential to significantly increase enclosed gas spaces, we demonstrated that in blood, xenon only marginally (approximately 17%) increased (compared to nitrogen) the volume of gas bubbles during CPB. This was in contrast to the in vitro study by Lockwood [12] where a 60% increase in an air microbubble was seen when studied in an aqueous solution exposed to xenon over a 3-min period. In that study, nitrous oxide predictably increased the bubble size significantly more. In contrast, the present study, which used blood instead of an artificial aqueous solution, demonstrated that an increase in the volume of the gas bubble (17%) occurred over a time span of 10 min. This is likely a much longer lifespan for an air bubble in the CPB circuit (and subsequent embolization to the cerebral microvascular) than is seen clinically. Based on observations from transoesophageal echocardiographic (TOE) detection of microbubbles being ejected from the beating heart [19], it is estimated that the transit time from the CPB circuit (i.e. aortic cannula) and/or the heart to the brain is a matter of seconds. As a result, the actual expansion of a microbubble during this embolic transit time is likely significantly less than the 10-min bubble equilibration time that we allowed. That said, once the microbubble reaches the cerebral microcirculation, its behaviour is less clearly known. While it dwells in the cerebral circulation, it would be exposed to similar expansion influences as it was during transit to the brain. This would likely be considerably longer than the few seconds it took to reach the brain and could lead to more significant expansion with subsequent prolonged obstruction of cerebral microvessels.

Further information on the behaviour of a bubble in the presence of xenon has come from Sta Maria and Eckmann [13]. In their study utilizing a mathematical model of a microbubble (50 nL), xenon theoretically increased the volume of this bubble. However, they did not model all the conditions of bypass, such as the fact that they performed their modelling assuming that 100% oxygen was being delivered during bypass, where more commonly, a lesser concentration of oxygen is typically used clinically. However, it provided valuable insights into how one can model bubble characteristics in vivo.

There were several limitations to this present study. Although an in vivo study, the bubbles we were studying were not exposed to a true vascular bed with the inherent lack of bubble-tissue interface, which would typically be seen clinically. In Sta Maria and Eckmann [13], the major flow of xenon into a bubble would be expected to occur at the blood-bubble interface, which we did model appropriately. In addition, we assumed that the gaseous composition of air bubbles during cardiac surgery is predominantly room air. However, it may be that the microbubbles contain a composition other than that found in room air, including the products of combustion resulting from the use of electrocautery. Another limitation was that the ‘bubble’ that we injected did not assume a completely spherical shape. This likely affected the interaction of the blood-gaseous interface. The air bubble that we used, although between 300 and 500 μL in size, had a dome shape to it with a flat surface at one end that represented the blood-gas interface (Fig. 1). This interface was only part of the actual surface area of the entire air bubble. It may be that if the entire surface of the bubble had been exposed to blood containing xenon or nitrous oxide, then the gas diffusion into the bubble may have been greater. In addition to the shape of the microbubble, its volume is likely to be larger than many of the bubbles observed clinically. This has implications again with respect to the surface area/volume ratio of the bubble. It is unclear what size of bubbles are experienced clinically; that is, clearly some are smaller and some are likely larger.

A final limitation to our modelling was the inclusion of the microbubbles on the venous side of the circulation. While bubbles on the arterial side would likely have greater clinical impact, venous bubbles are not innocuous. For example, air bubbles in the venous outflow of the CPB circuit have been shown to significantly increase the number of cerebral air bubbles as assessed using transcranial Doppler ultrasonography [20,21].

Whereas it is convenient to look at the effect of xenon on a microbubble and make extrapolations as to what may occur clinically, particularly with respect to its ability to augment cerebral injury through microembolization, it is important to remember that the dynamics of the in vivo situation of clinical CPB need to be taken into account. For example, xenon may well indeed increase the risk of cerebral injury by marginally increasing gaseous microbubble size. However, this must be weighed against its previously demonstrated neuroprotective effects in the setting of experimental CPB and cerebral ischaemia [7,17, 22,23]. It may be that the small increase in risk to cerebral injury due to marginally larger air bubbles could be offset by these neuroprotective abilities of xenon. It is also important to note that this study also reinforces the potential hazards of nitrous oxide use immediately before or after CPB.

In summary, we demonstrated that xenon resulted in a small increase in gas bubble size during CPB, which was significantly smaller than that seen with nitrous oxide. The impact of this small volume increase needs to be offset against the neuroprotective benefit that has been shown both in the setting of cardiac surgery as well as in other settings of cerebral injury.


We would like to acknowledge Cheryl Stetson for her skilled assistance in the preparation of this manuscript.


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. Goto T. Is there a future for xenon anesthesia? Can J Anaesth 2002; 49: 335-338.
3. Goto T, Nakata Y, Morita S. Will xenon be a stranger or a friend? The cost, benefit, and future of xenon anesthesia. Anesthesiology 2003; 98: 1-2.
4. Goto T, Saito H, Shinkai M, et al. Xenon provides faster emergence from anesthesia than does nitrous oxide- sevoflurane or nitrous oxide-isoflurane. Anesthesiology 1997; 86: 1273-1278.
5. Sanders RD, Franks NP, Maze M. Xenon: no stranger to anaesthesia. Br J Anaesth 2003; 91: 709-717.
6. Bedi A, McBride WT, Armstrong MA, Murray JM, Fee JP. Xenon has no effect on cytokine balance and adhesion molecule expression within an isolated cardiopulmonary bypass system. Br J Anaesth 2002; 89: 546-550.
7. Ma D, Yang H, Lynch J, Franks N, Maze M, Grocott H. Xenon attenuates cardiopulmonary bypass-induced neurologic and neurocognitive dysfunction in the rat. Anesthesiology 2003; 98: 690-698.
8. Lynch III C, Baum J, Tenbrinck R, Weiskopf R. Xenon anesthesia. Anesthesiology 2000; 92: 865-868.
9. Dingley J, Findlay GP, Foex BA, et al. A closed xenon anesthesia delivery system. Anesthesiology 2001; 94: 173-176.
10. Reinelt H, Schirmer U, Marx T, Topalidis P, Schmidt M. Diffusion of xenon and nitrous oxide into the bowel. Anesthesiology 2001; 94: 475-477.
11. Muth CM, Shank ES. Gas embolism. New Engl J Med 2000; 342: 476-482.
12. Lockwood G. Expansion of air bubbles in aqueous solutions of nitrous oxide or xenon. Br J Anaesth 2002; 89: 282-286.
13. Sta Maria N, Eckmann DM. Model predictions of gas embolism growth and reabsorption during xenon anesthesia. Anesthesiology 2003; 99: 638-645.
14. Einot I, Gabriel KR. A study of the powers of several methods of multiple comparisons (in applications). J Am Stat Assoc 1975; 70: 574-583.
15. Cullen SC, Gross EG. The anesthetic properties of xenon in animals and human beings, with additional observations on Krypton. Science 1951; 113: 580-582.
16. Preckel B, Mullenheim J, Moloschavij A, Thamer V, Schlack W. Xenon administration during early reperfusion reduces infarct size after regional ischemia in the rabbit heart in vivo. Anesth Analg 2000; 91: 1327-1332.
17. Homi HM, Yokoo N, Ma D, et al. The neuroprotective effect of xenon administration during transient middle cerebral artery occlusion in mice. Anesthesiology 2003; 99: 876-881.
18. Pugsley W, Klinger L, Paschalis C, Treasure T, Harrison M, Newman S. The impact of microemboli during cardiopulmonary bypass on neuropsychological functioning. Stroke 1994; 25: 1393-1399.
19. Grocott HP, Amory DW, Lowry E, Newman MF, Lowe JE, Clements FM. Cerebral embolization during transmyocardial laser revascularization. J Thorac Cardiovasc Surg 1997; 114: 856-858.
20. Willcox TW, Mitchell SJ, Gorman DF. Venous air in the bypass circuit: a source of arterial line emboli exacerbated by vacuum-assisted drainage. Ann Thorac Surg 1999; 68: 1285-1289.
21. Borger MA, Feindel CM. Cerebral emboli during cardiopulmonary bypass: effect of perfusionist interventions and aortic cannulas. J Extra Corp Technol 2002; 34: 29-33.
22. Ma D, Wilhelm S, Maze M, Franks NP. Neuroprotective and neurotoxic properties of the ‘inert’ gas, xenon. Br J Anaesth 2002; 89: 739-746.
23. Wilhelm S, Ma D, Maze M, Franks NP. Effects of xenon on in vitro and in vivo models of neuronal injury. Anesthesiology 2002; 96: 1485-1491.


© 2005 European Society of Anaesthesiology