Best Practice & Research Clinical Anaesthesiology

Vol. 15, No. 3, pp. 491±503, 2001

doi:10.1053/bean.2001.0179, available online at http://www.idealibrary.com on

16

Xenon as a replacement for nitrous oxide?

Peter H. Tonner*
Associate Professor of Anaesthesiology
Department of Anaesthesiology and Intensive Care, University Hospital Kiel, Schwanenweg 21, 24105 Kiel,
Germany

Kathrin Bangert
Resident in Anaesthesia
Department of Anaesthesiology, University Hospital Eppendorf, Hamburg, Germany

Jens Scholz
Professor and Chair
Department of Anaesthesiology, University Hospital Eppendorf, Hamburg, Germany

Although nitrous oxide (N2O) has been used routinely since the beginning of the modern era
of anaesthesia, some of its adverse e€ects have only been discovered during the last decades.
Thus there are some who advocate abandoning the use of N2O for anaesthesia. However, if the
use of N2O is stopped, the anaesthetic regimen will have to be changed in order to substitute
for a loss in potency. Thus, xenon has been suggested as a replacement for N2O. N2O and
xenon share some clinical and physicochemical properties. For example, both of them are only
weak anaesthetics but are potent analgesics. However, in many respects xenon appears to be
advantageous compared to N2O. Xenon exerts a high degree of cardiovascular stability and
does not have teratogenic or fetotoxic e€ects. In contrast to N2O it does not support the
greenhouse e€ect. The costs of xenon are prohibitively high at the moment. Advanced
anaesthesia machines and recycling equipment have necessary in order use xenon in an
economically feasible way.

Key words: xenon; nitrous oxide anaesthesia; nitrous oxide analgesia; nitrous oxide adverse
e€ects.

Nitrous oxide (N2O) and xenon share some physicochemical and clinical properties.
Both are gases at atmospheric pressure and di€er from other volatile or intravenous
anaesthetics in that their respective potencies (MAC xenon: 70 vol%; minimal
alveolar concentration (MAC) N2O: 110 vol%) are weak. However, both of them are
not only weak anaesthetics but are also good analgesics and this is one of the main
reasons for the administration of N2O in clinical practice to this day. In contrast to
xenon, N2O is `ancient' as an anaesthetic agent. The ®rst use of N2O in public dates
back to 1842, at a time just before the advent of modern anaesthesia in 1846 with the
*All correspondence to: Peter H. Tonner. Tel: ‡49-431-597-2990; Fax: ‡49-431-597-3002, E-mail:
tonner@anaesthesie.uni-kiel.de
1521±6896/01/030491‡13 $35.00/00 *
c 2001 Harcourt Publishers Ltd.
492 P. H. Tonner et al

public demonstration of ether as a potent anaesthetic at the Massachusetts General

Hospital in Boston, USA. However, as already detailed in previous chapters of this
book, the use of N2O is bene®cial but there are also drawbacks.
So, if xenon is the `new kid on the block' with many reported advantages compared
to N2O, could xenon therefore be a replacement for N2O? We will try to summarize
the arguments that have been produced both for and against the replacement of N2O by
xenon. However, we will not do so without citing a remark made by Professor Jan Baum
from Damme, Germany at a recent meeting on the pros and cons of N2O: `Nitrous
oxide does not need a replacement by xenon or any other anaesthetic, in fact, the use of
nitrous oxide in modern anaesthesia is so super¯uous that it does not need a
replacement at all'. Again, previous chapters have dealt with N2O in depth. In this
review, we will focus on a direct comparison of N2O with xenon and not on the use of
N2O itself.

HISTORY

Xenon was discovered later than N2O in 1898, by W Ramsay and M W Travers, after the
evaporation of the components of liquid air. It is thought to be an inert gas because of its
unique electron structure. In 1962 it was categorized with the noble gases (helium,
neon, argon, krypton and radon), all of them being found in the troposphere in very
small amounts. Of these, xenon is the heaviest stable compound with a molecular
weight of 131.3. With a relative concentration of 0.0000087 vol% it is rather rare. After a
demonstration of its analgesic e€ects in mice1, the ®rst use of xenon as an anaesthetic in
humans can be traced back to 1951 and was reported by S C Cullen and E G Gross.2
They described a rapid loss of consciousness and early recovery using the gas at a
concentration of 80 vol%.

CHEMICAL REACTIVITY

As a noble gas, xenon is chemically almost inert. It is non-explosive. However, under

certain conditions, it may form chemical compounds with very reactive elements, such
as oxygen or ¯uorine. During anaesthesia it is extremely unlikely that xenon
participates in any chemical reaction. Although biotransformation of xenon has not
been demonstrated, it is conceivable that xenon is involved in biochemical processes.
Similarly there is currently no known pathway for the metabolism of N2O in the
human body. In contrast to xenon however, N2O is chemically more reactive and may
form other derivatives under special conditions. Theoretically the content of an N2O
bottle may be contaminated with NOx compounds, which may react with water
forming nitric acid. A severe lung oedema and, in addition, a methaemoglobinaemia
may occur after administration to patients.

ACTIONS OF XENON AND N2O AT SUB-CELLULAR AND

CELLULAR LEVELS

Although the analgesic and anaesthetic properties of xenon and N2O have been known
for more than 5 and 15 decades, respectively, the exact mechanism of action at the
cellular or molecular level are not yet known. However, a number of studies have been
 Xenon as a replacement for N2O 493

Nitrogen
101

Nitrous oxide
MAC (atm) 100 Xenon

10–1 Desflurane
Diethyl ether
Sevoflurane Isoflurane
10–2 Enflurane Halothane

10–3
0.1 1.0 10.0 100.0
Oil–gas partition coefficient
Figure 1. Meyer±Overton correlation for various inhalational anaesthetics. Like nitrous oxide (N2O), xenon
is not very lipid soluble. As predicted by the Meyer±Overton correlation the corresponding anaesthetic
potency is also weak. The adherence to the Meyer±Overton correlation indicates an unspeci®c site of action
for xenon.

performed to de®ne parameters that are relevant for the action of these compounds
(Figure 1). It is known that many anaesthetics exert their actions at a variety of cellular
targets. Among these, the super-family of ligand-gated ion channels has been shown to
be more likely to be in¯uenced in their function, at clinical concentrations, than other
receptor or channel proteins. It has been demonstrated that most anaesthetics such as
volatile anaesthetics act primarily on g-aminobutyric acid A (GABA-A) receptors.
However, it was recently shown that N2O as well as xenon might be di€erent in their
mechanism of action since it was shown that both anaesthetics potently inhibit
N-methyl-D-aspartate (NMDA) receptor function.3±6 A recent study compared the
e€ects of N2O, xenon, iso¯urane and ethanol directly on nine di€erent recombined
ligand-gated ion channels including GABA-A, NMDA, serotonin (5HT3) and acetyl-
choline (Ach) receptors, by employing a voltage clamp technique.7 N2O and xenon both
acted in a similar fashion at these channels. Both exerted a stronger e€ect on NMDA-
receptors than on GABA-A-receptors, rendering these two anaesthetics
distinctly di€erent from iso¯urane and ethanol, which had their highest potencies at
GABA-A receptors. Studies looking at ether, N2O or xenon had already hinted that
GABA-A receptors are only weakly potentiated in their function in the presence of
N2O. Xenon had almost no e€ect on GABA-A receptors.3±6,8 In this respect the two
gaseous anaesthetics are markedly di€erent from volatile anaesthetics. However, if
one looks at the action at homologous glycine and GABA-A receptors, the action of
N2O is comparable to that of volatile anaesthetics suggesting a similar type of
interaction.9,10 In a study looking at a whole range of receptors, N2O inhibited NMDA
receptors as well as kainate receptors.7 By contrast, volatile anaesthetics enhanced the
function of kainate receptors and a transmembrane site has been shown to be involved
in this action,11 suggesting that N2O acts at a site that is di€erent to that at which volatile
anaesthetics act.
Another molecular target site that is shared by both N2O and xenon with gaseous
and volatile anaesthetics is the nicotinic acetylcholine receptor (nAChR). It has been
demonstrated that central nervous system nAChRs, especially those composed of the
b2-subunit, are inhibited in their function by many anaesthetics at relevant con-
centrations.12±15 Both N2O and xenon are potent analgesics at subanaesthetic
494 P. H. Tonner et al

1.0

0.8

0.6

FA/FI
0.4

0.2

0.0
0 10 20 30
Time (min)

Figure 2. Uptake of various inhalational anaesthetics. In contrast to the volatile anaesthetics, nitrous oxide
(N2O) and xenon possess a lower blood±gas partition coecient (see Table 1), thus their uptake is more
rapid. Xenon even surpasses N2O; however, the di€erence is clinically not relevant. ÐjÐ, halothane; ÐmÐ,
iso¯urane; Ð Ð, sevo¯urane; ÐrÐ, des¯urane; ÐdÐ, N2O; ÐjÐ, Xenon. FA/F1; alveolar fraction/
m
inspiratory fraction. Data from Yasuda et al68 and FroÈba et al.69

concentrations.16,17 It has been speculated that this anti-nociceptive e€ect is due to an

action on glutamate receptors, namely the NMDA receptor in a fashion similar to
ketamine.18

ANAESTHESIA AND MAC

The anaesthetic properties of xenon have been known for more than 50 years.2 Like
N2O, xenon has been reported to exert minimal e€ects on cardiac or vascular function
in either healthy or sick subjects as demonstrated in humans and in animals.19,20 The
rapid induction and emergence in combination with a higher potency than N2O
renders xenon a nearly ideal anaesthetic gas (Figure 2).
However, like N2O the anaesthetic potency of xenon is too weak for use as a single
anaesthetic. Although the MAC value of xenon is lower than that of N2O, it is not
possible to deliver more than 1 MAC of xenon at an inspiratory oxygen fraction FiO2
of 0.3, which is the standard setting, at least in Germany, for healthy patients. At a
concentration of 1 MAC, however, it is impossible to safely perform a surgical
operation. Under similar conditions, the MAC value of N2O is 0.7 MAC, which is
even lower than that of xenon. So for both compounds other anaesthetics are
required in combination in order to achieve surgical anaesthesia. The MAC of volatile
anaesthetics has been reported to be decreased when N2O or xenon are co-
administered. For sevo¯urane it has been shown, in patients, that it failed to attenuate
the cardiovascular response when administered by itself.21 However, in combination
with either N2O or xenon the cardiovascular response to skin incision was
completely blocked. A follow-up study evaluated whether there was any di€erence in
the e€ect of N2O on xenon in combination with sevo¯urane. A total of 43 patients
undergoing elective surgery were studied and the MAC was determined according to
the `up-and-down' method.22 It was shown that the suppression of cardiovascular
responses by xenon and N2O was similar at 70 vol%, possibly re¯ecting their nearly
identical analgesic properties.
 Xenon as a replacement for N2O 495

EFFECTS ON THE CARDIOVASCULAR SYSTEM

Xenon has been reported to possess a remarkably safe haemodynamic pro®le. In pigs it
has been demonstrated that haemodynamic parameters, as well as plasma catecholamine
levels, remained within the normal limits during xenon anaesthesia. Even at sub-
anaesthetic concentrations of xenon a signi®cant decrease of adrenaline was observed
possibly re¯ecting the high analgesic potency of xenon.23 In a study of the haemo-
dynamic e€ects of xenon versus N2O, 32 adult patients undergoing gynaecological,
plastic or orthopaedic surgery were studied.24 In each group anaesthesia was maintained
using either 70 vol% N2O or 70 vol% xenon. Fentanyl was added when the blood
pressure increased by more than 20% from baseline. The anaesthetist was unaware of
which gas was being used. In this setting it was found that incremental fentanyl was
necessary in all patients receiving N2O and only in 50% of those patients receiving
xenon, while awareness did not occur in both groups. No di€erences in blood pressure
or heart rate during surgery were observed between the groups. In contrast to these
clinical parameters the adrenaline levels showed a di€erence between the N2O and the
xenon group. In the xenon group the adrenaline level remained below the baseline
levels throughout the operation and only returned to control values at the end of

Figure 3. Examples of trans-oesophageal echocardiography in a patient treated with xenon. Trans-gastric

short axis view of end-systolic (a and c) and end-diastolic (b and d) area before the administration of xenon (a
and b) and after the administration of xenon (c and d). There was no clinically detectable di€erence after
xenon administration.
496 P. H. Tonner et al

surgery while in the N2O group a continuous increase in plasma adrenaline concentra-
tion was seen resulting in a signi®cant di€erence at the end of the operation. In addition,
the N2O group showed increased adrenaline values immediately post-operatively,
returning to baseline only after about 6 h. The increase in adrenaline levels during N2O
anaesthesia has repeatedly been reported and is probably due to an increase in
sympathetic tone caused by N2O.25,26 There was no apparent di€erence in recovery
time or quality of recovery and the attending anaesthetists were not able to tell which
gas had been used.24 In a subsequent study examining otherwise healthy American
Society of Anesthesiologists (ASA) 1 patients undergoing open cholecystectomy or
hysterectomy, the e€ect of xenon was studied by echocardiography. No alterations in
mean fractional area change were observed during induction with xenon.19 In contrast
to xenon, N2O has been demonstrated experimentally to possess negative inotropic
e€ects27,28; however, clinically these e€ects may be in¯uenced by the activation of the
sympathetic nervous system after N2O administration as well as by the co-
administration of other anaesthetics.25 In volunteers a modest increase of heart rate
has been observed (Figure 3).29

EFFECT ON CEREBRAL CIRCULATION

The e€ect of xenon on cerebral blood ¯ow seems to vary according to the model used
and the species studied.30±33 In radiology, radioactive xenon isotopes have been in use
for a long time for the determination of cerebral blood ¯ow. In volunteers, the
administration of 33 vol% xenon led to a signi®cant increase in cerebral perfusion.
Increases in perfusion were also noted in patients with traumatic brain injury. Thus,
some authors have recommended the prophylactic use of hyperventilation in patients
receiving xenon for diagnostic purposes.34±37 Studies using transcranial Doppler
ultrasonography found di€ering results. Increases as well as non-signi®cant changes
in cerebral blood ¯ow velocities have been reported.19,38±40 N2O is well known to
increase cerebral perfusion by a vasodilation of cerebral vessels.41±44 Its use is
contraindicated in patients with elevated intracranial pressure. In addition to the
increase in cerebral blood ¯ow, N2O also increases cerebral metabolism, possibly due to
its stimulation of the sympathetic tone. The concurrent increase in oxygen consumption
together with an elevated intracranial pressure may lead to increases in cell damage and
cellular death.

EFFECTS ON THE GASTROINTESTINAL TRACT

Because of its low solubility N2O di€uses readily from blood into any space within the
body that is ®lled with gas. Like N2O, xenon has a low blood±gas partition coecient
of 0.12±0.14.45,46 A typical side e€ect of N2O is the di€usion into the bowel leading to
its extension and, in abdominal surgery, this may lead to diculties with abdominal
closure. In dogs an increase in bowel volume of 100±200% has been reported after 4 h
of N2O anaesthesia (see Figure 4).47 A recent publication compared the e€ects of
xenon and N2O directly under well-controlled conditions.48 In 21 pigs anaesthesia was
induced and maintained intravenously. Three groups of animals were studied, one
receiving 75 vol% xenon, one 75 vol% N2O and one 75 vol% nitrogen for 240 min.
After a laparotomy was performed, 15 cm long segments of bowel were isolated in a
gas-tight manner, ®lled with 30 ml of air and the pressure was measured through an
inserted catheter. In both the N2O and xenon group the volume of the bowel
 Xenon as a replacement for N2O 497

200

Increase in bowel volume (%)

150

100

50

0
Baseline 100 200
Time (min)
Figure 4. Bowel distension after administration of halothane/oxygen (h) or nitrous oxide/oxygen (j). By
200 min after the start of N2O administration, bowel volume had increased to about 200% of that with
halothane. Reproduced with permission, from Eger and Saidman.47

increased compared to the nitrogen control group. However, in the N2O group there
was an increase of up to 88 ml, whereas in the xenon group the volume increased only
to 39 ml (21 ml in the control group). There was also a signi®cant increase in
intraluminal pressure in the N2O group compared with the xenon and control groups.
No di€erence in intraluminal pressure was found when the xenon and control groups
were compared.
Xenon has a fourfold lower blood±gas partition coecient compared to N2O. The
lower partition coecient predicts a lower transport capacity of the blood and a lower
rate of di€usion of xenon into gas ®lled volumes. However, the ability of various
substances to di€use into adjacent space is also in¯uenced by their respective molecular
diameters (xenon ˆ 4.0055 AÊ, N2O ˆ 3.879 AÊ; see Table 1) and their di€usion

Table 1. Physiochemical properties of xenon and nitrous oxide.

Xenon Nitrous oxide

Molecular weight 44.0 131.1
Molecular diameter (AÊ) 3.879 4.055
Blood±gas partition coecient 0.14 0.47
Oil±gas partition coecient 1.9 1.4
MAC (in humans) 0.71 1.05
Greenhouse e€ect ÿ ‡
Freezing point (8C) ÿ111.9 ÿ90.0
Boiling point (8C) ÿ108.2 ÿ88.5
Density (g/l) 5.40 1.81
Viscosity (micropoises) 226 145
Thermal conductivity (mWcmÿ1Kÿ1) 0.057 0.173
Thermal capacity (calKÿ1molÿ1) 4.97 9.19

AÊ; angstrom.
498 P. H. Tonner et al

coecients, which may be modi®ed by a number of factors such as the polarity of the
substance. Compared to other volatile anaesthetics, which decrease bowel motility,
N2O increases bowel motility thus further increasing pressure within bowel segments.49
In addition, no impairment of bowel function has been found in patients treated with
N2O.49±55 It may be concluded from these studies that xenon can be used as a substitute
for N2O in those cases of abdominal surgery in which N2O would have been used when
other factors such as haemodynamic control are of concern.

XENON AND RESPIRATORY MECHANICS

The physical parameters of xenon render this gas distinctly di€erent from the other
gases used in anaesthesia. In particular its high density and viscosity alter the physical
properties of any gas mixture used for anaesthesia when a high fraction of xenon is
used. There are two implications of this fact that have to be dealt with by the
anaesthesiologist: (i) the behaviour of monitoring devices for respiratory monitoring
may be in¯uenced, (ii) the respiratory mechanics of the patient may be altered,
especially in the case of patients whose lung function is compromised.
Flow meters in particular may be a€ected by the di€erent physical properties of
xenon. Although most ¯ow meters are designed to correct their readings according to
the composition of the gas used, most of them were not designed for use with xenon
and may, therefore, give a false reading without any compensation for possible errors.
In a recent study on the practicability of ¯ow meters for xenon anaesthesia it was
demonstrated that, of four types of ¯ow meters tested, only rotating vanes gave
suciently accurate readings when xenon was used.56
In addition, in patients with lung diseases the physical properties of a gas may a€ect
the respiratory mechanics.57±59 Both the density of xenon and its viscosity are higher
than those of N2O, an e€ect that led one anaesthetic apparatus manufacturer to redesign
one of their machines for xenon administration because its ventilator broke after only
short periods of use. The e€ect of xenon on respiratory mechanics was studied in
comparison to N2O in pigs.60 Groups of eight pigs were ventilated with either 70 vol%
N2O or 70 vol% xenon and airway pressure as well as resistance were measured. Both
groups of animals were then subjected to a metacholine infusion for the induction of
bronchoconstriction and the measurements were repeated. Both under normal
conditions and throughout bronchoconstriction the airway resistance was signi®cantly
higher in the xenon group compared to the N2O group. Airway pressure did not di€er
during normal airway conditions. However, during bronchoconstriction, there was a
signi®cant increase in airway pressure in the xenon group compared to the N2O group
(Ppeak xenon ˆ 33.2 (+5.5); nitrous oxide ˆ 28.4 (+5.7) cmH2O).60
Another study looked at the e€ect of xenon on respiratory mechanics both under
normal conditions and under bronchoconstriction.61 In contrast to the previously cited
study, the authors did not administer 70 vol% xenon throughout the study but instead
used 50 vol% y-xenon during bronchoconstriction. This change in experimental
conditions possibly led to the ®nding that there was no di€erence in airway resistance
between normal ventilation and ventilation during bronchoconstriction.61 In addition,
airway resistance in dogs is di€erent to that in pigs and conditions are also di€erent if
airway resistance is measured in an open chest situation. The main factor responsible
for the change in airway resistances if one switches from N2O to xenon appears to be
the change in both the viscosity and density of the gas, as predicted by the laws of ¯uid
dynamics.62 However, gas exchange was not a€ected during ventilation with xenon.60
 Xenon as a replacement for N2O 499

Calzia et al60 concluded that although there was a change in respiratory mechanics
during xenon anaesthesia compared to N2O, these changes were not of clinical
relevance for the following reasons:
1. During general anaesthesia the ventilator and not the patient has to overcome
increases in airway resistance.
2. The increase in airway resistance is small under healthy conditions and only
moderate during bronchoconstriction.
3. Gas exchange does not deteriorate during xenon anaesthesia.

DIFFUSION HYPOXIA

Since the inert gas volume moving between capillary blood and alveolar space di€ers
with varying gas solubility, speci®c gases may become concentrated or diluted during
uptake or elimination. This is the reason why an outpouring of gas occurs at the end of
an N2O-based anaesthesia, an e€ect called di€usion anoxia by Fink.63 The solubility of
N2O is higher (0.47) than that of nitrogen (0.015), thus the concentration of N2O rises
during wash-out in the alveolar space. The mechanism of hypoxia is twofold: (i) oxygen
may be displaced and oxygenation inhibited, (ii) carbon dioxide may be diluted in the
alveoli, thus decreasing respiratory drive and ventilation. Since the solubility of xenon is
lower than that of N2O but higher than that of nitrogen, it may be concluded that a
similar e€ect also occurs during xenon anaesthesia but it may be less pronounced.
Recently, a study in pigs was performed addressing the question of di€usion hypoxia
after N2O versus xenon anaesthesia.64 The lungs of pigs were ventilated either with a
mixture of 30 vol% oxygen and 70% N2O or 70 vol% xenon for 30 min. At the end of
this period all animals were ventilated with 30 vol% oxygen and 70 vol% nitrogen.
Arterial oxygen partial pressures were determined during nitrogen wash-in. The DpO2
of 17 mmHg in the N2O group was comparable to values found previously.59 In the pigs
receiving xenon the DpO2 was much smaller (6 mmHg), thus con®rming the idea that
inert gas exchange is mainly determined by the blood±gas partition coecient
(solubility in lipids). Although the lipid solubility of xenon is greater than that of N2O,
more xenon may be taken up during an anaesthetic event than N2O. However, only that
part of the gas that is actually dissolved in the blood and passes to the lungs may di€use
into the alveolar space so that lipid solubility is only of minor importance to these
e€ects.

TERATOGENIC AND TOXIC EFFECTS

In rats, a teratogenic e€ect of N2O has been demonstrated.65 In a study comparing the
teratogenic e€ects of N2O and xenon, pregnant rats were exposed to either 70 vol%
N2O or xenon in 30 vol% oxygen.66 Twenty days after exposure the fetuses were
evaluated for anomalies. The rats from the N2O group showed a high rate (37%) of fetal
absorption, skeletal anomalies and macroscopic anomalies such as encephalocele,
anophthalmia, microphthalmia and gastroschisis. By contrast, in the xenon group a low
rate for anomalies of only 3% was found. Thus, it was concluded that xenon in direct
comparison to N2O does not exert teratogenic e€ects.66
In order to determine the possible adverse e€ects of a long-term exposure to
xenon, dogs were exposed to 80 vol% xenon in oxygen every third day for 2 weeks.67
500 P. H. Tonner et al

At the end of the study no side e€ects of xenon were found in the morphological,
haematological or biochemical parameters studied. By contrast to long-term exposure
to N2O after xenon no e€ects were noted on methionine synthetase and vitamin B12
metabolism.

ECONOMICAL AND ECOLOGICAL ASPECTS

Currently xenon is routinely used only in Russia. In Germany, xenon is currently

undergoing Phase III trials and may thus be clinically available within a few years.
However, xenon is much more expensive than N2O. A litre of xenon gas currently
costs about 10±15 =C (in Germany) and is thus about 2000 times more expensive than
N2O. The xenon requirement for a routine anaesthetic is about 40 l, so an average
xenon anaesthesia will cost about 400±600 =C . The main competitors for the use of
xenon are the space agencies, in particular the National American Space Agency
(NASA). Since the Iridium satellite project, for which a large amount of xenon was
needed, ended recently, it may be that the price of xenon on the world market will be
cut by half or more. There are many projects under way that aim to reduce the
amount of xenon needed for anaesthesia. The use of closed systems in anaesthesia
machines such as the Physio¯ex will allow the use of less than half the volume of xenon
normally used in standard anaesthesia machines. Recycling systems may allow the
capture of xenon from the anaesthesia system at the end of the procedure and can thus
reduce costs even further.

SUMMARY

There are currently advocates for abandoning the clinical use of nitrous oxide (N2O).
One of the possible replacements may be xenon. Xenon shares many clinical as well as
physicochemical properties with N2O, while producing fewer side e€ects. However,
xenon is not abundantly available. Thus, its cost may prohibit the substitution of N2O
with xenon.

Practice points
. xenon is currently clinically available only in Russia, in Phase III studies xenon has
been demonstrated to o€er rapid induction and emergence from anaesthesia. In
order to use xenon in an economically feasible fashion, anaesthesiologists will have
to be able to manage low-¯ow or closed system anaesthesia

Research agenda
. more studies are needed to directly compare xenon with N2O
. the clinical e€ects of xenon need to be studied in more detail. Most of the data
known today were derived from animal work
. large multicentre studies are necessary to assess the e€ects of xenon in di€erent
clinical settings and population subgroups, e.g. high-risk cardiovascular patients
undergoing cardiac surgery or children at di€erent ages
 Xenon as a replacement for N2O 501

REFERENCES

1. Lawrence JH, Loomis WF, Tobias CA & Turpin FH. Preliminary observations on the narcotic e€ect of
xenon with review of values for solubilities of gases in water and oils. Journal of Physiology 1946; 105:
197±201.
2. Cullen SC & Gross EG. The anesthetic properties of xenon in animals and human beings, with additional
observations on krypton. Science 1951; 113: 580±581.
3. Jevtovic-Todorovic V, Todorovic SM, Mennerick S et al. Nitrous oxide (laughing gas) is an NMDA
antagonist, neuroprotectant and neurotoxin. Nature Medicine 1998; 4: 460±463.
4. Franks NP, Dickinson R, de Sousa SL, Hall AC & Lieb WR. How does xenon produce anaesthesia? Nature
1998; 396: 324.
5. Mennerick S, Jevtovic-Todorovic V, Todorovic SM et al. E€ect of nitrous oxide on excitatory and
inhibitory synaptic transmission in hippocampal cultures. Journal of Neuroscience 1998; 18: 9716±9726.
6. de Sousa SL, Dickinson R, Lieb WR & Franks NP. Contrasting synaptic actions of the inhalational general
anesthetics iso¯urane and xenon. Anesthesiology 2000; 92: 1055±1066.
7. Yamakura T & Harris RA. E€ects of gaseous anesthetics nitrous oxide and xenon on ligand-gated ion
channels. Comparison with iso¯urane and ethanol. Anesthesiology 2000; 93: 1095±1101.
8. Dzoljic M & Van Duijn B. Nitrous oxide-induced enhancement of gamma-aminobutyric acid A-mediated
chloride currents in acutely dissociated hippocampal neurons. Anesthesiology 1998; 88: 473±480.
9. Daniels S & Roberts RJ. Post-synaptic inhibitory mechanisms of anaesthesia; glycine receptors. Toxicology
Letters 1998; 100±101: 71±76.
10. Mihic SJ, Ye Q, Wick MJ et al. Sites of alcohol and volatile anaesthetic action on GABA(A) and glycine
receptors. Nature 1997; 389: 385±389.
11. Minami K, Wick MJ, Stern-Bach Y et al. Sites of volatile anesthetic action on kainate (glutamate
receptor 6) receptors. Journal of Biological Chemistry 1998; 273: 8248±8255.
12. Violet JM, Downie DL, Nakisa RC, Lieb WR & Franks NP. Di€erential sensitivities of mammalian
neuronal and muscle nicotinic acetylcholine receptors to general anesthetics. Anesthesiology 1997; 86:
866±874.
13. Flood P, Ramirez-Latorre J & Role L. Alpha 4 beta 2 neuronal nicotinic acetylcholine receptors in the
central nervous system are inhibited by iso¯urane and propofol, but alpha 7-type nicotinic acetylcholine
receptors are una€ected. Anesthesiology 1997; 86: 859±865.
14. Cardoso RA, Yamakura T, Brozowski SJ, Chavez-Noriega LE & Harris RA. Human neuronal nicotinic
acetylcholine receptors expressed in Xenopus oocytes predict ecacy of halogenated compounds that
disobey the Meyer-Overton rule. Anesthesiology 1999; 91: 1370±1377.
15. Yamakura T, Chavez-Noriega LE & Harris RA. Subunit-dependent inhibition of human neuronal
nicotinic acetylcholine receptors and other ligand-gated ion channels by dissociative anesthetics ketamine
and dizocilpine. Anesthesiology 2000; 92: 1144±1153.
16. Petersen-Felix S, Luginbuhl M, Schnider TW et al. Comparison of the analgesic potency of xenon and
nitrous oxide in humans evaluated by experimental pain. British Journal of Anaesthesia 1998; 81: 742±747.
17. Lynch C III, Baum J & Tenbrinck R. Xenon anesthesia. Anesthesiology 2000; 92: 865±868.
18. Hudspith MJ. Glutamate: a role in normal brain function, anaesthesia, analgesia and CNS injury. British
Journal of Anaesthesia 1997; 78: 731±747.
19. Luttropp HH, Romner B, Perhag L et al. Left ventricular performance and cerebral haemodynamics
during xenon anaesthesia. A transoesophageal echocardiography and transcranial Doppler sonography
study. Anaesthesia 1993; 48: 1045±1049.
20. Hettrick DA, Pagel PS, Kersten JR et al. Cardiovascular e€ects of xenon in iso¯urane-anesthetized dogs
with dilated cardiomyopathy. Anesthesiology 1998; 89: 1166±1173.
21. Nakata Y, Goto T, Ishiguro Y et al. Anesthetic doses of sevo¯urane to block cardiovascular responses to
incision when administered with xenon or nitrous oxide. Anesthesiology 1999; 91: 369±373.
22. Dixon WJ. The up-and-down method for small samples. Journal of American Statistics Association 1965; 60:
967±978.
23. Marx T, Froeba G, Wagner D et al. E€ects on haemodynamics and catecholamine release of xenon
anaesthesia compared with total i.v. anaesthesia in the pig. British Journal of Anaesthesia 1997; 78: 326±327.
24. Boomsma F, Rupreht J, Man in 't Veld AJ et al. Haemodynamic and neurohumoral e€ects of xenon
anaesthesia. A comparison with nitrous oxide. Anaesthesia 1990; 45: 273±278.
25. Smith NT, Eger EI II, Stoelting RK et al. The cardiovascular and sympathomimetic responses to the
addition of nitrous oxide to halothane in man. Anesthesiology 1970; 32: 410±421.
26. Hamberger B & Jarnberg PO. Plasma catecholamines during surgical stress: di€erences between
neurolept and en¯urane anaesthesia. Acta Anaesthesiological Scandinavica 1983; 27: 307±310.
502 P. H. Tonner et al

27. Price HL. Myocardial depression by nitrous oxide and its reversal by Ca‡ ‡ . Anesthesiology 1976; 44:
211±214.
28. Goldberg AH, Sohn YZ & Phear WP. Direct myocardial e€ects of nitrous oxide. Anesthesiology 1972; 37:
373±375.
29. Eisele JH & Smith NT. Cardiovascular e€ects of 40 percent nitrous oxide in man. Anesthesia and Analgesia
1972; 51: 956±961.
30. Fink H, Blobner M, Bogdanski R et al. E€ects of xenon on cerebral blood ¯ow and autoregulation: an
experimental study in pigs. British Journal of Anaesthesia 2000; 84: 221±225.
31. Obrist WD & Wilkinson WE. Regional cerebral blood ¯ow measurement in humans by xenon-133
clearance. Cerebrovascular and Brain Metabolism Reviews 1990; 2: 283±327.
32. Plougmann J, Astrup J, Pedersen J & Gyldensted C. E€ect of stable xenon inhalation on intracranial
pressure during measurement of cerebral blood ¯ow in head injury. Journal of Neurosurgery 1994; 81:
822±828.
33. Yao LP, Bandres J, Nemoto EM et al. E€ect of 33% xenon inhalation on whole-brain blood ¯ow and
metabolism in awake and fentanyl-anesthetized monkeys. Stroke 1992; 23: 69±74.
34. Darby JM, Yonas H, Pentheny S & Marion D. Intracranial pressure response to stable xenon inhalation in
patients with head injury. Surgical Neurology 1989; 32: 343±345.
35. Hartmann A, Dettmers C, Schuier FJ, Wassmann HD & Schumacher HW. E€ect of stable xenon on
regional cerebral blood ¯ow and the electroencephalogram in normal volunteers. Stroke 1991; 22:
182±189.
36. Kashiwagi S, Yamashita T, Nakano S et al. The wash-in/washout protocol in stable xenon CT cerebral
blood ¯ow studies. American Journal of Neuroradiology 1992; 13: 49±53.
37. Latchaw RE, Yonas H, Pentheny SL & Gur D. Adverse reactions to xenon-enhanced CT cerebral blood
¯ow determination. Radiology 1987; 163: 251±254.
38. Stapelfeldt CK, Hahn CP, Tonner PH, Scholz J & Schulte am Esch J. E€ect of xenon on carbon dioxide
reactivity in humans. Anesthesiology 2000; 93: A347.
39. Tonner PH, Stapelfeldt CK, Hahn CP, Scholz J & Schulte am Esch J. Cerebral blood ¯ow velocities during
xenon anesthesia. Anesthesiology 2000; 93: A377.
40. Stapelfeldt CK, Tonner PH & Scholz J. Xenon vs. iso¯urane: e€ects on cerebral blood ¯ow and cerebral
autoregulation in humans. Anesthesiology 2001; 95: A336.
41. Kolbitsch C, Lorenz I, Keller C et al. The in¯uence of increasing concentrations of nitrous oxide on
cerebral blood ¯ow velocity in hypocapnic patients with brain tumours. European Journal Anaesthesiology
1999; 16: 543±546.
42. Aono M, Sato J & Nishino T. Nitrous oxide increases normocapnic cerebral blood ¯ow velocity but does
not a€ect the dynamic cerebrovascular response to step changes in end-tidal P(CO2) in humans.
Anesthesia and Analgesia 1999; 89: 684±689.
43. Watts AD, Luney SR, Lee D & Gelb AW. E€ect of nitrous oxide on cerebral blood ¯ow velocity after
induction of hypocapnia. Journal of Neurosurgical Anesthesiology 1998; 10: 142±145.
44. Dirnhuber MJ & Davies NJ. E€ect of nitrous oxide on cerebral blood ¯ow. British Journal of Anaesthesia
1993; 71: 460.
45. Goto T, Suwa K, Uezono S et al. The blood-gas partition coecient of xenon may be lower than
generally accepted. British Journal of Anaesthesia 1998; 80: 255±256.
46. Froeba G, Baeder S, Calzia E, Georgie€ M & Marx T. Xenon washin and washout time during controlled
mechanical ventilation in an animal model. Applied Cardiopulmonary Pathophysiology 1998; 7: 157±160.
47. Eger EI II & Saidman LJ. Hazards of nitrous oxide anesthesia in bowel obstruction and pneumothorax.
Anesthesiology 1965; 26: 61±66.
48. Reinelt H, Schirmer U, Marx T, Topalidis P & Schmidt M. Di€usion of xenon and nitrous oxide into the
bowel. Anesthesiology 2001; 94: 475±477 (Discussion 6A).
49. Condon RE, Cowles V, Ekbom GA, Schulte WJ & Hess G. E€ects of halothane, en¯urane, and nitrous
oxide on colon motility. Surgery 1987; 101: 81±85.
50. Aitkenhead AR. Anaesthesia and bowel surgery. British Journal of Anaesthesia 1984; 56: 95±101.
51. Ryan P, Schweitzer SA & Woods RJ. E€ect of epidural and general anaesthesia compared with general
anaesthesia alone in large bowel anastomoses. A prospective study. European Journal of Surgery 1992; 158:
45±49.
52. Pedersen FM, Wilken-Jensen C, Knudsen F, Lindekaer AL & Svare EI. The in¯uence of nitrous oxide on
recovery of bowel function after abdominal hysterectomy. Acta Anaesthesiologica Scandinavica 1993; 37:
692±696.
53. Ste€ey EP, Johnson BH, Eger EI II & Howland D Jr. Nitrous oxide: e€ect on accumulation rate and
uptake of bowel gases. Anesthesia and Analgesia 1979; 58: 405±408.
54. Taylor E, Feinstein R, White PF & Soper N. Anesthesia for laparoscopic cholecystectomy. Is nitrous oxide
contraindicated? Anesthesiology 1992; 76: 541±543.
 Xenon as a replacement for N2O 503

55. Krogh B, Jorn Jensen P, Henneberg SW, Hole P & Kronborg O. Nitrous oxide does not in¯uence
operating conditions or postoperative course in colonic surgery. British Journal of Anaesthesia 1994; 72:
55±57.
56. Goto T, Saito H, Nakata Y et al. E€ects of xenon on the performance of various respiratory ¯owmeters.
Anesthesiology 1999; 90: 555±563.
57. McIlroy MB, Mead J, Selverstone NJ & Radford EP. Measurement of lung tissue viscous resistance using
gases of equal kinematic viscosity. Journal of Applied Physiology 1954; 7: 485±490.
58. Drazen JM, Loring SH & Ingram RH Jr. Distribution of pulmonary resistance: e€ects of gas density,
viscosity, and ¯ow rate. Journal of Applied Physiology 1976; 41: 388±395.
59. Wood LD, Engel LA, Grin P, Despas P & Macklem PT. E€ect of gas physical properties and ¯ow on
lower pulmonary resistance. Journal of Applied Physiology 1976; 41: 234±244.
60. Calzia E, Stahl W, Handschuh T et al. Respiratory mechanics during xenon anesthesia in pigs: comparison
with nitrous oxide. Anesthesiology 1999; 91: 1378±1386.
61. Zhang P, Ohara A, Mashimo T et al. Pulmonary resistance in dogs: a comparison of xenon with nitrous
oxide. Canadian Journal of Anaesthesia 1995; 42: 547±553.
62. Pedley TJ & Drazen JM. Aerodynamic theory. In Fishman AP, Macklem PT, Mead J & Geiger SR (eds)
Handbook of Physiology. Bethesda: Williams and Wilkins, 1986.
63. Fink BR. Di€usion anoxia. Anesthesiology 1955; 16: 511±519.
64. Calzia E, Stahl W & Handschuh T. Continuous arterial P(O2) and P(CO2) measurements in swine during
nitrous oxide and xenon elimination: prevention of di€usion hypoxia. Anesthesiology 1999; 90: 829±834.
65. Cohen EN, Belville JW & Brown BW. Anesthesia, pregnancy and miscarriage, a study of operating room
nurses and anesthetists. Anesthesiology 1971; 35: 343±347.
66. Lane GA, Nahrwold ML, Tait AR et al. Anesthetics as teratogens: nitrous oxide is fetotoxic, xenon is
not. Science 1980; 210: 899±901.
67. Burov NY, Kornienko LY, Arzamastev YV, Korotich VN & Golubykh VL. Study of xenon toxicity in a
subchronic experiment. Anestheziologiia i Reanimatologiia 1998; 3: 58±60.
68. Yasuda N, Weiskopf RB, Cahalan MK et al. Does des¯urane modify circulatory responses to stimulation
in humans? Anesthesia and Analgesia 1991; 73: 175±179.
69. FroÈba G. [Xenon as inhalation anesthetic]. Anaesthesiology Intensivimed Notfallmed Schmerzther 1997; 32:
48±51.