Secondary Logo

Journal Logo

Monitoring and technology

A xenon recirculating ventilator for the newborn piglet: developing clinical applications of xenon for neonates

Faulkner, Stuart D.; Downie, Neil A.; Mercer, Christopher J.; Kerr, Stuart A.; Sanders, Robert D.; Robertson, Nicola J.

Author Information
European Journal of Anaesthesiology: December 2012 - Volume 29 - Issue 12 - p 577-585
doi: 10.1097/EJA.0b013e3283583c4b
  • Free



Xenon is an anaesthetic gas with attractive properties including cardiostability, a swift induction and reversal profile,1 a lack of toxicity,2 substantial cardioprotection, renal protection3 and neuroprotection4–8 in experimental and clinical settings. Xenon is particularly attractive for the neonate as it attenuates isoflurane neurotoxicity,9,10 suggesting that it may improve the safety of neonatal anaesthesia in the future.11 Perhaps of even greater significance, xenon provides additive neuroprotection when combined with hypothermia following hypoxic ischaemic injury in some newborn animal models.4,6,7,11,12 Adjunct therapies are needed to optimise hypothermic neuroprotection, driving the need to develop xenon neuroprotection.

Xenon is currently in use in adults for anaesthesia, particularly in high-risk patients, and has been used in neonates to measure cerebral blood flow.13 However, so far the high cost of xenon (currently approximately €13 per litre) (Knop F, personal communication) and scarcity (0.0087 ppm in air) has limited its use to small numbers of experimental and clinical trials.14–18

Xenon has a swift induction and emergence profile due to its relative insolubility with patient uptake relatively low at 2.4 to 4 l h−1, favouring its use in a closed breathing system.19 Specialised recirculating equipment is required to reduce the cost of xenon use. In the past, the Physioflex (Drager, Lubeck, Schleswig-Holstein, Germany) and the Zeus (Drager) revolutionised closed and semiclosed ventilation; these ventilators used electronically controlled gas flow analysers and gas filling technologies. Briefly, one such ventilator, PhysioFlex-Xe, had closed circuit xenon delivery capabilities.20 However, they fell out of use due to complex software and specialist parts. Because the neuroprotective potential of xenon has been recognised, several semi-closed and closed circuit custom-built systems have been developed.14,18,21,22 A particular challenge of developing a xenon ventilator is presented by the physical properties of the gas23,24 (Table 1)23,25 and it is necessary to design a novel ventilator rather than adapt existing neonatal ventilators.

Table 1
Table 1:
Densities and viscosities of air and air/xenon mixtures at ambient pressure and temperature.24,25

The objective of this study was to design and utilise a stand-alone closed circuit xenon recirculating system for ventilation of an anaesthetised piglet remotely located in the bore of a magnetic resonance (MR) spectrometer for 24 h using an automated control mixture of xenon, air, oxygen and isoflurane. Development of this ventilator utilised four piglets to test the efficacy and safety of the xenon recirculating ventilator (XRV) and to facilitate future neonatal applications of xenon including anaesthesia and neuroprotection. This equipment was subsequently used in a preclinical study using 18 piglets with MR spectroscopy and immunohistochemical outcome biomarkers to assess safety and efficacy of combined xenon and therapeutic hypothermia.26



Ethics approval was granted to perform experiments under UK Home Office Guidelines (Animals for Scientific Procedures Act 1986) and University College London Institutional Animal Care and Use Committee (July 2007).

Two complete XRV systems were built for this study. One was used for operations and the other for testing and validating modifications prior to acceptance. Neither was designed to be magnetic-resonance compatible, but for remote delivery of xenon to a piglet which was located in a MR spectrometer.

Ventilator technical specifications

The XRV was constructed using experience from a previous prototype xenon recirculator ventilator built for a phase I trial in adult humans undergoing elective coronary artery bypass graft surgery.14 The authors reported no adverse clinical consequences or increased embolic load; the latter is associated with increased neurological impairment.27

During the design and validation process, unlike ventilators designed by Westenskow et al.28 and the ‘Physioflex’, we used an internal recirculation and closed mixing loop alongside an external loop to the patient (Fig. 1), and one pump for circulation and ventilation. To avoid substantial software development, the XRV used analogue electronics and pneumatic components and could be adapted by exchange of readily available components to other animals and purposes.

Fig. 1
Fig. 1:
No captions available.

Internal circuit

The internal circuit is made up of an oxygen-compatible diaphragm pump (Techno Takatsuki Co, Takatsuki, Osaka, Japan) leading to a regulator, then an adjustable valve to deliver a variable positive end-expiratory pressure (PEEP). This valve leads to two fuel cell oxygen sensors (Teledyne Co, Thousand Oaks, CA, USA) which lead to a bellows, which provides the adjustable system volume (Fig. 2a). After the bellows, xenon concentration (%) is measured based on the speed of ultrasound pulses and the gas returns via a restricting valve, which allows the system to deliver negative PEEP to the pump inlet. An additional small flow is circulated via an infrared cell to monitor isoflurane concentration (Analytical Development, Hoddeson, Herts, UK) and an infrared cell to monitor carbon dioxide concentration (E2V, Chelmsford, Essex, UK) in the circuit.

Fig. 2
Fig. 2:
No captions available.

A small volume of air-oxygen, isoflurane and xenon is added to the internal circuit prior to the bellows (maximum rate of 3 l min−1) in volumes proportional to those consumed by the piglet or lost through leaks. Fresh gases are governed by thermal mass flow controllers (MFCs) driven by differential instrumentation amplifiers (Fig. 2b). The system can be filled with gas mixtures in approximately 1 min under manual control. The rate of change of gas concentrations is governed by maximum fresh gas flow rates and the system volume. The latter is dominated by the bellows volume (0.6 l) and the CO2 absorber canister.

One of the oxygen sensors triggers the addition of oxygen via its MFC, at a rate determined by the difference between target and achieved oxygen concentration. Similarly, xenon is added when the system volume is lower than target. In addition to electronic flow controls, there are visual rotameter flowmeters (Platon, Domont Cedex, France) on the internal circuit, external circuit and fresh gas flows.

A second MFC in the internal circuit controls accurate isoflurane delivery in the XRV internal circuit by diverting airflow via a sealed glass reservoir containing isoflurane and the vapour circulated within the breathing circuit (Fig. 2a, b). The amount of flow diverted is determined by another differential amplifier.

External circuit

The circuit gas is diverted from the internal circuit to the animal (external circuit) via a humidifier. This humidified gas is then taken via 4-m long pipes, enclosed by a warm water jacket, into the MR spectrometer where the animal is located remotely from the XRV (Fig. 2c). A solenoid valve is positioned between the internal and external parts of the circuit; this valve is activated by a simple electronic timer controlling the inspiratory and expiratory phases of ventilation. A magnahelic pressure gauge provides visual confirmation of ventilation.

The exhaust gas mixture from the piglet containing carbon dioxide returns from the external circuit via a condensing copper coil, thermoelectric (Peltier) cooler and then a water trap. After the water trap, the exhaust gas passes through a soda-lime carbon dioxide absorber and back to the XRV internal circuit.

Air, xenon and isoflurane delivery

Actual and preset gas concentrations (air, xenon, isoflurane) are displayed digitally. A toggle switch and rotary potentiometers allow preset gas concentration selection. The bellows level is indicated on a vertical linear LED display, and the bellows level set point is controlled by a linear potentiometer. The ventilator oscillator used gives continuous mandatory ventilation provided by a timer with numerical ‘countdown’ display (Omron H8GN) allowing exact inspiration and expiration times (10-ms intervals) and ratios to be set in percentage points. The peak inspiratory pressure (PIP) is provided by an adjustable pressure regulator, whereas PEEP is adjusted by a manual valve.

Xenon is added according to an algorithm programmed to maintain a constant system volume. As the piglet consumes oxygen and the system volume decreases, xenon is added, limited only by the preset value of oxygen (minimum 21%) and residual nitrogen within the circuit (minimum approximately 10%). An ultrasonic device situated within the flow circuit measures the xenon concentration. This has previously been validated with a mass spectrometer.14

Isoflurane was used prior to and during xenon administration at 1.8 to 2% [0.9 to 1.1 minimum alveolar concentration (MAC) without xenon].29 As the MAC of xenon in the pig is 119%30 (vs. 63 to 71% in humans31), 50% xenon could not be used alone without additional sedation.

Avoidance of hypoxia

Delivery of fresh oxygen to the circuit is controlled automatically, avoiding any user error. If the concentration of oxygen is less than target, oxygen is injected at a rate proportional to the difference between the two. The oxygen MFC was capable of higher flow rates than other gases, so oxygen would always have priority over other gases. A second, independent fuel cell (Vandagraph, Keighley, Yorkshire, UK) measuring oxygen confirms accuracy of the primary analyser and has high and low oxygen alarms.


The humidifier has high/low humidity alarms. An LED and a solenoid valve provide visual and auditory indicators for the inspiration cycle on the front panel and an audio alarm for high/low oxygen concentration. In the event of power or ventilator failure, the XRV can be disconnected from the animal by pulling out two quick-fit connections, and a backup ventilator connected.

Isoflurane and xenon calibration

Unlike most anaesthetic gases, xenon does not absorb in the near infrared spectrum nor can it be measured electrochemically. Therefore, it is measured via its ultrasonic speed. Isoflurane influences this speed of sound value and, therefore, the readout concentration of xenon. A specific xenon–isoflurane calibration curve was required to give a ‘xenon readout error’. At circuit flow rates of 3 l min−1 and isoflurane concentrations of 0 to 4% (volume/volume; v/v), xenon readout values were recorded in triplicate and a ‘xenon readout error’ curve created.

Xenon use

Because more than 99% of inhaled xenon is exhaled and only a small volume of oxygen is consumed by the piglet, more than 95% of fresh gas added to the circuit should constitute that lost to leaks, with 50% of that volume comprising xenon gas.

Additionally, arterial blood samples were collected after 12 h of xenon administration and blood:gas partition levels of xenon (estimated by head-space gas chromatography) were assessed.26 To further assess the integrity of the XRV for 24-h ventilation in a large number of piglets over 12 months, the total volume of xenon gas used was measured from 18 piglets (also four piglets in pilot studies).26


Prior to and after xenon inhalation, piglets were ventilated on a pressure-controlled neonatal open circuit ventilator (SLE ltd, Croydon, UK). Fresh gas flow rate was 5 to 6 l min−1 with PIP of 14 to 18 cmH2O and PEEP of 0 to 5 cmH2O, using an air (nitrogen approximately 79% and oxygen approximately 21%)–oxygen mixture with isoflurane at 2% v/v; fractional inspired oxygen concentration (FiO2) was adjusted to maintain oxygen saturation at more than 95%.

Circuit denitrogenation on the XRV with approximately 70% oxygen allowed for an optimised wash in of xenon gas; excess oxygen was replaced gradually with xenon (50%) and finally isoflurane (approximately 1%) was added. The circuit was tested for leaks by observing any change in bellows height during a 3-min period during which no fresh gas was added. Any leak was rectified before continuing.

Changing between ventilator circuits

Gas flow to the internal XRV circuit was briefly closed, inhalation and exhalation tubes from the piglet were disconnected from the open circuit ventilator and reconnected to the XRV and the internal circuit reopened. Physiological parameters were monitored closely for 5 min after switching to the XRV with the backup ventilator always situated next to the XRV.

Experimental methods

Four male piglets (1.95 kg ± 0.1 SD) within 24 h of birth were anaesthetised (isoflurane 1 to 4% v/v), a tracheostomy performed and an uncuffed endotracheal tube sutured in place. The piglet was ventilated mechanically to maintain arterial partial pressures of oxygen (PaO2) and carbon dioxide (PaCO2) at 8 to 13 and 4.5 to 6.5 kPa, respectively, allowing for temperature correction of the arterial blood sample. An umbilical venous catheter was inserted for infusion of maintenance fluids (10% dextrose, 60 ml kg−1 per 24 h), morphine (0.05 mg kg−1 h−1) and antibiotics (benzylpenicillin 50 mg kg−1 and gentamicin 2.5 mg kg−1, every 12 h). An umbilical arterial catheter was inserted for continuous heart rate (HR) and arterial blood pressure monitoring, and intermittent blood sampling was used to measure PaO2, PaCO2 and pH, and concentrations of electrolytes, glucose and lactate (Abbott Laboratories, Maidenhead, UK). Bolus infusions of colloid (Gelofusine, B. Braun Medical, Emmenbrucke, Switzerland) and dopamine (5 to 20 μg kg−1 min−1) were used to maintain mean arterial blood pressure (MABP) more than 40 mmHg. Rectal temperature was maintained using a warm water mattress positioned below the animal and circulating warm air.

Piglets were placed in a plastic pod and the head was fixed in a stereotactic frame. Transient hypoxic ischaemia was induced by remote occlusion of both common carotid arteries using inflatable balloon occluders and reducing the FiO2 to 12% for approximately 25 min.26 Two hours after hypoxic ischaemic, ventilation was switched to the XRV administering 50% xenon and isoflurane (approximately 1%) for 24 h (from 2 to 26 h following hypoxic ischaemic).


Isoflurane and xenon calibration

The ‘xenon readout error’ at different isoflurane concentrations (0 to 4% v/v) demonstrated a linear relationship (R2 = 0.98). At the required level of isoflurane (1%) and xenon (50%), the ‘readout error’ was approximately 5% (Fig. 3).

Fig. 3
Fig. 3:
No captions available.

Pilot results summary

In the first two pilot studies, hypercapnia occurred within the first 2 h following transfer to the XRV; PaCO2 increased from 5.4 to 11 kPa, indicating insufficient PIP. Different solenoid valves were fitted to improve the effective PIP/PEEP range, allowing subsequent easy adjustment of PaCO2 at 4.5 to 6.3 kPa for 24 h of XRV ventilation in subsequent pilot studies.

In the third pilot study, the introduction of 50% xenon on a background of isoflurane led to a brief tachycardia and hypotension. The HR increased from 165 beats per minute (SD 15) to 204 (22) beats per minute and MABP decreased from 48 (2) to 35 (5) mmHg. Although xenon remained at 50%, further reductions in MABP and HR occurred on each occasion when isoflurane was increased from 1 to 1.5% (Fig. 4). To counteract these events, in the fourth pilot study, xenon was introduced in a stepwise fashion starting at 20% and increasing to 50% over 1 h, adjusting the isoflurane concentration to maintain an overall total MAC of 1 (Fig. 5).

Fig. 4
Fig. 4:
No captions available.
Fig. 5
Fig. 5:
No captions available.

Changes to the isoflurane concentration required venting of the bellows. Xenon loss was reduced by directing the circuit flow to a nanoporous coconut-based activated carbon isoflurane scavenger (Chemviron: Aquacarb 207C in granules >6 <12 mesh) on the return circuit. By the addition of two manual plug valves on the outward and return parts of the external circuit, a brief closure of the internal circuit was possible during suctioning. The scavenger was reactivated by heating to 20°C in a dry nitrogen stream for 24 h.

A pulsed infrared beam arrangement (mk120 Velleman Nv, Belgium) was added to the side of the bellows and set to maximum and minimum bellows height to prevent nonventilation in situations of full or empty bellows. Electronic output ports (Bayonet Neill–Concelman) were added to the rear panel of the XRV to allow ‘real time’ gas concentrations to be viewed onscreen and saved to file on a data logger (Grant instruments, Cambridge, UK) for retrospective analysis (Fig. 6).

Fig. 6
Fig. 6:
No captions available.

Xenon use

Ventilation leakage over 24 h was minimal. Sources of leak included space around the tracheostomy tube, transfer of the piglet between circuits, manual vents and isoflurane changes. The overall rate of xenon use was 10.4 ml min−1 which equates to approximately 0.6 l h−1 or approximately €8 h−1. With 50% xenon, this equates to a 20.8 ml min−1 total circuit volume leak rate. Circuit concentrations of carbon dioxide were 0.15 to 0.16%, indicating that carbon dioxide scavenging was efficient and oxygen concentration was adjusted to maintain arterial oxygen saturation more than 95%. Blood xenon concentrations during the 24-h delivery (postwash-in phase) in these pilots were consistent at approximately 50%. Additionally, in a subset of 18 piglets in which blood gas xenon concentration was measured by head-space gas chromatography, the blood levels were consistent with the numerical readout on the xenon ventilator at the time of sampling, at approximately 50%.26


Closed circuit ventilators have been commercially available for over 80 years. There is now renewed interest in these ventilators as demand for xenon gas in anaesthetic, neuroprotective and cardioprotective roles increases. Our XRV was designed as a stand-alone ventilator and xenon-closed circuit delivery system for large animal experimental use. Compared to other closed circuit ventilators, the XRV was simpler to build as it required just one pump for both circulation and ventilation, with bellows providing an adjustable system volume22 corrected by MFC. The choice of analogue circuitry avoided the need for software quality assurance and testing. In addition, several safety alarm systems were added, for example for low oxygen concentration and ventilation failure.

The use of a clinical bellows as an automated ‘reservoir’ for gas mixing allowed accurate measurement and electronic adjustment of gas mixtures. Automated xenon (and other gas) addition allowed precise control and stability of the concentration of xenon in the circuit. Other circuits use manual addition of xenon gas via small, pressurised gas cartridges every 2 h.21 Although these smaller volumes preclude large xenon waste from a pressurised cylinder, they are a distracting user interaction and an additional consumable. Uptake of xenon is swift32 and minimal additions are needed via the MFC during ventilation. However, a proportion of the circuit volume had to be vented if large concentration changes in xenon or isoflurane were required. Overall, the total circuit leak rate was minimal with xenon utilisation low, estimated at only 0.6 l h−1. The higher xenon use in our study compared to 0.18 l h−1 reported by Chakkarapani et al.21 can be explained by the longer duration of xenon ventilation in our study (24 vs. 16 h), switching between circuits and use of isoflurane, all of which resulted in additional losses of xenon. The nylon tubing, metal and glass components are not expected to have any leakage or adsorption, although a small number of elastomer ‘O’ rings and rubber bellows would have some adsorption and leakage of xenon. The measured leak values probably consisted of a combination of (very small) adsorption effects, diffusion leaks and actual leaks due to small clearance-type gaps in a few components. Continuous ventilation with xenon gas for 24 h equated to €8 h−1 or €195 per piglet compared to more than €6500 per piglet on an open circuit. These costs are similar to those of anaesthetics such as sevoflurane (€10.50 h−1 for a 2-l fresh gas flow rate).33 The blood:gas partition coefficient of xenon is low and equilibration between inspired and blood concentrations is expected to be reached shortly after induction. This stability of xenon blood and circuit gas was demonstrated in the XRV with blood and circuit xenon concentrations both measuring approximately 50%. Thus, the major component of the (very small) changes in circuit xenon concentration was probably a combination of minor leaks and component adsorption rather than any notable changes in uptake by the piglets.

The greater density of xenon (4.5 that of air) might be expected to lead to decreased gas flow in the ventilator tubing and piglet lungs. However, because most flow here is laminar and xenon is only 28% more viscous than air, xenon has little effect on gas flow or lung compliance.23 The hypercapnia which occurred initially after switching ventilation to the XRV was easily corrected and not seen in subsequent pilot studies.

Haemodynamic changes which followed the addition of 50% xenon to isoflurane could be due to potential myocardial depressant effects of opioids and anaesthetics. These effects were reduced by introducing xenon at 20% and gradually increasing to 50% over 1 h whereas simultaneously reducing the concentration of isoflurane to achieve an overall MAC of 1. The gradual xenon wash-in was economical22 because only one vent was required to achieve the initial 20% xenon concentration and the concentration was stable at 50% thereafter. For clinical anaesthesia this approach is unlikely to be feasible, as it would take too long to achieve an adequate depth of anaesthesia. However, in the clinical setting, we anticipate that xenon may not require co-administration of a volatile agent, reducing the haemodynamic burden of two anaesthetic agents and, therefore, allowing faster wash-in of xenon.

Limitations and improvements

The XRV was custom-built for a preclinical experimental series and not intended as a commercial device. It lacked the direct adaptation of an existing neonatal ventilator such as SLE or Drager. However, the ‘dialup’ system for gas was similar to current neonatal ventilators and switching between ventilators was swift, uncomplicated and negated a longer wash-in phase that would be required using the same machine for standard and xenon ventilation. In future models, it would be possible to internalise many of the flowmeters and reduce the unit size by approximately 30%. The XRV can be adapted easily to operate in standard and xenon modes.

For neuroprotection, co-administration of isoflurane with xenon may not be required in neonates. Clinically, xenon has yet to be used clinically in combination with other volatile anaesthetics. Thus, some clinical XRVs (without co-administration of isoflurane) could be far simpler and xenon loss minimised further. Nonetheless, the facility to combine volatile and xenon anaesthesia may have utility to reduce xenon dosing, and, therefore, cost, if lower doses of xenon are found to be protective.

Although a slow wash-in of xenon over 1 h was economical, the circuit required venting to remove excess xenon for each new study animal. However, this approach is unlikely to be suitable for clinical anaesthesia unless only very low levels of xenon (requiring minimal wash-in time) are required to provide neuroprotection against injury. Recapturing xenon from the vent port would be the most viable option for reducing xenon waste because the wash-in phase can consume approximately 5 l h−1 (approximately €40 h−1) of xenon.18

A microcuff endotracheal tube could have been used to reduce xenon loss. We did not use one here, but instead secured the endotracheal tube tightly to the trachea with sutures. In the clinical situation, we would recommend the use of cuffed tubes during the administration of xenon; such tubes are already in safe use in some centres.34

The safety of the XRV could be enhanced further by a xenon concentration limiter and xenon concentration alarms. A safe power supply, internal battery and audio alarm for power failure would be vital for translation to clinical practice and could be added in future.

In summary, we designed, built and utilised a closed circuit xenon and anaesthetic ventilator which was suitable for prolonged use and was economical (€8 h−1), in a preclinical asphyxia piglet model. The XRV contained several important safety features to prevent hypoxia, failed ventilation and excessive noxious gases. Several concepts from the development of this XRV are relevant to future preclinical studies in large animal models using other noble gas mixtures and the development of neonatal xenon ventilators for the current phase II clinical trial of xenon-augmented hypothermia (TOBY Xe; NCT00934700).


Assistance with the study: we thank Professor Nick Franks for expert guidance and use of gas chromatography equipment to measure the concentration of xenon dissolved in blood.

Financial support and sponsorship: financial assistance consisted of part funding from Air Products PLC, Basingstoke, England and from the UK Department of Health's National Institute for Health Research Biomedical Research Centres funding scheme.

Conflicts of interest: RDS has acted as a consultant for Air Liquide on the development of medical gases including xenon. He has no shares or royalties in the company. The other authors have no conflicts of interest.

Patents: Downie NA, Kerr SA (2006) ‘Monitoring medical gas xenon concentration using ultrasonic gas analyser.’ US Patent 7434580.

Downie NA, Kerr SA (2006) ‘Medical gas recirculation system.’ ‘WO worldwide placeholder patent’ WO03092776.


1. Goto T, Suwa K, Uezono S, et al. The blood-gas partition coefficient of xenon may be lower than generally accepted. Br J Anaesth 1998; 80:255–256.
2. Ma M, Zhang C, Liu Y, et al. The influence of hypoxia inducible factor-1alpha expression on apoptosis and proliferation of T lymphocyte in the Peyer's patches after burn with delayed fluid resuscitation in rats at high altitude. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue 2009; 21:296–299.
3. Ma D, Lim T, Xu J, et al. Xenon preconditioning protects against renal ischemic-reperfusion injury via HIF-1alpha activation. J Am Soc Nephrol 2009; 20:713–720.
4. Dingley J, Tooley J, Porter H, Thoresen M. Xenon provides short-term neuroprotection in neonatal rats when administered after hypoxia-ischemia. Stroke 2006; 37:501–506.
5. 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.
6. Ma D, Hossain M, Chow A, et al. Xenon and hypothermia combine to provide neuroprotection from neonatal asphyxia. Ann Neurol 2005; 58:182–193.
7. Ma D, Hossain M, Rajakumaraswamy N, et al. Combination of xenon and isoflurane produces a synergistic protective effect against oxygen-glucose deprivation injury in a neuronal-glial co-culture model. Anesthesiology 2003; 99:748–751.
8. 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.
9. Sanders RD, Davidson A. Anesthetic-induced neurotoxicity of the neonate: time for clinical guidelines? Paediatr Anaesth 2009; 19:1141–1146.
10. Sanders RD, Ma D, Brooks P, Maze M. Balancing paediatric anaesthesia: preclinical insights into analgesia, hypnosis, neuroprotection, and neurotoxicity. Br J Anaesth 2008; 101:597–609.
11. Ma D, Williamson P, Januszewski A, et al. Xenon mitigates isoflurane-induced neuronal apoptosis in the developing rodent brain. Anesthesiology 2007; 106:746–753.
12. Hobbs C, Thoresen M, Tucker A, et al. Xenon and hypothermia combine additively, offering long-term functional and histopathologic neuroprotection after neonatal hypoxia/ischemia. Stroke 2008; 39:1307–1313.
13. Jayasinghe D, Gill AB, Levene MI. CBF reactivity in hypotensive and normotensive preterm infants. Pediatr Res 2003; 54:848–853.
14. Lockwood GG, Franks NP, Downie NA, et al. Feasibility and safety of delivering xenon to patients undergoing coronary artery bypass graft surgery while on cardiopulmonary bypass: phase I study. Anesthesiology 2006; 104:458–465.
15. Kunitz O, Baumert JH, Hecker K, et al. Xenon does not modify mivacurium induced neuromuscular block. Can J Anaesth 2005; 52:940–943.
16. Coburn M, Kunitz O, Baumert JH, et al. Randomized controlled trial of the haemodynamic and recovery effects of xenon or propofol anaesthesia. Br J Anaesth 2005; 94:198–202.
17. 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.
18. Rawat S, Dingley J. Closed-circuit xenon delivery using a standard anesthesia workstation. Anesth Analg 2010; 110:101–109.
19. Luttropp HH, Thomasson R, Dahm S, et al. Clinical experience with minimal flow xenon anesthesia. Acta Anaesthesiol Scand 1994; 38:121–125.
20. Nathan N, Sperandio M, Erdmann W, et al. PhysioFlex: a target-controlled self-regulating closed-circuit inhalation anesthesia regulator. Ann Fr Anesth Reanim 1997; 16:534–540.
21. Chakkarapani E, Thoresen M, Hobbs CE, et al. A closed-circuit neonatal xenon delivery system: a technical and practical neuroprotection feasibility study in newborn pigs. Anesth Analg 2009; 109:451–460.
22. Roehl AB, Goetzenich A, Rossaint R, et al. A practical rule for optimal flows for xenon anaesthesia in a semi-closed anaesthesia circuit. Eur J Anaesthesiol 2010; 27:660–665.
23. Baumert JH, Reyle-Hahn M, Hecker K, et al. Increased airway resistance during xenon anaesthesia in pigs is attributed to physical properties of the gas. Br J Anaesth 2002; 88:540–545.
24. Katz I, Caillibotte G, Martin AR, Arpentinier P. Property value estimation for inhaled therapeutic binary gas mixtures: He, Xe, N2O, and N2 with O2. Med Gas Res 2011; 1:28.
25. Baumert JH, Hein M, Gerets C, et al. The effect of xenon anesthesia on the size of experimental myocardial infarction. Anesth Analg 2007; 105:1200–1206.
26. Faulkner S, Bainbridge A, Kato T, et al. Xenon augmented hypothermia reduces early lactate/N-acetylaspartate and cell death in perinatal asphyxia. Ann Neurol 2011; 70:133–150.
27. Stroobant N, Van Nooten G, Van Belleghem Y, Vingerhoets G. Relation between neurocognitive impairment, embolic load, and cerebrovascular reactivity following on- and off-pump coronary artery bypass grafting. Chest 2005; 127:1967–1976.
28. Westenskow DR, Johnson CC, Jordan WS, Gehmlich DK. Instrumentation for measuring continuous oxygen consumption of surgical patients. IEEE Trans Biomed Eng 1977; 24:331–337.
29. Hecker KE, Reyle-Hahn M, Baumert JH, et al. Minimum alveolar anesthetic concentration of isoflurane with different xenon concentrations in swine. Anesth Analg 2003; 96:119–124.
30. Hecker KE, Horn N, Baumert JH, et al. Minimum alveolar concentration (MAC) of xenon in intubated swine. Br J Anaesth 2004; 92:421–424.
31. Saito H, Saito M, Goto T, Morita S. Priming of anesthesia circuit with xenon for closed circuit anesthesia. Artif Organs 1997; 21:70–72.
32. Hanne P, Marx T, Musati S, et al. Xenon: uptake and costs. Int Anesthesiol Clin 2001; 39:43–61.
33. Lerman J. Gas flow in the upper airway: turbulent or laminar? Paediatr Anaesth 2009; 19:1241.
34. Weiss M, Gerber AC. Cuffed tracheal tubes in children: things have changed. Paediatr Anaesth 2006; 16:1005–1007.

anaesthesia; closed circuit ventilation; neonate; neuroprotection; xenon

© 2012 European Society of Anaesthesiology