Anticonvulsant effect of xenon on neonatal asphyxial seizures
- Denis Azzopardi1,2,
- Nicola J Robertson3,
- Andrew Kapetanakis1,
- James Griffiths4,
- Janet M Rennie5,
- Sean R Mathieson3,
- A David Edwards1
- 1Centre for the Developing Brain, King's College London, London, UK
- 2Institute of Clinical Sciences, Imperial College London, London, UK
- 3Institute for Women's Health, University College London, London, UK
- 4National Perinatal Epidemiology Unit, University of Oxford, Oxford, UK
- 5Department of Neonatal Medicine, University College Hospital, London, UK
- Correspondence to Professor Denis Azzopardi, Centre for the Developing Brain, St Thomas's Hospital, Westminster Bridge Road, London SE1 7EH, UK;
- Received 30 January 2013
- Revised 13 March 2013
- Accepted 14 March 2013
- Published Online First 9 April 2013
Xenon, a monoatomic gas with very high tissue solubility, is a non-competitive inhibitor of N-methyl-D-aspartate (NMDA) glutamate receptor, has antiapoptotic effects and is neuroprotective following hypoxic ischaemic injury in animals. Xenon may be expected to have anticonvulsant effects through glutamate receptor blockade, but this has not previously been demonstrated clinically. We examined seizure activity on the real time and amplitude integrated EEG records of 14 full-term infants with perinatal asphyxial encephalopathy treated within 12 h of birth with 30% inhaled xenon for 24 h combined with 72 h of moderate systemic hypothermia. Seizures were identified on 5 of 14 infants. Seizures stopped during xenon therapy but recurred within a few minutes of withdrawing xenon and stopped again after xenon was restarted. Our data show that subanaesthetic levels of xenon may have an anticonvulsant effect. Inhaled xenon may be a valuable new therapy in this hard-to-treat population.
What is known about this topic
Xenon is a non-competitive blocker of the NMDA receptor and an anaesthetic with very high tissue solubility. It also has antiapoptotic effects and is neuroprotective in animal studies. Early phase neuroprotective clinical studies of inhaled xenon combined with moderate hypothermia for perinatal asphyxial encephalopathy are in progress using specialised recirculating delivery devices.
What this study adds
Subanaesthetic levels of inhaled xenon have anticonvulsant and EEG suppressant effects in infants with asphyxial encephalopathy. Xenon may be a promising anticonvulsant in this hard-to-treat condition.
Xenon, a monoatomic gas with very high tissue solubility, is a non-competitive inhibitor of the N-methyl-D-aspartate (NMDA) glutamate receptor and has antiapoptotic effects.1 It is an anaesthetic with a minimum alveolar concentration of approximately 60% in humans and lower concentrations of xenon are neuroprotective following hypoxic ischaemic injury in animals.2 ,3 Xenon may be expected to have anticonvulsant effects through glutamate receptor blockade, but this has not previously been demonstrated clinically. We now report an anticonvulsant effect of xenon in infants suffering perinatal asphyxial encephalopathy, a condition in which seizures are common and notoriously hard to treat.
We examined seizure activity on the real-time and amplitude-integrated EEG (aEEG) records of 14 full-term infants with perinatal asphyxial encephalopathy who were treated within 12 h of birth with 30% inhaled xenon for 24 h combined with 72 h of moderate systemic hypothermia during an ongoing randomised neuroprotection trial, the TOBY Xe trial (trial number NCT00934700). The selection criteria for this trial are similar to those of the previous TOBY hypothermia neuroprotection trial: gestation between 36 and 43 weeks, evidence of asphyxia at birth, and the presence of moderate or severe encephalopathy and abnormal aEEG. In addition, treatment with cooling must have been initiated within 6 h of birth. The TOBY Xe protocol was approved by the National Research Ethics Committee and the local research ethics committee of each participating hospital. Conduct of the study is overseen by an independent trial steering committee with advice from an independent data monitoring and ethics committee. The TOBY Xe trial protocol is available at http://www.npeu.ox.ac.uk/tobyxe.
The EEG/aEEG was recorded from one channel (two biparietal electrodes) in five infants and from multiple channels in the others, and recording was started within 6 h after birth and continued to at least 6 h after rewarming was completed. The presence of seizure activity on the EEG/aEEG record was assessed visually and by automated seizure detection software, which was available in nine infants. Seizure activity was defined as a change in pattern of the aEEG consistent with previous descriptions of aEEG seizure activity, usually a sudden increase and narrowing of the amplitude followed by a fall in amplitude when the seizure terminated. In all cases, seizure activity was confirmed by the occurrence of repetitive spike and wave discharge on the raw EEG lasting at least 20 s.
The severity of suppression of the EEG/aEEG was also assessed visually before starting, during and after discontinuation of xenon therapy. Automated software to analyse EEG spectral power was available in four cases using the NicoletOne trends software (Carefusion, Middleton, Wisconsin, USA). This software was used to segment the EEG into 2 min epochs and uses Fourier analysis to derive total power between 0.5 and 30 Hz. Further smoothing of the power spectrum to remove high power peaks from patient handling was performed by taking the median value of the 2 min epochs for each hour.
Seizures were identified on 5 of 14 infants, who received treatment with phenytoin and/or phenobarbital. In these infants, the aEEG initially was severely suppressed and seizures occurred on recovery of electrical activity. Seizures stopped during xenon therapy but recurred within a few minutes of withdrawing xenon. Further anticonvulsant therapy was administered and the seizures ceased, recurring transiently in one infant following rewarming. In one infant with meconium aspiration and pulmonary hypertension, xenon was temporarily discontinued because of increasing oxygen requirements due to worsening respiratory complications; seizures recurred within 10 min of discontinuing xenon and they ceased again a few minutes after restarting xenon (figure 1).
Seizures were not present in 9 of 14 records. In three of these nine records, the EEG initially was intermittent, recovered during xenon therapy, and did not change following termination of xenon. In the other 6 of the 9 records without seizures, the EEG initially was severely suppressed, recovered slightly during xenon therapy, but no seizures were noted. The aEEG amplitude became more suppressed on starting xenon inhalation in two infants and there was an increase in amplitude of the aEEG on cessation of xenon in four infants; these changes were noticeable within a few minutes of starting or stopping xenon. Total power showed a marked increase following discontinuation of Xenon therapy in two of the four cases where power spectral analysis was available. Figure 2 shows a smoothed power trend of one case during and post-xenon therapy. The two remaining cases showed a highly suppressed EEG during and post-xenon administration.
Our observation of a paucity of seizures in encephalopathic infants during xenon therapy and their recurrence on discontinuing xenon suggests that xenon has anticonvulsant effects. We observed these EEG suppressant effects in all five infants with seizures treated with xenon and saw an on/off response to xenon when xenon was transiently discontinued. Accumulation of extracellular glutamate, an excitatory neurotransmitter, is considered important in the aetiology of prolonged seizures following perinatal hypoxic–ischaemic injury, and so, xenon would be a suitable potential anticonvulsant. Our preliminary observations are consistent with this hypothesis and with the experimental data showing NMDA glutamate receptor inhibition by xenon.
Seizures are common during asphyxial encephalopathy and are often resistant to standard anticonvulsant therapy. Frequent or prolonged seizures result in depletion of cerebral high-energy phosphates and elevation of cerebral lactate and may contribute to cerebral injury following asphyxia.4 Suppression of seizures, therefore, may be an additional mechanism contributing to the neuroprotective effect of xenon observed in animal studies.
Inhibition of the NMDA receptor by xenon is concentration-dependent: maximum inhibition occurs with 80% xenon, which is above anaesthetic levels, and a dose that is not clinically practicable. Our data indicate that the anticonvulsant and EEG depressant effect of xenon occurs at substantially lower than anaesthetic levels, close to hypnotic levels. Experimental studies show that the neuroprotective effect of xenon is augmented by additional agents that block glutamate activity, allowing lower doses of xenon to be used.5 Purposely designed recirculating xenon delivery devices with minimal gas wastage have been developed for use in newborn infants making this therapy feasible and cost effective.6 Inhaled xenon may be a valuable new therapy in this hard-to-treat population.
Collaborators Aniko Deierl; Nazakat Merchant; Latha Srinivasan; Isabelle Viac; Giles Kendall; Mary Dinan; Cristina Uria; Ratnaval Nandiran, Mark Turner; Virginia Wallace; Nicholas Franks; Mervyn Maze; Gianlorenzo Fagiolo; Joseph Hajnal, Geoff Charles-Edwards; Ernest Cady; Lawrence Abernethy.
Contributors DA is the chief investigator of the TOBY Xe study who analysed the data and wrote the initial draft of the report. JG is the co-ordinator of the TOBY Xe study, and helped draft the report. DA, NJR, AK, ADE contributed to collection and analysis of data and helped draft the report. JMR and SRM helped analyse the data and draft the report.
Funding The TOBY Xe randomised trial “Neuroprotective Effects of Hypothermia Combined With Inhaled Xenon Following Perinatal Asphyxia” is supported by the Medical Research Council (UK). Air Products PLC (Walton on Thames, Surrey, UK) provided the xenon gas. The xenon delivery device was manufactured by SLE Ltd (Croydon UK). Grant number G0701714//1, supported by the National Institute for Health Research King's College London, Imperial College London and University College London Hospitals Biomedical Research Centres.
Competing interests None.
Ethics approval National Research Ethics Service.
Provenance and peer review Not commissioned; externally peer reviewed.