Epileptiform activity during inert gas euthanasia of mice

Abstract

Carbon dioxide (CO2) is one of the most commonly used euthanasia agents for mice, yet it is highly aversive and nociceptive. Inert gases are a possible alternative, however there are qualitative reports of seizures resulting from exposure. Here we evaluate epileptiform activity caused by inert gases (N2, He, Ar and Xe) and CO2 in mice chronically instrumented for EEG/EMG undergoing single-gas euthanasia. We found that N2, He and Ar caused epileptiform activity in all animals, CO2 in half of animals and no epileptiform activity produced by Xe. Atmospheric O2 concentrations at epileptiform activity onset were significantly higher for CO2 than for all other gases and occurred soon after loss of motion, whereas N2 and Ar epileptiform activity occurred at cessation of neocortical activity. Helium caused the longest epileptiform activity and these commenced significantly before isoelectric EEG. We did not detect any epileptiform activity during active behaviour. Taken together, these results demonstrate that whilst epileptiform activity from inert gases and particularly Ar and N2 are more prevalent than for CO2, their occurrence at the onset of an isoelectric EEG is unlikely to impact on the welfare of the animal. Epileptiform activity from these gases should not preclude them from further investigation as euthanasia agents. The genesis of epileptiform activity from CO2 is unlikely to result from hypoxia as with the inert gases. Helium caused epileptiform activity before cessation of neocortical activity and for a longer duration and is therefore less suitable as an alternative to CO2.

Fig 1. Experimental design.

(A) Graphic representation of the experimental apparatus. The gas flow controller was calibrated to deliver precise amounts of each gas used and to switch from 21% oxygen at the end of the baseline period. (B) Timeline of the experimental procedure. (C) Graphic demonstrating the typical visual appearance of epileptiform activity. Not all elements were omnipresent, however lateral recumbency and hind limb movements were exhibited by all animals during epileptiform activity. (D) Example of electrophysiological appearance of an epileptiform event, taken from an Ar recording. Note the low muscle tone before onset. Epileptiform activity was characterised by high amplitude, highly synchronous bursting in the EEG. Note in this example the bursts are interspersed with very low EEG activity as the animal approaches cessation of neocortical activity.

Fig 2. Demographics of epileptiform activity.

(A) Prevalence of epileptiform events resulting from exposure to each gas. Note that electrophysiological epileptiform activity was exhibited by CO₂; however, only one out of six animals demonstrated physical signs of epileptiform activity. Only Xe did not result in any epileptiform activity. (B) Duration of epileptiform events. CO₂ produced the shortest epileptiform events, and He the longest. There was no difference between N₂ and Ar.

Fig 3. Timing of epileptiform event onset.

(A) Representative EEG/EMG trace showing the criteria for determining LOM. Note the change in EEG from a low amplitude fast (awake) pattern to high amplitude slower rhythm. Note also that changes in EMG activity (LOM) occur several seconds before EEG activity changes. (B) Representative example from Xe recording (i.e.: no epileptiform activity) of the criteria for cessation of neocortical activity: defined as the point of onset of consolidated isoelectric EEG. (C) Time delay from LOM to epileptiform activity onset. Epileptiform events from CO₂ occurred soon after cessation of neocortical activity. There was a significant delay for other gases, with He induced epileptiform activity occurring latest. (D) Time of epileptiform activity onset before cessation of neocortical activity. N₂ and Ar epileptiform activity occurred at the point of cessation of neocortical activity. (E), (F) Representative EEG/EMG traces of the onset of epileptiform activity and cessation of neocortical activity for N₂ and CO₂ respectively. Note the EMG tone occurring after cessation of neocortical activity, which indicates the continued activity of spinal and brainstem reflexes. (G) Normalised power spectra of EEG for 15s after LOM and also natural sleep (NREM). CO₂ and He resulted in brain activation whereas other gases reduced cortical arousal compared to sleep.

Fig 4. Oxygen concentration at epileptiform event onset.

(A) Average oxygen titration curves during gas recordings, starting with a 5-minute baseline at 21% oxygen. There was no difference between groups. (B) Oxygen concentrations at epileptiform event onset. CO₂ epileptiform events started at higher oxygen concentrations than other gases. Note that He epileptiform activity started at lower oxygen concentrations than N₂ and Ar.