Decompression killing of laboratory rodents: a humane alternative to carbon dioxide?

Rodents are the most widely used laboratory animals worldwide, and millions of mice and rats are used annually in scientific and medical research. The vast majority of these are killed either during or after the research, and the most common method is to expose them to a rising concentration of carbon dioxide (CO2). However, killing rodents in this way gives rise to important welfare concerns. Studies have shown that CO2 exposure is aversive to rodents because it induces anxiety, breathlessness and above certain concentrations, pain. While researchers have examined different ways of exposing rodents to carbon dioxide with the aim of reducing the welfare impact of this process, there appears to be very limited scope for meaningful refinement. There is an urgent unmet need to identify a humane alternative to this method of killing.

Induction of unconsciousness by hypoxia (a deficiency in oxygen reaching body tissues) is considered to be humane in many species, and there is growing interest in the potential of hypobaric hypoxia (a process equivalent to rapidly ascending to high altitude) as a high welfare method of killing. During gradual decompression, animals are rendered unconscious by hypobaric hypoxia as air pressure and therefore oxygen availability is reduced. Although there is evidence that rodents also find hypoxia aversive, this work has been done with inert gases (e.g. Argon) and rapid hypoxia exposure rates. Gradual decompression offers the means to finely control hypobaric hypoxic exposure over variable time periods, improving the likelihood of a humane death. We have encouraging pilot data that suggests that hypobaric hypoxia has the potential to be a method of killing for laboratory rodents that is superior, in welfare terms, to gradual exposure to CO2. The aim of this project is to systematically investigate whether hypobaric hypoxia could be a humane, reliable and efficient method of killing for laboratory rodents.

Initially, we will conduct experiments to determine suitable and effective decompression rates for mice and rats, using anaesthetised animals to avoid any welfare harms. We will then work on conscious mice to determine the time to loss of consciousness during exposure to candidate decompression curves and specifically look for signs of pain, anxiety and aversion to the process. These studies will be based on measurement of physiological responses and spontaneous behaviour, and also on a test where we will train mice to use an escape door if they wish to leave an environment that they find aversive. The responses of mice undergoing decompression will be compared to mice exposed to CO2 in the same way as current laboratory practice, to evaluate which (if any) decompression rates are demonstrably more humane. We expect these experiments to clearly show not just the extent of aversion to the different killing methods but to allow us to identify the potential reasons for aversion. If our studies show that hypobaric hypoxia has promise as a humane approach to killing of mice, we will determine an optimal decompression curve, defined as one which causes less pain, anxiety and aversion than gradual exposure to CO2. If our results show that gradual decompression is not better than CO2, the project will provide a definitive answer on hypobaric hypoxia and the results will also provide novel information on the welfare impact of CO2.

As a potentially humane alternative, wide adoption of hypobaric hypoxia by the laboratory animal science community would have a profoundly positive effect, enabling millions (and potentially billions) of laboratory animals to be reliably killed with low welfare costs. This outcome would also benefit the operatives whose legal and ethical responsibility it is to kill laboratory animals as humanely as possible. In addition, the killing of laboratory rodents in the most humane way possible will also help to enhance public acceptance of in-vivo animal research.

Gradual exposure to carbon dioxide (CO2) remains the most common method of killing laboratory rodents, despite considerable evidence that it is aversive because it induces anxiety, dyspnoea and at high concentrations, nociceptive activation. Attempts to refine exposure to CO2 as a euthanasia method have not solved these issues, and there is an urgent unmet need to improve this aspect of laboratory animal welfare. There is increasing interest in progressive hypobaric hypoxia (gradual decompression) as a high welfare method of killing animals. Although hypoxia is also aversive to rodents, studies are limited and gradual decompression offers the control necessary to achieve slow rates of hypoxia (which cannot be achieved by replacement of air by inert gas), improving the likelihood of a humane death. We propose to systematically investigate the feasibility of hypobaric hypoxia as a humane and efficient method of killing for laboratory rodents.

Our specific objectives are to
(1) Determine candidate decompression curves by assessing the effectiveness of hypobaric hypoxia by characterising physiological, behavioural and pathological responses in anaesthetised mice and rats.
(2) Assess the welfare impact of candidate decompression curves, compared to gradual exposure to CO2, by determining the time to loss of consciousness via electroencephalogram recordings, assessing cardiac and respiratory responses, recording spontaneous behavioural responses and the effects of analgesic and anxiolytic drug interventions, and testing aversion to each killing method via an operant motivational trade off paradigm.

We aim to determine an optimal decompression curve, defined as one which causes less pain, anxiety and aversion than gradual exposure to CO2. This work has the potential to be hugely impactful, providing the possibility to refine the most common procedure carried out on laboratory rodents, improving the welfare of millions of animals at the time of their death.

A conservative estimate is that over 35 million mice and over 8 million rats are used in scientific research annually and the majority of these animals will be killed as part of the research or because they are surplus to requirements. This could represent an underestimate as it may not fully account for genetically modified rodents that are not used in further procedures or those animals killed by breeders as being surplus. The majority of these animals are killed using the Schedule 1 method of gradual exposure to carbon dioxide. Despite the research focus on alternative approaches to the humane killing of laboratory rodents, carbon dioxide exposure remains the standard method yet there is increasing evidence that it causes significant welfare harms including aversion, anxiety and fear-related behaviours. It is an ethical and legal requirement that laboratory animals are killed as humanely as possible, thus there is an urgent unmet need to refine this aspect of laboratory animal use. Our aim is to determine whether hypobaric hypoxia (achieved by exposure to gradual decompression) could represent a high welfare alternative method of killing for laboratory rodents with CO2. In line with the BBSRC (and NC3Rs) priority to improve the welfare of managed animals, this research will provide crucial proof of principle data on whether hypobaric hypoxia has the potential to provide a better death for laboratory mice and rats, based on direct comparison to carbon dioxide exposure. An iterative approach will allow for the determination of optimal decompression rates, evidenced by a comprehensive and corroborative approach to rodent welfare assessment. This information is needed to underpin the generation of novel policy, practice and legislation safeguarding welfare during the killing of laboratory rodents. This has been the case in our recent work on poultry, where our findings have directly informed a favourable European Food Safety Authority opinion allowing gradual decompression (also known as low atmospheric pressure stunning) to be added to the permitted approaches in Regulation (EC) no. 1099/2009 On the Protection of Animals at the Time of Killing. As a potentially humane alternative, wide adoption of hypobaric hypoxia by the laboratory animal science community would enable laboratory animals to be reliably killed with low welfare costs, benefitting many millions (and potentially billions) of animals worldwide. Application of a fixed hypobaric hypoxic dose via a purpose built device would provide greatly improved standardisation of killing, removing human error or variability in application, further protecting welfare. Such consistency in the killing process could also enhance tissue harvesting possibilities and improve scientific outcomes, potentially reducing animal numbers. It is well recognised that there is a potential negative psychological impact for technical staff that must carry out killing, and the provision of a standardised and humane decompression device is likely to improve staff experience when dealing with this difficult area, thus benefitting the wellbeing of technical and scientific staff. While killing methods are a difficult topic with which to engage the general public, societal trust is important for the financial support and advancement of science and wide uptake of a demonstrably improved approach to killing of laboratory animals is likely to enhance public acceptance of this problematic area of animal use.