Ethics of Research involving animals
Reduction - continuation IV
Refining experimental procedures
12.17 As we have illustrated in Chapters 5-9, a very wide variety of experimental procedures is applied to laboratory animals, from those that are relatively minor, such as blood sampling, through to major surgery. The procedures themselves may cause adverse effects (paragraph 9.28) and there may also be adverse effects as a result of a procedure (paragraph 4.54). For this reason it is crucial to consider what opportunities there are to refine every aspect of each procedure from start to finish. One particular category of procedures where there is great potential for Refinement is the administration of substances to animals.23 Such procedures are required for many experiments, for example to create a disease in order to study it, to test the effectiveness of a new medicine, or to assess the toxicity of a chemical.
There is a variety of techniques employed for such purposes, and in each case it is important to think about Refinement with respect to the animal’s immediate experience of the administration method and all that it entails. This assessment should include any distress from necessary handling and restraint (see paragraphs 4.44–4.47), as well as from the administration method itself. The substance administered can also have a profound effect on the animal in the short and long term. For example, it may irritate the animal’s nose or stomach, or cause nausea or seizures.
12.18 The potential for Refinement may be understood better in considering a specific example, such as the injection of a substance into an animal’s joint to study arthritis (see paragraph 6.7) or to ascertain the efficacy of medicines to treat the disease. This procedure can be very painful and has the potential to cause swelling, inflammation and infection of the joint, and consequent lameness. Refinement of the technique encompasses several elements. The needles used for injection must be the smallest size possible and the volume of the substance given and frequency of dosing should also be kept to a minimum so as not to distend the joint. The animal needs to be kept calm and held very still and the operator has to have a good knowledge of the anatomy of the joint. The procedure should only be done once and to one joint only. If all these Refinement aspects are fully implemented, the animals will suffer far less pain. The guiding principle in this, and in any other aspect of Refinement, is never to assume that current practice is best practice, and to review all the techniques and protocols that are used at regular intervals. One helpful approach in devising possible improvements can be to think about a technique from the animal’s point of view and to ask how a specific procedure would feel if it were applied to oneself. While this suggestion is not intended to encourage uncritical anthropomorphism (see paragraph 4.3) it can be a helpful tool in reviewing current practice in view of species-specific needs.
Refining the management of pain
12.19 Reducing any pain associated with experiments is another important aspect of Refinement. Success depends critically on the ability of those dealing with the animals to recognise and assess pain and suffering (paragraphs 4.18–4.30). Most people involved with animal use are confident of their ability to detect the relevant signs, but some staff are insufficiently trained and lack the relevant expertise.24 Special training is required because many laboratory animals are adapted to conceal signs of pain or distress (see paragraph 4.12). Most people can recognise and respond to overt clinical signs of moderate to severe pain in laboratory animals, but it can be more difficult to recognise indicators of mild to moderate discomfort, pain or distress, which can be very subtle and hard to detect. For example, audible vocalisation is still often cited as a sign that rats are in pain, yet it is now widely known that rats usually vocalise ultrasonically. For truly effective Refinement, these subtle signs of suffering also need to be identified so that staff can become familiar with them. For example, in the case of rats undergoing abdominal surgery, recent research has shown that behaviours such as flank twitching can be used to identify whether rats require more pain relief.25 Until this research was carried out, few if any guidelines on pain assessment for rats mentioned this behaviour, yet it occurs regularly and is highly diagnostic. This approach requires rigorous evaluation of animal behaviour and an open mind.
12.20 It is important to appreciate that Refinement is a continuous process and not a static formula that is only applied at one stage. Ideally, research establishments should have a framework in place for regularly reviewing the way in which experiments are conducted, and comparing current practice with new evidence emerging from research on animal behaviour. This can allow for the development of improved methods of managing pain. A proactive establishment would provide any or all of the following measures as appropriate in its pain management programme:
- pre-emptive pain relief as well as post-operative pain relief;
- multi-modal pain therapy using different pain relieving medicines, which work in different ways and therefore achieve improved control of pain;26
- husbandry and care in the spirit of critical anthropomorphism, which addresses speciesspecific needs (paragraph 4.30);
- staffing (of appropriate expertise) at such levels as will enable the need for intervention (whether treatment or euthanasia) to be anticipated.
Refining endpoints
12.21 The vast majority of animals are killed at the end of the experiment, either because their tissues are required as part of the experiment, or because the scientific objectives have been achieved and the animal can no longer be used. However, under UK law there is provision for limited and tightly controlled re-use, or release of animals to the wild, or a home, where this is appropriate for the individual animal.27 If the experiment leads to an increasing amount of suffering during its course then it is best for the animals to be killed as early as possible. This approach is described as operating ‘humane endpoints’ and requires indicators of likely suffering to be detected at an early stage. For example, if it is known that particular clinical signs such as decreased body temperature lead to a specific outcome such as death, then animals can be killed as soon as these signs appear. Other markers that can be used to define humane endpoints include flank twitching (paragraph 12.19) and chemical and haematological changes in the blood.28
12.22 But in some cases the only way to determine the clinical signs indicating that animals should be euthanised may be to allow some animals to suffer considerably or even die, carefully recording the clinical signs throughout their lives so that a retrospective analysis can be undertaken. This approach has been used successfully to refine some of the protocols required for toxicity testing of vaccines.29 Humane endpoints should generally be easier to define within safety-assessment programmes because routine procedures are used and the only variable is, for example, the batch of vaccine.30 next page
Footnotes23 Morton DB, Jennings M, Buckwell A et al. (2001) Refining procedures for the administration of substances Lab Anim 35: 1–42.
24 See, for example, Flecknell P and Karas A (2004) Assessing and Managing Pain and Distress for Ethics Committees ATLA 32
Supplement 1: 265–6.
25 Roughan JV and Flecknell PA (2001) Behavioural effects of laparotomy and analgesic effects of ketoprofen and carprofen in
rats Pain 90: 65–74.
26 Multimodal pain therapy is the combined administration of opioid and non-opioid pain relief.
27 See A(SP)A Schedule 14 and 15, the options of release of animals to the wild, an abattoir or a home are chosen very rarely.