Forum on the Use of Analgesia in Animal Research
Forum on the use of analgesia in animal research
Date: Tuesday 27thAugust 2013, 10h00-12h00
Venue: Groote Schuur OMB, Venue 12
Speakers: Dr Bert Mohr (FHS Veterinarian; Director, UCT Research Animal Facility)
The potential detrimental effects of pain on research outcomes; and an overview of available options for analgesia
A/Professor Jo-Ann Passmore (Head of Mucosal Immunology Laboratory, Division of Medical Virology, IIDMM)
Anti-inflammatories in immunological research
Professor David Benatar (Professor and HOD of Philosophy; Director, UCT Bioethics Centre)
Can the withholding of analgesia from (animal) research subjects be morally justified?
Chair: Associate Professor Anne Pope
Aide memoire
Attendees were welcomed and the topic contextualised. As the notice of the event indicated
It is currently UCT policy that ‘experiments accompanied by pain [are] allowed only if they are shown to be for the purpose of producing results that could benefit humans, and where the benefits are clearly commensurate with the amount of pain caused. In such cases, analgesia must be used to alleviate pain and distress. In special circumstances, the requirement for analgesics may be relaxed.’ (See C (c) of the FHS Research Animal Facility Use of Animals in Research and Teaching SOP June 2013)
The SAEC wishes to examine the aspect of the policy that permits relaxation of the requirement for analgesics more carefully, and seeks to gauge the views of the wider university community, especially researchers who use animals, on this topic. You are cordially invited to attend and provide input on this issue.
The objective of the meeting was to gather information and to hear views from three colleagues in the first instance and then to invite comments and queries from the floor.
Presentations
Dr Mohr gave a presentation on pain in animals that explained that non-human animals are sentient and do experience pain, albeit that they do not necessarily show it in ways that humans can easily recognize; that the UCT Policy emphasises that pain, amongst other aspects, must be avoided if possible or minimised; that controlled laboratory experiments try to limit confounding factors that may interfere with results; and that it is increasingly recognized that pain itself can be a significant confounding factor. This point is relevant to arguments that support withholding analgesia on the basis that some drugs may be confounding factors. Unalleviated pain can be a potent physiological stressor that can disturb homeostasis, cause dysregulation of the immune system, lead to the release of stress hormones, mental depression, neuronal remodeling, weight loss and a catabolic state. As an uncontrolled study variable, pain itself can decrease the stability of animal models. If a subset of animals in a study experience more severe pain than others, conclusions may be flawed.
It became clear that insufficient supporting information is available to make withholding of analgesia justifiable in general terms. However, it is also clear that the roles of pain and analgesia seem to be very specific which makes investigation complicated. Evidence-based science is needed. It was pointed out that different types of analgesic drugs have very different physiologic effects; that only a subset of analgesics has the mechanism of action that interrupts inflammatory pathways; and that some types of analgesics seem not to significantly affect immune responses. Several examples exist of pain-killers, e.g. buprenorphine, being successfully used in immune and infectious studies.
However, in the absence of good guiding information, policy choices have to be made, and UCT’s policy choice is an example. Other institutions, e.g. in northern hemisphere, may choose to make analgesia mandatory. A changing moral view is evident: more people seem to think that use of animals in research is not acceptable when unalleviated pain is inflicted. This changing view will affect the ethics policies and the scientific framework. But this happens fairly slowly. In the interim, responsible informed deliberation is required. This deliberation should be inclusive so that clinicians with human immunological-suppression experience and pharmacologists are part of the discussions.
Associate Professor Passmore spoke about her expertise in human genital tract inflammation and HIV susceptibility, pointing out that inflammation is necessary for the susceptibility to HIV to exist. In this context, she emphasised that the translation of this research focus was the use of anti-inflammatories or analgesics in women to reduce their inflammation and hence susceptibility to HIV infection. Regarding analgesic use in animal experimentation that involve infectious diseases, she suggested that there is unlikely to be single solution – at a scientific technical level, as many of the classes of analgesics target core pathways in the immune inflammatory pathway, the mechanisms of analgesics to control inflammation and pain may fundamentally affect the primary endpoint of the study, i.e. inflammation is an integral part of the immune response to infectious diseases (including TB, malaria, HIV, listeria, and others) and to minimize this discomfort and pain necessarily diminishes the inflammation. There are many studies and meta-analyses that look at the mechanisms by which analgesics dampen immune responses and inflammation although there are no clear trends in terms of impact on immune responses. Some have shown no impact on antibody responses, while others have shown broad impact on NK cells and macrophages and multiple arms of the immune response. Because of the (1) complexity and redundencies intrinsic to the host immune response to different infectious diseases, (2) the fact that different classes of analgescis act on different pathways within this complex network differentially, (3) that not all infectious diseases rely on the same immune response, and (4) different animals have slightly different responses to infectious diseases; makes a one-size-fits-all recommendation to use of analgesics in infectious disease models in animals. All of this makes it very difficult to extrapolate general principles. There are very few systematic reviews of studies. These might be useful for purposes of better understanding the issues.
Professor Benatar posed the question whether withholding of analgesia in (animal) research can be justified morally. His discussion focussed on trying to answer the question. Two assumptions informed the discussion: that the pain under consideration is non-negligible; and that inflicting that pain would not be permissible if the research participant were a human.
He said that if one thinks that the answer to the question depends on the species of the research subject, one needs to show that the species difference is relevant (in the way in which race, sex, religion, ethnicity, etc is not). Otherwise one would be guilty of specieism, an analogue of racism and sexism in which one treats one being worse than another merely on the grounds of its species.
He then considered and rejected various arguments for why the species difference is relevant:
First, he considered the suggestion that humans’ higher cognitive capacity might make them suffer more. Two problems confront this view: (a) not all humans have greater cognitive capacity than animals – and it would surely not be acceptable to withhold analgesia from such humans; and (b) reduced cognitive capacity can sometimes make suffering worse because the being does not understand what is happening.
Second, he considered the suggestion that humans with lower cognitive capacities should enjoy greater protection than animals because they (the humans) belong to a species with greater cognitive capacity. The problem with this argument is that our treatment of beings should be based on their individual attributes rather than on the attributes of the group to which they belong.
Third, he considered the argument that we should treat cognitive impaired humans better than animals because they are our fellow humans. However, this argument merely begs the question by assuming that conspecificity is relevant. (The racist, for example, assumes that he should treat fellow members of his own race better.)
He then considered whether withholding analgesia could be justified on the grounds of “necessity” – that is, the research could not be done if analgesia were provided because of the confounding effect of analgesia (assuming that this claim is accurate). He considered a number of responses to this, but the most basic one is that if the necessity argument worked it would also justify doing the painful research on humans. Indeed, it would more readily justify such research because humans would be a much better model of human disease than would animals. This ‘thought experiment’ is a useful tool for seeing the issues more clearly.
Discussion
A question was asked about whether animals might be able to suppress pain rather than just not show it in a recognizable way. Dr Mohr responded that in clinical veterinary practice, if no pain is confirmed but after a dose of analgesia a positive behavioural/physiologic change is seen, then this is regarded as indicative that pain was indeed present.
A point was made about animals not being able to consent to research participation and how this is unacceptable; that blue sky research should not be permitted with animals if research is being done just to further career development. It was pointed out that career development is a legitimate human goal; and that not all humans who are research participants are able to consent. Consequently, the tensions must be acknowledged in principled manner to guide deliberations about these matters.
Someone commented that, since the translatability rate of blue sky research into products that benefit humans is less than 0.1%, the argument favouring use of analgesia is strong. It is difficult to understand a justification to cause pain to animals in basic research. Furthermore, since the deliberative process for Animal Ethics Committees (AECs) purportedly relies on utilitarian principles in respect of cost/benefit analysis, it seems reasonable that the experience of pain by humans and other animals should be treated similarly.
If all harm was impermissible, then certain types of research would not be possible. At present, society is not expressing this view through the prevailing policies. Dr Mohr pointed out that the AEC performs a harms/benefit analysis in terms of which the potential life-time harm as well as the degree of such potential harm to the animals is weighed against the potential benefit (to humans/animals/the environment) anticipated from the research. However, he conceded that this exercise is inherently difficult because e.g. the potential benefit cannot really be known. If it were, the research would be unnecessary. Furthermore, weighing harms against benefits is very inexact – they are not identical or directly comparable.
Discussed revealed that, if a radical approach were to be taken, questions to ask include whether painful blue sky research should be permitted; whether potential for translatability should be a guiding factor in approving studies without analgesia; and whether animal models are actually useful and predictive for use in humans.
A/Prof Passmore commented that although the number of lead ideas that go to clinical trial stage is small, it is important to focus on successful processes and to appreciate the importance of the animal studies that were part of that process. She pointed out that the view that all research should be done with human participants does not take account of the fact that data from human studies is largely confounded by the fact that the environment in which humans function and their genetic makeup (which impacts heavily on immunity and inflammation) is difficult to control, as is the confounding substances that humans consume. This leads to the practice of going to reductionist models in animals where potential confounders can be better managed. As a result, addition of analgesics without a solid scientific knowledge of their impact on all the different parts of the immune response would complicate and possibly confound the model. Dr Mohr conceded that the complexity is great and that confusion is unpleasant but pointed out that simplification is possible. A good understanding of pain itself is required because it also causes ‘noise’ in research results. Some experiences of pain will be significant and others perhaps not so. The degree of interference is thus important. Similarly, the degree of interference of a given analgesic could vary from minimal to major. The animal ethics committee spends considerable time weighing things as best possible. However, it is critical that researchers provide sufficient information in order for these deliberations to be optimal. Recommendations that a study should begin slowly i.e. in a pilot format are not always well received. This is because the view is that pilot studies are not funded or do not introduce new approaches. Consequently, costs become an issue to negotiate. A researcher pointed out that the post-intervention analgesic administration of animals by researchers used to be three days but is now five days for some models, which has cost implications for the study.
It is evident that many of these logistical issues require discussion and planning. Dr Mohr noted that refinement (one of the three Rs of animal ethics policies) is a tool for improving design of studies. However, this too has cost implications e.g. increased monitoring/analgesia requriements. A/Prof Passmore recommended that the study of the impact of analgesics on immune responses in infectious disease models is a field in its own right. Rather than recommend that each animal research include extra controls in their studies to addresss the impact of the drug class on their models, it would be more scientifically sound to take this on within the animal unit or set up a national forum of animal ethics committees to address this. Much of the published literature cited by A/Prof Passmore drew from research in anaesthetics and linkage or recommendations with such research groups would be important to consider. Dr Mohr advised that obtaining funding for such studies would be important; in some countries there are national centres that fund such research.
Professor Benatar, in response to a comment, distinguished between different senses of the word ‘ethical’. It might mean “morally right” or it might mean “compliant with current research codes”. He noted that research might be ethical in the second sense without being ethical in the first sense.
In response to another comment he reminded all that weighing up harms and benefits in animal experimentation is not as hard as people think. The process is assisted by asking how the calculation would be made if humans were the prospective research subjects. This provides a frame of reference that reveals the issues clearly. The resort to reductionism is understandable. It provides methodological strength but also has inherent weakness because of the low incidence of translatability to humans. It is very clear that it is very complex to do research with animals. Pragmatism is also part of the mix: the obligation to do research is clear; the guidelines expect particular standards to be maintained, while honouring the ethical boundaries of the current paradigm. However, equal interests ought to be treated equally, which means that the utilitarian model used to work through practical problems must be applied honestly.