3R Research Foundation Switzerland - Front page

3R-Project 71-00

Detection and characterisation of signs of pain in mice by the combined use of a telemetry model and behavioural observations

Margrit Arras
Vetsuisse Faculty, Institute of Laboratory Animal Sciences, University of Zürich, 8091 Zürich, Switzerland.

Keywords: mice; analgesia; ethology: laboratory animals; telemetry; refinement

Duration: 3 years Project Completion: 2005

Background and Aim
Pain in animals is commonly assessed by observing behavioural aberrations. The diagnosis of pain can be confirmed by measuring physiological parameters (heart rate, respiration rate, bodily tension). An evaluation of pain is feasible in animals such as monkeys, dogs, cats and pigs, because these species show clear behavioural signs of pain.
In contrast, mice live in constant fear of falling prey to their enemies thus, the intention is probably to show as few signs of disease, suffering or weakness as possible. Accordingly, during animal experiments, or even when a person is simply present in the room, the mouse will hide almost all signs of slight or intermediate pain. Additionally, intrinsic analgetic sys-tems that are well developed in mice for protecting them from intolerable fear and pain may influence their responses and make their pain less recognisable.
In consequence, there are no reliable indicators to detect low and intermediate pain in the mouse. Only when an animal is near death can we see that it is suffering pain. As a result, demands for adequate pain therapy are often ignored, or met with the anthropomorphic atti-tude "as long as the mouse shows no pain, it must feel no pain". Another consequence of the dilemma is that the success of a pain therapy is not visible and therefore doubts about the efficacy of many analgesic regimes came up - which were again counteracted by those, who claim, that every suspicious manipulation (as little as it might be) must be treated by a standardized pain therapy in mice even if this never would be done in other species or in man.

The aim of the project is to identify indicators for low or intermediate pain in mice using a combined analysis of physiological and behavioural parameters. This goal is becoming in-creasingly urgent, since, due to the increasing popularity with the advent of genetically modi-fied mice lines, this species is being used in increasingly complex biomedical models and procedures (e.g. the induction of stroke or ischemic heart disease, organ transplants, et al.).

Method and Results
Measurements of physiology and behaviour must be free of artefacts caused by external in-fluences such as the presence of the investigator. Therefore, the physiological data were collected by a telemetric system. For this, mice were implanted with transmitters able to measure ECG, body temperature and activity on a permanent basis. The transmitters are small enough not to be noticeable by the animal. This system allows data to be collected in the conscious mouse without the presence of the investigator in the room. Monitoring the animal over a number of days documents the time course of body core temperature, locomo-tion and heart rate/ECG. From this, the circadian rhythm of these parameters are drawn and heart rate variability and ECG wave forms can be analysed.

Aberrant circadian rhythms are indicative for disturbances in the animal’s welfare, whereas the elevation of heart rate and certain changes of heart rate variability and ECG wave forms are signs of pain, stress or an activation of the sympathetic nervous system. Data are com-pared with the daily food and water consumption and the body weight progression because these parameters are widely used clinical investigations for a rough assessment of the physi-cal condition as well as the aspects of the mouse`s welfare in biomedical experiments.

After instrumentation and recovery, mice underwent a defined surgical intervention to induce a stage of postoperative pain that is suggested to be of low to intermediate graduation. Pain was either untreated or treated with two different widely used analgetic regimes.
A few hours after the operation all animals, whether treated with analgetics or not, exhibited not overt signs of pain: their fur and wounds were cleaned and the animals ate, drank and showed the spectrum of their normal behaviour in their home cage, such as walking around, climbing underneath the cage lid, digging or nest building activities.
The telemetrically measured locomotion activity counts are nearly unchanged indicating that the motorical behaviour is not depressed or even increased (e.g. restlessness), which con-firms the assumption of slight pain and not that animals suffered from severe or intolerable pain. Body core temperature showed only a marginal increase, suggesting, that post-surgical inflammation was of no influence.
Only untreated animals showed significant post-operative changes of the heart's actions (heart rate, heart rate variability, ECG wave forms) which suggests pain and sympathetic activation.
From the clinical parameters, the food intake was significantly diminished after surgery in untreated animals; body weight was only slightly reduced, also showing, that animals did not suffer from serious or long-lasting pain.
The analgetic regimes both were able to prevent all changes in cardial action. Post-operative food intake and body weight also remained constant when animals were under post-operative pain therapy.
Control experiments confirmed, that the anaesthetic and analgetic substances had no effect on the measured parameters per se.

The results show, that
i.) the model was able to identify and objectify signs of low to inter-mediate pain in mice, which could not be clearly detected otherwise in this species
ii.) analgetic regimes acted successfully in relief of low to intermediate post-surgical pain in mice.

Behavioural observations, data and score sheets are under investigation.

In the future, the model will be used for the assessment of pain in mice in frequently used, standardized biomedical experiments (e.g. antibody production), particularly for methods, in which the degree of pain and the benefit of pain therapy is a matter of diversity.
Simultaneously, the behavioural studies will go on with the aim to develop a score sheet for the detection of pain under routine laboratory conditions.

Conclusions and Relevance for 3R
The model will be a valuable tool for
i) objective classification of the degree of pain,
ii) pre-estimating the necessity of pain therapy in commonly used biomedical methods and
iii) the proof of the efficacy of various analgetic regimes used on mice.
The results will have a great impact on the refinement of animal experiments.

(see also 3R-INFO-BULLETIN Nr. 31)
Published updated Version 31/2007 (pdf)

Arras, M., Rettich, A., Cinelli, P., Kasermann, H. P., Buerki, K., (2007) Assessment of post-labarotomy pain in laboratory mice by telemetric recording of heart rate and heart rate variability.BMC Veterinary Research 2007, 3:16
see open access article