University of Virginia Guidelines and Policy for Determination of Humane Endpoints for Animals Used in Research


The obligation to minimize or eliminate unnecessary pain and distress that animals may experience when used in research, research training, and biological testing activities is clearly stated in the USDA Animal Welfare Act and the PHS Policy on Humane Care and Use of Laboratory Animals 1. To that end, the Institutional Animal Care and Use Committee (IACUC) at the University of Virginia reviews all proposed and current animal use to ensure that we, as an institution, meet both our legal and ethical obligations with regard to animal research.

Animal models used for the study of infectious diseases pose particular problems as these animals may experience significant pain and/or distress as a consequence of the disease process being studied. The IACUC recognizes the difficulty in choosing appropriate humane endpoints for animals in these types of studies. Seeking earlier, more humane endpoints, prior to impending death, can result in significant reductions in the potential pain and distress experienced by these animals. However, if these endpoints do not accurately predict the outcome of the experiment, they may vastly alter the research results obtained 2. For example, animals becoming sick in an infection challenge/vaccine efficacy trial might be euthanized for humane reasons when in fact they could have survived the challenge, proving that the new vaccine was an effective treatment.

The following policy has been generated to provide guidelines to investigators performing these types of experiments. The specific intent is to minimize animal pain and distress while allowing the researcher to collect data that is meaningful and reliable. The decision to euthanize an animal must be made with appropriate clinical judgment, taking into account the severity of the condition, the amount of pain or distress, and the potential loss of valuable data. To accomplish this, the IACUC has developed suggestions for clinically scoring these animals and using humane endpoints to reliably predict morbidity. It is not possible to have a list of all clinical signs that will occur in all experiments. Moreover, clinical signs can vary between species, strain, and the method of recording. The use of pilot studies for individual cases is thus required 3.


A. Definitions

The stages leading to death can be characterized as:

Predictable Death:
presence of clinical signs indicative of death before the planned end of the experiment, for example, inability to reach water or food.
Impending Death:
when moribund state or death is expected prior to the next planned time of observation. Signs indicative of this state in rodents could include convulsions, recumbency, difficulty breathing, unwillingness to move around cage, tremors.
being in state of dying or inability to survive, even if treated.

B. Endpoints

There are a number of studies in the literature that have evaluated various endpoints for their ability to reliably predict morbidity. These should serve as useful guidelines, but each study will entail developing its own set of clinical parameters to be followed and scored to determine which physiologic or behavioral changes best predict morbidity for that particular study. The decision to euthanize an animal must be made with appropriate clinical judgment, taking into account the severity of the condition, the amount of pain or distress, the prognosis, and the potential loss of valuable data.

  1. Body Temperature Change:
    Hypothermia has been shown to be an important indicator of a deteriorating condition in several infectious disease and toxic states. Septic animals, especially rodents, lose their ability to maintain body temperature. A decrease of 4-6 °C has correlated with death as an outcome in several infectious disease models 4-7. Therefore, monitoring of body temperature should be an important part of clinical monitoring for any infectious disease or toxic shock model. This can be accomplished using infrared scanners 8, or implanted thermomistor microchips 4.
  2. Weight Loss:
    One effect of cytokines on animals is the loss of appetite. In some models, weight loss may be a cardinal indicator of the severity of infection. The total amount of weight lost, as well as its duration and consistency, should be used to determine the endpoint for infectious disease animal models. Animals that are recumbent and have lost the ability to reach food and water will lose weight rapidly. Weight loss should be compared to age-matched controls. A difference (loss of weight) of more than 20% from control animals is a commonly used criterion for euthanasia.
  3. Other Behavioral and Physiological Changes:
    Decreased activity (lethargy) and alertness, a rough hair coat, and hunched posture are other direct signs of illness, pain, or distress. Detailed observational checklists (see Appendices 1 and 2 for examples) should be devised for each individual study in consultation with the veterinarian. Clinical scoring sheets should be used to help assess the animal’s condition and determine appropriate and humane endpoints.
  4. Acute Phase Proteins:
    Increases in serum levels of cytokines and acute phase proteins (APPs) occur early in response to infection, before severe behavioral and physiologic changes are seen. Recent studies have evaluated the ability to use APPs as predictors of severity and outcome in infectious disease animal models 9,10. APPs are normally at very low or undetectable levels in the plasma, and rise rapidly after cytokine up-regulation. The serum profile of APPs seems to be species-specific, e.g., the major APP in rats is α2-macroglobulin, whereas in mice it is serum amyloid A. More research is needed before APP measurements become a practical method for predicting disease outcome. The specific level of APP beyond which the animal cannot recover should be determined, and standardized rapid tests for APPs using small blood volumes is also necessary.

Basic considerations that should be followed in all animal experiments:

  • The earliest possible endpoints that are indicators of severe pain or impending death should be used as indications for euthanizing an animal. These endpoints should be determined during the pilot study and should serve two purposes: 1) to determine humane endpoints and 2) to provide reasonable assurance to the investigator that early euthanasia is not altering final outcome (i.e. are animals being euthanized that would have ultimately survived without intervention?).
  • Studies should be terminated prior to their planned termination time if the objectives of the study have been satisfied, or it is obvious that they will not be achieved.
  • Studies should build on existing knowledge about the substance to be tested. This enables better prediction of the likely signs and timing of adverse effects, and allows those conducting the study to incorporate these endpoints into the experimental protocol.
  • The successful application of humane endpoints is dependent on the involvement of all members of the study team. Personnel must be adequately trained and aware of their individual roles and responsibilities.


Any investigator proposing a study involving the induction of a diseased state where pain and/or distress is anticipated must do the following:

  1. Be familiar with the species-specific clinical signs that may be indicative of impending morbidity. The “Policy on Recognition and Assessment of Pain, Stress, and Distress in Laboratory Animals” contains information on species-specific signs of pain and suggested analgesics.
  2. Consult with one of the laboratory animal veterinarians to devise a clinical scoring sheet that will be used at every animal observation time point. Parameters to be included should be based on prior knowledge of the condition of these animals after induction of disease. The use of pilot studies is required. Submission of a pilot study plan should be submitted as part of an animal use protocol to the IACUC for review and approval. If an approved protocol already exists, the pilot study may be added as a minor modification. This pilot study should be used to develop and test a scoring system that establishes endpoints for your animal experiments that are both effective and humane. Results of the pilot study should be reported back to the IACUC, with a proposal for the full study.
  3. Animals should be monitored at least 3 times daily during the window of time in which illness and mortality are anticipated. This window should be predetermined as in #2 above. If the time course of the disease is not known and the researcher does not wish to perform a pilot study, then he/she is obligated to ensure that the animals are observed and clinically scored at least once every 8 hours for the length of the study.
  4. Humane endpoints must be established and clearly defined in the animal use protocol. These endpoints should be based on the clinical scoring system. A total score greater than a predetermined amount will be a criterion for euthanasia.
  5. Any study to be performed that includes animals for which unrelieved pain or distress is anticipated and scientifically unavoidable will be categorized as USDA pain level ‘E’. Unavoidable pain or distress must be scientifically justified within the animal use protocol.


List of clinical signs and conditions where euthanasia may be appropriate:

  1. Any condition resulting in a prolonged or irreversible inability to eat or drink, e.g., prolonged immobility, obstruction of the oral cavity, missing or abnormal teeth. Rodents should not go more than 24 hours without access to food or water, unless scientifically justified.
  2. Diseases or conditions indicating severe pain, distress, or suffering, e.g., fractures, self-induced trauma, abnormal posture or movements, open wounds or ulcers, or abnormal vocalization that is not relieved by anti-anxiety or analgesic agents.
  3. Rapid or continuous weight loss, e.g., 20% or greater body weight over a few days, or gradual but continued weight loss.
  4. Generalized decrease in grooming and abnormal appearance over an extended time period, e.g., rough hair coat, extensive alopecia, prolonged diarrhea, urine-stained hair coat, rectal prolapse, swollen limbs, paralysis, or other neurological disturbances (convulsions, abnormal head carriage, circling behavior, prostration).
  5. Severe or continuing respiratory distress, e.g., coughing, sneezing, bloody nasal discharge, increased respiratory rate, labored breathing.
  6. Frank bleeding that is uncontrollable.
  7. Evidence of microbial infections or other diseases that interfere with the experimental protocol or cause any of the above and based on veterinary judgment are not treatable.


Example of response variables and scoring system in rodents:

The total score should be tallied with every observation time point and a score equal to or above a certain predetermined amount would warrant euthanasia or evaluation by a laboratory animal veterinarian. If all of the criteria listed below were included in a clinical scoring sheet, a total combined score above 8 would warrant euthanasia. Any mice with 2 or more individual maximum scores (e.g., a 3 for activity and a 3 for weight) should also be euthanized. Notes for technicians should be included at the bottom of all score sheets, providing guidance on how to record qualitative clinical signs, criteria for humane endpoints, and instructions for euthanasia and carcass disposal, as well as emergency phone numbers for laboratory personnel.

Body weight changes

0- Normal
1- <10% weight loss
2- 10-15% weight loss
3- >20% weight loss

Physical condition

Haircoat Eyes and Nose
0- Normal, well-groomed 0- Normal
1- Rough haircoat 1- Eyes closed or squinted – no discharge
2- Rough coat, hair loss, ungroomed 2- Eyes closed or squinted – discharge or porphryin staining


Activity Posture
0- Normal 0- Normal
1- Decreased activity, locomotion after slight stimulation 1- Sitting in hunched up position
2- Inactive, less alert, locomotion after moderate stimulation 2- Hunched posture/ head on cage floor
3- Self-mutilation, either very restless or immobile, or no locomotion after moderate stimulation 3- Lying prone on cage floor


    1. Stokes WS. Reducing Unrelieved Pain and Distress in Laboratory Animals Using Humane Endpoints. ILAR. 2000;41:59-61.
    2. Olfert ED, Godson DL. Humane Endpoints for Infectious Disease Animal Models. ILAR. 2000;41:99-104.
    3. CCAC [Canadian Council on Animal Care]. CCAC guidelines on choosing an appropriate endpoint in experiments using animals for research, testing, and teaching. Ottawa, Canada; 1998.
    4. Kort WJ, Hekking-Weijma JM, TenKate MT, Sorm V, R. V. A microchip implant system as a method to determine body temperature of terminally ill rats and mice. Lab Animal. 1998;32:260-269.
    5. Siems JJ, Allen SD. Early euthanasia as an alternative to death in chronic infectious disease studies using a systemic Candida albicans model. In: Abstr. 89th Annual Meeting of the American Society for Microbiology; 1989.
    6. Soothill JS, Morton DB, Ahmad A. The HID50 (hypothermia-inducing dose 50): An alternative to the LD50 for the measurement of bacterial virulence. Int J Exp pathol. 1992;75:95-98.
    7. Wong JP, Saravolac EG, Clement JG, Nagata LP. Development of a murine hypothermia model for study of respiratory tract influenza virus infection. Lab Animal Science. 1997;47:143-147.
    8. Love JA, Booth CD, Boyd J. Remote temperature sensing for determining humane endpoints. In: 2nd World Congress on Alternatives and Animal Use in the Life Sciences. Utrecht, The Netherlands; 1996.
    9. Conner JG, Eckersall PD, Wiseman A, Bain RK, Douglas TA. Acute phase response in calves following infection with Pasteurella hemolytica, Ostertagia ostertagi, and endotoxin administration. Res Vet Sci. 1989;47:203-207.
    10. Eckersall PD. Acute phase proteins as markers of inflammatory lesions. Comp Hematol Int. 1995;5:93-97.