Written and maintained by CASRAI Editorial Board
Last updated
The AVMA Guidelines for the Euthanasia of Animals: 2020 Edition, published by the American Veterinary Medical Association, is the reference standard that Institutional Animal Care and Use Committees (IACUCs) use to evaluate every euthanasia method proposed in a laboratory animal research protocol. It is not itself a federal regulation, but the Public Health Service Policy and USDA Animal Welfare Act regulations both require institutions to follow “the most recent” AVMA Guidelines unless a deviation is scientifically justified, which makes the document functionally binding on any IACUC-approved protocol.
What the AVMA Guidelines Actually Are
The Guidelines define euthanasia as ending the life of an individual animal in a way that minimizes or eliminates pain and distress, using techniques intended to produce the most rapid, painless, and distress-free death possible. The 2020 Edition is the current version cited across recent veterinary and animal-science literature as the operative standard, superseding the AVMA’s earlier Panel on Euthanasia reports and the 2013 edition. It covers methods and species far beyond the laboratory setting — companion animals, food and fiber production species, wildlife, zoo and aquarium animals, and animals used in research and teaching all fall within its scope — but the laboratory research chapters are what an IACUC protocol reviewer actually applies.
Method entries throughout the document are organized into tiers commonly summarized as acceptable, conditionally acceptable, and not recommended (unacceptable). “Acceptable” methods reliably and consistently produce a humane death when used correctly. “Conditionally acceptable” methods can produce humane death but only under specific circumstances, with particular equipment, training, or species restrictions — used outside those conditions, the same method can cause a poor welfare outcome. “Not recommended” methods are excluded because they cause unacceptable pain, distress, or a documented risk of the animal regaining consciousness before death is confirmed. This tiering is the mechanism that makes the document usable as a review standard rather than a flat list: a protocol reviewer isn’t just checking whether a method appears in the document, but whether it is being proposed under the conditions the Guidelines attach to it.
How the IACUC Uses the Guidelines as the Reference Standard
Every vertebrate animal protocol that ends in euthanasia has to specify the method, and the IACUC protocol review process checks that method against the current Guidelines before approval. In practice this means the IACUC, often in consultation with the institution’s Attending Veterinarian, confirms three things: the proposed method sits in the acceptable or conditionally acceptable tier for the species and life stage involved, any conditions attached to a conditionally acceptable method (equipment calibration, operator training, agent concentration) are actually met, and a secondary method of ensuring death is specified where the primary method doesn’t itself guarantee it. Personnel performing euthanasia have to be trained and competent in the specific technique they’ll use — a protocol that names a method but not who is qualified to perform it, or how that competency was verified, is incomplete. This is the same personnel-qualification expectation covered in more detail for training requirements generally.
NIH grant applications carry a parallel requirement in the Vertebrate Animals Section, where the method of euthanasia and its consistency with AVMA recommendations has to be described explicitly. A protocol that is silent on method, or that names a method without tying it to the Guidelines’ tier and conditions, is a routine reason IACUCs send protocols back for revision.
Method Choice Depends on Species and Study Endpoint
No single method is acceptable across every species and every experimental context, which is why the Guidelines are organized by species group rather than as one universal method list. A method’s tier can also shift depending on the animal’s age and physiological state: neonatal and fetal animals of several species tolerate hypoxia far better than adults, which is why methods relying on inducing hypoxia are evaluated separately, and often more restrictively, for very young animals of a given species than for adults of the same species.
Study endpoint matters just as much as species. If the research question requires tissue, blood, or organs that a given method would compromise — through agent residue, tissue damage at the injection or exposure site, or physiological changes in the minutes before death — that is a legitimate scientific reason to select a different acceptable or conditionally acceptable method, and the protocol should say so explicitly rather than leaving the reviewer to infer it. This is also where refinement work under the 3Rs concentrates: choosing the least distressing method that is still compatible with the study’s scientific requirements, and pairing it with pre-euthanasia sedation or analgesia where the Guidelines and the species’ biology support it.
Carbon Dioxide as a Case Study
Carbon dioxide inhalation is a useful illustration of why method choice is conditional rather than absolute. CO2 is treated as acceptable or conditionally acceptable for many adult laboratory rodent species when delivered under the flow-rate, chamber, and pre-charging conditions the Guidelines specify for that method, but it is evaluated more restrictively — and can be inappropriate — for neonates and for species or strains with documented aversion responses to CO2 exposure. A protocol proposing CO2 for a species or age group outside the conditions the Guidelines describe needs to justify that choice on scientific grounds, the same as any other deviation.
Secondary Physical Methods: Confirming Death
A number of methods the Guidelines classify as acceptable or conditionally acceptable are not, by themselves, a reliable guarantee that death has occurred by the time the animal is removed from care. Where that’s the case, the protocol has to specify a secondary physical method — commonly cervical dislocation, decapitation, bilateral thoracotomy, or exsanguination — performed once the animal is unconscious, specifically to eliminate any possibility of recovery. Cervical dislocation in mice is the method most CASRAI readers will recognize as a secondary confirmation step, and it carries its own operator-competence requirements the IACUC has to verify independently of whatever primary method preceded it.
Justifying a Deviation From the Guidelines
The Guidelines explicitly anticipate that a researcher may need a method they don’t list, or need to use a listed method outside its usual conditions, because the scientific question genuinely requires it. That deviation is permitted, but only with documented scientific justification in the protocol — the reviewer needs to see why no acceptable or conditionally acceptable method would answer the research question, not just that the investigator prefers an alternative. The Attending Veterinarian’s input is typically part of that justification, since the deviation is being weighed against the same welfare standard the Guidelines exist to protect. An undocumented deviation, or one justified only by convenience or cost, is not something an IACUC can approve consistent with PHS Policy or Animal Welfare Act regulations, both of which point back to the current AVMA Guidelines as the baseline a deviation departs from.
The 3Rs and ARRIVE 2.0
Euthanasia method selection sits squarely in Refinement, the third of the 3Rs: given that the endpoint of the study requires the animal’s death, refinement is about minimizing pain and distress in how that death is carried out, and about setting humane endpoints so animals are euthanized before suffering becomes severe rather than only once a study’s planned endpoint is reached. It intersects with Reduction where a less invasive euthanasia or sampling approach lets tissue be collected without adding animals to a study, and with Replacement only at the margins, where an in vitro or ex vivo alternative removes the need for terminal procedures altogether — a distinct question from method selection covered elsewhere on this site under replacement methods.
The ARRIVE 2.0 guidelines bear on euthanasia in two practical ways for a protocol and its eventual manuscript. First, ARRIVE 2.0 expects the method of euthanasia to be reported explicitly, including any deviation from a standard method and the reason for it — the same justification the IACUC required at protocol approval should reappear in the published methods section. Second, where euthanasia timing is itself a study variable (humane endpoint triggers, survival-surgery endpoints), applying the same pre-specified criteria and method consistently across experimental groups, rather than at the discretion of whoever is present, is part of avoiding the kind of unblinded, non-randomized assessment ARRIVE 2.0 flags as a source of bias in outcome measurement.
What This Means for Protocol Writers
A euthanasia section in an IACUC protocol should specify: the exact method and, if applicable, agent and dose; the species, strain, and age range it applies to; the secondary physical method used to confirm death, if the primary method requires one; the personnel qualified to perform it and how that competency is documented; and, where the method or its conditions fall outside the Guidelines’ acceptable or conditionally acceptable tier for that species and age, the scientific justification for the deviation. Reviewers checking against the Guide for the Care and Use of Laboratory Animals and the current AVMA Guidelines together, rather than either document alone, is standard IACUC practice, since the Guide incorporates the AVMA Guidelines by reference for euthanasia specifically.
Frequently Asked Questions
Is the AVMA Guidelines document itself a federal regulation?
No. It’s a professional veterinary standard, but PHS Policy and USDA Animal Welfare Act regulations both require institutions to follow the current edition unless a deviation is scientifically justified in the protocol, which is what gives it regulatory weight in practice.
What edition of the AVMA Guidelines is current?
The 2020 Edition is the current version, cited as the operative standard across recent veterinary and animal-science literature and superseding the 2013 edition.
Can an IACUC approve a euthanasia method not listed in the Guidelines?
Yes, but only with documented scientific justification showing why no acceptable or conditionally acceptable method would meet the study’s requirements — convenience or cost is not sufficient justification.
Why do some acceptable methods still require a secondary physical method?
Because the primary method, used correctly, produces unconsciousness and death reliably in most animals but doesn’t itself provide certainty of death for every individual. A secondary physical method performed after unconsciousness removes any possibility of recovery.
Does the same euthanasia method apply across all life stages of a species?
Not necessarily. Some methods evaluated as acceptable for adults of a species are evaluated more restrictively, or are not recommended, for neonatal or fetal animals of the same species because of physiological differences such as hypoxia tolerance — the protocol needs to specify the age range a given method applies to.
Which of the 3Rs does euthanasia method selection primarily serve?
Refinement — minimizing pain and distress in how a required endpoint is carried out and setting humane endpoints that trigger euthanasia before suffering becomes severe.
