Sage Journals: Discover world-class research

Abstract

German

Spanish

French

Appropriate end-points are integral to the refinement of laboratory animal experiments. Our recent experience has highlighted that ambiguity around end-points is hampering their adoption in experiments that cause severe suffering to fish. In toxicology, the term endpoint (single word) refers to the response variable to the treatment that is measured and analysed. This differs to usage within laboratory animal experimentation, where end-point (hyphenated) refers to the time-point when exposure of the animal(s) to the treatment (and suffering) ends. Within laboratory animal experimentation, standardised terminology is needed for different types of early end-point which are based on the condition of the animal(s) or progress of the experiment. We propose that those involved in regulating and conducting animal experiments consider seven distinct types of early end-point (aim, technical error, biological error, mortality, moribundity, prognostic humane, non-prognostic humane) in addition to the planned experimental end-point (i.e. maximum duration). Moribundity (not morbidity) refers to an animal in a severely debilitated state close to death. Moribundity in fish is not yet defined, so we propose identification via a lack of response to external stimuli, loss of equilibrium (i.e. loss of righting reflex), and a slow opercular ventilation rate. As these clinical signs equate to those of deep/surgical anaesthesia, this moribundity end-point cannot be considered a humane end-point as the fish is likely to be unconscious and have passed the point of maximum suffering. We believe that identification of earlier humane end-points based on clinical signs and wider recognition of other types of early end-point can reduce suffering in experiments.

Keywords

Ethics review toxicity regulatory test refinement humane end-points

Introduction

Users of animals in laboratory experiments are legally and/or ethically obliged to minimise suffering in experimental animals via consideration and application of the 3Rs: replacement, reduction, refinement.^1–3 Prescribed regulatory tests for fish, such as those assessing the acute toxicity of chemicals⁴ and the efficacy of vaccines,⁵ are still based around a mortality end-point and are amongst the most severe experiments that use fish.⁶ As opportunities for replacement and reduction in such prescribed tests are limited or have already been addressed,^7–9 the current focus is on refinement.

A key refinement required by European animal experimentation legislation is that ‘Death as the end-point of a procedure shall be avoided as far as possible and replaced by early and humane end-points’.² The need to identify and apply such end-points to reduce suffering has been highlighted.¹⁰ However, terminology for different types of early end-point is variable^11,12 and potentially confusing. Although the recently published PREPARE guidelines¹³ highlight the need for consideration of humane end-points during study planning, potential other early end-points are overlooked.

Here we aim to clarify different types of early end-point for animal experiments. We believe that such elucidation is needed, based upon our recent experience: involvement in fish disease studies at the Cefas Weymouth Laboratory;¹⁴ participation in the latest review of the Fish, Acute Toxicity Test (OECD Test Guideline 203, TG203);^4,8 and attendance at three recent meetings (Severity and Humane Endpoints in Fish Research, Bergen, 4 October 2019; Third RSPCA international meeting, Focus on severe suffering: Avoiding mortality, Athens, 10–11 October 2019; Establishing score sheets and defining endpoints in fish experiments, Bergen, 28 January 2020). The benefits of clarifying end-points are likely to extend beyond fish to other experimental animals.

What is an end-point?

The hyphenated term end-point is used synonymously with the single word (endpoint) and un-hyphenated term (end point).^{2,3,10,13,15–19} The term is being used with two distinct meanings:

Within laboratory animal experimentation, end-point refers to the point at which exposure of the animal(s) to the treatment (and associated suffering) ends (sensu Wolfensohn and Lloyd).¹⁶

In toxicology, endpoint is a long-standing term referring to the response variable measured to assess treatment effects.^19,20 Acute toxicology studies (< 7 d) have traditionally assessed mortality as the endpoint to determine LC₅₀ or LD₅₀ (i.e. the lethal dose/concentration of a compound required to kill half the organisms in the study) but may also assess lesions/injuries to the skin or eye as endpoints.^19,20Chronic (longer) toxicology studies typically assess multiple endpoints, which can be either dichotomous or continuous variables, and various descriptive prefixes have been used, for example mechanistic, sensitive, non-lethal, sublethal, apical, endocrine-specific endpoints.²¹ Response variables in health studies may similarly be termed endpoints, surrogate endpoints or clinical biomarkers.^22,23

Parallel usage of the two distinct meanings has developed because: mortality and moribundity end-points refer to both the response variable and the point at which exposure to the treatment ends;²⁴ and the responses to a treatment (adverse effects) being used to determine the point at which exposure of the animal ends.²⁵ For future clarity, we recommend that:

the hyphenated term ‘end-point’ is used to refer to points at which exposure of the animal(s) to the treatment ends (consistent with the EU),² and differentiated from

the single word ‘endpoint’ which is restricted to response variables, as originally and still commonly used.

What categories of laboratory animal experimentation end-point are there?

In their handbook of laboratory animal welfare, Wolfensohn and Lloyd¹⁶ refer to three different types of end-point: experimental, error and humane. Morton¹² suggested five different categories of (humane) end-point. End-points can therefore be based on time, the progress of the experiment, or the condition of the animals. However, there does not appear to be standardised terminology and a wide range of other idioms is being used to refer to different types of end-point, for example objective-achieved, surrogate (pre-lethal), absolute, predictive, clinical, predetermined operational, welfare and scientific.^11,26,27 We suggest differentiation of eight distinct categories of end-point (Table 1).

Table 1.

Eight different end-points to be considered in animal experiments, developed from Dennison and Ryder,¹¹ Morton,¹² Wolfensohn and Lloyd¹⁶, Dennison²⁶ and Lelliott²⁷.

End-point		Applies to	Description
Experimental end-point		All experimental units	Termination when the experiment has run its planned duration
Early end-points	Aim end-point	All experimental units	Termination as soon as the experimental aim has been demonstrated (when contemporaneous data available)
	Biological error end-point	Individual animal	Animal removed prematurely if recognised as unsuitable, based upon clinical signs (e.g. of infection, injury, deformity)
	Technical error end-point	All or individual experimental units	Premature termination when recognised that results will be invalid/compromised or cannot continue due to operational issues
	Mortality end-point	Individual animal	Animal death due to treatment during experiment
	Moribundity end-point	Individual animal	Animal removed before death, when classed as moribund based upon clinical signs
	Prognostic humane end-point	Individual animal	Animal removed at a pre-defined state, based upon clinical signs, predictive of a more severe end-point used historically
	Non-prognostic humane end-point	Individual animal	Animal removed at a pre-defined state based upon clinical signs to limit suffering

What is the experimental end-point?

Following Wolfensohn and Lloyd,¹⁶ the experimental end-point is the pre-defined duration of the experiment, based on time and/or the progress of responses to the treatment. An example of an experimental end-point is the 96 h planned duration of TG203.⁴ All other types of end-point occur before the experimental end-point and can therefore be considered as categories of early end-point.

Please note that the National Academy of Sciences²⁸ terms the experimental end-point when ‘the scientific aims and objectives have been reached’. This matches what we propose calling the aim end-point (see below) and suggest that use of experimental end-point is restricted to define the maximum duration.

What is the aim end-point?

Experimental study plans should have a clear aim, as well as a maximum duration. Once the aim has been achieved (e.g. demonstration of protection from disease), researchers are obliged to intervene¹¹ to end the experiment rather than continuing to the experimental end-point. This will only be possible when response data is available contemporaneously.

TG203 aims to assess the toxicity of a chemical over an experimental end-point (duration) of 96 h, but requires calculation of LC₅₀ values for 24, 48, 72 and 96 h.⁴ If treatment-related mortality occurs before 96 h and an earlier LC₅₀ value indicative of toxicity can be calculated (e.g. LC₅₀ at 24 h), then has the aim been achieved obliging early termination of the remaining fish?

What are error end-points?

Wolfensohn and Lloyd¹⁶ described the error end-point as early termination when it is recognised that the experimental result will be invalid. We suggest dividing into two types: biological and technical. Biological error end-points are when individual animals develop a physiological (or psychological state) that may affect the response to the treatment,¹² such as infection with a non-target pathogen, injury or distressed behaviour that then may affect physiology or dosing. Examples of technical error end-point are: human error resulting in incorrect treatment doses being administered; system failures due to loss of power, food or water supply; insufficient animals remaining to conduct a statistically valid test. Prior consideration of action to take in the event of such occurrences will reduce any delay in interventions.

What is the mortality end-point?

If an animal dies due to the treatment before the experimental end-point, it has reached the mortality end-point. The mortality end-point was developed nearly 100 years ago, associated with the LD₅₀ method of standardising drug toxicity assessment.²⁰ The mortality end-point has proved popular because ‘it is objective, is not time-consuming, and can be monitored by nonskilled staff’.²⁹ Regulatory toxicity tests involving mammals no longer require the mortality end-point which has been refined by use of humane end-points on ethical grounds.^3,20 However, TG203 still requires the mortality end-point for fish, to enable calculation of LC₅₀ values.⁴

Death in fish can be difficult to confirm, so it has been defined as ‘no visible movement (e.g. gill movements) and if touching the caudal peduncle produces no reaction’.⁴ TG203 requires that ‘mortalities are recorded, and dead fish are removed as soon as they are observed’, but does not state that death should be confirmed prior to disposal. To conform to the principles of humane experimentation, it is recommended that when such fish are removed, death is confirmed, for example, by destruction of the brain or presence of rigor mortis, as required for completing euthanasia.²

What is the moribundity end-point?

Moribundity (i.e. the state of being moribund) generally refers to an animal ‘close to death’¹⁸ or in a ‘severely debilitated state that precedes imminent death’.^7,30 (Moribundity has also been defined as ‘being in state of dying or inability to survive, even if treated’³ which is a broader state and reflects progression beyond the point of no return (see below).)

The term moribundity must not be confused with morbidity, which is also mentioned in relation to end-points, for example ‘use morbidity rather than mortality as the endpoint’;¹⁸ ‘Death or severe morbidity as an endpoint should always be challenged in any field of research’.³¹ Morbidity has a very different meaning, referring to ‘a departure… from a state of physical or mental well-being’.³²

There is a lack of clarity around moribundity in fish (as with higher vertebrates) and a definition is needed.^4,24,30 In a discussion of regulatory testing of fish vaccines, Midtlyng et al.⁷ suggest that the ‘moribund state is relatively easy to identify’ and that clinical signs in infected fish progress with:

very sick fish becoming unresponsiveness to external stimuli, and usually float at the surface or sink to the bottom of the tank. This can progress to a loss of equilibrium, with fish floating ‘belly up’ as respiration rate slows. At this stage mortality usually occurs.

As these clinical signs are also reported in a discussion of regulatory toxicity testing,²⁴ we suggest that moribundity in fish is identified by a lack of response to external stimuli, and a loss of equilibrium (i.e. loss of righting reflex potentially resulting in ‘belly up’ orientation), and a slow opercular ventilation rate. These signs may be accompanied by loss of buoyancy control (floating at the surface or sinking to the bottom).¹⁴

When considering moribundity in fish, it is pertinent to consider the progressive clinical signs of anaesthesia in fish (Table 2). The behavioural clinical signs we propose to identify the moribundity end-point match those of the deep/surgical anaesthesia stage in fish, which (imminently) precede the medullary collapse and death stages during overdose (Table 2). Fish in deep/surgical anaesthesia are considered unconscious, due to the loss of reaction to stimuli. Moribund fish are therefore similarly likely to be unconscious and incapable of suffering,^18,30 and the moribundity end-point cannot be considered as a refinement of the mortality end-point. Although it is recommended that moribund fish are removed for humane killing, they can be assumed to have already experienced maximal suffering.^7,30

Table 2.

Progression of clinical signs through anaesthesia stages in fish, collated from Brown,³³ Martins et al.,³⁴ Ross et al.,³⁵ and Tytler and Hawkins.³⁶ Please note that classification of stages and associated signs differ slightly between sources. Indication of assumed level of consciousness added, based on response to stimuli.

Anaesthesia stage	Reaction to visual and tactile stimuli	Equilibrium	Opercular ventilation rate	Swimming activity	Muscle tone	Heart rate	Analgesia	Assumed consciousness
Normal	Normal	Normal	Normal	Normal	Normal	Normal	None	Normal
Light sedation	Slight loss of reaction to stimuli			Voluntary swimming continues but reduced				Slight loss
Deep sedation	No reaction to minor stimuli		Slight decrease	Voluntary swimming ceases	Slight decrease		Some
Light narcosis	(Increased reaction to stimuli - excitement phase may occur)	Partial loss. Weak responses to postural change	Irregular (and/or Increase)	Swimming capability seriously disrupted	Decrease
Deep narcosis	No reaction to minor stimuli	Total loss	Decrease (below normal)				Good
Light anaesthesia	No reaction to major stimuli		Further decrease		Complete loss	Decrease		Complete loss
Surgical anaesthesia	No reaction to massive stimuli		Very low			Slow
Medullary collapse			None		Flaccid	Slow to none - cardiac arrest
Death

Early end-points based upon clinical signs, have been applied within TG203 in those countries where animal experimentation legislation requires that the mortality end-point is avoided.⁴ Rufli²⁴ assessed the available data on clinical signs and five alternative definitions for the moribundity end-point (Table 3). However, the use of the term moribundity to classify these states appears to be a misnomer because: a clinical sign that is present for 2 d is not indicative of imminent death within a 4 d experiment; strong ventilation, and severe scores for swimming behaviour which include hyper-activity and abnormal swimming patterns, do not indicate a severely debilitated state. We therefore consider that the use and analysis of such early end-points by Rufli²⁴ represents progression towards introducing humane (rather than moribundity) end-points.

Table 3.

The clinical signs proposed previously by Rufli²⁴ for five different definitions of moribundity end-point assessed for the fish acute toxicity test TG203.

	Swimming behaviour		Loss of equilibrium		Strong ventilation		Pigmentation
A	Severe score for ≥ 48 h	and	Severe score for ≥ 48 h
B	Severe score for ≥ 48 h	and	Single Severe score	or	Single Severe score
C	Severe score for ≥ 48 h	and	Single Severe score	or	Single Severe score	or	Single Severe score
D	Single Severe score	and	Single Severe score	or	Single Severe score
E	Single Severe score	and	Single Severe score	or	Single Severe score	or	Single Severe score

What is a humane end-point?

A humane end-point is the point when suffering of an individual animal due to a treatment is ended (typically by humane killing), based upon its clinical signs reaching pre-defined criteria that precede those of moribundity and mortality (sensu Olfert and Godson).³⁷ Given their importance to refinement, it is surprising that there is not already a consistent definition of humane end-point:

• the point at which an experimental animal’s pain and/or distress can be terminated, minimised, or reduced by actions such as killing the animal humanely, terminating a painful procedure, or providing treatment to relieve pain and/or distress;^11,18

• the earliest indicator in an animal experiment of severe pain, severe distress, suffering, or impending death;³

• the animal is killed at a defined point to limit its suffering;¹⁶

• clear, predictable and irreversible criteria which substitute for more severe experimental outcomes such as advanced pathology or death;³⁸

• criteria that can serve as the basis for ending a test procedure earlier in order to terminate or avoid pain and distress associated with the traditional study endpoint, while still allowing attainment of study objectives;³⁹

• the earliest indicator in an animal experiment of (potential) pain and/or distress that, within the context of moral justification and scientific endpoints to be met, can be used to avoid or limit pain and/or distress by taking actions such as humane killing or terminating or alleviating the pain and distress.⁴⁰

Based on these variable definitions and our understanding, we consider that humane end-points can be divided into two distinct categories: non-prognostic and prognostic (Table 1). Non-prognostic humane end-points reflect cases where the researcher accepts that the experimental aim can be achieved without animals progressing to a more severe state associated with greater suffering. Prognostic humane end-points reflect cases where clinical sign data are analysed to identify those that are predictive of a historically used, more severe end-point;³⁹ Morton¹² termed such cases surrogate humane end-points. EU researchers are required to employ an experimental design and observations to minimise the animal suffering whilst achieving the experimental aim.² Humane end-points therefore need to be considered for all experimental procedures, regardless of severity.

A conceptual model is presented to illustrate the theoretical application of a humane end-point in a severe acute toxicity test where the conventional end-point was mortality (Figure 1); analogous stages can be considered for a severe disease challenge. There are various points and stages in the time-line leading to death of a fish (Figure 1(a)):

start of exposure to a toxic dose of a chemical;

an initial period where there are likely to be physiological and molecular responses to the chemical, but no changes in appearance or behaviour due to toxicity are visible. Fish may however show transient aversive behaviour or hyperactivity if the chemical is a chemosensory irritant.⁴¹

A period when clinical signs (visible abnormalities in fish behaviour or appearance) occur due to toxicity which are assumed to intensify with duration of exposure due to increased toxic damage;

the point of no return (PNR),¹⁶ that is, a theoretical point when damage to the fish exceeds the capacity to recover, and death is therefore inevitable at a future point;

moribundity;

death.

Figure 1.

Conceptual model to illustrate the application of a humane end-point during the time course of a severe acute toxicity test, adapted from Wolfensohn and Lloyd,¹⁶ and^, Brønstad.⁴² (a) Time-line of stages between initiation of exposure and death. (b) Suffering of individual fish at any time-point. Please note that severity bands have been included for illustrative purposes only. See text for further explanation.

This conceptual model can be expanded to visualise suffering (Figure 1(b)) based upon the assumptions that the level of suffering at any time-point increases as the test progresses, correlating with the degree of abnormality of the clinical signs, and moribund fish are past the point of suffering as discussed above.

A humane end-point, based upon visible abnormalities in behaviour and appearance (clinical signs), can therefore be applied between the PNR and moribundity, which reduces the suffering experienced. The theoretical model also illustrates that consideration should be given to continual refinement, that is, to reduce suffering by moving towards earlier, more humane end-points within the intervention window where possible.

Please note that the theoretical model (Figure 1(b)) illustrates the level of suffering at any time-point; the area under the line indicates the integrated suffering experienced to that point.¹⁶ There is some ambiguity on severity assessment within the EU Directive² due to the requirement to consider ‘cumulative suffering within a procedure’. Our understanding is that: it should be based upon ‘the most severe effects likely to be experienced’, that is the maximum level of suffering at any time-point, and not the total suffering integrated over time; the requirement to consider ‘cumulative’ suffering is meant to reflect (usually) additive causes of suffering at any time-point, for example: handling to select a defined size range for the experiment + food deprivation + transfer to a barren test tank + abnormal social grouping + chemical exposure.

Discussion

End-points are integral to ethical animal experimentation, and there is often a legal duty to apply appropriate end-points when animals are used for scientific purposes.² However, terminology around end-points is confusing as highlighted in the Introduction. Our aim in writing this paper is to clarify the variable terminology to avoid further misunderstanding and encourage application. We have suggested division of early end-points into seven different types (Table 1). We hope that this framework and associated text does clarify, rather than further confuse, and its utility will be determined by uptake. We also expect our proposed framework to evolve further. For example, when clinical signs reach a specific stage in some animal models, it may be possible to make therapeutic interventions (e.g. fluid replacement, pain relief) to reduce suffering without ending the experiment.²⁹ Another example is the appearance of an unexpected phenotype in genetically modified animals;²⁸ rather than being identified as a biological error end-point and euthanised, these may be held under modified conditions due to intrinsic scientific value. As use of the animal continues in such cases, separate terminology may need to be developed for intervention points.⁴³

Prior consideration of the various early end-points has been introduced as part of study planning and ethical review process in our establishment; this reinforces the need to consider refinement and contributes to our culture of care. Application of the aim end-point has reduced suffering in our research studies of disease transmission and treatment efficacy. Moribundity, humane and biological error end-points are routinely applied by study staff via checks for general and treatment-specific clinical signs.¹⁴ Wider appreciation of different early end-points by regulators and experimenters is likely to have broader refinement benefits for fish and other species.

There is growing acceptance that regulatory tests requiring mortality and severe suffering are ethically unacceptable.²² With specific reference to TG203, discussions have been ongoing for a number of years to introduce an early end-point to refine the mortality end-point.^4,24 The focus of debate was to use clinical signs to identify a prognostic humane end-point. Due to perceived reluctance to move substantially away from the mortality end-point (which reflects the environmental protection goal of acute toxicity testing) the term moribundity end-point was introduced. However, this term generated confusion and debate, which demonstrates that clarity in terminology is needed for progress at international level. Our aim in clarifying the moribundity end-point is to enable progression towards genuine humane end-points in regulatory tests using fish.

Momentum is building to implement humane end-points in severe eco-toxicity and fish disease experiments.^7,8 To identify humane end-points, score sheets are needed for recording clinical signs,⁴⁴ and have been proposed for fish (Annex 4 in TG203;⁴ Appendix B in Midtlyng et al.⁷). Laboratory fish researchers and technicians should already be trained and competent to recognise clinical signs¹⁸ due to familiarity with the progressive stages of anaesthesia in fish (Table 2). This combination provides an important foundation from which to identify and implement humane end-points in fish. Katsiadaki et al.⁸ discuss practical and regulatory issues associated with implementation of humane end-points in fish toxicity testing. The refinement discussions for TG203 have also expanded to include addressing a separate early end-point, the aim end-point, that is early termination where a toxicity result is derived before the 96 h experimental end-point.⁸

Footnotes

Acknowledgements

The manuscript benefited from comments from Hans Rufli, Chris Green, three anonymous referees and the journal editors.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Defra, International Chemicals (How can we produce internationally standardised guidance for assessing sublethal clinical signs of toxicity in fish to improve animal welfare in chemical safety tests?); the Veterinary Medicines Directorate (VM0519: Can humane end-points be introduced into fish disease challenges for fish medicine development?).

ORCID iD

Tim Ellis

References

Russell

WMS

Burch

The Principles of Humane Experimental Technique. London: Methuen, 1959.

EU. EU Directive 2010/63 on the protection of animals used for scientific purposes, https://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2010:276:0033:0079:en:PDF (2010, accessed September 2020).

OECD. Guidance document on the recognition, assessment, and use of clinical signs as humane endpoints for experimental animals used in safety evaluation, https://www.oecd-ilibrary.org/environment/guidance-document-on-the-recognition-assessment-and-use-of-clinical-signs-as-human-endpoints-for-experimental-animals-used-in-safety-evaluation_9789264078376-en (2000, accessed September 2020).

OECD. Test Guideline No. 203 Fish, Acute Toxicity Testing, https://www.oecd-ilibrary.org/environment/test-no-203-fish-acute-toxicity-test_9789264069961-en (2019, accessed September 2020).

EDQM (European Directorate for the Quality of Medicines & Healthcare). European Pharmacopoeia, Vol. 1, 9th ed. Strasbourg: Council of Europe, 2016.

Hawkins

Dennison

Goodman

, et al. Guidance on the severity classification of scientific procedures involving fish: report of a Working Group appointed by the Norwegian Consensus-Platform for the Replacement, Reduction and Refinement of animal experiments (Norecopa). Lab Anim 2011; 45: 219–24.

Midtlyng

Hendriksen

Balks

, et al. Three Rs approaches in the production and quality control of fish vaccines. Biologicals 2011; 39: 117–28.

Katsiadaki

Ellis

Andersen

, et al. Dying for change: a roadmap to refine the fish acute toxicity test after 40 years of suffering. Ecotoxicol Environ Saf In prep

Rufli

Springer

TA.

Can we reduce the number of fish in the OECD acute toxicity test?

Environ Toxicol Chem 2011; 30: 1006–1011.

10.

Jirkof

Jarvis

Riederer

Collection on score sheets, severity assessment and humane end points: invitation to submit. Lab Anim 2020; 54: 149.

11.

Dennison

Ryder

The challenges of using humane endpoints in fish research, https://norecopa.no/media/6272/abstract-ryder-endpoints.pdf (2009, accessed September 2020).

12.

Morton

DB.

Humane endpoints in animal experimentation for biomedical research: ethical, legal and practical aspects.

In: Coenraad F, Hendriksen C and Morton DB (eds) Humane endpoints in animal experiments for biomedical research. London: Royal Society of Medicine Press, 1999. p. 5–12.

13.

Smith

Clutton

Lilley

, et al. PREPARE : Guidelines for planning animal research and testing. Lab Anim 2018; 52: 135–141.

14.

Ellis

Rimmer

GSE

Parker

, et al. In-tank underwater cameras can refine monitoring of laboratory fish. Anim Welf 2019; 28: 191–203.

15.

UK Home Office . Guidance on the Operation of the Animals (Scientific Procedures) Act 1986 [Internet]. Home Office, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/662364/Guidance_on_the_Operation_of_ASPA.pdf (2014, accessed September 2020).

16.

Wolfensohn

Lloyd

Handbook of Laboratory Animal Management and Welfare. Oxford: Blackwell, 2003.

17.

Dolan

Laboratory Animal Law. Oxford: Blackwell Science Ltd, 2000.

18.

CCAC. CCAC Guidelines On: Choosing an appropriate endpoint in experiments using animals for research, teaching and testing, https://www.ccac.ca/Documents/Standards/Guidelines/Appropriate_endpoint.pdf (1998, accessed September 2020).

19.

Walker

Hopkin

Siblly

, et al. Principles of Ecotoxicology. London: Taylor & Francis Ltd, 1997.

20.

Sass

Humane endpoints and acute toxicity testing. ILAR J 2000; 41: 114–123.

21.

OECD. Fish toxicity testing framework, https://www.oecd.org/env/fish-toxicity-testing-framework-9789264221437-en.htm (2012, accessed September 2020).

22.

Olsson

IAS

Nicol

Niemi SM , et al. From unpleasant to unbearable — why and how to implement an upper limit to pain and other forms of suffering in research with animals. ILAR J 2020; ilz018.

23.

Wallace

Humane endpoints and cancer research. ILAR J 2000; 41: 87–93.

24.

Rufli

Introduction of moribund category to OECD fish acute test and its effect on suffering and LC50 values. Environ Toxicol Chem 2012; 31: 1107–1112.

25.

Dennis

MB.

Humane endpoints for genetically engineered animal models. ILAR J 2000; 41: 94–98.

26.

Dennison

Avoiding mortality – what to monitor and how. In: 3rd RSPCA international meeting -- Focus on severe suffering: avoiding mortality. Athens, Greece, 10--11 October 2019.

27.

Lelliott

Using a data mining approach to avoiding mortality when using large numbers of animals. In: 3rd RSPCA international meeting -- Focus on severe suffering: avoiding mortality. Athens, Greece, 10--11 October 2019.

28.

National Academy of Sciences. Guide for the care and use of laboratory animals. 8th ed. Washington: The National Academies Press, 2011. https://grants.nih.gov/grants/olaw/guide-for-the-care-and-use-of-laboratory-animals.pdf (accessed September 2020).

29.

Hendriksen

CFM

Steen

Refinement of vaccine potency testing with the use of humane endpoints. ILAR J 2000; 41: 105–113.

30.

Toth

LA.

Defining the moribund condition as an experimental endpoint for animal research. ILAR J 2000; 41: 72–79.

31.

RSPCA. Humane Endpoints, https://science.rspca.org.uk/sciencegroup/severesuffering/reduce/humaneendpoints (accessed September 2020).

32.

OECD. Glossary of statistical terms., https://stats.oecd.org/glossary/detail.asp?ID=1281 (2001, accessed September 2020).

33.

Brown

LA.

Anesthesia and restraint. In: Stoskopf MK (ed.) Fish Medicine. Philadelphia: W. B. Saunders Company, 1993, pp. 79–90.

34.

Martins

Valentim

Pereira

, et al. Anaesthesia and analgesia in laboratory adult zebrafish: A question of refinement. Lab Anim 2016; 50: 476–488.

35.

Ross

Anaesthetic and Sedative Techniques for Aquatic Animals. 3rd ed. Oxford: Blackwell Publishing Ltd, 2008.

36.

Tytler

Hawkins

AD.

Vivisection, anaesthetics and minor surgery. In: Hawkins AD (ed.) Aquarium Systems. London: Academic Press, 1981, pp. 247–278.

37.

Olfert

Godson

DL.

Humane endpoints for infectious disease animal models. ILAR J 2000; 41: 99–104.

38.

NC3Rs. Humane endpoints, https://www.nc3rs.org.uk/humane-endpoints (2014, accessed September 2020).

39.

Stokes

Marsman

DS.

Animal welfare considerations in biomedical research and testing. In: Bayne K and Turner P (eds) Laboratory Animal Welfare. Cambridge, MA: Academic Press, 2014, pp. 115–140.

40.

3Rs-Centre Utrecht Life Sciences. What are humane endpoints? https://www.humane-endpoints.info/en/what-are-humane-endpoints (accessed September 2020).

41.

Drummond

Russom

Geiger

, et al. Behavioral and morphological changes in fathead minnow (Pimephales promelas) as diagnostic endpoints for screening chemicals according to mode of action. Aquat Toxicol Environ Fate 1986; 9: 415–435.

42.

Brønstad

Severity and humane endpoints in fish research. In: Severity and humane endpoints in fish research [Internet]. Fondazione Guido Bernardini, https://www.fondazioneguidobernardini.org/usermedia/file/materiale_corsi/fondazione/FarmFish2019/Intro HEP severity in FISH.pdf (2019, accessed September 2020).

43.

Stokes

WS.

Reducing unrelieved pain and distress in laboratory animals using humane endpoints. ILAR J 2000; 41: 59–61.

44.

Morton

DB.

A systematic approach for establishing humane endpoints. ILAR J 2000; 41: 80–86.

Clarification of early end-points for refinement of animal experiments,with specific reference to fish

Abstract

Keywords

Introduction

What is an end-point?

What categories of laboratory animal experimentation end-point are there?

What is the experimental end-point?

What is the aim end-point?

What are error end-points?

What is the mortality end-point?

What is the moribundity end-point?

What is a humane end-point?

Discussion

Footnotes

Acknowledgements

Declaration of Conflicting Interests

Funding

ORCID iD

References