Volume 50 Number 6 December 2016 ISSN 0023-6772 ......la.sagepub.com Published on behalf of...

la.sagepub.com

Published on behalf of Laboratory Animals Ltd by

SAGE Publications Press Ltd

Volume 50 Number 6 December 2016 ISSN 0023-6772

Offi cial Journal of AFSTAL, ESLAV, FELASA, GV-SOLAS, ILAF, LASA, NVP, SECAL, SGV, SPCAL

THE INTERNATIONAL JOURNAL OF LABORATORY ANIMAL SCIENCE AND WELFARE

Laboratory Animals

Special Issue: Score Sheets and Postoperative AnalgesiaGuest Editors: H Golledge and P Jirkof

ON LABORATORY ANIMAL SCIENCE 22-24 MARCH 2017 | BERLIN, GERMANY

8T H E U R O P E A N

EMPOWER YOUR PROFESSIONAL SELFThe Charles River Short Course is a leading educational event that brings together an acclaimed panel of speakers with a programme which will allow you to gain an in-depth understanding of relevant subjects that are shaping the ever-changing world of life science and research. Whether you’re a seasoned veteran or just starting out, the Short Course is an excellent opportunity for you to further your knowledge, obtain continuing education credits and network with fellow researchers, facility managers, veterinarians and technicians. Space is limited, so don’t delay.

Visit our website for full details at http://eushortcourse.criver.com

http://eushortcourse.criver.com

ab oratory

l i m i t e d

lan imals

Volume 50 Number 6 December 2016

Contents

Special Issue: Score Sheets and Postoperative AnalgesiaGuest Editors: H Golledge and P Jirkof

Thanks to Reviewers 409

Special Editorial

Editorial: Score sheets and analgesia 411H Golledge and P Jirkof

Special Articles

What the literature tells us about score sheet design 414P Bugnon, M Heimann and M Thallmair

Analgesia in clinically relevant rodent models of sepsis 418V Jeger, T Hauffe, F Nicholls-Vuille, D Bettex and A Rudiger

The more the merrier? Scoring, statistics and animal welfare in experimental autoimmuneencephalomyelitis 427P Palle, FM Ferreira, A Methner and T Buch

Osteotomy models – the current status on pain scoring and management in small rodents 433A Lang, A Schulz, A Ellinghaus and K Schmidt-Bleek

Severity assessment and scoring for neurosurgical models in rodents 442S Pinkernell, K Becker and U Lindauer

Morbidity scoring after abdominal surgery 453R Graf, P Cinelli and M Arras

Recommendation for severity assessment following liver resection and liver transplantationin rats: Part I 459S Kanzler, A Rix, Z Czigany, H Tanaka, K Fukushima, B Kogel, K Pawlowsky and RH Tolba

Severity assessment in rabbits after partial hepatectomy: Part II 468N Drude, K Pawlowsky, H Tanaka, K Fukushima, B Kogel and RH Tolba

Anaesthesia and analgesia in laboratory adult zebrafish: a question of refinement 476T Martins, AM Valentim, N Pereira and LM Antunes

News

Iniciativa espanola para mejorar la transparencia en el uso de animales de experimentacion 489J Guillen

Calendar of events/Index to advertisers xxviii

Catalogue 2017OUR BRAND NEW CATALOGUE 2017 IS NOW AVAILABLE!

Request your free copy at: finescience.de or tel.: +49 (0) 6221 905050

FINE SURGICAL INSTRUMENTS FOR RESEARCHTM

.............................................................................................................................................................................................

Laboratory Animals

Editorial Board

Editor-in-Chief B RiedererDeputy Editors A Bomzon, J RodgersEditorial Assistant N Baehler

Section Section Editors

Anaesthesia and Analgesia P Flecknell, S Robertson,M Leach

Anatomy and Neuroscience B RiedererBehaviour M Bateson, D PreissmannBiostatistics and ExperimentalDesign

R-D Gosselin

Clinical Chemistry P O’BrienEducation P VergaraEquine Models T MorrisGenetic Engineering T RuelickeImaging Techniques L van der WeerdLarge Animal Models M Jensen-Waern,

D AndersonManagement of AnimalFacilities

J-B Prins

Microbiology M DennisMolecular genetics,pain & distress

P Cinelli

Pathology P Clements, D SalvatoriPhysiology C Gilbert, M SommersPrimates G Rainer3Rs and ethics G Griffin, A OlssonReproductive Biology H HedrichSmall Animal Models M BerardSurgical procedures R TolbaSystematic Review M Ritskes-HoitingaToxicology F RuttenVeterinary Medicine E Rivera, J Sanchez-

Morgado, L Whitfield

Comment and correspondence relating to editorial mattersmay be sent to the Chairman of the Editorial Board by email:[email protected]; or post: LAL, PO Box 373, Eye,Suffolk, IP22 9BS, UK.See also http://www.lal.org.uk

Laboratory Animals, (ISSN 0023-6772) is published anddistributed bimonthly (February, April, June, August,October, December) in both print and electronic form bySAGE Publications Ltd, 1 Oliver’s Yard, 55 City Road,London EC1Y 1SP, UK.

All manuscripts submitted for publication shouldbe prepared in accordance with the Guidelinesfor Authors which can be found online atla.sagepub.com/. Please submit your paper online atmc.manuscriptcentral.com/la, contributions for newsitems can also be made via manuscript central.

Subscription information

Annual subscription (2017) including postage: AALASMember, (print and electronic) [£113/US$144].Combined Institutional Rate (print and electronic)[£288/US$533]. Electronic only and print only

subscriptions are available for institutions at a discountedrate. Note VAT is applicable at the appropriate local rate.Visit sagepublishing.com for more subscription details.To activate your subscription (institutions only) visitonline.sagepub.com. Abstracts, tables of contents andcontents alerts are available on this site free of chargefor all. Student discounts and single issue rates are avail-able from SAGE Publications Ltd, 1 Oliver’s Yard, 55 CityRoad, London EC1Y 1SP, UK, tel. þ44 (0)20 7324 8500,email [email protected] and in North America,SAGE Publications Inc, PO Box 5096, Thousand Oaks,CA 91320, USA.

Advertisement Managers

PRC Associates Ltd, 1st Floor Offices, 115 Roebuck Road,Chessington, Surrey KT9 1JZ, UK; Tel: +44 (0) 20 8337 3749;Fax: +44 (0) 20 8337 7346; Email: [email protected]

Laboratory Animals Ltd

Laboratory Animals Ltd is a company limited byguarantee and has no share capital. The Memorandumof Association obliges the company to apply all its resourcesto the advancement of public education in laboratory animalscience, technology and welfare. It is a registered charity(Registered Charity Number 261047) and none of its direc-tors may receive any fee or remuneration.

Registered Office:Laboratory Animals Ltd, 44 Springfield Road, Horsham,West Sussex, RH12 2PD, UK

Council of Management

Chairman J-B PrinsSecretary E WeirTreasurer J Gregory

L Antunes J HelpiK Applebee P NowlanV Baumans J OrellanaM Berard B RiedererM Dorsch M Ritkes-HoitingaN Ezov A ShortlandA Forslid S WellsC Gilbert M WilkinsonJ Guillen

� 2016 Laboratory Animals Ltd. Apart from any fair dealing for the

purposes of research or private study, or criticism or review, as permitted

under the UK Copyright, Designs and Patents Act, 1988, no part of this

publication may be reproduced, stored, or transmitted, in any form or by

any means, without the prior permission of the publishers, or in the case

of reprographic reproduction in accordance with the terms of licences

issued by the Copyright Licensing Agency in the UK, or in accordance

with the terms of licences issued by the appropriate Reproduction Rights

Organization outside the UK. Enquiries concerning reproduction outside

the terms stated here should be sent to SAGE at the address.

Whilst every effort is made to ensure that no inaccurate or misleading

data, opinion or statement appears in the journal, Laboratory Animals Ltd

wish to make it clear that the data and opinions appearing in the articles

and advertisements herein are the responsibility of the contributor or

advertiser concerned. Accordingly, Laboratory Animals Ltd and their offi-

cers and agents accept no liability whatsoever for the consequences of any

such inaccurate or misleading data, opinion or statement.

Printed in Great Britain

AFSTALAssociation Francaise des Sciences etTechniques de I’Animal de LaboratoirePresidentSebastian Paturance

Vice PresidentElodie Bouchoux

Secretariat: 28, rue Saint Dominique,75007, Paris, France(www.afstal.com)

ESLAVEuropean Society of LaboratoryAnimal VeterinariansPresidentNicolas Dudoignon

SecretaryKaterina Marinou

Secretariat: Sanofi-Aventis R&D,1, Avenue Pierre Brossolette91385 Chilly-Mazarin Cedex France(www.eslav-eclam.org)

FELASAFederation of EuropeanLaboratory Animal ScienceAssociationsPresidentHeinz Brandstetter

Past-PresidentJan-Bas Prins

Secretariat: PO Box 951,Needham MarketIpswich IP6 8WH, UK(www.felasa.eu)

GV-SOLASGesellschaft fur Versuchstierkunde(Society for Laboratory Animal Science)

PresidentRene Tolba

SecretaryNicole LinklaterFaculty of BiologyPhilipps UniversityKarl-von-Frisch Str. 835043 MarburgGermany(www.gv-solas.de)

ILAFIsraeli Laboratory Animal ForumPresidentAmir Rosner

SecretaryDavid CastelNeufeld Cardiac ResearchInstituteSheba Medical CenterTel Hashomer 52621Israel(www.ilaf.org.il)

LASALaboratory Animal ScienceAssociationPresidentDavid Anderson

Honorary SecretaryKathleen Mathers

PO Box 524, Hull,HU9 9HE, UK(www.lasa.co.uk)

NVPNederlandse Vereniging voorProefdierkunde(Dutch Association for LaboratoryAnimal Science)

PresidentMartje Fentener van Vlissingen

SecretaryJan LangemansBPRCLange Kleiweg 1392288 GJ RijswikThe Netherlands(www.proefdierkunde.nl)

SECALSociedad Espanola para las Cienciasdel Animal de Laboratorio(Spanish Society for Laboratory AnimalScience)

PresidentAntonio Martınez Escandell

Vice PresidentMarıa Teresa Rodrıgo Calduch

SecretaryAngel Naranjo Pino

TreasurerCarlota Largo Aramburu

Secretariat: c/Maestro Ripoll, 8,28006 Madrid,Spain(www.secal.es)

SGVSchweizerische Gesellschaft furVersuchstierkundeSociete Suisse pour la Science desAnimaux de Laboratoire (SwissLaboratory Animal ScienceAssociation)

PresidentDr. Birgit Ledermann

SecretaryDr Brunhilde Illgen-WilckeMicroBios GmbHMicrobiological ServiceChristoph-Merian-Ring 31ACH-4153 Reinach, Switzerland(www.scienzenaturali.ch)

SPCALSociedade Portuguesa de Cienciasem Animais de Laboratorio(Portuguese Society for LaboratoryAnimal Science)

PresidentIsabel Vitoria Figueiredo

Vice-PresidentRicardo Afonso

SecretaryCatarina Pinto Reis

Secretariat: Laboratorio deFarmacologiaFaculdade de FarmaciaLargo de D. Dinis3000 CoimbraPortugal(www.spcal.pt)

vi...............................................................................................................................................................

ab oratory

l i m i t e d

lan imals

Thanks to Reviewers

Thanks to Reviewers

The Editor-in-Chief, Editorial Board and Publisher would like to thank the following reviewers for their contri-bution to Laboratory Animals in 2016.

Alcorn, JaneAllen, KyleAnderson, DavidArany, PraveenArras, MargareteArts, JohannaAvey, MarcBakker, JacoBalafas, EvangelosBalan, MariaBaligand, CelineBaneux, PhillipeBaruch, YaacovBatkai, SandorBaumans, VeraBaumgartner, BernhardBayans, ToniBeckmann, NicolauBekkevold, ChristineBendtsen, KatjaBenga, LaurentiuBergadano, AlessandraBerset, CorinaBertrand, HenriBlau, ChristophBollen, PeterBomzon, AriehBoor, PeterBotting, KimBrown, JustinBugnon, PhilippeBussell, JamesButcher, GeoffBosze, ZsuzsannaCapuano, SaverioCarbone, LarryChourbaji, SabineCinelli, PaoloCrim, MarcusDahlborn, KristinaDahmen, UtaDeeny, Adrian

Ehall, HelmutEklof, Ann ChristineFinger-Baier, KarinFlecknell, PaulFoley, TrishFont, EnriqueFrance, MalcolmGilbert, ColinGlowka, TimGobbi, AlbertoGolledge, HuwGomez de Segura, IgnacioGrefhorst, AldoGuarnieri, MichaelGunnarsson, StefanGomez -Mantilla, Jose DavidHaberstroh, JorgHackbarth, HansjoachimHall, YperHalsey, LewisHamilton, LindsayHankenson, ClaireHansen, AxelHansson, KerstinHasenau, JohnHaubro Andersen, PiaHedenqvist, PatriciaHeimann, MaikeHobbs, TheodoreHonickel, MarkusHowells, DavidHuneke, RichardJacobson, MagdalenaJensen-Waern, MarianneJirkof, PaulinKalliokoski, OttoKaras, AliciaKara, FiratKlarenbeek, SjoerdKolawole, AbimbolaKonno, KenjiroKostomitsopoulos, Nikos

Kramer, KlaasKramer, MatthewKuchel, TimKunter, UtaLee, Jae IlLarsson, AndersLeach, MatthewLilley, ElliotLofgren, JennieMacFarlane, PaulMahabir, EstherMahler, MichaelMakowska, JoannaManell, ElinMarques, JoanaMartin, LaurenMartinsen, OrjanMcDonell, WayneMcKeegan, DorothyMiller, AmyMiller, ManuelMocho, Jean-PhilippeMogil, JeffreyMohr, BertMoon, PaulaMorton, DavidMuhlhausler, BevMule, FlaviaMusk, GabrielleNemzek, JeanO’Brien, PeterOgden, BryanOtto, KlausPang, DanielParasuraman, SubramaniParker, MatthewRayner, EmmaRobertson, SheilahRodgers, JanetRoughan, JohnnySalvatori, DanielaScavizzi, Ferdinando

Laboratory Animals

2016, Vol. 50(6) 409–410

! The Author(s) 2016

Reprints and permissions:

sagepub.co.uk/

journalsPermissions.nav

DOI: 10.1177/0023677216680962

la.sagepub.com

Schlabritz-Lutsevich, NataliaSchofield, JohnSchuppli, CatherineShofty, RonaShortland, AmitaSidener, HeatherSinger, KanakadurgaSingh, SanilSotocinal, Susana

Sztein, JorgeSørensen, DorteThielebein, JensTobin, GrahamTolba, ReneTreuting, Piper M.Tsai, Ping-PingVan Loo, PascalleVerbeek, Sjef

Vreman, SandraWaisman, AriWallner, BarbaraWeese, MariaWhittaker, AlexandraYao, RutaoZaretsky, Asaph

410 Laboratory Animals 50(6)

Influenza-Free Ferrets Available at Marshall BioResources

Marshall’s history began with the ferret, and we continue to make it our future.

Barrier-bred, specific pathogen free ferrets are raised in the United Kingdom, and in the United States.

Marshall has been raising ferrets since 1939, and we are known as the leading supplier of ferrets worldwide. Our new barrier facilities combined with our decades of experience allow us to provide ferrets with the best health status and temperament for the development of new therapeutics and vaccines.

Don’t hesitate to visit our new website at www.marshallbio.com to learn more about everything we can offer.

Japan+81 29 875 [email protected]

Continental Europe+33 (0) 4 72 56 98 [email protected]

United Kingdom+44 (0) 1964 [email protected]

China+86 10 8492 [email protected]

North America +1 315 587 2295 [email protected]

CUSTOM DIETS

REFERENCE DIETS

HIGH FAT DIETS, ENRICHED WITH SUGARS

HYPERCHOLESTEROLIZING DIETS

DIETS RICH IN CELLULOSE

MACRONUTRIENT DIETS: PROTEINS, FATS

ALTERED MINERAL OR MICRONUTRIENT DIETS

ALTERED VITAMIN DIETS

DIETS INDUCING A PHENOTYPE

DIETS WITH YOUR OWN INGREDIENTS

MEDICATED DIETS

LOW FLUORESCENCE DIETS

1

2

3

4

5

6

7

8

9

10

11

DIETS CUSTOM DIETS BEDDINGS & ENRICHMENT

SAFE • Route de Saint Bris 89290, Augy • FranceT: +33(0)3 86 53 76 90 • Fax: +33(0)3 86 53 35 [email protected] •www.safe-diets.com

BABY FO

ODNO

A

DD IT IVES

THE TAILORED DIET SERVICE

ab oratory

l i m i t e d

lan imals

Special Editorial

Score sheets and analgesia

This special issue of Laboratory Animals presents anoverview and some detailed examples of the use ofscore sheets for monitoring welfare of animals undergo-ing scientific procedures. Score sheets formalize andstandardize the assessment of welfare and make it pos-sible to record the impacts of scientific procedures.Ideally, the data obtained from the keeping of scoresheets should also drive the development and facilitatethe implementation of measures such as refined anal-gesia to improve the welfare of animals used inresearch. However, there is no one-size-fits-allapproach to scoring animal welfare, and ongoing devel-opment is crucial to increase the accuracy and specifi-city of scoring systems. For this reason we have soughtto collate score sheets from different scientific fields inthis special issue.

It should be remembered that the papers in this spe-cial issue provide examples of the use of score sheets forvery specific animal models or scientific procedures;and, as several of the authors have emphasized, ascore sheet needs to be tailored in a number of ways.Firstly, score sheets need to be specific to the experi-mental model used. The impact of different procedureswill lead to different signs of compromised welfare, andtherefore score sheets designed for one model areunlikely to be generalizable to different experimentalprocedures. Score sheets should contain signs thatmeasure the actual, specific impact on the system ororgan targeted by the experimental manipulation aswell as more general welfare measures. This mayinclude changes in gait in osteotomy models asdescribed in the contribution by Lang et al.,1 jaundiceas a possible consequence of manipulations of the liverseen in the paper by Graf et al.,2 or neurological symp-toms in models affecting the central nervous system asdetailed by Pinkernell et al.3 and Palle et al.4

Additionally, proper observation of these signs mayalso contribute to scientific data, for example in proto-cols of experimental autoimmune encephalomyelitis.Secondly, a score sheet must be adapted for the animalswhich will be scored (species, strain, genotype, sex andage may all have pronounced effects on the relevance ofvarious welfare indicators). Signs of pain or distress inmice are not the same as those in rabbits or even rats,even when the experimental procedure or disease modelis the same. Finally, score sheets must be designed with

the purpose in mind for which they will be used. It isimportant that the user of the score sheet is clear aboutwhether it will be used simply to record the welfareimpact of the procedure on the animal (a process nowmandatory in many jurisdictions for the purpose ofreporting retrospective severity, and also very import-ant for comparing the severity of procedures and thusfor testing whether refinements are effective), or used asa trigger for some intervention. In the case where scoresheets are used as a clinical tool for deciding whether toemploy an intervention to treat compromised welfare itis important that welfare-relevant changes in the con-dition of the animals are identified reliably and early tominimize possible suffering. All possible actions to betaken in this case (and the threshold scores which willtrigger them) should be established in advance – e.g.treatment of the negative effects on the animal (forinstance by providing appropriate analgesia), termin-ation of the procedure, or euthanasia (when a prede-fined endpoint is reached). These considerations areimportant: score sheets are only useful if the informa-tion recorded within them is used in a meaningful wayto reduce animal suffering, either of the animals cur-rently being studied or future animals used in similarprocedures. Reductions of future suffering can bebrought about either by refining procedures or bydeciding that the harms quantified by the score sheetshow that the use of animals cannot be justified in aharm/benefit analysis (see also the report of theAALAS–FELASA working group on Harm–BenefitAnalysis5).

It should also be remembered that score sheets areunlikely to encapsulate all possible signs of poorwelfare. Therefore the expertise of animal caretakers,veterinarians and other expert staff should not be over-looked if they identify signs of poor welfare which arenot part of the scoring system. Score sheets should ide-ally include the possibility of recording new signs whichmay emerge during the study, and be reviewed continu-ously for improvement.

Taking the above considerations into accountresearchers can begin to consider which measures toinclude on a score sheet tailored to their study.

The indicators included within a score sheet shouldideally be scientifically validated for the experimentalmodel and species/strain/sex/genotype/age of the

Laboratory Animals

2016, Vol. 50(6) 411–413



sagepub.co.uk/


DOI: 10.1177/0023677216675387

la.sagepub.com

animal for which they are used. Indicators on a scoresheet can be treated rather like clinical tests for a par-ticular disease, the criteria for a good clinical test arethe same as those for a good welfare indicator. Welfareindicators should be tested for their sensitivity (the abil-ity to correctly identify animals with poor welfare) andspecificity (the ability to identify those not sufferingpoor welfare). The relative importance of specificityand sensitivity will depend on the use to which theinformation will be put. If the test is to be used todecide to end a study and euthanize the animal (inother words, to set an endpoint), it is crucial that thetest is both highly specific and sensitive to exclude thepossibility of animals undergoing unnecessary sufferingdue to inadequate sensitivity (false-negative scores) orunnecessary wastage of animals due to inadequate spe-cificity (false-positive scores). On the other hand, if anintervention is available which will potentially amelior-ate poor welfare and which does not affect study out-come it is more important that the test is sensitiverather than specific; treating animals which do notneed treatment (due to false-positives) is better thanfailing to offer treatment to ameliorate the sufferingof animals which have been misidentified as havinggood welfare (false-negatives). If the score sheet isused solely to characterize the impacts on welfare forthe purposes of retrospective assessment, then it isbetter in animal welfare terms if the severity is over-rather than underestimated. The worst case scenariois the use of measures which are neither specific norsensitive as this can lead to poor welfare goinguntreated or the underestimation of reported severitydue to false-negative scores. Every effort should bemade to test the validity in terms of both specificityand sensitivity of measures included on score sheetsto ensure that they do not contribute to false-negatives.Significant progress has been made recently in thedevelopment of behavioral and physiological indicatorsof pain, suffering and distress in laboratory animals.Nevertheless, many of the newly developed signs ofnegative, or positive, welfare have still to be evaluatedsystematically for different animal models, and uncer-tainty regarding the actual implication of such signsexists – a concern also expressed by many of theauthors in this special issue. As many of the articlesrecognize, there is a need for dedicated studies to iden-tify and evaluate reliable, sensitive and specific indica-tors as well as a need for evidence-based interventionprotocols, such as efficient analgesia administration fol-lowing the detection of reduced well-being. Ongoingresearch into both these areas is imperative. Where suf-ficient data exist we would encourage the use of system-atic reviews and meta-analyses to confirm the validity

of welfare indicators included on score sheets as well asthe efficacy of interventions. Ideally, a score sheetshould rely on a number of indicators to improve theoverall specificity and sensitivity, such that any onemeasure is not relied upon, thus reducing the chanceof false-negatives. The measures included on scoresheets can be refined over time and we would urgethose using score sheets to keep them under review.Ongoing developments mean that new measures mayemerge or older measures may be invalidated by newevidence. Improvements to score sheets over time arelikely to improve their validity as a welfare indicator.

Once data are collected from score sheets it isimportant that these data are shared so that otheranimal users can understand the welfare impacts ofvarious models before using them and new refinementscan be developed. Similarly if new welfare indicators orseverity reducing measures, e.g. improved analgesia andprotocols, are developed these should also be shared viascientific publications. We hope that this special issuegoes some way towards that aim and that LaboratoryAnimals will serve as a repository for future publica-tions highlighting new developments in score sheets.

We are aware that at the moment, for example in thelight of the EU Directive 2010/63, extensive efforts arebeing made to develop better measures of welfare tofacilitate retrospective severity assessment. There hasalso been a recent recognition of the role of cumulativeseverity upon the welfare of animals used in research.The effect of procedures on animals may change overtime and it is important that this is recognized, particu-larly where animals become sensitized to a particularprocedure. Score sheets can also play an important rolein identifying cumulative suffering, for instance, byidentifying an increased response to a particular pro-cedure when it is applied more than once.

We hope that this special issue will encourage morewidespread use of score sheets and the increased shar-ing of the knowledge gained from their use.

We thank all the authors and reviewers who havecontributed their time and expertise to this issue.

References

1. Lang A, Schulz A, Ellinghaus A and Schmidt-Bleek K.

Osteotomy models – the current status on pain scoring

and management in small rodents. Lab Anim 2016; 50:

433–441.2. Graf R, Cinelli P and Arras M. Morbidity scoring after

abdominal surgery. Lab Anim 2016; 50: 453–458.3. Pinkernell S, Becker K and Lindauer U. Severity assess-

ment and scoring for neurosurgical models in rodents. Lab

Anim 2016; 50: 442–452.


4. Palle P, Ferreira FM, Methner A and Buch T. The morethe merrier? Scoring, statistics and animal welfare inexperimental autoimmune encephalomyelitis. Lab Anim

2016; 50: 427–432.5. Special Issue: Report from the AALAS–FELASA work-

ing group on Harm–Benefit Analysis. Parts 1 & 2. LabAnim 2016; 50(1 Suppl): 1–20 & 21–42.

Huw Golledge1 and Paulin Jirkof21Universities Federation for Animal Welfare,Wheathampstead, Hertfordshire, UK2Division of Surgical Research, University HospitalZurich, University of Zurich, Zurich, Switzerland

Corresponding author:Paulin Jirkof, Division of Surgical Research, UniversityHospital Zurich, Sternwartstrasse 6, CH-8091, Zurich,Switzerland.Email: [email protected]

Special Editorial 413

ab oratory

l i m i t e d

lan imals

Special Article

What the literature tells us aboutscore sheet design

Philippe Bugnon1, Maike Heimann2 and Michaela Thallmair3

AbstractScore sheets are an essential tool of animal welfare. They allow transparent assessments to be made ofanimal health and behavior during animal experiments and they define interventions when deviations fromnormal status are detected. As such, score sheets help to refine animal experiments as part of the 3R(replacement, reduction and refinement) concept. This mini review aims at summarizing the scarce literatureavailable on score sheet design.

Keywordsscore sheet, 3Rs, pain assessment, refinement

This review focuses on score sheet design and hencealso on refinement as part of the 3R (replacement,reduction and refinement) concept originally proposedby Russell and Burch.1 Score sheets are especiallyimportant in animal experiments that cause pain, suf-fering, distress or harm. When animals undergo stress-ful or painful procedures, the researcher must ensurethat the animals are used in the least stressful wayand that their welfare is maximized. On these grounds,the animal needs to be assessed carefully by theresearcher, who then has to decide about possible inter-ventions. This is usually done using a scoring system,also called ‘score sheet’ or ‘welfare assessment proto-col’, as proposed by Morton and Griffiths in 1985.2

A carefully designed score sheet for a given experi-mental set-up ensures a reproducible and standardizedassessment of the experimental animal by trained per-sonnel, clear guidelines for interventions, a consistentevaluation of the efficacy of a given intervention, and afull traceability of all actions. This not only helps toincrease validity and reliability of an animal experiment,but also helps to improve the welfare of the animals.

As discussed below, the challenge is to design a scoresheet that is efficient and easy to follow. The sheetshould contain a reasonable number of meaningful par-ameters, while the use of excessive numbers of, espe-cially subjective, parameters should be avoided toreduce difficulties in correct evaluation by personnel.

The animals are assessed at a predefined frequencyaccording to the situation and clinical symptoms, and

deviations from the normal state are recorded.This enables the researcher to monitor the animalsaccurately and effectively in a consistent way by focus-ing on the relevant and accessible symptoms and par-ameters and to decide about the requiredinterventions.3–6

Information on score sheets in theliterature

The researcher typically faces three obstacles whendesigning a score sheet:

1. How to choose easy-to-understand parameters thatallow an observer to detect changes in animal well-being in a timely manner.

2. How to score these parameters.3. How to decide what refinement interventions – trig-

gered by the scores assigned – are needed to

1Institute of Laboratory Animal Science, University of Zurich,Zurich, Switzerland2Responsible for training and education in laboratory science, ETHZurich, Zurich, Switzerland3Animal Welfare, University of Zurich, Zurich, Switzerland

Corresponding author:Philippe Bugnon, Institute of Laboratory Animal Science,University of Zurich, Winterthurerstrasse 190, CH-8057 Zurich,Switzerland.Email: [email protected]

Laboratory Animals

2016, Vol. 50(6) 414–417



sagepub.co.uk/


DOI: 10.1177/0023677216671552

la.sagepub.com

guarantee the best possible condition for the animalin an experiment.

Publicly available score sheets for experimental condi-tions found by this literature search should not beadopted without prior assessment and adaptationto the specific situation. If an existing score sheet issimply duplicated, one may miss specific symptoms,or monitor parameters that are not meaningful forthe experimental animal in question. The adaptationof a score sheet should consider such issues as theanimal species, strain, sex, procedures or interventions,application routes, volumes and frequencies, as well astimelines. In the event that no experience with theexperimental model exists, a pilot study may help toreveal relevant symptoms and critical phases.3,7–9

The first step when designing a score sheet is to iden-tify the parameters that will serve to detect deviationsfrom the normal state. Very few publications providecritical information on clinical symptoms for com-monly used experimental models and few help to iden-tify relevant parameters.3,5,6,8,10–12 Some publicationsalso describe how to use these parameters: e.g.Flecknell, Miller, Leach and Baran focus on painassessment and Van der Meer describes the monitoringof rodent pups.6,13–19

After defining the parameters, the scientist mustevaluate all possible factors that can influence theseparameters. For instance, body weight is one of thecommonly used parameters for assessing the health ofan animal in many experimental settings. Comparisonscan be made to the correct reference weight of theanimal in question, taking into account developmentalchanges in body weight, the animal line, sex, age,pregnancy status, etc.5,20 Under certain experimentalconditions, e.g. when animals gain body fluids ortissue mass (e.g. tumor growth), the interpretation ofbody weight may be difficult since body weight lossmay be disguised.4,5,20 Therefore, for laboratoryrodents some publications recommend using bodycondition score instead of measuring body weight,which is a system that has long been used for farmanimals.4,20–23

When addressing the composition and use of scoresheets, many authors emphasized that the assessmentof pain and distress is challenging and can be problem-atic due to anthropocentric assumptions (exam-ples3,8,14). Although perhaps self-evident, someauthors pointed out that the normal behavior of ananimal must be known to be able to detect abnormalconditions.3,5,7,8 Thus, the researcher using a scoresheet must be well trained and be familiar with thenormal behavior of the species of interest; althoughfor very specific set-ups (e.g. assessment of the effectsof acute post-operative pain in rats undergoing ventral

abdominal procedures), Flecknell and Roughan foundthat inexperienced observers can rapidly learn to scoreanimals correctly.24

The second step is to define the method of scoringthe parameters. This may be a simple binary system(yes/no; present/not present), or a numerical scoringthat weighs the symptoms.3,7,8,10,16,22 The binarysystem may be less sensitive for subjective evaluation,and its interpretation in terms of a retrospective ana-lysis is likely to be more difficult.7 Since the numericaland binary systems each have their strengths and weak-nesses some authors provide a table to support theresearchers in choosing the best system in a specificcontext or a combination of both systems.3,7,8,25

For the numerical scoring system one needs to con-sider carefully whether symptoms sum up and so resultin a higher severity (cumulative score), and whether oneneeds to implement measures based on cumulativescore and/or also single scores.5,8 Application ofbadly designed score sheets may mean that an animalachieves a score that does not require action, yet experi-ences severe suffering (see above example on ‘bodyweight’). Empathy and common sense of the assessoris the only way to protect the animal from avoidablesuffering in such cases.8

Whatever scoring system is chosen, it must facilitatethe assessment of severity accurately in order that suf-fering, pain and other harm are prevented.7 The fre-quency and subject matter of the monitoring mustalways be adapted to the situation. In certain phasesof an animal experiment, e.g. in a post-operativeperiod, the frequency of monitoring should be higherto allow for timely responses to sudden changes affect-ing the welfare of the animal.7

The assessment of an animal starts with observingundisturbed behavior and appearance (which maymean initially observing at a distance) and ends withprovoked behavior/reactions and/or manipulation ofthe animal (e.g. for clinical examination or assessingskin turgor, weight or temperature).5,7 This chrono-logical sequence should be reflected on the score sheetto enable a simple and easy-to-follow procedure.Unfortunately, most score sheets found in the literaturedo not follow this simple rule, thus hampering the qual-ity of observations.

As a last step, the interventions that are appliedin the event of animal suffering need to be defined,including humane endpoints. The aim here is to keepthe degree of severity as low as possible. Examples ofinterventions are analgesia, hydration, provision offood and water gel on cage floor, warming or euthan-asia. These interventions can be based on the rankof one of the parameters or, in the case of a numer-ical scoring system, on the sum of some or allparameters.3–5,7,8

Bugnon et al. 415

It is important to assess unscheduled observations andto record such adverse events in open slots on the scoresheet.10 In addition, it can be helpful to add an‘NAD’(nothing abnormal diagnosed) box to save time.3,7

While score sheets are planned according to theexpected symptoms at the beginning of a study, it isimportant constantly to re-evaluate the parametersand the scoring system, and to adapt them when neces-sary. It may also be useful to remove parameters thatnever show a change in a particular experiment, and toadd missing parameters to cover symptoms that ani-mals have actually shown.10

In case an animal has to be euthanized prematurelybecause it has reached a humane endpoint, thescore sheet should give guidance of what actions areto be taken (i.e. the killing method), which samplesare to be retrieved, and how the samples should beprocessed and stored.7 This practice may allow validscientific data to be retrieved from the animal and isthus in line with the 3R principle.

Conclusion

In the ethical framework of the 3Rs it is the responsi-bility of the researchers to use a well-designed scoresheet that is adapted to the specific animal experimentand that enables any changes of normal behavior to bedetected in a timely manner, thus avoiding anyunnecessary suffering.

In general, publications reporting on in vivo studiesdo not provide information on whether or not a scoresheet was used, or how scoring systems were set up.Only a few publications have been found that specific-ally describe aspects of score sheet design in detail.More elaborated descriptions can be found in threelaboratory animal science books.3,6,7 This importantprint information, however, may be missed, since litera-ture searches are now typically made online. Whendesigning a score sheet, knowledge exchanged withother research groups working with the same or similarexperimental models or consultation of publications isadvised.3,26 In addition, consultations with involvedanimal caretakers, researchers and veterinarians are abasic necessity.3,6,7 Score sheets must be adapted to theindividual needs of a specific experiment and model.Step-by-step instructions on score sheet design applic-able in all situations and species should be possible andwould be desirable.

Two very important points must be highlighted: thescore sheet is an essential tool for reporting the degreeof severity of an experiment retrospectively. Also, byproviding an objective score, it is an invaluable aid indeciding whether the predetermined humane endpointhas been reached.

Literature search

The authors of this review used the following data-bases/search machines for the literature search:

. PubMed (http://www.ncbi.nlm.nih.gov/pubmed/)

. Google Scholar (https://scholar.google.ch/)

. Science Direct (http://www.sciencedirect.com/)

. Web of Knowledge (http://apps.webofknowledge.com/UA_GeneralSearch_input.do?product¼UA&search_mode¼GeneralSearch&SID¼W2dSKJQqdBquomDRpWJ&preferencesSaved¼)

. Scopus (https://www.scopus.com/)

. Embase (http://www.embase.com)

The following multiple keywords were chosen and usedin different combinations: pain, scoring, score, moni-toring, distress, animal, rodent, mouse/mice, rat,behavior, stress, evaluation, welfare assessment,humane endpoints, clinical sign(s), symptom(s), bodycondition score.

Acknowledgements

We like to thank Dr Sabine Specht and Prof Dr ThorstenBuch for their constructive feedback on this manuscript andProf Dr Kevan Martin for his proofreading.

Declaration of Conflicting Interests

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

article.

Funding

The author(s) received no financial support for the research,authorship, and/or publication of this article.

References

1. Russell W and Burch R. The principles of humane experi-

mental technique. London: Methuen, 1959.2. Morton DB and Griffiths PH. Guidelines on the recogni-

tion of pain, distress and discomfort in experimental ani-mals and an hypothesis for assessment. Vet Rec 1985; 116:

431–436.3. Hendriksen C, Morton DB and Cussler K. Use of humane

endpoints to minimize suffering. In: Howard B,

Nevalainen T and Perretta G (eds) The COST manual of

animal care and use: refinement, reduction and research.Boca Raton: CRC Press, 2010, pp.338–342.

4. Foltz CJ and Ullmann-Cullere M. Guidelines for assessingthe health and condition of mice. Lab Anim Sci 1999; 28:

28–32.5. Arbeitskreis der Berliner Tierschutzbeauftragten.

Empfehlungen der Berliner Tierschutzbeauftragten zu

Score Sheets und Abbruchkriterien. 2013.6. Wolfensohn S and Lloyd M. Pain, stress and humane end

points. In: Wolfensohn S and Lloyd M (eds) Handbook of


laboratory animal management and welfare. Oxford:Blackwell Publishing Ltd, 2003, pp.65–73.

7. Morton DB and Hau J. Welfare assessment and humane

end points. In: Hau J and Schapiro SJ (eds) Handbook oflaboratory animal science, 3rd ed. Boca Raton: CRCPress, 2010, pp.555–559.

8. Hawkins P, Morton DB, Burman O, et al. A guide to

defining and implementing protocols for the welfareassessment of laboratory animals: eleventh report of theBVAAWF/FRAME/RSPCA/UFAW Joint Working

Group on Refinement. Lab Anim 2011; 45: 1–13.9. Lloyd MH and Wolfensohn SE. Practical use of distress

scoring systems in the application of humane endpoints.

In: International conference on humane endpoints (HEP)in animal experiments for biomedical research (ed CHendricksen, B Steen, K Cussler and DB Morton),

Zeist, The Netherlands 22–25 November 1998.10. Fentener van Vlissingen J, Borrens M, Girod A, Lelovas

P, Morrison F and Torres YS. The reporting of clinicalsigns in laboratory animals: FELASA Working Group

Report. Lab Anim 2015; 49: 267–283.11. Cussler K, Morton DB and Hendriksen C. Humane end-

points in vaccine research and quality control. In:

International conference on humane endpoints (HEP) inanimal experiments for biomedical research (ed CHendricksen, B Steen, K Cussler and DB Morton),

Zeist, The Netherlands 22–25 November 1998.12. Jones HRP, Oates J and Trussell BA. An applied

approach to the assessment of severity. In: Internationalconference on humane endpoints (HEP) in animal experi-

ments for biomedical research (ed C Hendricksen, B Steen,K Cussler and DB Morton), Zeist, The Netherlands 22–25 November 1998.

13. Baran S, Johnson E and Perret-Gentil M. Fundamentalsof pain assessment in rodents. ALN Magazine 2010.

14. Flecknell PA. Refinement of animal use – assessment and

alleviation of pain and distress. Lab Anim 1994; 28:222–231.

15. Leach MC, Klaus K, Miller AL, Scotto di Perrotolo M,

Sotocinal SG and Flecknell PA. The assessment of post-vasectomy pain in mice using behaviour and the mousegrimace scale. PLoS One 2012; 7: e35656.

16. van der Meer M, Rolls A, Baumans V, Olivier B and vanZutphen LFM. Use of score sheets for welfare assessmentof transgenic mice. Lab Anim 2001; 35: 379–389.

17. Langford DJ, Bailey AL, Chanda ML, et al. Coding offacial expressions of pain in the laboratory mouse. NatMethods 2010; 7: 447–449.

18. Miller AL and Leach MC. The mouse grimace scale: a

clinically useful tool? PLoS One 2015; 10: e0136000.19. Sotocinal S, Sorge R, Zaloum A, et al. The rat grimace

scale: a partially automated method for quantifying pain

in the laboratory rat via facial expressions. Mol Pain2011; 7: 55.

20. Ullmann-Cullere MH and Foltz CJ. Body condition scor-

ing: a rapid and accurate method for assessing healthstatus in mice. Lab Anim Sci 1999; 49: 319–323.

21. Hickman DL and Swan M. Use of a body condition score

technique to assess health status in a rat model of poly-cystic kidney disease. J Am Assoc Lab Anim Sci 2010; 49:155–159.

22. Office of Regulatory Affairs. IACUC Guideline: Humane

intervention and endpoints for laboratory animal species.Philadelphia: Office of Regulatory Affairs, University ofPennsylvania, 2011.

23. Stober M. Kennzeichen, anamnese, grundregeln, allge-meine untersuchung. In: Dirksen G, Grunder HD andStober M (eds) Die klinische Untersuchung des Rindes,

3rd ed. Singhofen: Paul Parey Verlag, 1990, p.126.24. Flecknell PA and Roughan JV. Training in behaviour-

based post-operative pain scoring in rats – an evaluationbased on improved recognition of analgesic requirements.

Appl Anim Behav Sci 2006; 96: 327–342.25. Morton DB. Humane endpoints in animal experimenta-

tion for biomedical research: ethical, legal and practical

aspects. In: International conference on humane endpoints(HEP) in animal experiments for biomedical research (edC Hendricksen, B Steen, K Cussler and DB Morton),

Zeist, The Netherlands 22–25 November 1998.26. Jennings M (ed). Guiding principles on good practice for

animal welfare and ethical review bodies. A report by the

RSPCA Research Animals Department and LASAEducation, Training and Ethics Section. 3rd ed. RSPCAand LASA, 2015.

Bugnon et al. 417

ab oratory

l i m i t e d

lan imals

Special Article

Analgesia in clinically relevant rodentmodels of sepsis

Victor Jeger1,2, Till Hauffe2, Flora Nicholls-Vuille3,Dominique Bettex1 and Alain Rudiger1

AbstractPostoperative analgesia in rodent sepsis models has been considerably neglected in the past. However,intentions to model clinical practice, increasing awareness of animal ethics, efforts to apply the 3Rs (replace-ment, reduction, refinement), and stricter legislation argue for a change in this respect. In this review, wedescribe different concepts of analgesia in rodent models of sepsis focusing on opioid agonists as well asnon-opioid analgesics. Advantages and pitfalls in study design and side-effects are discussed. Score sheetsshould be used to adapt analgesia or to terminate experiments using humane endpoints. Further research isneeded to differentiate behavioral changes caused by sepsis and pain or as a consequence of analgesia.Information on the efficacy of analgesia in sepsis models is scarce. Hence, studies are needed to identifythe best ways to reduce suffering of research animals and thereby optimize the clinically relevant rodentmodels of sepsis.

Keywordsanimal model, sepsis, rat, analgesia, nalbuphine, buprenorphine, 3Rs, score sheet

Animal model

Over recent decades, animal models of sepsis have beenrefined continuously. Among the wide range of existingmodels, the majority are focused on abdominal sepsis inrodents, as this has been shown to be both valid andreproducible.1,2 Advantages and pitfalls of rodentsepsis models have been discussed extensively.1–4 Onemain issue is the difficult translation from rodent tohuman studies, which partly results from the complex-ity of human sepsis and the interplay of an unknownnumber of confounders. The traditional model of fecalperitonitis is cecal ligation and puncture (CLP), whichcauses subacute abdominal sepsis. Mice are the pre-dominant species used in sepsis research.5 However,the use of mouse models has been criticized mainlydue to the pathobiological differences between mouseand human sepsis.6 By contrast to humans and rats,sepsis in mice lowers body temperature and oxygenconsumption.7 The debate about the usefulness ofmurine sepsis models is still ongoing. In this currentreview the focus is on rat sepsis models as we believethat they resemble human patients more closely.Characteristics of a clinically relevant rat model ofsepsis are given in Table 1 and Figure 1.

After induction of polymicrobial sepsis (by CLP orintraperitoneal injection of fecal slurry), animals areusually awake during the experiment. In general, anes-thesia (e.g. isoflurane) is only applied during instrumen-tation and induction of sepsis.1 These experiments arehighest in the severity grade (in Switzerland class III,European severity category ‘severe’) and require appro-priate animal ethics permission, which may differaccording to the legislation of the country where theexperiments are performed. In Switzerland, the regula-tion of ‘animal’s dignity’ in legislation influences thedesign of animal models even more strictly, which hasbeen discussed elsewhere.8

1Institute for Anesthesiology, University and University HospitalZurich, Switzerland2Department of Medicine, University and University HospitalZurich, Switzerland3Research Unit, Department of Surgery, University and UniversityHospital Zurich, Zurich, Switzerland

Corresponding author:Alain Rudiger, University Hospital Zurich, Raemistrasse 100, CH-8091, Zurich, Switzerland.Email: [email protected]

Laboratory Animals

2016, Vol. 50(6) 418–426



sagepub.co.uk/


DOI: 10.1177/0023677216675009

la.sagepub.com

In the early days of sepsis research, the aim was toobserve the effect of the infection itself, therefore nei-ther fluids nor antibiotics were administered. Duringthe last decade, there has been a paradigm shift tomodel clinical practice of human sepsis in animalsinstead of reproducing animal sepsis.1 Therefore theuse of fluid resuscitation during septic shock is manda-tory and the use of antibiotics is widely applied.1,2,9

By contrast to the clinical reality of intensive careunits, the application of analgesia or even analgoseda-tion has been neglected for a long time in animalresearch. In comprehensive reviews analgesia is barelymentioned.1,2 In 2012, only a few research groups(15%, 7 out of 45 publications) used analgesia,although pain and distress were likely in all animalsundergoing experimentally-induced sepsis.10 Marshallet al. have expressed concern about the interactionbetween analgesia and the pathobiology of sepsis, espe-cially with opioid-induced immunosuppression.3

Assessment of welfare and pain

As indicated above, animal welfare in sepsis studies hasnot been a major topic of concern in sepsis research.10

However, pain anddistress are important confounders inanimal models and may influence behavior and the ani-mal’s immune system.10 The induction of peritonitis maycause pain and will influence the behavior of ani-mals.11,12 To improve animal welfare, researchers haveto identify animal distress and should therefore know thebehavior of healthy animals beforehand. Behaviorchanges and distress can be quantified using clinicalscores.13–16 Scoring may then be used to either refinethe animal model by using adequate analgesia or toimplement termination criteria (also called humane end-points) to the sepsis model. Huet et al. have describedand validated a clinical scoring system for a murinesepsismodel.13 It was based on behavioral signs (activity,posture, reaction to stimulation), fur aspect, respirationand chest sounds as well as body weight. The authorshave shown that severe clinical scores corresponded tohigh cytokine and reactive oxygen species levels.13

In order to improve animal welfare, termination cri-teria should be designed such that not only moribundanimals close to death are recognized.17 Early clinicalsigns with a high sensitivity and specificity to predictthe animals’ demise should be defined so that experi-ments can be terminated early. Good predictors willalso allow comparisons between potential survivorsand non-survivors, thereby enabling comparisonsbetween adaptive mechanisms of survival and maladap-tive mechanisms of death.

Even more subtle indicators may be used to assesspain in animals, as applied in the rat grimace scale vali-dated for acute pain.18 There is no report of its use in

Table 1. Suggestions for a clinically relevant rat model offecal peritonitis.

Interventions Advantages

Use of live bacteriainstead of endotoxin

� Clinically relevant type ofsepsis

Injection of fecal slurryto induce peritonitis

� No laparotomy (less pain,less surgical trauma) ascompared with CLP

� Identical insult in a batchof animals as comparedwith CLP

Intravenous fluidresuscitation withcrystalloids

� Sufficient preload in caseof distributive shock

� Adherence to clinicalpractice

Observation time atleast 48 h

� Clinical relevant investi-gation of various sepsisphases

Expected mortalityrange between 10and 50%

� Clinically relevantoutcome

Continuous administra-tion of intravenousopioids

� Opioid guarantees effect-ive analgesia

� Infusion guarantees con-stant drug plasma levels

� High clinical relevance� No disturbance of sick

animals by repeatedinjections

CLP: cecal ligation and puncture.

Figure 1. Representative picture of a rat equipped with acentral venous line on a swivel–tether system.

Jeger et al. 419

acute/subacute sepsis models. Kawano et al. haverecently applied the rat grimace scale using video stills,and have used a blinded evaluator to explore the role oflipopolysaccharide (LPS) on an incisional pain modelretrospectively.14However, sepsismodels shouldbe eval-uated in real-time and not retrospectively in order toidentify relevant termination criteria. Furthermore, therat grimace scale has not yet been validated to differen-tiate sepsis-induced distress from acute pain. Our experi-ence in a rodent model of fecal peritonitis reveals thatorbital tightening and ear changes are obvious.However,identification and quantification of nose/cheek flatteningand whisker changes are difficult if not impossible evenfor experienced investigators. Other clinical signs such asinactivity, piloerection, eye discharge, sunken flanks orback aching have also been used,19 but it is unclear howtheymight accurately predict severity of illness in generalor abdominal pain in particular.

The implementation of physiological parameterssuch as changes in water intake, body weight, body

temperature or heart rate may be more reliable, andtherefore better discriminators for the outcome.12,17,20

We have described that clinical scores during earlysepsis might not prognosticate outcome, whereas echo-cardiography-derived stroke–volume and heart rate canpredict death with a good sensitivity and specificity asearly as 6 h after the septic insult,21 and which cantherefore lead to shorter observation times and robustand early termination criteria. The use of implantableand wireless telemetry devices may further help todetermine quantitative cut-off values, first describedas sickness behavior by Bauhofer et al.12 Figure 2shows an example of telemetry recordings in twoseptic Wistar rats. In a recent murine sepsis model, pre-defined deterioration thresholds of heart rate and bodytemperature assessed by telemetry, independent of thedelay since sepsis induction, were superior to datasampling at predefined time points. This led to morehomogenous study groups.22 The authors stated thatit might be possible to predict the outcome by defining

Figure 2. Example of a rat ECG recording based on the implanted telemeter electrode. ECG: electrocardiogram; mV:amplitude in mV; �C: body temperature in degree Celsius; BPM: beats per minute.


adequate thresholds based on telemetry data.22

Although not mentioned by them, these thresholdsmay be used to define termination criteria. However,pilot studies are needed for each species and sepsismodel to define thresholds based on vital signs.

Analgesia should reduce stress levels and therebyincrease reliability of the model. One important diffi-culty is that sepsis and analgosedation may cause simi-lar behavior changes in rats.23 In particular inactivityand apathy may be caused by the disease as well as by(over-)sedation caused by opioid analgesia. Reductionof vertical movement of rats – as a marker of behaviorchange – in response to pain was similarly observed inhealthy animals receiving high doses of morphine.23

Therefore behavior changes induced by analgesia haveto be distinguished from sepsis-induced distress andpain in further studies.

Refinement opportunities – analgesicsin sepsis models

Opioid agonists/antagonists

Nalbuphine is a well-known opioid analgesic and actsas a kappa-receptor agonist and mu-receptor

antagonist. In humans, it is commonly used in pediatricanesthesia due to its ceiling effect and good safety pro-file.24 Nalbuphine is not regulated by narcotics law(compared with buprenorphine, fentanyl or morphine).Therefore ordering and storing of this drug do notrequire any special restriction, which facilitates its use.Its application in mice as a postoperative analgesicagainst visceral and complex inflammatory pain hasbeen reviewed recently.25 Nalbuphine has been used asa continuous analgesic during CLP-induced sepsis inrats, but no details have been revealed regarding theefficacy of the drug in this particular condition.26 Thedrug has also been evaluated in pain research using dif-ferent rat strains.27,28 It has been shown that the effect ofnalbuphine is not only strain-specific, but also gender-specific.29,30 Contrary to humans, male rats respondbetter to nalbuphine than females.31 These strain andgender differences underline the need for pilot experi-ments to determine the exact amount of nalbuphineneeded in postoperative analgesia. An overview ofdoses used in the literature is provided in Table 2.In rats, nalbuphine seems to have less analgesic effectcompared with buprenorphine or to pure mu-agonistssuch as fentanyl or morphine,28 but a better safetyprofile as described below. When injected

Table 2. Publications on nalbuphine in pain studies and postoperative settings.

Drug Route Time Dose Keywords Reference

Nalbuphine ID? Single 0.5–1mg/kg Pain, gender-difference Khasar SG, et al. Neurosci Lett2003; 345: 165–168

Nalbuphine SC Single 0.1–30mg/kg Pain, gender-difference Craft RM, et al. Drug AlcoholDepend 2001; 63: 215–228

Nalbuphine IM Single 0.25–250mcmol/kg Pain Chu KS, et al. Anesth Analg 2003;97: 806–809

Nalbuphine IP Single 1–2mg/kg Pain, reversal of operativeanalgosedation, postoperativeanalgesia

Hu C, et al. Lab Anim 1992; 26:15–22

Nalbuphine IP Single 0.01–30mg/kg Gastric emptying,gastrointestinal passage

Asai T, et al. Br J Anaesth 1998;80: 814–819

Nalbuphine IP Single 0.01–100mg/kg Pain, strain-difference Morgan D, et al. J Pharmacol ExpTher 1999; 289: 965–975

Nalbuphine IP Single 0.01–100mg/kg Pain, strain- andgender-difference

Terner JM, et al. Pain 2003; 106:381–391

Nalbuphine IP Bolus andcontinuous

Bolus: 40mcg/kg, infusionrates 10–200mcg/kg/min

Respiratory depression DiFazio CA, et al. Anesth Analg1981; 60: 629–633

Nalbuphine IV Single 1mg/kg Pain, strain-difference Avsaroglu H, et. al. Lab Anim2007; 41: 337–344

Nalbuphine IV Continuous 0.2mg/kg/h Sepsis, CLP Kimmoun A, et al. Crit Care Med2015; 43: e332–e340

CLP: cecal ligation and puncture, ID: intradermal, SC: subcutaneous, IP: intraperitoneal, IV: intravenous, MAC: minimum alveolarconcentration.

Jeger et al. 421

intramuscularly it has an effect of around 2h, and a 100-fold dose increase might only provide a 2.7-fold increasein duration of action.32 In most postoperative pain stu-dies, nalbuphine is administered as a single injection.29,33

This approach may be adequate if only a limited obser-vation time is available.

There are two major advantages of nalbuphine com-pared with pure mu-opioids: it only weakly inhibitsgastric emptying and gastrointestinal passage,34 anddoes not cause marked respiratory depression.35,36

Nalbuphine also has an effect on hemodynamics.37

These side-effects may interfere with sepsis-inducedhypotension or tachycardia, which has to be takeninto account in an experimental design.

Buprenorphine is a similar partial agonist/antagonistwith a stronger analgesic effect compared withnalbuphine.28 It has been evaluated in postoperativeanalgesia in rats.38–41 It is usually administered

intraperitoneally or subcutaneously, but it can alsobe consumed voluntarily by rats (mixed in a sweetNutella� hazelnut spread) as demonstrated byGoldkuhl et al.39 Repeated subcutaneous injections ofbuprenorphine have been described in experimentalsepsis to achieve adequate postoperative analgesia inCL57BL/6 mice.42 An overview of doses used in theliterature is provided in Table 3. It has been comparedwith tramadol or placebo in a murine sepsis model(female ICR mice) to assess the immune-modulatingeffect.43 High-dose tramadol was associated with ahigher mortality in sepsis compared with the placeboor buprenorphine.43 In septic CL57BL/6 mice, bupre-norphine had an adverse impact on mortality in malesbut not in females, which could be prevented by dosereduction.44 Buprenorphine had only a few effects oncell counts and cytokines in both genders.44 In a similarsetting, it had little effect on the outcome of septic

Table 3. Publications on buprenorphine in pain studies and postoperative settings.

Drug Route Time Dose Keywords Reference

Buprenorphine SC Repeated 0.6mg/kg (all 8 h) Sepsis, mice,postoperative

Albuszies G, et al. CritCare Med 2005; 33:2332–2338

Buprenorphine SC Single 0.1mg/kg Sepsis, mice,immunosuppression

Hugunin KM, et al. Shock2010; 34:250–260

Buprenorphine SC Repeated 0.1 or 0.05mg/kg Sepsis, CLP, mice,immunosuppression,gender-difference

Cotroneo TM, et al. J AmAssoc Lab Anim Sci.2012; 51: 357–365.

Buprenorphine SC orvoluntaryingestion

Repeated 0.05mg/kg SC or0.4mg/kg in 2 g/kgNutella� VI

Postoperative, pain,corticosterone levels

Goldkuhl R, et al. Lab Anim2010; 44: 337–343

Buprenorphine SC Repeated 0.05mg/kg (all 12 h) Pain, postoperative McKeon GP, et al. J AmAssoc Lab Anim Sci2011; 50: 192–197

Buprenorphine SC Repeated 0.05mg/kg (all 8 hor 12 h) in additionwith 0.2mg/kgmeloxicam

Postoperative, pain,dosing interval

Schaap MW, et al.Lab Anim 2012; 46:287–292

Buprenorphine IP Single 0.001–1.0mg/kg Pain, strain-difference Morgan D, et al.J Pharmacol Exp Ther1999; 289: 965–975

Buprenorphine IP Single 0.01–10mg/kg Pain, strain- andgender-difference

Terner JM, et al. Pain2003; 106: 381–391

Buprenorphine IV single(3 – 5min)

10, 30 and100mcg/kg

Postoperative,isoflurane–MAC

Criado AB, et al. Lab Anim2000; 34: 252–259

Buprenorphine IV single 0.05mg/kg Pain, strain-difference Avsaroglu H, et. al.Lab Anim 2007; 41:337–344

CLP: cecal ligation and puncture, SC: subcutaneous, IP: intraperitoneal, IV: intravenous, VI: voluntary ingestion, MAC: minimum alveolarconcentration.


female mice (BALB/c mice), however the stage theywere at in their estrous cycle influenced their immuneresponse to CLP.45 There are conflicting reports of anti-inflammatory activity or exacerbated inflammation indifferent arthritis models in combination with bupre-norphine analgesia.46 The immune-modulating effectof opioids is still a topic of controversy, concerningwhich there is no clear evidence, and data in humansare contradictory.47

Butorphanol may be an alternative partial agonist/antagonist and has been used in rat postoperative painmodels.48,49 However, there are no data so far regard-ing its use in sepsis.

Pure mu-agonists

Morphine and fentanyl are potent mu-agonists, whichare used in adult intensive care every day. In largeranimals, where a safe airway and controlled ventilationcan be achieved, fentanyl may be the favorite choice.50

In rodents, where long-term (>6 h) controlled ventila-tion is difficult, if not impossible, continuous fentanyladministration has to be evaluated carefully due torespiratory depression.51,52 Other strategies maybe the use of epidural analgesia53 which again may beeasier to apply in larger animals than in rodents. Ofnote, the immune-modulating effects of morphineaffect mechanistic studies in sepsis models.54

Non-opioid analgesia

Non-steroidal anti-inflammatory drugs (NSAIDs) havebeen used in postoperative analgesia in rats.55,56 Theuse of NSAIDs in sepsis has been discussed, and atrend might exist towards reduced acute lung injury.57

In clinical practice with septic patients, the use ofNSAIDs is avoided due to the risk of renal impairmentand bleeding. Furthermore their anti-inflammatoryeffect, mainly on prostaglandin synthesis,25 may inter-fere with the sepsis model itself and could thereby influ-ence the pathobiology of the disease.

Local anesthetics such as lidocaine and ropivacainemight be used as further adjuncts to reduce intra- andpostoperative pain.58–60 We use local anesthesia in add-ition to isoflurane for surgical instrumentation. So farno data exist on whether a prolonged or repeated appli-cation of local anesthetics may improve analgesia insepsis models.

Whatever the choice of drug or the modality ofadministration, almost all studies use a fixed regimenof analgesia. However, Goldkuhl et al. have demon-strated that in a postoperative pain model the plasmacorticosterone levels were significantly lower when ratscould choose their appropriate buprenorphine dose(mixed with Nutella�) compared with a fixed regimenof subcutaneous buprenorphine administration.39 Thisraises the issue of adaptive analgesia. However, oralself-administration cannot be applied in sepsis modelsas the animals tend to reduce their food intake.61

Frequent screening of sepsis-induced pain and discom-fort using standardized score sheets may therefore beuseful to enable individual analgesic dose adjustments.An example of a score sheet used in our sepsis model isgiven in Table 4. However, adaptive analgesia may leadto inconsistent administration of analgesics, whichmay introduce additional confounders (respiratoryand cardiovascular depression, immunosuppression)to the model.

As discussed above, assessment of pain and distressmay be challenging in sepsis models. To measure the

Table 4. Score sheet to assess sepsis-induced behavior changes.

Rat grimace scale Clinical scoring*

Orbital tightening 0 1 2 Reduced activity 0 1 2

Nose/cheek flattening 0 1 2 Sunken flanks 0 1 2

Back arching 0 1 2 Ear changes 0 1 2

Whisker changes 0 1 2 Piloerection 0 1 2

Bloated abdomen 0 1 2

Chromodacryorrhea

(= eye discharge)

0 1 2

0: not present, 1: moderate, 2: severe. The score sheet was adapted from Ref 18 and www.ahwla.org.uk. *Six or more points in the clinical scoring lead to termination of the experiment byeuthanizing the animal with one milliliter of phenobarbital intravenously.

Jeger et al. 423

effectiveness of analgesia may be even more difficult asanalgesics themselves interfere with normal animalbehavior.23 In a postoperative laparotomy rat model,efficacy of analgesia was assessed by changes in foodintake, body weight, and by measurement of painthreshold using a paw-flick latency test.62 However,sepsis itself may have an effect on these parameters.61

Furthermore LPS increases hyperalgesia in incisionalpain models14 but hyperalgesia in polymicrobial sepsismodels has not been investigated so far. Assessment ofpain thresholds together with measurements of plasmalevels may improve our understanding of pain therapyin sepsis models. Plasma drug levels from human stu-dies could serve as a reference for animals.46 To ourknowledge, there are no reports on plasma concentra-tions of analgesics in sepsis models. In addition, sepsis-induced organ dysfunctions might influence drugmetabolism and could lead to accumulation of activecompounds and metabolites.

Further refinement opportunities other than anal-gesia could be achieved using less invasive sepsismodels and/or different housing conditions. For exam-ple, peritoneal injection of fecal slurry does not requirepainful laparotomy, which is necessary in CLP models.Furthermore, animals have to be singly-housed if aswivel–tether system is used. Lilley and colleaguesfrom an expert working group on 3Rs in sepsis modelshave mentioned the idea of co-housing healthy, non-instrumented animals with instrumented animals.63

This may be an advantage in controlling body tempera-ture in these social animals.63 However, co-housingcould also induce stress to the healthy animals due tothe presence of their sick cage mates, as it is well knownthat housing and experimentation within the same roommay cause stress to the untreated animals.64

Recommendations and further 3R research

Recently, recommendations on applying the 3Rs insepsis research have been published, providing an excel-lent overview of potential refinement of sepsis models.63

The authors conclude that ‘applying the ‘‘R’’ inRefinement of animal studies can be a highly effectiveway to reduce suffering and improve scientific qual-ity’.63 Osuchowski et al. have mentioned in their editor-ial that the 3Rs should also be interpreted as ‘relevance,robustness and reproducibility’.65 Humane endpointswere proposed more than a decade ago in sepsisresearch,12,16 however we still lack preclinical studieswith the primary aim of assessing and improving anal-gesia and termination criteria in sepsis. More researchshould therefore focus on the translational approach ofsepsis models and the ‘refinement’ aspect. More studiesare also needed to investigate the pharmacokinetics andpharmacodynamics of different types of analgesia,

behavior and physiological alterations in sick animals,and the implementation of humane endpoints to reduceunnecessary suffering.


The author(s) declared no potential conflicts of interest withrespect to the research, authorship, and/or publication of thisarticle.

Funding

The author(s) received no financial support for the research,

authorship, and/or publication of this article..

References

1. Deitch EA. Rodent models of intra-abdominal infection.

Shock 2005; 24(Suppl. 1): 19–23.2. Dyson A and Singer M. Animal models of sepsis: why

does preclinical efficacy fail to translate to the clinicalsetting? Crit Care Med 2009; 37(1 Suppl): S30–S37.

3. Marshall JC, Deitch E,Moldawer LL, Opal S, Redl H and

van der Poll T. Preclinical models of shock and sepsis:what can they tell us? Shock 2005; 24(Suppl. 1): 1–6.

4. Rittirsch D, Hoesel LM and Ward PA. The disconnectbetween animal models of sepsis and human sepsis.

J Leukoc Biol 2007; 81: 137–143.5. Lewis AJ, Seymour CW and Rosengart MR. Current

murine models of sepsis. Surg Infect (Larchmt) 2016;

17: 385–393.6. Efron PA, Mohr AM, Moore FA and Moldawer LL. The

future of murine sepsis and trauma research models.

J Leukoc Biol 2015; 98: 945–952.7. Zolfaghari PS, Pinto BB, Dyson A and Singer M. The

metabolic phenotype of rodent sepsis: cause for concern?

Intensive Care Med Exp 2013; 1: 25.8. Schindler S. The animal’s dignity in Swiss Animal

Welfare Legislation – challenges and opportunities. EurJ Pharm Biopharm 2013; 84: 251–254.

9. Dyson A, Rudiger A and Singer M. Temporal changes in

tissue cardiorespiratory function during faecal peritonitis.Intensive Care Med 2011; 37: 1192–1200.

10. Bara M and Joffe AR. The ethical dimension in publishedanimal research in critical care: the public face of science.

Crit Care 2014; 18: R15.11. Ozcan PE, Senturk E, Orhun G, et al. Effects of intra-

venous immunoglobulin therapy on behavior deficits and

functions in sepsis model. Ann Intensive Care 2015; 5: 62.12. Bauhofer A, Witte K, Celik I, Pummer S, Lemmer B and

Lorenz W. Sickness behaviour, an animal equivalent to

human quality of life, is improved in septic rats by G-

CSF and antibiotic prophylaxis. Langenbecks Arch Surg2001; 386: 132–140.

13. Huet O, Ramsey D, Miljavec S, et al. Ensuring animalwelfare while meeting scientific aims using a murine

pneumonia model of septic shock. Shock 2013; 39:

488–494.14. Kawano T, Eguchi S, Iwata H, et al. Effects and under-

lying mechanisms of endotoxemia on post-incisional pain

in rats. Life Sci 2016; 148: 145–153.


15. Morton DB. A systematic approach for establishinghumane endpoints. ILAR J 2000; 41: 80–86.

16. Nemzek JA, Xiao HY, Minard AE, Bolgos GL and

Remick DG. Humane endpoints in shock research.Shock 2004; 21: 17–25.

17. Franco NH, Correia-Neves M and Olsson IA. How‘humane’ is your endpoint? Refining the science-driven

approach for termination of animal studies of chronicinfection. PLoS Pathog 2012; 8: e1002399.

18. Sotocinal SG, Sorge RE, Zaloum A, et al. The rat grim-

ace scale: a partially automated method for quantifyingpain in the laboratory rat via facial expressions. Mol Pain2011; 7: 55.

19. Protti A, Carre J, Frost MT, et al. Succinate recoversmitochondrial oxygen consumption in septic rat skeletalmuscle. Crit Care Med 2007; 35: 2150–2155.

20. Dellavalle B, Kirchhoff J, Maretty L, Castberg FC andKurtzhals JA. Implementation of minimally invasive andobjective humane endpoints in the study of murinePlasmodium infections. Parasitology 2014; 1–7.

21. Rudiger A, Dyson A, Felsmann K, et al. Early functionaland transcriptomic changes in the myocardium predictoutcome in a long-term rat model of sepsis. Clin Sci

(Lond) 2013; 124: 391–401.22. Lewis AJ, Yuan D, Zhang X, Angus DC, Rosengart MR

and Seymour CW. Use of biotelemetry to define physiol-

ogy-based deterioration thresholds in a murine cecal liga-tion and puncture model of sepsis. Crit Care Med 2016;44: e420–e431.

23. Matson DJ, Broom DC, Carson SR, Baldassari J, Kehne

J and Cortright DN. Inflammation-induced reduction ofspontaneous activity by adjuvant: a novel model to studythe effect of analgesics in rats. J Pharmacol Exp Ther

2007; 320: 194–201.24. Shin D, Kim S, Kim CS and Kim HS. Postoperative pain

management using intravenous patient-controlled anal-

gesia for pediatric patients. J Craniofac Surg 2001; 12:129–133.

25. Narver HL. Nalbuphine, a non-controlled opioid anal-

gesic, and its potential use in research mice. Lab Anim(NY) 2015; 44: 106–110.

26. Kimmoun A, Louis H, Al Kattani N, et al. Beta1-adrenergic inhibition improves cardiac and vascular func-

tion in experimental septic shock. Crit Care Med 2015;43: e332–e340.

27. Avsaroglu H, van der Sar AS, van Lith HA, van Zutphen

LF and Hellebrekers LJ. Differences in response toanaesthetics and analgesics between inbred rat strains.Lab Anim 2007; 41: 337–344.

28. Morgan D, Cook CD and Picker MJ. Sensitivity tothe discriminative stimulus and antinociceptive effectsof mu opioids: role of strain of rat, stimulus intensity,and intrinsic efficacy at the mu opioid receptor.

J Pharmacol Exp Ther 1999; 289: 965–975.29. Craft RM and Bernal SA. Sex differences in opioid anti-

nociception: kappa and ‘mixed action’ agonists. Drug

Alcohol Depend 2001; 63: 215–228.30. Terner JM, Lomas LM, Smith ES, Barrett AC and Picker

MJ. Pharmacogenetic analysis of sex differences in opioid

antinociception in rats. Pain 2003; 106: 381–391.

31. Khasar SG, Gear RW and Levine JD. Absence of nalbu-phine anti-analgesia in the rat. Neurosci Lett 2003; 345:165–168.

32. Chu KS, Wang JJ, Hu OY, Ho ST and Chen YW. Theantinociceptive effect of nalbuphine and its long-actingesters in rats. Anesth Analg 2003; 97: 806–809.

33. Hu C, Flecknell PA and Liles JH. Fentanyl and medeto-

midine anaesthesia in the rat and its reversal using atipa-mazole and either nalbuphine or butorphanol. Lab Anim1992; 26: 15–22.

34. Asai T, Mapleson WW and Power I. Effects of nalbu-phine, pentazocine and U50488H on gastric emptyingand gastrointestinal transit in the rat. Br J Anaesth

1998; 80: 814–819.35. DiFazio CA, Moscicki JC and Magruder MR. Anesthetic

potency of nalbuphine and interaction with morphine in

rats. Anesth Analg 1981; 60: 629–633.36. Flecknell PA, Liles JH and Wootton R. Reversal of fen-

tanyl/fluanisone neuroleptanalgesia in the rabbit usingmixed agonist/antagonist opioids. Lab Anim 1989; 23:

147–155.37. Muldoon SM, McKenzie JE and Collins FJ. Pressor effect

of nalbuphine in hemorrhagic shock is dependent on the

sympathoadrenal system. Circ Shock 1988; 26: 89–98.38. Criado AB, Gomez de Segura IA, Tendillo FJ and

Marsico F. Reduction of isoflurane MAC with

buprenorphine and morphine in rats. Lab Anim 2000;34: 252–259.

39. Goldkuhl R, Jacobsen KR, Kalliokoski O, Hau J andAbelson KS. Plasma concentrations of corticosterone

and buprenorphine in rats subjected to jugular vein cath-eterization. Lab Anim 2010; 44: 337–343.

40. Schaap MW, Uilenreef JJ, Mitsogiannis MD, van ‘t

Klooster JG, Arndt SS and Hellebrekers LJ. Optimizingthe dosing interval of buprenorphine in a multimodalpostoperative analgesic strategy in the rat: minimizing

side-effects without affecting weight gain and foodintake. Lab Anim 2012; 46: 287–292.

41. McKeon GP, Pacharinsak C, Long CT, et al. Analgesic

effects of tramadol, tramadol-gabapentin, and buprenor-phine in an incisional model of pain in rats (Rattus nor-vegicus). J Am Assoc Lab Anim Sci 2011; 50: 192–197.

42. Albuszies G, Radermacher P, Vogt J, et al. Effect of

increased cardiac output on hepatic and intestinal micro-circulatory blood flow, oxygenation, and metabolism inhyperdynamic murine septic shock. Crit Care Med 2005;

33: 2332–2338.43. Hugunin KM, Fry C, Shuster K and Nemzek JA. Effects

of tramadol and buprenorphine on select immunologic

factors in a cecal ligation and puncture model. Shock2010; 34: 250–260.

44. Cotroneo TM, Hugunin KM, Shuster KA, Hwang HJ,Kakaraparthi BN and Nemzek-Hamlin JA. Effects of

buprenorphine on a cecal ligation and puncture modelin C57BL/6 mice. J Am Assoc Lab Anim Sci 2012; 51:357–365.

45. Kennedy LH, Hwang H, Wolfe AM, Hauptman J andNemzek-Hamlin JA. Effects of buprenorphine andestrous cycle in a murine model of cecal ligation and

puncture. Comp Med 2014; 64: 270–282.

Jeger et al. 425

46. Guarnieri M, Brayton C, DeTolla L, Forbes-McBean N,Sarabia-Estrada R and Zadnik P. Safety and efficacy ofbuprenorphine for analgesia in laboratory mice and rats.

Lab Anim (NY) 2012; 41: 337–343.47. Rittner HL, Roewer N and Brack A. The clinical (ir)rele-

vance of opioid-induced immune suppression. Curr OpinAnaesthesiol 2010; 23: 588–592.

48. Abreu M, Aguado D, Benito J and Gomez de Segura IA.Reduction of the sevoflurane minimum alveolar concen-tration induced by methadone, tramadol, butorphanol

and morphine in rats. Lab Anim 2012; 46: 200–206.49. Liles JH and Flecknell PA. The effects of buprenorphine,

nalbuphine and butorphanol alone or following halo-

thane anaesthesia on food and water consumption andlocomotor movement in rats. Lab Anim 1992; 26:180–189.

50. Correa TD, Jeger V, Pereira AJ, Takala J, Djafarzadeh Sand Jakob SM. Angiotensin II in septic shock: effects ontissue perfusion, organ function, and mitochondrial res-piration in a porcine model of fecal peritonitis. Crit Care

Med 2014; 42: e550–e559.51. Baechtold F, Cavadas C, Gasser D, et al. Cardiovascular

effects of fentanyl in conscious rats. Pflugers Arch 2001;

443: 155–162.52. Nardi GM, Bet AC, Sordi R, Fernandes D and Assreuy

J. Opioid analgesics in experimental sepsis: effects on

physiological, biochemical, and haemodynamic param-eters. Fundam Clin Pharmacol 2013; 27: 347–353.

53. Daudel F, Ertmer C, Stubbe HD, et al. Hemodynamiceffects of thoracic epidural analgesia in ovine hyperdy-

namic endotoxemia. Reg Anesth Pain Med 2007; 32:311–316.

54. Breslow JM, Monroy MA, Daly JM, et al. Morphine, but

not trauma, sensitizes to systemic Acinetobacter bauman-nii infection. J Neuroimmune Pharmacol 2011; 6: 551–565.

55. Araki Y, Kaibori M, Matsumura S, Kwon AH and Ito S.

Novel strategy for the control of postoperative pain:long-lasting effect of an implanted analgesic hydrogel in

a rat model of postoperative pain. Anesth Analg 2012;114: 1338–1345.

56. Kawano T, Takahashi T, Iwata H, et al. Effects of keto-

profen for prevention of postoperative cognitive dysfunc-tion in aged rats. J Anesth 2014; 28: 932–936.

57. Eisen DP. Manifold beneficial effects of acetyl salicylicacid and nonsteroidal anti-inflammatory drugs on sepsis.

Intensive Care Med 2012; 38: 1249–1257.58. Castel D, Willentz E, Doron O, Brenner O and Meilin S.

Characterization of a porcine model of post-operative

pain. Eur J Pain 2014; 18: 496–505.59. Charlet A, Rodeau JL and Poisbeau P. Radiotelemetric

and symptomatic evaluation of pain in the rat after lapar-

otomy: long-term benefits of perioperative ropivacainecare. J Pain 2011; 12: 246–256.

60. Lu L, Zhang W, Wu X, et al. A novel ropivacaine-loaded

in situ forming implant prolongs the effect of local anal-gesia in rats. Arch Med Sci 2013; 9: 614–621.

61. Bauhofer A, Witte K, Lemmer B, Middeke M, Lorenz Wand Celik I. Quality of life in animals as a new outcome

for surgical research: G-CSF as a quality of life improv-ing factor. Eur Surg Res 2002; 34: 22–29.

62. Shavit Y, Fish G, Wolf G, et al. The effects of periopera-

tive pain management techniques on food consumptionand body weight after laparotomy in rats. Anesth Analg2005; 101: 1112–1116.

63. Lilley E, Armstrong R, Clark N, et al. Refinement ofanimal models of sepsis and septic shock. Shock 2015;43: 304–316.

64. Pitman DL, Ottenweller JE and Natelson BH. Plasma

corticosterone levels during repeated presentation oftwo intensities of restraint stress: chronic stress andhabituation. Physiol Behav 1988; 43: 47–55.

65. Osuchowski M, Redl H and Radermacher P.Implementing refinements in preclinical sepsis modeling:walking a fine line between what ethics condones and the

gaping investigative holes call for. Shock 2015; 43:422–423.


ab oratory

l i m i t e d

lan imals

Special Article

The more the merrier? Scoring, statisticsand animal welfare in experimentalautoimmune encephalomyelitis

Pushpalatha Palle1,*, Filipa M Ferreira1,*, Axel Methner2

and Thorsten Buch1

AbstractExperimental autoimmune encephalomyelitis (EAE) is a frequently used animal model for the investigation ofautoimmune processes in the central nervous system. As such, EAE is useful for modelling certain aspects ofmultiple sclerosis, a human autoimmune disease that leads to demyelination and axonal destruction. It is animportant tool for investigating pathobiology, identifying drug targets and testing drug candidates. Eventhough EAE is routinely used in many laboratories and is often part of the routine assessment of knockoutsand transgenes, scoring of the disease course has not become standardized in the community, with at least83 published scoring variants. Varying scales with differing parameters are used and thus limit comparabilityof experiments. Incorrect use of statistical analysis tools to assess EAE data is commonplace. In experimentalpractice the clinical score is used not only as an experimental readout, but also as a parameter to determineanimal welfare actions. Often overlooked factors such as the animal’s ability to sense its compromisedmotoric abilities, drastic though transient weight loss, and also the possibility of neuropathic pain, makethe assessment of severity a difficult task and pose a problem for experimental refinement.

KeywordsEAE, scoring scales, animal welfare, humane endpoints, refinement

EAE: a model for multiple sclerosis

Experimental autoimmune encephalomyelitis (EAE) isthe most commonly used animal model for multiplesclerosis (MS), an autoimmune demyelinating disorderof the human central nervous system (CNS). MS affects2.5 million people worldwide with a preponderance inhigher latitudes and developed countries. Usually MScommences in early adulthood and is more common infemales. Afflicted individuals develop motor impair-ment and cognitive dysfunction.1,2 Disease severitycan be assessed using the expanded disability statusscale (EDSS),3 a scale based mainly on a combinationof functional systems and ambulation. It serves todocument the course of the disease and to escalate ther-apy if necessary. It is also used in clinical trials whereit helps to assess the efficacy of the therapeutic agent.To ensure inter-rater standardization, training in theuse of the EDSS score is delivered by an independentonline platform.4 Although MS is a uniquely human

disorder not observed spontaneously in other species,animal models have helped greatly in increasing ourknowledge of MS. They served as useful tools in inves-tigating the dynamics of both the immune system andthe CNS during neuroinflammation. Accordingly,many of the MS drugs in use and under testing inhumans have been developed on the basis of experi-mental data coming from EAE.5 EAE is, however,not a single model but consists of a family of animal

1Institute of Laboratory Animal Science, University of Zurich,Zurich, Switzerland2Focus Program Translational Neuroscience (FTN), Rhine MainNeuroscience Network (rmn2), Johannes Gutenberg UniversityMedical Center Mainz, Department of Neurology, Mainz, Germany

Corresponding author:Thorsten Buch, Institute of Laboratory Animal Science, Universityof Zurich, Wagistr. 12, 8952 Schlieren, Switzerland.Email: [email protected]

*These authors contributed equally.

Laboratory Animals

2016, Vol. 50(6) 427–432



sagepub.co.uk/


DOI: 10.1177/0023677216675008

la.sagepub.com

models induced through different protocols, eachserving a different experimental purpose. It was firstdescribed in the 1930s while investigating the neuro-logical complications arising after the rabies vaccin-ation.6 After a series of studies that showed myelindestruction and perivascular infiltration in the CNS7

in 1947, similarities between EAE and human MSwere described.8 Since then, EAE was established in avariety of mammals such as monkeys, guinea pigs, cats,goats, primates, rats and mice, and was used to inves-tigate the pathobiology of MS.9 In mice, active EAE isinduced by subcutaneous immunization with myelincomponents and adjuvants. Self-tolerance is brokenand encephalitogenic effector T cells migrate into theCNS to attack the myelin sheath.10 The first clinicalsigns of disease, characterized by ascending flaccid par-alysis, appear 7–8 days after immunization, with dis-ease peaking often between days 14 and 15. The mostcommonly used antigens, derived from myelin, are pro-teolipid protein (PLP), myelin oligodendrocyte glyco-protein (MOG), and myelin basic protein (MBP). Inboth MS and EAE, the CNS is infiltrated by T cells,B cells and macrophages.11 Nevertheless, other aspectsof the disease differ between patients and can be mod-eled in EAE. For instance, induction in C57BL/6 miceusing MOG35-55 peptide emulsified in completeFreund’s adjuvant (CFA) and followed by Pertussistoxin (PT) injection usually results in chronic disease.12

On the other hand, induction in SJL/J mice usingPLP131-151 peptide in CFA results in a relapsing–remitting pattern. Adoptive transfer EAE, or passiveEAE, is a model in which encephalitogenic T cells aretransferred from myelin-immunized or diseased mice tonaıve recipient mice.10 This method allows the directassessment of the effector phase of EAE, or the particu-lar study of transferred cells types or hosts’ back-grounds. Apart from active and passive EAE,transgenic models were developed in which EAE devel-ops spontaneously (reviewed by Croxford and col-leagues).13 One example are T cell receptor transgenicmice crossed with IgH knock-in mice, both specific forMOG,14 which develop disease within 28 days.

Scoring scales for EAE

Active, passive and spontaneous EAE, even though pre-senting occasionally with different symptoms and signs,are usually assessed through some similar type of ‘EAEscoring scale’. Nevertheless, the term ‘EAE scoring scale’does not refer to a cohesive scoring scheme. In 2010, ameta-analysis of EAE studies showed that 126 manu-scripts have used 83 different clinical EAE scoringscales, mostly without giving any explanation on whya particular system was chosen.15 EAE scoring usuallyserves two purposes: (1) assessment of disease severity as

outcome value of the scientific study, and (2) providing aparameter for the determination of animal welfareactions. The EAE scales used range from 0 (no clinicalsigns) to between 4 and 10. The highest number usuallycorresponds to the death of the animal. Strikingly, evenscales that have the same range often do not have thesame increment, or the same increment does not corres-pond to the same clinical description. The scoring scalesof Miller (2007)16 and Bachmann (1999)17 range from0 to 5 (Table 1); but the first is a five-point scoringscale whereas the second is a 10-point scale. Adding tothe confusion, the same number of identifiers maydescribe different signs, e.g. in the Kalyvas (2004)scale, a score 5 comprehends both hind limb paralysiswith forelimb weakness, and moribund states,18 two dis-tinct conditions that are separated as scores 4 and 5,respectively in most papers. In general, it is not or notsufficiently discussed why the used scale fits the respect-ive study. Ten-point scoring scales may be superior byallowing a more accurate description of symptoms, andprovide a better distinction between recovering andrelapsing stages. They may therefore contribute to ahigher statistical power and lead to improved assess-ment of changes in EAE progression. Such moreextended scales would also overcome partial scoring(e.g. score 1.5 in Miller’s scale16), another issuein EAE clinical monitoring which is often reportedwithout being appropriately accounted for in theMethods sections. As it relies on the researcher’s experi-ence, it is context-dependent and therefore subjective.Nevertheless, one must note that while a larger rangewithin the scale (i.e. the number of identifiers tochoose from) is scientifically superior, the chances ofinter-observer and intra-observer variability will beenhanced. In this regard, blinding experimentalgroups to the observer is crucial. Finally, EAE inmice carrying certain gene deficiencies, leads to atypicalEAE,19 which is often more severe and progressive.20

Atypical EAE involves axial-rotatory movements dueto infiltration and demyelination of the cerebellum andbrainstem, instead of the spinal cord, thus requiring adifferent scoring scheme (Table 1).20,21

Assessment of welfare

In most studies the EAE score, alone or in combinationwith other parameters such as weight, defines animalwelfare actions. These commonly constitute provisionof food and water on the cage floor, mostly in jellifiedform, and termination by euthanasia. In EAE, animalill-being results from a combination of neurologicaldeficits such as reduced motor control, the possibilityof nausea or neuropathic pain, and features associatedwith any severe disease like weight loss and dehydra-tion. Increasing loss of motor function clearly impairs


the animals’ access to food and water, participation insocial activities and their ability to fend off cage mates.Whether the animal realizes its disability and sufferspurely from comprehending this remains an open andprobably unanswerable question. In humans, hedonisticadaptation appears to allow paraplegic patients a similarquality of life as the healthy population.22 Weight lossand dehydration, which are both easy to assess duringdaily monitoring of the mice, have direct welfare rele-vance and their assessment is often required by theresponsible animal welfare authorities. Mice can adaptto a deprivation of up to 50% of water for one week.23

Nevertheless, due to daily scoring in EAE, dehydrationcould be detected rather acutely by adopting the scaleestablished by Bekkevold and colleagues.23 In case ofdehydration, an intraperitoneal administration of saline(maximum 80mL/kg)24 should be enough for full recov-ery. In contrast to dehydration, critical weight loss, oftendefined as a reduction of 20% in body weight, leads totermination of the experiment because mice must beeuthanized to abide with humane endpoints. Althoughto the best of our knowledge there is no direct causalrelationship between weight loss and disease progres-sion; in EAE mice frequently lose weight transiently,correlating with higher scores and paralysis.25 Thus,they are able to recover their weight when disease ameli-orates. This raises the question of whether a 20% weightloss constitutes a good termination point. A report inABH mice shows that these are highly susceptible toweight loss without any corresponding increase in dis-ease severity.26 In this study, mice lost around 26% ofweight during acute disease, followed by almost com-plete recovery. Thus, if the authors had rigidly appliedthe frequently pre-set guidelines of 20% body weight lossas the endpoint, they would have killed most of theirexperimental subjects, without gaining insight into therelapsing–remitting phase of the disease that is charac-teristic of this strain and which can be a source of valu-able clinical information.26 Consequently, in EAEcritical weight loss may have to be defined on a case-by-case basis in order to overcome the problem of losingstatistical power and scientifically important data. In thisregard, a European Union document with practicalguidelines on how to implement Directive 2010/63/EUin EAE studies suggests using 35% weight loss as thehumane endpoint, whenever applicable, in order tomaximize 3R practice.27 For ABH mice, an endpointbased on body temperature was suggested in onestudy: when body temperature decreases below 31�C,recovery is unlikely.26 An even more stringent sugges-tion was made in a pertussis infection/vaccinationstudy, where a lower body temperature limit of 34.5�Cwas shown to be a humane endpoint. Nevertheless,future research in EAE using C57/BL6 also has toshow whether temperature could be a useful endpoint,

and whether it justifies the additional handling ofanimals.28

Refinement opportunities

Current clinical scoring of EAE suffers from some obvi-ous problems, most of which can be easily overcome.Firstly, the community needs to come to agreement ona commonly accepted scoring scheme that allows com-parison of experiments. This may well be based on afrequently practised 10-step scale. Also, since data fromclinical EAE scoring are generated within non-linearscales, they must be analysed using non-parametricalstatistical tests such as the Mann–Whitney U orWilcoxon rank sum test. Even though this is long-known, 50% of reports feature parametrical statistics.29

Application of correct tests and the correspondingpower calculations would strongly increase reproduci-bility of EAE experiments. Blinding of EAE studies israrely reported, making inadvertent biases highly likely.Since circadian rhythm affects immune responses andvice versa, induction and scoring should be performedat identical times during an experiment.30,31 Thoughonly suitable for mice not yet paralysed, the use ofquantitative motor function tests such as the gripstrength and rotarod have been shown to help decreasebias.25 Such motor assessment facilitates statistical ana-lysis with parametric tests and thus increases power.Treating weight loss as a humane endpoint criterionshould be questioned critically, as discussed above.

Littermate controls should be preferred over wild-type mice bought from commercial suppliers to ensurethe genetic background differs only by the factor instudy. Lastly, environmental stress and gut microbiomemay influence disease outcome,32 and should thus besimilar between experimental groups. Hence, preferenceshould be given to mixing mice of different experimentalgroups in the same cage from early age on and allowingthem to adapt to housing conditions.33 In conclusion, aunifying system capable of efficiently and objectivelyinducing and accessing disease progression and animalwelfare without causing extra discomfort to the animalsshould be sought.


A common measurement in most EAE studies is theassessment of clinical symptoms using scoring scales,which not only yield experimental data but also definewelfare actions. EAE experiments would profit greatly,if researchers were to implement the following points:

(a) One common scale.(b) Induction and measurement of disease at identical

time of the day.

Palle et al. 429

Table

1.Examplesofexp

erimentalautoim

muneence

phalomyelitis(EAE)scoringsystems.

Miller(2007)16

Bach

mann(1999)17

Axial-rotatory

EAE20

Bebo(1998)34

Bittner(2014)35

Exp

andeddisability

statusscale

(MS

inpatients)3

0Noclinicalsigns

Noclinicalsigns

Noclinicalsigns

0Noclinicalsigns

Noclinicalsigns

Noclinicalsigns

0.5

Distallimptail

1Minim

alhindlimb

weakness

Partiallimptail

Noim

pairment

1Lim

ptailorhindlimb

weakness

Lim

ptail

Mildtiltingofthe

head

2Moderate

hindlimb

weakness

ormildataxia

Paralyse

dtail

Minim

alim

pairment

1.5

Lim

ptailandhindlimb

weakness

3Moderate

severe

hind

limbweakness

Hindlimbparesis

Moderate

impairment

2Both

limptailandlimb

weakness

Unilateralpartialhind

limbparalysis

Markedtiltingofthe

head

4Severe

hindlimbweak-

ness

ormildforelimb

weakness

ormoderate

ataxia

Hindlimbparaplegia

Severe

impairment

2.5

Bilateralpartialhind

limbparalysis

5Paraplegia

withmoder-

ate

forelimbweakness

Both

hindlimbs

paralyse

dWalkingrestrictedto

<200m

3Partialhindlimb

paralysis

Complete

bilateralhind

limbparalysis

Tiltingofthebody

6Paraplegia

withse

vere

forelimbweakness

or

severe

ataxia

Quadriparesis

Constantass

istance

3.5

Complete

bilateralhind

limbparalysisandpar-

tialforelimbparalysis

71forelimbparalyse

dWheelchair

bound

4Complete

hindlimb

paralysis

Totalparalysisofhind

andforelimbs

Continuousaxial

rotation

8Quadriplegia

Bedbound

4.5

Moribund

9Moribund

Helpless

bedpatient

5Death

Death

Death

10

Death

Death

MS:multiple

sclerosis.


(c) Induction and measurement of disease in a blindedfashion.

(d) Use of littermate controls of the same geneticbackground and hosting similar microbiomes.

(e) Use of non-parametric statistics for data analysisand power calculation during experimentalplanning.

(f) Allowance of up to 35% transient weight loss,according to characteristics of the strain, EAEinduction paradigm and aim of the study.

(g) Consistent use of jellified food/water as welfareaction.

(h) Assessment of dehydration paired with respectiveactions.

When the aim of a study is to describe small differ-ences at low scores, motor tests such as rotarod andgrip strength should be used and may increase thepower of the study. Future research has to showwhether neuropathic pain constitutes a relevantanimal welfare problem in EAE. In conclusion, EAEscoring scales are a good example of tools that repre-sent well-established and common practice, but whichneed to be re-evaluated with a critical eye. Currently,the variety of scoring scales and their analysis may con-tribute to irreproducibility and failure in translation ofanimal experiments.

Acknowledgement

We thank Phillipe Bugnon for his helpful comments on the

manuscript.


The author(s) declared no potential conflicts of interest with

respect to the research, authorship, and/or publication of thisarticle.

Funding

The author(s) disclosed receipt of the following financial sup-port for the research, authorship, and/or publication of this

article: This work was supported by the Hertie Foundation,grant number P1140090, but otherwise received no specificgrant from any funding agency in the public, commercial,or not-for-profit sectors.

References

1. Goldenberg MM. Multiple sclerosis review. PT 2012; 37:

175–184.

2. Pinkston JB and Alekseeva N. Neuropsychiatric manifest-

ations of multiple sclerosis. Neurol Res 2006; 28: 284–290.3. Kurtze JF. Rating neurologic impairment in multiple scler-

osis: an expanded disability status scale (EDSS). Neurology

1983; 33: 9.

4. Lechner-Scott J, Huber S and Kappos L. Expanded dis-

ability status scale (EDSS) training for MS multi-center

trials. J Neurol 1997; 244(Suppl. 3): S25. www.neurosta-

tus.net (1997, accessed 14 March 2016)).

5. Constantinescu CS, Farooqi N, O’Brien K and Gran B.

Experimental autoimmune encephalomyelitis (EAE) as a

model for multiple sclerosis (MS). Br J Pharmacol 2011;

164: 1079–1106.6. Stuart G and Krikorian K. A fatal neuro-paralytic acci-

dent of antirabies treatment. Lancet 1930; 215:

1123–1125.

7. Kabat EA, Wolf A and Bezer AE. Rapid production of

acute disseminated encephalomyelitis in rhesus monkeys

by injection of brain tissue with adjuvants. Science 1946;

104: 362.8. Wolf A, Kabat EA and Bezer AE. The pathology of acute

disseminated encephalomyelitis produced experimentally

in the rhesus monkey and its resemblance to human

demyelinating disease. J Neuropathol Exp Neurol 1947;

6: 333–357.9. Baxter AG. The origin and application of experimental

autoimmune encephalomyelitis. Nat Rev Immunol 2007;

7: 904–912.10. Denic A, Johnson AJ, Bieber AJ, Warrington AE,

Rodriguez M and Pirko I. The relevance of animal

models in multiple sclerosis research. Pathophysiology

2011; 18: 21–29.11. McCarthy DP, Richards MH and Miller SD. Mouse

models of multiple sclerosis: experimental autoimmune

encephalomyelitis and Theiler’s virus-induced demyeli-

nating disease. Methods Mol Biol 2012; 900: 381–401.

12. Mendel I, Kerlero de Rosbo N and Ben-Nun A. A myelin

oligodendrocyte glycoprotein peptide induces typical

chronic experimental autoimmune encephalomyelitis in

H-2b mice: fine specificity and T cell receptor V beta

expression of encephalitogenic T cells. Eur J Immunol

1995; 25: 1951–1959.13. Croxford AL, Kurschus FC and Waisman A. Mouse

models for multiple sclerosis: historical facts and

future implications. Biochim Biophys Acta 2011; 1812:

177–183.14. Krishnamoorthy G, Lassmann H, Wekerle H and Holz

A. Spontaneous opticospinal encephalomyelitis in a

double-transgenic mouse model of autoimmune T cell/B

cell cooperation. J Clin Invest 2006; 116: 2385–2392.15. Vesterinen HM, Sena ES, ffrench-Constant C, Williams

A, Chandran S and Macleod MR. Improving the trans-

lational hit of experimental treatments in multiple scler-

osis. Mult Scler 2010; 16: 1044–1055.

16. Miller SD, Karpus WJ and Davidson TS. Experimental

autoimmune encephalomyelitis in the mouse. Curr Protoc

Immunol 2007; Unit 15.1.17. Bachmann R, Eugster H-P, Frei K, Fontana A and

Lassmann H. Impairment of TNF-receptor-1 signaling

but not fas signaling diminishes T-cell apoptosis in

myelin oligodendrocyte glycoprotein peptide-induced

chronic demyelinating autoimmune encephalomyelitis in

mice. Am J Pathol 1999; 154: 1417–1422.18. Kalyvas A and David S. Cytosolic phospholipase A2

plays a key role in the pathogenesis of multiple sclero-

sis-like disease. Neuron 2004; 41: 323–335.

Palle et al. 431

19. Krakowski M and Owens T. Interferon gamma confersresistance to experimental allergic encephalomyelitis. EurJ Immunol 1996; 26: 1641–1646.

20. Abromson-Leeman S, Bronson R, Luo Y, et al. T-cellproperties determine disease site, clinical presentation,and cellular pathology of experimental autoimmuneencephalomyelitis. Am J Pathol 2004; 165: 1519–1533.

21. Wensky AK, Furtado GC, Garibaldi Marcondes MC,et al. IFN-gamma determines distinct clinical outcomesin autoimmune encephalomyelitis. J Immunol 2005; 174:

1416–1423.22. Brickman P and Coates D. Lottery winners and accident

victims: is hapiness relative? J Pers Soc Psychol 1978; 36:

917–927.23. Bekkevold CM, Robertson KL, Reinhard MK, Battles

AH and Rowland NE. Dehydration parameters and

standards for laboratory mice. J Am Assoc Lab AnimSci 2013; 52: 233–239.

24. Hawk CT, Leary SL and Morris TJ. Formulary forlaboratory animals, 3rd ed. Ames, IA: Blackwell

Publishing, 2005.25. van den Berg R, Laman JD, van Meurs M, Hintzen RQ

and Hoogenraad CC. Rotarod motor performance and

advanced spinal cord lesion image analysis refine assess-ment of neurodegeneration in experimental autoimmuneencephalomyelitis. J Neurosci Methods 2016; 262: 66–76.

26. Al-Izki S, Pryce G, O’Neill JK, et al. Practical guide tothe induction of relapsing progressive experimental auto-immune encephalomyelitis in the Biozzi ABH mouse.Mult Scler Relat Disord 2012; 1: 29–38.

27. Examples to illustrate the process of severity classification,day-to-day assessment and actual severity assessment.

http://ec.europa.eu/environment/chemicals/lab_animals/pdf/examples.pdf (2013, accessed 11 October 2016).

28. Hendriksen CFM, Steen B, Visser B, Cussler K, Morton

D and Strijger F. The evaluation of humane endpoints inpertussis vaccine potency testing. London: Royal Societyof Medicine Press Limited, 1999.

29. Fleming KK, Bovaird JA, Mosier MC, Emerson MR,

LeVine SM and Marquis JG. Statistical analysis of datafrom studies on experimental autoimmune encephalomy-elitis. J Neuroimmunol 2005; 170: 71–84.

30. Silver AC, Arjona A, Walker WE and Fikrig E. The circa-dian clock controls toll-like receptor 9-mediated innateand adaptive immunity. Immunity 2012; 36: 251–261.

31. Buenafe AC. Diurnal rhythms are altered in a mousemodel of multiple sclerosis. J Neuroimmunol 2012; 243:12–17.

32. Berer K, Mues M, Koutrolos M, et al. Commensalmicrobiota and myelin autoantigen cooperate to triggerautoimmune demyelination. Nature 2011; 479: 538–541.

33. Wu GD, Chen J, Hoffmann C, et al. Linking long-term

dietary patterns with gut microbial enterotypes. Science2011; 334: 105–108.

34. Bebo BF Jr, Schuster JC, Vandenbark AA and Offner H.

Gender differences in experimental autoimmune enceph-alomyelitis develop during the induction of the immuneresponse to encephalitogenic peptides. J Neurosci Res

1998; 52: 420–426.35. Bittner S, Afzali AM, Wiendl H and Meuth SG. Myelin

oligodendrocyte glycoprotein (MOG35-55) inducedexperimental autoimmune encephalomyelitis (EAE) in

C57BL/6 mice. J Vis Exp 2014; (86): e51275.


ab oratory

l i m i t e d

lan imals

Special Article

Osteotomy models – the currentstatus on pain scoring and managementin small rodents

Annemarie Lang1,2,3, Anja Schulz3, Agnes Ellinghaus4 andKatharina Schmidt-Bleek4,5

AbstractFracture healing is a complex regeneration process which produces new bone tissue without scar formation.However, fracture healing disorders occur in approximately 10% of human patients and cause severe pain andreduced quality of life. Recently, the development of more standardized, sophisticated and commerciallyavailable osteosynthesis techniques reflecting clinical approaches has increased the use of small rodentssuch as rats and mice in bone healing research dramatically. Nevertheless, there is no standard for painassessment, especially in these species, and consequently limited information regarding the welfare aspectsof osteotomy models. Moreover, the selection of analgesics is restricted for osteotomy models since non-steroidal anti-inflammatory drugs (NSAIDs) are known to affect the initial, inflammatory phase of bonehealing. Therefore, opioids such as buprenorphine and tramadol are often used. However, dosage data inthe literature are varied. Within this review, we clarify the background of osteotomy models, explain thecurrent status and challenges of animal welfare assessment, and provide an example score sheet includingmodel specific parameters. Furthermore, we summarize current refinement options and present a briefoutlook on further 3R research.

Keywordsanalgesia, animal model, laboratory animal welfare, pain assessment, pain reduction

Osteotomy models

Fracture healing is a complex regeneration process,starting with a fracture hematoma that is characterizedby closely regulated inflammatory events and a hypoxicmicroenvironment which initiates the healing cascade.1

The following healing phases include the formation of afibrocartilaginous callus, the transformation to miner-alized cartilage, the restructuring with woven bone, andfinally the remodeling. The first phase of bone healing isparticularly important, and includes the inflammatoryprocesses within the fracture hematoma as well as opti-mal fixation stability.2 To develop and test new thera-peutic strategies, animal models are needed in bonehealing research. Besides the physiological direct orindirect fracture healing, clinical human scenariossuch as delayed union, established hypertrophic oratrophic non-union (pseudarthrosis), segmental bonedefects, and multiple traumata are of great interest.

Rodent models are often used because of theiradvantages such as the ready availability of geneti-cally-modified animals, high reproductive rates, lowcosts, easy housing, the existence of readily availableresearch kits, and their known genome sequence.

1Department of Rheumatology and Clinical Immunology, Charite-Universitatsmedizin, Berlin, Germany2Berlin Brandenburg School for Regenerative Therapies, Charite-Universitatsmedizin, Berlin, Germany3German Rheumatism Research Centre Berlin, Berlin, Germany4Julius Wolff Institute and Center for Musculoskeletal Surgery,Charite-Universitatsmedizin, Berlin, Germany5Berlin Brandenburg Center for Regenerative Therapies, Charite-Universitatsmedizin, Berlin, Germany

Corresponding author:Annemarie Lang, Department of Rheumatology and ClinicalImmunology, Charite-Universitatsmedizin, Chariteplatz 1, 10117Berlin, Germany.Email: [email protected]

Laboratory Animals

2016, Vol. 50(6) 433–441



sagepub.co.uk/


DOI: 10.1177/0023677216675007

la.sagepub.com

Nevertheless, there is no single animal model thatreplicates all the key events of a human patient, andtherefore animal models should be adapted to theresearch question. More than half of the animals usedin orthopedic research are rats (38%) or mice (15%).3

The development of more sophisticated osteosynthesistechniques with more clinically relevant stabilizationhas increased the number of small rodents (includingrabbits; together >80%) used in bone healingresearch.4,5 Most animal species show slight analogiesto the human bone macro- and microstructure. Themain differences in mice concern the permanent open-ing of the growth plate in the epiphyses of long bonesleading to a lifelong skeleton modeling, the lack of aHaversian system and low cancellous bone content atthe epiphyses of long bones.5,6 Mice are appropriate forbasic research (transgenic modifications), while rats aremore suitable for pharmacological interventions, toxi-cological studies and, due to their bigger size, for bio-material approaches.7,8

Well-established osteotomy models of long bones inrodents exist with stabilization possibilities using exter-nal fixators, plates or intramedullary pin/lockingnails.9–11 The most common localization is the femur,while models also exist for the tibia. Each different sta-bilization possibility bears the potential of different gapsizes varying between 1–8mm in rats and 0.1–4mm inmice.11,12 Pathophysiological processes such as bonehealing disorders are created by forming an unstablefixation or by creating critical size defects, cauterizationof the periosteum or removal of the bone marrow.Nevertheless, for any defined experiment each modelhas its own advantages and disadvantages. Whilethe external fixation prevents the direct interaction ofthe device with the defect and interferes with the bonemarrow only through the pins, it is especially suitablefor cell-based approaches and biomaterial testing. Thedisadvantage is its position outside the body thattempts the animal to manipulate the device andincrease the risk of infection due to direct communica-tion between the outside environment and internalstructures. Therefore, the plate features a closure offascia and skin over the device that prevents manipu-lation and decreases the susceptibility to infections.Even more important is the increase in mechanical sta-bility due to a lower offset of the fixation plate.Therefore, it is more suitable for larger defects, whichprevent osteolysis at the pins. However, the plateaffects the periosteum and changes the bone healingprocess at the site of the mounted plate. By contrastwith the other two models, the intramedullary pin/lock-ing nail more or less replaces the bone marrow with allits cells and vessels. Current understanding is that thisdestruction diminishes the autologous healing potentialof the bone.13 However, it is the most adequate model

for closed fractures, in comparison to the clinicalroutine.14

The mouse osteotomy model

The small size of mice has long been considered ahindrance in bone healing research. However fixatorsystems are available today for various applications,which allow experimenters to mimic clinically used sta-bilization techniques with systems designed and opti-mized for the small bones of mice. This still leaves thechallenge of the small sample size of the harvested spe-cimens, a fracture hematoma of 0.7� 1.4mm or 200 mLof blood serum after terminal cardiac blood sampling.However, with the high sensitivity of current analyticaltools this has become a surmountable challenge.15

External fixation in the femur of the mouse can beaccomplished with a gap size of 0.1mm and up to2mm in a rigid or semi-rigid setting, respectively. Forour research, we have established a model that does notcompletely heal within a time period of 21 days (oste-otomy gap 0.7mm) in the control group so that animprovement in the healing process can be easilydetected in treated groups. This model only works infemale animals as the bone healing process is slower infemales than in males,16 and we only use the C57BL/6strain in our studies. This strain has good bone healingqualities: bone growth is ‘adult’ at around 12 weeks andbone reaches its peak bone mass at around 16 weeks.The external fixation system (MouseExFix; RISystem,Davos Platz, Switzerland) which we now use in multiplesettings has proved to be highly reliable so that we havevery few dropouts in our studies, allowing for smallgroup sizes in our study design which reduces animalnumbers (Figure 1). Surgery time for an experiencedoperator is around 15min per animal. This reducesthe time of anesthesia and thus the stress for theanimals.17

Severity assessment

Within EU regulations, Directive 2010/63/EU listsexamples to aid in the assignment of severity classifica-tions to specific procedures (Annex VIII). On this basis,severity depends on the degree of pain, suffering,distress or lasting harm expected to be experienced byan individual animal during the course of the proced-ure. Levels range from (i) mild (short-term mild pain,suffering or distress; no significant impairment of thewell-being or general condition), to (ii) moderate(short-term moderate pain, suffering or distress, orlong-lasting mild pain, suffering or distress; moderateimpairment of the well-being or general condition), andto (iii) severe (severe pain, suffering or distress, or long-lasting moderate pain, suffering or distress; severe


impairment of the well-being or general condition).According to Directive 2010/63/EU, the level ofseverity of osteotomy as a procedure depends on thedegree of fixation stability. Hence, we consider a stablefixation as having moderate severity for up tothree days post-surgery and low severity for an ongoing7–10 days until the wounds are healed. Unstable fix-ation without restoration of functionality is normallyclassified as severe.


A distinct recognition of pain in experimental animalsis essential for evaluation of biological effects as well asfor ethical reasons. Pain can lead to wound healing andblood flow disorders, immunosuppression andincreased infection risks.18 Poorly-managed pain canhave an impact on the success of a study as the animalschange the load-bearing of the affected limb.19,20

Therefore, welfare and pain assessment in small rodentsis of great importance. Unfortunately, there is no ‘goldstandard’ for measuring or assessing animal pain,especially in rats and mice.21 In bone research, well-described systems only exist for large animal modelssuch as lambs and dogs.22,23 General pain indicatorsin rats and mice are worsening of appearance (such asneglected fur care), reduced condition (e.g. loss of bodyweight), prolonged re-convalescence, reduced food andwater uptake, conspicuous posture, pain face (closedeyes, flattened ears and nose hairs), decreased activity,abnormal movement, vocalization, and absence of nestbuilding.24 The daily observation by trained personnel,video surveillance with automatic analysis and telemet-ric methods (e.g. heart frequency, body temperature,breath frequency) can support pain assessment.25,26

Animals should always be observed in their familiarenvironment and before any kind of manipulation.Furthermore, it must be recognized that rodents arenocturnal and less active during the day. Weight lossis frequently used as an indicator of poor animal wel-fare.27 But severe weight loss will always be accompa-nied by other symptoms such as dehydration, apathy or

separation from the group.28 On the other hand, slowweight loss over days with no other symptoms might becaused by transportation, change of environment,handling or recovery from anesthesia and surgery andconsequent reduction in food uptake. In such cases, theanimals can be fed with soaked food, apples or jelly ifthey have been used as food beforehand. Rats and miceare unlikely to eat novel food when in a state of poorwelfare, therefore cases of unexplained weight lossshould be carefully investigated.

To evaluate pain with respect to bone injury inrodents, only models of post-fracture pain evaluationand bone cancer models have so far been used.29–31 Thedifference between post-fracture pain models and oste-otomy models is that in these studies no pain medica-tion is given; also with bone cancer models, pain willoften exceed what is to be expected in a stabilizedfracture. Nevertheless, the methods to evaluate paincould be applied to osteotomy models in rodents. Toclassify pain levels the following methods can be distin-guished: (i) subjective assessment of spontaneous orpalpation-provoked behavior of the affected limb, e.g.reduced load or prevention of use or abnormal pose ofthe limb (measured as guarding time span and amountof flinching);30,32 and (ii) methods such as the hot plate(thermal nociception) or the von Frey test (mechanicalnociception) which indicate the development of hyper-algesia (exaggerated sensitivity to pain) that can occurtemporarily after damage of nociceptors or peripheralnerves.29–31,33–35 However, both these listed methodsare invasive and are associated with additional stressand pain burden for the animal. Therefore, the selectionof pain assessment methods should be carefullyweighed. To date, no study has been published onpain assessment in osteotomy models which providesevidence-based data on behavioral changes in miceand rats. Such studies are desperately needed andshould be included in basic research. In addition, gen-eral methods to assess pain such as clinical scoring (e.g.the mouse grimace scale), and to determine well-beingsuch as nesting score or time-to-incorporate-into-nestscore (TINT), could be beneficial but need to be

Figure 1. (A) External fixation on the left femur. (B) MouseExFix from RISystem. (C) X-ray of 0.7mm gap post-operation.(D) Mouse 10min after surgery, still under the red heat light with close supervision in the surgical theater, and alreadyrearing. (E) Mouse with external fixator after observation time of 21 days, no irritation of the surrounding tissue at theexternal fixator, and the mouse shows all the signs of a healthy animal.

Lang et al. 435

validated for osteotomy models. Further methods suchas testing of burrowing behavior have to be consideredin view of the fixation system and the possible risk ofinjury – for example the possibility of jamming of theexternal fixator on plastic housing.36

Score sheet

As mentioned above, there is no standardized scoresheet for osteotomy models in rats and mice. Table 1represents an example score sheet that compares gen-eral physiological behavior with specific indicators seenin unwell animals that is used in our facility. In general,the responsible scientist must leave his/her contactdetails to be available if necessary. The score sheet isplaced next to the cage, and must be recognizable andunderstandable for everyone. In addition, the animals’health is assessed daily and they are weighed at leastonce a week. Signs of grade 1 must be discussed duringthe daily ward round with the responsible veterinarian.Grade 2 signs require the immediate notification of theon-call responsible veterinarian. After determination ofthe symptoms, the veterinarian will initiate the treatmenttaking into account the expected success of the treatmentas well as the scientific data value. Consultation withthe responsible scientist is recommended.

Prescribed interventions depend on the diagnosis(for example: pain¼ analgesia, infection¼ antibiotics,circulatory disorders¼ infusions).

For general indicators of animal welfare, a humaneendpoint would be applicable if two or more of thelisted points reached grade 3 in this model scoresheet. In such cases the animal must be killed painlesslyand immediately. Such an animal would already beunder specific observation upon reaching grade 2. Theappropriate therapeutic intervention should have beenapplied once the animal had reached grade 2 with oneor more signs, which therefore implies that interven-tions had failed and the endpoint has been reached.If one of the specific indicators reaches grade 3 of thescore sheet, this is sufficient reason to terminate theanimal immediately to prevent suffering, as eithera humane endpoint has been reached or the animalcan no longer be used in the experimental setting.

In any case, signs of grade 3 are undesirable andshould be avoided.

Scores that are expected during a well conductedosteotomy in mice or rats from transport to sacrifice,assuming no unexpected complications, are summar-ized in Figure 2.


A rigorous and established pain management plan isessential when considering refinement and animal

welfare. Perioperative analgesia should begin with apre-emptive dose of an appropriate analgesic givenprior to surgical incision or pain stimulus. Therefore,the present pain is alleviated, and pain developmentand pain memory are avoided, resulting in a reductionin the intensity of postoperative pain. Many analgesicsenable a distinct intraoperative dosage reduction of theanesthetics and therefore reduces recovery time fromthe anaesthesia. Postoperative analgesia enhances painrelief, accelerates convalescence, as well as preventingpain-induced symptoms.37,38 Criteria for the choice ofanalgesics are:

. Preoperative: type of expected pain, influence onresults, length of effectiveness, effectiveness of painrelief, time point of administration.

. Intraoperative: possible dosage reduction of anes-thetics, possible additional dosage, monitoring ofanimals.

. Postoperative: length of effectiveness, effectiveness ofpain relief, minimal invasive application strategies.

At first, it needs to be pointed out that there are sub-stantial deficits in knowledge about pain treatment insmall rodents (e.g. uncertainty regarding dosages, routeof administration, efficacy, and duration of action, i.e.injection interval). The selection of suitable analgesicsfor pain management in osteotomy models is limited,since concerns have been raised about potential adverseeffects of non-steroidal anti-inflammatory drugs(NSAIDs) during the initial and inflammatory phaseof bone healing.39 Therefore, researchers normally usemorphine, buprenorphine or tramadol in these models.Nevertheless, empirical data on the effectiveness ofthese analgesics for osteotomy models in rats andmice are missing. That is why varying dosage details,application methods (injection vs drinking water andfood) and intervals as well as different treatment dur-ations are found in the literature. With our studies,osteotomized mice and rats are treated for at leastthree days with pain medication postoperatively.There are only a few studies on fracture pain showingevidence for morphine as a short-term treatment.30,32,40

Buprenorphine is a fast acting and potent opioid, andto date it is the most commonly used analgesic in treat-ing postoperative pain in small animals.41–44 Differentdosages have been specified for subcutaneous or intra-peritoneal injection ranging between 0.05 and0.75mg/kg for repeated doses, respectively,45–47 or foradministration via drinking water (9mg/L) for mice orvia food (jelly) at 0.15mg/kg.48,49 There are new studieson the application interval of buprenorphine whichindicate that the pharmacokinetic half-life is short(<6 h).50 In order to avoid additional stress for theanimals from regular handling during injection,


Table

1.Example

score

sheetforosteotomymodels

inrats

andmice.

Signs

01

23

General

Indicators

Appearance

Norm

al

Reduce

dgrooming

Scruffy,

unkemptfur

Piloerection,excrement

soiling

Generalco

ndition

Norm

al

Dim

inishedgeneral

condition

Distinct

dim

inished

generalco

ndition

Apathy,

forcedbreathing

Facialexp

ress

ion

Norm

al

Slightlyclose

d,wetor

dulledeyes

Sunken,cloudedeyes

Close

dandclottedeyes

Bodyposture

Norm

al

Slightco

ntorted

posture

Contortedposture

Hunch

edposture

Foodandwaterintake

Norm

al

Decrease

dfor24h

Constantlydecrease

dNointake

Fece

s,urine

Norm

al

Slightch

angesin

color,

quantity

andco

nsist-

ency

(aloneorin

combination)

Abnorm

alin

color,

quantity

andco

nsist-

ency

(alone)

Abnorm

alin

color,

quantity

andco

nsist-

ency

(inco

mbination)

Bodyweight

Norm

al

Weightloss

<5%

Weightloss

5–15%

20%

weightloss

Specific

indicators

Skin

suture

Skin

suture

dry,co

hesive

Stitchespartly

removed

butsk

insu

ture

still

cohesive

Skin

suture

notco

hesive

Complete

failure

ofsk

insu

ture

Activity

Norm

al

Isolated,noclim

bing

Inactive

Moribund

Loadonlimb

Fullload,padwithco

m-

plete

contact

toground

Notfullload,pad<50%

contact

toground

Notfullload,padwith

rare

contact

tothe

ground

Noload,noco

ntact

totheground

Standandwalk

Norm

al,hindlimbstand

Reduce

dhindlimb

stand,reduce

dwalk

Lim

ping,nohindlimb

stand

Nowalk

Fixationsystem

Stable

fixation

Externalfixator/pins

breakthroughsk

in,

pin/intramedullary

nail/loose

ningoffixa-

tor/fracture/m

alposi-

tionoffracture

endings>10�

Lang et al. 437

non-invasive application over drinking water and foodis preferable. This is of special interest for tramadol, asynthetic opioid analog, which has an extremely often-ignored short half-life (<2 h).51 Therefore, tramadolshould only be applied via drinking water. TheGerman Society for Laboratory Animals Science rec-ommends 1 g/L (the former recommendation was25mg/L)52,53 for the application via drinking waterwithout any data-based justification. Therefore, theneed for evidence-based and empirical data for thedosage and effectiveness of analgesics in bone researchis apparent and further studies need to be carried out.In our osteotomy models, we administer 0.03mg/kgBuprenorphine s.c. pre-operatively and 25–100 mg/L(mice) and 25 mg/L Tramadol (rats) via the drinkingwater for three days post-operatively.17,20 To provideevidence-based data for our pain management regimen,we are now performing a specific study on this topicand results will be published soon. Table 2 lists thenormally used pain medications in small rodent osteot-omy models. Nevertheless, it should be taken intoaccount that pain management regimens can varybetween different mouse and rat strains, and shouldbe adjusted accordingly.

Furthermore, an effective pain management regimenshould include: (i) knowledge of the pain pathology,(ii) knowledge of the pharmacological and non-pharmacological analgesic strategies, and (iii) the abil-ity to identify pain.21 Therefore, researchers should beinformed and be adequately trained to recognize painand choose the appropriate treatment. A short surgerytime without additional infliction of injury is essentialto avoid stress and pain in research animals, thereforesurgeons should be properly trained. Surgical trainingcan be acquired at a first stage while performing together

with an experienced surgeon using animal cadavers fromother experiments, and at a second stage through assist-ing with an actual surgery. Once the handling andthe familiarity with the drill and saw are assured, thetrainee can perform the surgery independently.The surgical procedure should include standard precau-tions such as gentle handling and positioning as well asheat supply. Furthermore, the use of standardizedand validated fixation systems is of great importanceand reduces failure rates. For rodents, it is always advis-able to use single stitches to prevent reopening of thewound as the animals are liable to nibble at the sutures,and eventually remove them themselves.

Additionally, animals should be kept under perman-ent surveillance after surgery until they recover theirmobility fully. Housing should be in groups with twoindividuals that are familiar (same group as before) andof the same character (activity).61 Cages should be well-filled with soft bedding and must not include houses orpipes due to the risk of injury if an external fixator hasbeen used. Nesting material (e.g. tissues, towels orpaper shavings) has the potential to improve rodentwell-being. The nesting material should be providedprior to surgery so the animal can build a good qualitynest before they may be disabled and unable to engagein effective nest building behavior.62,63 Easy to reachsoft, palatable food (in the form of a gel) and water(via a longer sipper tube) can be quite suitable for post-operative animals that are reluctant to move or putweight on their hind limbs to engage in rearing pos-itions. Furthermore, note that it is not necessarily tooperate on all planned animals in order to achieve astatistically relevant number of animals. Pilot studiescan provide first evidence of the effect and continuousanalysis of the results during experimentation, which

Figure 2. Scheme of scores to be expected in a well conducted osteotomy in mice or rats from transport to sacrifice,assuming no unexpected complications. SPF: specific pathogen-free.


can lead to a reduction in the number of animals used ifstatistical relevance is achieved before the full group isused. Nevertheless, it is important to avoid underpow-ering, and to control the type 1 error rate and followstatistical recommendations.64

For further reading, we recommend the review byAuer et al. (2007) as well as the web-based search forgeneral refinement approaches from specific institutions(e.g. NC3R, Norecopa) that can be adopted to the usedanimal model.21

Recommendations and further 3Rresearch

Before starting a study using the osteotomy model inrats or mice, the following considerations are stronglyrecommended: (i) perform an extensive literatureresearch, (ii) choose the most suitable species and fix-ation system to answer the clearly formulated researchquestion, (iii) define the pain management strategy afterconsultation with experienced researchers in this field,(iv) determine the appropriate score sheet (includinghumane endpoints), (v) instruct competent animal kee-pers of the goal of your experiment and keep theminvolved during the duration of the study, and (vi) care-fully train the surgeon before starting the trial. Afteraccomplishing the study, it is recommended to describein detail the strategies of pain assessment and analgesiaemployed as well as unusual findings and failures in thepublications according to the ARRIVE guidelines.

Nevertheless, further 3R research, especially inrefinement is needed. To mention one example: ratsare known to gnaw at the external fixator, necessitatingthe use of a suitable collar to prevent this. Such a collaris associated with stress for the animals. The develop-ment of a bite block that is fitted onto the externalfixator could solve this problem without impeding theanimal.

In addition, the standardization and harmonizationof pain assessment strategies in rats and mice are neces-sary as well as the enrolment of efficacy studies foranalgesics. This could be realized without the use ofadditional animals if these studies are embedded inbasic experimental studies.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest withrespect to the research, authorship, and/or publication of thisarticle.

Funding

The author(s) disclosed receipt of the following financial sup-port for the research, authorship, and/or publication of thisarticle: Our work is supported by the DFG (SCHM2977),

BfR (RefineMOMo), BCRT and BSRT.Table

2.Summary

ofexample

pain

managementregim

enuse

din

osteotomymodels.

Substance

Applica

tionmethod

Dose

Applica

tioninterval

Comment

Evidence

-base

d?

Mouse

Morphine

SC

>10mg/kg30,54

Acu

tedose

;effectivefor<4h

Only

use

din

fracture

pain

models

andtumor

models

Yes

Buprenorphine

SC

0.1mg/kg55

4–6hfor3days

50

Pharm

aco

kinetichalf-lifesh

ort

No

Food(e.g.jelly)

0.15g/kg48

Adlibitum

for2days

Anim

als

should

beuse

dto

thefoodpreoperative

No

Tramadol

Drinkingwater

25mg/L

17,56–60

100mg/L

17

1g/L

53

Adlibitum

for3days

Pharm

aco

kinetichalf-lifetoosh

ort

forSC

applica

tion(�

2h)

No

Rat Buprenorphine

SC

0.1mg/kg53

4–6hfor3days

Pharm

aco

kinetichalf-lifesh

ort

No

Tramadol

Drinkingwater

25mg/L

19,20

500mg/L

53

Adlibitum

for3days

Pharm

aco

kinetichalf-lifetoosh

ort

forSC

applica

tion(�

2h)

No

SC:su

bcu

taneous.

Lang et al. 439

References

1. Kolar P, Gaber T, Perka C, Duda GN and Buttgereit F.

Human early fracture hematoma is characterized by

inflammation and hypoxia. Clin Orthop Relat Research

2011; 469: 3118–3126.

2. Schmidt-Bleek K, Petersen A, Dienelt A, Schwarz C and

Duda GN. Initiation and early control of tissue regener-

ation – bone healing as a model system for tissue regen-

eration. Expert Opin Biol Ther 2014; 14: 247–259.3. O’Loughlin PF, Morr S, Bogunovic L, Kim AD, Park B

and Lane JM. Selection and development of preclinical

models in fracture-healing research. J Bone Joint Surg Am

2008; 90(Suppl 1): 79–84.4. Garcia P, Holstein JH, Maier S, et al. Development of a

reliable non-union model in mice. J Surg Res 2008; 147:

84–91.5. Peric M, Dumic-Cule I, Grcevic D, et al. The rational use

of animal models in the evaluation of novel bone regen-

erative therapies. Bone 2015; 70: 73–86.6. Gomes PS and Fernandes MH. Rodent models in bone-

related research: the relevance of calvarial defects in the

assessment of bone regeneration strategies. Lab Anim

2011; 45: 14–24.7. Thompson DD, Simmons HA, Pirie CM and Ke HZ.

FDA guidelines and animal models for osteoporosis.

Bone 1995; 17: 125s–133s.8. Tibbitts J, Cavagnaro JA, Haller CA, Marafino B,

Andrews PA and Sullivan JT. Practical approaches to

dose selection for first-in-human clinical trials with

novel biopharmaceuticals. Regul Toxicol Pharmacol

2010; 58: 243–251.9. Holstein JH, Menger MD, Culemann U, Meier C and

Pohlemann T. Development of a locking femur nail for

mice. J Biomech 2007; 40: 215–219.

10. Jager M, Sager M, Lensing-Hohn S and Krauspe R. The

critical size bony defect in a small animal for bone healing

studies (II): implant evolution and surgical technique on a

rat’s femur. Biomed Tech Biomed Eng (Berl) 2005; 50:

137–142.11. Kaspar K, Schell H, Toben D, Matziolis G and Bail HJ.

An easily reproducible and biomechanically standardized

model to investigate bone healing in rats, using external

fixation. Biomed Technik Biomed Eng (Berl) 2007; 52:

383–390.12. Foo T, Reagan J, Watson JT, Moed BR and Zhang Z.

External fixation of femoral defects in athymic rats:

applications for human stem cell implantation and bone

regeneration. J Tissue Eng 2013; 4: 2041731413486368.13. Histing T, Garcia P, Holstein JH, et al. Small animal

bone healing models: standards, tips, and pitfalls results

of a consensus meeting. Bone 2011; 49: 591–599.14. An Y, Friedman RJ, Parent T and Draughn RA.

Production of a standard closed fracture in the rat

tibia. J Orthop Trauma 1994; 8: 111–115.

15. Garcia P, Histing T, Holstein JH, et al. Rodent animal

models of delayed bone healing and non-union forma-

tion: a comprehensive review. Eur Cell Mater 2013; 26:

1–12; discussion 12–14.

16. Mehta M, Schell H, Schwarz C, et al. A 5-mm femoral

defect in female but not in male rats leads to a reprodu-

cible atrophic non-union. Arch Orthop Trauma Surg

2011; 131: 121–129.17. Schlundt C, El Khassawna T, Serra A, et al.

Macrophages in bone fracture healing: their essential

role in endochondral ossification. Bone, Epub ahead of

print 31 October 2015. DOI: 10.1016/j.bone.2015.10.019.18. Guo S and Dipietro LA. Factors affecting wound heal-

ing. J Dent Res 2010; 89: 219–229.19. Birkhold AI, Razi H, Duda GN, Weinkamer R, Checa S

and Willie BM. Mineralizing surface is the main target of

mechanical stimulation independent of age: 3D dynamic

in vivo morphometry. Bone 2014; 66: 15–25.20. Schwarz C, Wulsten D, Ellinghaus A, Lienau J, Willie

BM and Duda GN. Mechanical load modulates the

stimulatory effect of BMP2 in a rat nonunion model.

Tissue Eng Part A 2013; 19: 247–254.21. Auer JA, Goodship A, Arnoczky S, et al. Refining animal

models in fracture research: seeking consensus in optimis-

ing both animal welfare and scientific validity for appro-

priate biomedical use. BMC Musculoskelet Disord 2007;

8: 72.22. Firth AM and Haldane SL. Development of a scale to

evaluate postoperative pain in dogs. J Am Vet Med Assoc

1999; 214: 651–659.23. Graham MJ, Kent JE and Molony V. The influence of

the site of application on the behavioural responses of

lambs to tail docking by rubber ring. Vet J 2002; 164:

240–243.24. Jirkof P, Fleischmann T, Cesarovic N, Rettich A, Vogel J

and Arras M. Assessment of postsurgical distress and

pain in laboratory mice by nest complexity scoring. Lab

Anim 2013; 47: 153–161.25. Arras M, Glauser DL, Jirkof P, et al. Multiparameter

telemetry as a sensitive screening method to detect vac-

cine reactogenicity in mice. PLoS One 2012; 7: e29726.

26. Arras M, Rettich A, Cinelli P, Kasermann HP and Burki

K. Assessment of post-laparotomy pain in laboratory

mice by telemetric recording of heart rate and heart

rate variability. BMC Vet Res 2007; 3: 16.27. Burkholder T, Foltz C, Karlsson E, Linton CG and

Smith JM. Health evaluation of experimental laboratory

mice. Curr Protoc Mouse Biol 2012; 2: 145–165.

28. Ullman-Cullere MH and Foltz CJ. Body condition scor-

ing: a rapid and accurate method for assessing health

status in mice. Lab Anim Sci 1999; 49: 319–323.29. Freeman KT, Koewler NJ, Jimenez-Andrade JM, et al. A

fracture pain model in the rat: adaptation of a closed

femur fracture model to study skeletal pain.

Anesthesiology 2008; 108: 473–483.30. Minville V, Laffosse JM, Fourcade O, Girolami JP and

Tack I. Mouse model of fracture pain. Anesthesiology

2008; 108: 467–472.

31. Sabino MA, Luger NM, Mach DB, Rogers SD, Schwei

MJ and Mantyh PW. Different tumors in bone each give

rise to a distinct pattern of skeletal destruction, bone

cancer-related pain behaviors and neurochemical changes


in the central nervous system. Int J Cancer 2003; 104:550–558.

32. Minville V, Fourcade O, Mazoit JX, Girolami JP and

Tack I. Ondansetron does not block paracetamol-induced analgesia in a mouse model of fracture pain. BrJ Anaesth 2011; 106: 112–118.

33. Piel MJ, Kroin JS, van Wijnen AJ, Kc R and Im HJ. Pain

assessment in animal models of osteoarthritis. Gene 2014;537: 184–188.

34. Houghton AK, Hewitt E and Westlund KN. Enhanced

withdrawal responses to mechanical and thermal stimuliafter bone injury. Pain 1997; 73: 325–337.

35. Jimenez-Andrade JM, Martin CD, Koewler NJ, et al.

Nerve growth factor sequestering therapy attenuatesnon-malignant skeletal pain following fracture. Pain2007; 133: 183–196.

36. Jirkof P, Leucht K, Cesarovic N, et al. Burrowing is asensitive behavioural assay for monitoring general well-being during dextran sulfate sodium colitis in laboratorymice. Lab Anim 2013; 47: 274–283.

37. Hedenqvist P, Roughan JV and Flecknell PA. Effects ofrepeated anaesthesia with ketamine/medetomidine and ofpre-anaesthetic administration of buprenorphine in rats.

Lab Anim 2000; 34: 207–211.38. Santos M, Kuncar V, Martinez-Taboada F and Tendillo

FJ. Large concentrations of nitrous oxide decrease the

isoflurane minimum alveolar concentration sparingeffect of morphine in the rat. Anesth Analg 2005; 100:404–408.

39. Radi ZA and Khan NK. Effects of cyclooxygenase inhib-

ition on bone, tendon, and ligament healing. Inflamm Res2005; 54: 358–366.

40. Brewster D, Humphrey MJ and McLeavy MA. The sys-

temic bioavailability of buprenorphine by various routesof administration. J Pharm Pharmacol 1981; 33: 500–506.

41. Blaha MD and Leon LR. Effects of indomethacin and

buprenorphine analgesia on the postoperative recovery ofmice. J Am Assoc Lab Anim Sci 2008; 47: 8–19.

42. Bomzon A. Are repeated doses of buprenorphine detri-

mental to postoperative recovery after laparotomy inrats? Comp Med 2006; 56: 114–118.

43. Cowan A. Buprenorphine: the basic pharmacology revis-ited. J Addict Med 2007; 1: 68–72.

44. Roughan JV and Flecknell PA. Buprenorphine: a reap-praisal of its antinociceptive effects and therapeutic use inalleviating post-operative pain in animals. Lab Anim

2002; 36: 322–343.45. Carbone ET, Lindstrom KE, Diep S and Carbone L.

Duration of action of sustained-release buprenorphine

in 2 strains of mice. J Am Assoc Lab Anim Sci 2012; 51:815–819.

46. Matsumiya LC, Sorge RE, Sotocinal SG, et al. Using themouse grimace scale to reevaluate the efficacy of post-

operative analgesics in laboratory mice. J Am AssocLab Anim Sci 2012; 51: 42–49.

47. Yu G, Yue YJ, Cui MX and Gong ZH. Thienorphine is a

potent long-acting partial opioid agonist: a comparativestudy with buprenorphine. J Pharmacol Exp Therapeut2006; 318: 282–287.

48. Khan SN, DuRaine G, Virk SS, et al. The temporal roleof leptin within fracture healing and the effect of localapplication of recombinant leptin on fracture healing.

J Orthop Trauma 2013; 27: 656–662.49. Swenson J, Olgun S, Radjavi A, Kaur T and Reilly CM.

Clinical efficacy of buprenorphine to minimize distress inMRL/lpr mice. Eur J Pharmacol 2007; 567: 67–76.

50. Jirkof P, Tourvieille A, Cinelli P and Arras M.Buprenorphine for pain relief in mice: repeated injectionsvs sustained-release depot formulation. Lab Anim 2015;

49: 177–187.51. Matthiesen T, Wohrmann T, Coogan TP and Uragg H.

The experimental toxicology of tramadol: an overview.

Toxicol Lett 1998; 95: 63–71.

52. Committee on Anaesthesia. Expert Information: Painmanagement for laboratory animals. GV–SOLAS, 2010.

53. Committee on Anaesthesia. Expert Information: Painmanagement for laboratory animals. GV–SOLAS, 2015.

54. Luger NM, Sabino MA, Schwei MJ, et al. Efficacy ofsystemic morphine suggests a fundamental difference in

the mechanisms that generate bone cancer vs inflamma-tory pain. Pain 2002; 99: 397–406.

55. Sandberg O and Aspenberg P. Different effects of indo-

methacin on healing of shaft and metaphyseal fractures.Acta Orthop 2015; 86: 243–247.

56. Ehrnthaller C, Huber-Lang M, Nilsson P, et al.

Complement C3 and C5 deficiency affects fracture heal-ing. PLoS One 2013; 8: e81341.

57. Haffner-Luntzer M, Heilmann A, Rapp AE, et al.Midkine-deficiency delays chondrogenesis during the

early phase of fracture healing in mice. PLoS One 2014;9: e116282.

58. Heilmann A, Schinke T, Bindl R, et al. Systemic treat-

ment with the sphingosine-1-phosphate analog FTY720does not improve fracture healing in mice. J Orthop Res2013; 31: 1845–1850.

59. Heilmann A, Schinke T, Bindl R, et al. The Wnt serpen-tine receptor Frizzled-9 regulates new bone formation infracture healing. PLoS One 2013; 8: e84232.

60. Wehrle E, Wehner T, Heilmann A, et al. Distinct fre-quency dependent effects of whole-body vibration onnon-fractured bone and fracture healing in mice.J Orthop Res 2014; 32: 1006–1013.

61. Freund J, Brandmaier AM, Lewejohann L, et al.Emergence of individuality in genetically identical mice.Science 2013; 340: 756–759.

62. Hess SE, Rohr S, Dufour BD, Gaskill BN, Pajor EA andGarner JP. Home improvement: C57BL/6J mice givenmore naturalistic nesting materials build better nests.

J Am Assoc Lab Anim Sci 2008; 47: 25–31.63. Van de Weerd HA, Van Loo PL, Van Zutphen LF,

Koolhaas JM and Baumans V. Preferences for nestingmaterial as environmental enrichment for laboratory

mice. Lab Anim 1997; 31: 133–143.64. Simmons JP, Nelson LD and Simonsohn U. False-posi-

tive psychology: undisclosed flexibility in data collection

and analysis allows presenting anything as significant.Psychol Sci 2011; 22: 1359–1366.

Lang et al. 441

ab oratory

l i m i t e d

lan imals

Special Article

Severity assessment and scoring forneurosurgical models in rodents

Sarah Pinkernell, Katrin Becker and Ute Lindauer

AbstractThe most important acute neurological diseases seen at neurosurgery departments are traumatic braininjuries (TBI) and subarachnoid hemorrhages (SAH). In both diseases the pathophysiological sequela arecomplex and have not been fully understood up to now, and rodent models using rats and mice are mostsuitable for the investigation of the pathophysiological details. In both models, surgery is performed underanesthesia, followed by assessment of their functional outcome and behavioral testing before brain tissueanalysis after euthanasia. Postoperative analgesia is mandatory, and supplementary care is highly recom-mended for refinement purposes. Pain and stress assessment is mainly based on clinical and behavioralsigns, and further research is needed to improve the evaluation of severity in these models.

Keywordspain, rodent, severity, subarachnoid hemorrhage, traumatic brain injury

Besides spinal injury, the most important acute neuro-logical diseases that neurosurgery departments have todeal with are traumatic brain injuries (TBI) and hemor-rhaging into the subarachnoid space from acuteruptures of cerebral aneurysms (aneurysm-induced sub-arachnoid hemorrhages, SAH). In both diseases thepathophysiological sequela are complex and have notbeen fully understood up to now.1–3 For targeted ther-apy we have to remedy this lack of knowledge by thor-oughly investigating the pathophysiological cascadesusing well characterized animal models that mimickthe human diseases in total or in well-defined separatecomponents. The most commonly used species for TBIand SAH research are small rodents such as rats andmice,4 with rats being used more often in SAH studiesthan mice (evaluation up to 20075). We will thereforefocus our review on the severity assessment and scoringfor TBI and SAH models in rats and mice.

Animal models of TBI and SAH

Both diseases are characterized by an initial impact tothe cerebral tissue due to a physical insult (TBI) or amassive bleeding from a ruptured artery (SAH), andthe development of secondary damage which evolves

over hours and days or even weeks after the initialinsult, leading to the growth of a lesion. Dependingon the area of the damaged tissue in the brain, specificfunctional impairment occurs. To evaluate the degreeof cognitive, sensory or motor deficits, a battery offunctional tests are available which have been estab-lished in small rodents. By using the modified neurolo-gical severity score, the grade of impairment can bequantitatively assessed. Both basic reflexes (pinnareflex and corneal reflex) and somatomotor functions(paw flexion, head support, spontaneous and initiatedlocomotion), as well as postural responses and startleresponses are usually tested in mice and rats.6,7 In addi-tion to these simple tests, more specific assessments arepossible, though these are highly demanding of timeand experience (a review and descriptions of thesetests are beyond the scope of this paper but are

Translational Neurosurgery, Medical Faculty, RWTH AachenUniversity, Aachen, Germany

Corresponding author:Prof. Dr Ute Lindauer, Translational Neurosurgery, MedicalFaculty, RWTH Aachen University, Pauwelsstr. 30, 52074 Aachen,Germany.Email: [email protected]

Laboratory Animals

2016, Vol. 50(6) 442–452



sagepub.co.uk/


DOI: 10.1177/0023677216675010

la.sagepub.com

provided in detail elsewhere8–13). As a final outcomemeasure, the brains are usually analyzed by histomor-phological and molecular biological methods in detailafter sacrifice of the animals.

Induction of TBI

TBI is the leading cause of mortality and disability inthe younger population (up to the age of 45) world-wide.14 Within the pathophysiological course of TBI,we can differentiate between the primary damage,which rapidly evolves as a direct result of the physicalimpact on brain tissue, and secondary damage, whichenlarges the initial injury to the tissue.15

Several injury models have been used in TBIresearch, and detailed descriptions of the models mostoften applied in rodents and other species have recentlybeen published.4,8,16 Based on the information pro-vided, we have summarized important aspects of themost commonly used TBI models in Table 1a.17–22

Induction of SAH

In men, aneurysmal SAH is accompanied by a high riskof mortality immediately after or within the first fewhours after the initial bleeding. The early injury ismainly caused by a dramatic transient increase in intra-cranial pressure (ICP) followed by global cerebralischemia,23,24 and by the impact of the blood clot (indu-cing inflammation, vascular endothelial cell damage,etc.).9 Up to 30% of patients who survive the initialevent develop delayed neurological deficits caused bydelayed ischemic events.25 The early brain injury isnot only responsible for the early symptoms of the dis-ease but is also the starting point for the delayed dete-rioration.26,27 Two main approaches have been used inSAH model experiments using mice and rats which arecompiled in Table 1b.5,9,28–35

Additional procedure (both TBI and SAH)

Despite TBI or SAH in humans occurring while theyare awake, anesthesia is obligatory when performingTBI or SAH surgery in animals.

In both models, during the initial surgery (SAH) and/or at different time points after the insult (SAH, TBI), itis general practice to measure the ICP as an importantparameter of disease severity.36 This can be achieved byinsertion of a cannula in the cisterna magna connectedto a pressure transducer, or by using fiber-optic-basedsystems, where a small pressure transducer is placedbetween the skull and the dura mater. In both cases, asmall craniotomy has to be performed.

Anesthesia and surgical procedures may inducehypothermia in rodents, not only during the surgery

but also within hours after awakening. However nor-mothermia has to be ensured, since hypothermia pro-vides neuroprotection,37,38 and may also impair animalwelfare.39

Reasons for choosing the rodent model forTBI and SAH

Most TBI and SAH animal models were first estab-lished in larger animal species, however in recent dec-ades rodent models have most often been used. Besidesthe advantages of low cost (purchase and housing), easyhandling and high level of standardization, rodentsprovide the possibility of advanced assessment of func-tional outcome and analysis of the underlying complexmolecular cascades of cellular damage in the tissue aftereuthanasia. In addition genetically-modified mousemodels can advance our understanding of molecularpathways.40 As suggested in Tables 1a and 1b, thevalidity in all three common categories (construct,face and predictive) is not perfect for either TBI orSAH. This is not surprising since TBI in humans is ahighly heterogeneous and complicated disease. Thesame is true for SAH, where size and location of theaneurysm, and the amount of blood and pressure underwhich it is released during rupture, depend on eachindividual situation. In addition, comorbidities (suchas systemic hypertension or diabetes) make the situa-tion even more complex. It is therefore not possible tomodel all aspects of the diseases in a single animalmodel,4,16 and the selection of the model has to bebased on a clear definition of the respective aim to befollowed and hypothesis to be proven. Accordinglyrodent models for TBI and SAH are of extraordinaryhigh value, and have remained indispensable toadvancement of our yet incomplete understanding ofthe pathophysiological cascades in TBI and SAH, byelucidating important parts of the complex puzzle.

Current knowledge shows that the pathophysiologi-cal events in TBI and SAH are substantially compar-able between small rodents and humans; yet the humanbrain is structurally and functionally more complex,with significant differences in the white-to-grey matterratio. Therefore, in order to close the translational gapfor preclinical drug development, studies in largeranimal species with brains more closely resemblingthose of humans are recommended in order to makecomprehensive use of the findings on pathophysiologi-cal cascades achieved in rodent studies.

Severity of the model

As stated in the ‘Working Document on a SeverityAssessment Framework’,41 a standard procedureshould be followed for severity assessment, based on

Pinkernell et al. 443

Table

1a.Most

commonly

use

dtraumaticbrain

injury

(TBI)models

inrats

andmice.

TBImodel

Establish

ed

for

Main

proce

dure

,wayofim

pact

Reproducibility/

controllabilityof

impact

and

tiss

uedeform

ation

Foca

linjury

Diffuse

/axonal

injury

Ass

ess

mentaccordingto

a)Etiology(construct

validity)

b)Biologicalmarkers

(face

validity)

c)Treatm

entresp

onse

(predictive

validity)

Adva

ntage

Disadva

ntage

Fluid

percuss

ion

injury

model

(FPI)17,18

Rat,mouse

Craniotomy,

dura

materintact,

fluid

wave

Moderate

reproducibility:

controlofparameters

ofinitiationoffluid

wave,variabilityby

loca

tionofcraniotomy

anddispersionof

fluid

wave

33

a)Brain

disloca

tionco

mpar-

able

tothesituationafter

fallsin

humans

b)Foca

lco

rticalco

ntusion

withaxo

nalinjury,acu

teapnea,acu

teandch

ronic

seizures,

functionaldeficit/

behavioralim

pairment

c)Goodtreatm

entresp

onse

a)Craniotomy

b)Noregularbrainstem

damage–lack

ofunco

n-

sciousn

ess

,mortality

variable

(low

tohigh)

dependingonthedegree

ofpress

ure

pulse,rarely

monitoringofvital

physiologicalparameters

perform

ed

c)Translationinto

clinical

trials

failed

Controlledco

rtical

impact

(CCI)19,20

Rat,mouse

Craniotomy,

dura

materintact,

pneumaticdevice

Highreproducibility:

controloftimeof

compress

ion,design

andvelocity

of

impactor,

depth

ofim

pact

3–(only

minor)

b)Corticalco

ntusion,lesion

enlargementoverdays,

functionaldeficit/beha-

vioralim

pairment

c)Goodtreatm

entresp

onse

a)Craniotomy,

only

foca

lim

pact

b)Noaxo

nalinjury,large

contusioninjury,no

brainstem

damage–lack

of

unco

nsciousn

ess

,very

low

mortality,rarely

monitor-

ingofvitalphysiological

parameters

perform

ed

c)Translationinto

clinical

trials

failed

Weightdrop

impact

acceleration

model(W

DIA)

(‘Marm

arou

model)21,22

Rat,(m

ouse

only

as

repeatmild

model)

Skullintact,weight

droponmetal

diskfixe

dto

the

skull,headnot

fixe

dandlying

onacu

shion

withsp

ringforce

Moderate

reproducibility:

controlofweightdrop,

acceleration/

dece

lerationof

headnotco

ntrollable

–3

a)Strongdiffuse

impact

b)Brainstem

damage–

prolongedunco

nscious-

ness

,highermortality,

functionaldeficit/beha-

vioralim

pairment

b)Nofoca

llesionorce

lldeath

c)Modelrarely

use

din

pharm

aco

logicalevaluation


Table

1b.Most

commonly

use

dsu

barach

noid

hemorrhage(SAH)models

inrats

andmice.

SAH

model

Establish

ed

for

Main

proce

dure

Reproducibility/

controllabilityof

bleedingse

verity/

amountofblood

Early

injury

Delaye

dva

sosp

asm

/delaye

dneurological

deficits

Ass

ess

mentaccordingto

a)Etiology(construct

validity)

b)Biologicalmarkers

(face

validity)

c)Treatm

entresp

onse

(predictive

validity)

Adva

ntage

Disadva

ntage

Endovascular

perforation

model28–32

Rat,mouse

Smallfilamentorwire

introduce

dinto

theICA

(via

theCCAorECA)and

advance

dintracranially

toperforate

theCircle

ofWillisattheregionof

theanteriorcirculation

atthebifurcationofthe

anteriorce

rebralartery

andthemiddle

cerebral

artery

Moderate

reproducibility:

only

bydesignofthe

filamenttip/w

iresize,

variabilityco

nce

rning

exact

perforation

loca

tionandamount

ofbloodpouringout

3Vaso

spasm

oflarge

arteries3–7days

afterSAH/delayed

neurologicaldeficit

orlate

deterioration

notsystematica

lly

reported

a)Close

lyrese

mblesthe

situationofsp

ontaneous

rupture

oface

rebral

aneurysm

,distributionof

bloodin

thesu

barach

noid

space

b)Largeandrapid

increase

inICPwithtransientglobal

isch

emia;earlyco

nstriction

ofmicrocirculation;

inflammation,vascular

endothelialce

lldamage;

BBB

damage,parench

ymal

celldeath,functionaldefi-

cit/behavioralim

pairment,

mortality

comparable

tohumansituation

c)Goodtreatm

entresp

onse

a)Endovascularmanipu-

lationbythedevice

c)Translationinto

clinical

trials

failed

Bloodinjection

modelin

cisternamagna

orprech

iasm

atic

cistern

(only

once

or

repeated)9,28,29,33–35

Rat,mouse

Smallcraniotomy,

puncture

ofthecisterna

magnaorneedle

place

din

theprech

iasm

atic

cistern

Moderate

tohigh

reproducibility:

control

ofamountofblood

injectedandinjection

press

ure/velocity,

variabilityco

nce

rning

thedistributionofblood

within

thesu

barach

noid

space

3Vaso

spasm

oflarge

arteries3–7days

afterSAH/delayed

neurologicaldeficit

orlate

deterioration

notsystematica

lly

reported

a)Distributionofbloodin

the

subarach

noid

space

b)Largeandrapid

increase

inICPwithtransientglobal

isch

emia;earlyco

nstriction

ofmicrocirculation;

inflammation,vascular

endothelialce

lldamage;

BBB

damage,parench

ymal

celldeath,functionaldefi-

cit/behavioralim

pairment,

mortality

dependson

amountofbloodinjected–

inmore

severe

models

comparable

tohuman

situation

c)Goodtreatm

entresp

onse

a)Craniotomy,

norupture

oface

rebralvess

el–

bloodartificially

reach

essu

barach

noid

space

bycisternal

injection

c)Translationinto

clinical

trials

failed

BBB:blood–brain

barrier;

CCA:co

mmonca

rotidartery;ECA:externalca

rotidartery;ICA:internalca

rotidartery;ICP:intracranialpress

ure.


the thorough consideration of each step within themodel procedure, to result in an assessment tailoredfor each respective species and strain, and for the indi-vidual laboratory situation. Therefore we can only out-line the most basic considerations here and not refer todetails within the models and species.

In both TBI and SAH, ICP increase occurs acutelyor develops within hours. To measure ICP, which ishighly recommended, a small craniotomy has to beperformed. In addition, the skin and the underlyingsoft tissue at the head and/or the neck have to be cut.The brain itself has no pain receptors, but the neuro-cranium with its overlying periost and the meningesthat are mainly affected during physical impact, and/or the elevated ICP, are highly algesic. The level of paininduced by the increased ICP correlates with the degreeof the physical impact in TBI (which can be standar-dized) or the amount of blood in SAH (which can bestandardized only in the blood injection model). Afterawaking, animals develop neurological deficits, whichmay reduce their ability to take food and water (withconsequent risk of weight loss and dehydration) andinhibit normal behavior within their social environ-ment. Assessment of functional deficits will probablyalso be associated with stress for the animals affectedby the disease, particularly if they are not used to thetests. In addition, rodents are by nature prey animals,and as such fugitive,42 and it is not known whetherbeing unable to escape due to motor and/or cognitivedeficits itself induces stress in rats and mice. Post-sur-gical housing is typically individual housing, so as toavoid wound manipulation by cage mates; but isolationcan also be stressful for social species.43,44

Mortality in the controlled cortical impact (CCI)–TBI model is usually low (<10%, while controlled bythe degree of impact) because no damage to the brain-stem occurs. Mortality in fluid percussion injury (FPI)models depends strongly on the size of the pressurepulse, with mild to moderately severe impact resultingin mortality rates of 10–22% in rats17 and zero mortal-ity in mice.18 In the ‘Marmarou’ (weight drop impactacceleration [WDIA]) model, the originally describedmortality of up to 60% in spontaneously breathingrats is reduced to <10% by intubation and mechani-cally-assisted ventilation, which is recommended as thestandard procedure for this model.22 Very recently, theWDIA rat model has been redesigned for applicationin mice as a repeat mild TBI model, with zeromortality.45

In SAH, mortality depends strongly on the model. Itcan reach up to 60% in the endovascular perforationmodel in rats,9,46 while most of the animals usually dieimmediately after the perforation (while still underanesthesia) or within the first few hours after the bleed-ing,46 which resembles the early mortality rate in

human patients. Recently, a modification of the endo-vascular perforation model in rats has been described,using a tungsten wire instead of a polypropylene fila-ment, which induces a vessel perforation with a smallerdiameter, leading to reduced mortality rates of around20%.47–49 In mice, the endovascular perforation modelis accompanied by a mortality rate of around 30%.5,30

In the blood injection models, mortality depends on theamount of blood injected, and usually remains �10%.5

In summary, besides ensuring a thorough assessmentof pain and distress, the definition of humane endpointsis especially necessary when using the more clinicallyrelated models of TBI and SAH. Overall, based onthe cumulative consideration of the above-mentionedaspects, severity is usually graded as moderate forTBI (depending on the degree of physical impact) aswell as for SAH when using the modified versions ofthe models with reduced mortality rates, and whenimplementing all the refinement opportunities men-tioned in the text and table 3 below.


Expected signs of pain, stress, disease,constraint or signs of reduced welfare

In rats and mice, pain and suffering cannot be easilydetected. Generally, the burden induced by the pre-and postoperative handling, the surgery and thelocal effect of the lesion (most important of which isthe increase in ICP) on the one hand, and the impair-ment of normal behavior caused by the lesion-inducedfunctional deficit on the other hand, cannot be distin-guished by common measures of reduced welfare.Therefore, up to now, signs of reduced well-being inTBI and SAH studies have had to be considered asunspecific as well as model-specific. Rats and miceare social animals, and reduced welfare rapidly leadsto reduced spontaneous activity and isolation of theanimals from their cage mates, which can be easilydetected by observing the animals in their homecages. Furthermore, reduced activity can be tested bystimulating movement with a gentle push. Unusualbody posture (crooked back) or an ungroomedappearance indicate that the animal is reluctant tomove. Following TBI and SAH, any disregard ofgrooming (resulting in porphyrin deposits around theeyes and nose in rats) may be a sign of discomfort aswell as the result of functional impairment of the pawsor a deficit in motor coordination. Unusually aggres-sive behavior is another, yet more unspecific, sign ofreduced well-being or stress and pain. Another impor-tant marker of constraint, and possibly the mostimportant and scalable one, is reduced body weight,which is measured by weighing the animals daily. This


indicates reduced food and water intake, which iscaused by both unspecific illness as well as by func-tional impairment. More model-specific symptoms arethe development of convulsions and epilepticseizures, which add to the burden for the animals.Isolation and sitting with the head towards thecorner of the cage may be signs of head pain causedby increased ICP.

All these signs can be monitored by clinical exam-ination. Routine home cage observation is thereforeobligatory for welfare assessment, which howeverrequires thorough knowledge of the normal appear-ance and behavior of the respective species. Strain-dependent characteristics also have to beconsidered. Besides body weight, these measures arenot scalable and are, at least to some extent, observer-dependent.

This basic clinical observation and scoring systemhas been developed according to the standard proce-dure defined in the ‘Working Document on a SeverityAssessment Framework’,41 and is frequently used andaccepted as a routine procedure for experimental stu-dies of TBI and SAH. The main disadvantage of clin-ical scoring systems is the lack of scalability and thenecessity for the observer to be sufficiently experienced;and it is not clear to what extent minor stress and painor even subtle deviations from well-being can bedetected using these tests. In addition, illness-relateddeviation from normal behavior cannot be discrimi-nated from functional impairment-induced deficits,and it is not known whether (and if so, to whatextent) the sole occurrence of functional impairmentaffects the subjective well-being of the animals.Further research is needed on this issue.

The same is true for more advanced tests assessingspontaneous locomotion (for example from home cagevideo recording over longer time periods or perfor-mance of home cage wheel running), which are increas-ingly being used for assessing pain in rodent models.50

Overall, when applying tests requiring normal sensor-imotor and cognitive functions, it must be thoroughlyinvestigated whether abnormal test results are indeedinduced by pain and stress, or rather by cognitive, sen-sory or motor impairment induced by the brain lesionfollowing TBI or SAH, which could be independent ofpain and stress.

Score sheet

In the forefront of an experiment a score sheet based onexpected signs of reduced well-being has to be devel-oped. In Table 2, a typical score sheet is provided,which has been developed for SAH or TBI studiesand is relevant to all the main procedures (see above).It is equally applicable for rats and mice, however,

strain differences have to be considered, mainly con-cerning the overall level of normal activity andaggression.

Assessment of the ‘General condition, physicalappearance/Breathing/Spontaneous behavior/Reactionto handling/Wound healing, condition of suture’topics and of body weight is standard practice for anexperienced staff member familiar with normal rodentbehavior.

The simple neurological evaluation can be brieflyexplained. It serves as a rough measure of the lesion-induced functional deficit, which may lead to impairedfood and water intake and to deficits in grooming andactivity, and therefore adds to the summed score. Toevaluate forelimb flexion, the animals are carefullylifted by the tail with support of the lower body andadvanced to the surface of a table. Healthy animalssymmetrically stretch their limbs towards the ground,and animals with deficits in motor function and coor-dination of the limbs show asymmetric stretching orfailure of stretching. Asymmetric limb function alsoleads to circular locomotion instead of straightmovement.

Humane endpoints

In TBI and SAH models, pain or distress are unavoid-able, therefore humane endpoints must be obligatorilydefined by an upper limit of maximally tolerable scorepoints. Close-mesh monitoring of the animals through-out the experiment is mandatory for checking if theendpoint is reached, and reaching the humane endpointscore in TBI and SAH studies leads to euthanasia of theanimal.

In TBI and SAH models, the most likely problemsexpected to lead to endpoint scoring are:

. weight loss;

. pain due to headache (craniotomy/ICP increase) orsurgical manipulation;

. infection of surgical wound;

. immobility.

Thus, the most important symptoms, any one ofwhich can by itself lead to euthanasia, are:

. weight loss exceeding >20%;

. extreme pain which cannot be relieved by analgesics;

. immobility >12 h (assessed by close-mesh observa-tion in the home cage);

. status epilepticus, coma, complete paralysis.

In addition, a combination of several signs of painand stress may also lead to euthanasia, any one ofwhich on its own may not require euthanasia.


Table 2. Score sheet for traumatic brain injury (TBI) and subarachnoid hemorrhage (SAH).

Parameter Description Points

Loss of body weight 0–5% 0

5–10% 1

11–15% 5

16–<20% 10

20þ% 30

General condition,physical appearance

Well groomed; clean fur; body openings clean;clear, shiny eyes

0

Coat slightly unkempt; chromodacryorrhea (‘red tears’) 1

Coat shaggy and dirty; signs of dehydration(no relief of skin fold); moderate piloerection

5

Coat unkempt; marked piloerection 10

Breathing Normal breathing 0

Tachypnea 5

Dyspnea 10

Spontaneous behavior Normal locomotion, social contact 0

Reluctance to move, slightly abnormal gait 5

Lethargy, apathy, separation from cage mates,markedly abnormal gait

10

Significant mobility problems, intermittentimmobility <12 h

20

Immobility >12 h 30

Reaction to handling Normal curiosity, alert 0

Tense and nervous on handling 5

Markedly distressed on handling, e.g. shaking,vocalizing, aggressive

10

Wound healing,condition of suture

Wound clear and inconspicuous, no signs of infection,all sutures in place

0

Suture insufficiency, wound dehiscence withoutsigns of infection

5

Swelling, redness as first signs of infection 10

Massive infection, swelling, redness, purulent discharge 25

Simple neurologicevaluation

No observable deficit 0

Forelimb flexion 1

Decreased resistance to lateral push(and forelimb flexion) without circling

5

Decreased resistance to lateral push(and forelimb flexion) with circling

10

Seizure (status epilepticus), coma, complete paralysis 30

Actions: Score 0–9: frequency of monitoring as planned, in case of suture insufficiency: suturerevision

Score 10–20: provide supplementary care (e.g. extra fluids, wet mash), check analgesia protocol,increase monitoring frequency, in case of signs of wound infection: consultveterinarian

Score 21–29: review progress with veterinarianScore �30: implement humane endpoint.


Table 3. Refinement opportunities.

Subject Description Refinement To be considered

Anesthesia duringsurgery

Most commonly usedanesthesia:

Isoflurane52 þ fentanyl orbuprenorphine;

Ketamine þ xylazine;Midazolam þ medetomidine

þ fentanyl53

Isoflurane alone has onlylittle or no analgesia effect! add analgesic(fentanyl,54 buprenor-phine,55 ketamine or atleast local anesthetics56)

Effect of anesthetics on brainfunction: neuroprotection byreduction of spontaneous andevoked activity (all anes-thetics) and/or increase ofcerebral blood flow(isoflurane), or inhibition ofglutamate receptors(ketamine) (see review52)

Support of vitalfunctionduring surgery

Anesthesia incidents:Apnea in more severe SAH

models and TBI modelswith brainstem damage(WDIA);

Reduced breathing – hypoxia;Anesthesia-inducedhypothermia

Monitor vital parameters bypulse-oximetry and endexpiratory–CO2 monitorand adjust anesthesia levelaccordingly;30

Intubation and mechanicalventilation;22,30

Add oxygen to inspiration –air;

Monitor body temperatureand use a servo-controlledheating pad

Use non-invasive approaches tomeasure vital parameters;

Do not apply too lowconcentrations of anesthetics;

Do not apply to high levels ofoxygen;

Check proper function of heatingpad to avoid overheating

Postoperativeanalgesia

Pain reducing strategies areobligatory and highlyrecommended

Opioids more appropriatethan NSAIDs and thereforerecommended for post-operative pain reduction;sustained–release opioidspreferable due to avoid-ance of frequent handlingfor injection

Choice of the appropriateanalgesic not easy: inflam-mation is part of the patho-physiological cascade in TBIand SAH, and NSAIDs aretherefore not suitable due totheir COX-2 inhibition-mediated anti-inflammatoryeffect; opioids can modify thefunctional outcome52

Postoperativecare – bodytemperature

Hypothermia occurs duringand following anesthesiaand due to reduced activity,but has to be avoided notonly during but also aftersurgery due to its neuro-protective effect and forreasons of animal welfare

Keeping the animals in smallheating cabinets with con-trolled inside temperaturefor several hours after fin-ishing the surgery as bestchoice for body tempera-ture control in smallrodents

Check proper function andestablish correct heatingtemperature before first use;

Red-light warming lamps notrecommended/should beavoided due to inaccuratesetting of the temperature anddue to high temperaturegradient

Postoperativecare –food andwater intake

Animals often are too weakafter surgery for appropri-ate food and water intake,animals might avoid chew-ing hard rodent chow dueto headache;

Straightening up to the waterbottle might be hindered bypain at surgical wound andby neurological impairment

Offering water-gel pads andmashed food57 in a Petri-dish on the floor to encou-rage water and food intake;

Subcutaneous injections ofsaline or saline enrichedwith carbohydrates beforeawakening and daily duringrecovery phase58

Regular and frequent change ofdish with mashed food due tocontamination with beddingand to avoid mold formation;

Subcutaneous injections mayinduce stress

Handling duringlab routineand functionaltesting

Handling during housing careand laboratory routine59

and during functional testsmay induce stress, whichshould be minimized

Preoperative habituation ofthe animals to handlingand to the testsconditions60

Stress reducing effect ofhabituation to handling isless clear in mice61

SAH: subarachnoid hemorrhage, TBI: traumatic brain injury, WDIA: weight drop impact acceleration, NSAIDs: non-steroidalanti-inflammatory drugs, COX-2: cyclooxygenase isoform 2.



According to the ‘ARRIVE – Animals in Research:Reporting In Vivo Experiments’ guidelines,51 all rele-vant refinement strategies should be described in theMethods sections when publishing results. In additionit should be kept in mind that every strategy to reduce(avoidable) stress in animal experiments is not only amatter of animal welfare but also leads to more stableand reliable results.

Within recent years, several strategies for refinementhave been developed. Model-specific refinements aimedat reducing mortality rates have been described above.The general refinement of the effective training of sur-geons should be ensured. Further opportunities forrefinement have been identified in the areas of anesthe-sia and postoperative analgesia, care during surgery,and postoperative care and handling,22,30,52,53 whichare summarized in Table 3.


As described above, the evaluation of pain and stressfollowing TBI and SAH is not easy and up to now hasmainly been based on the observance of behavioral andclinical signs of reduced welfare. More advanced eva-luation is not commonly performed, but furtherresearch is highly recommended for enhancing ourknowledge of suffering by the models, and for refiningthe parameters monitored within the score sheetapplied here.

By thoroughly assessing facial expressions in rats andmice, the recently developed ‘grimace scale’ can in gen-eral be used to quantify pain.62,63 Whether it can beapplied to TBI and SAH has yet to be proven.64

Measurement of hormonal parameters such as prolactinand corticosterone in the blood can be useful for stressassessment.65,66 However it has to be considered thatthese hormones underlie circadian variations and showa high variance of basal levels, and sampling itself can bestressful. For reliable assessments, basal preoperativevalues have to be compared with postoperative values,and blood sampling has to be conducted under similarcircumstances (time of day, experimenter and samplingmethod). It has to be investigated whether analysis ofcorticosterone content in feces may be a useful alterna-tive to assessment in the blood. This provides an index ofstress reaction over several hours, in contrast to bloodsampling which provides an index of stress at one briefmoment in time.67 Monitoring of ultrasonic vocalizationmay also be considered. While experiencing fear oranxiety, adult rats emit alarm calls at a frequency of22 kHz.68 Yet in a recent study the suitability of using22kHz vocalization for pain assessment has beenquestioned.69

Compiling standard operating procedures for themost common models of TBI and SAH with implemen-tation of the above-mentioned refinement strategies, asis already available for the focal cerebral ischemiamodel of filament-induced middle cerebral artery occlu-sion in the mouse,57 would further help to enhancestandardization, reproducibility, and animal welfarein these models.



article.

Funding


References

1. Pearn ML, Niesman IR, Egawa J, et al. Pathophysiology

associated with traumatic brain injury: current treatments

and potential novel therapeutics. Cell Mol Neurobiol,

Epub ahead of print xx month July 2016. DOI:

10.1007/s10571-016-0400-1.2. Sehba FA and Friedrich V. Early events after aneurysmal

subarachnoid hemorrhage. Acta Neurochir Suppl 2015;

120: 23–28.3. Kooijman E, Nijboer CH, van Velthoven CT, Kavelaars

A, Kesecioglu J and Heijnen CJ. The rodent endovascu-

lar puncture model of subarachnoid hemorrhage: mech-

anisms of brain damage and therapeutic strategies.

J Neuroinflammation 2014; 11: 2.4. Marklund N and Hillered L. Animal modelling of trau-

matic brain injury in preclinical drug development: where

do we go from here?. Br J Pharmacol 2011; 164:

1207–1229.5. Marbacher S, Fandino J and Kitchen ND. Standard

intracranial in vivo animal models of delayed cerebral

vasospasm. Br J Neurosurg 2010; 24: 415–434.

6. Chen J, Sanberg PR, Li Y, et al. Intravenous administra-

tion of human umbilical cord blood reduces behavioral

deficits after stroke in rats. Stroke 2001; 32: 2682–2688.

7. Thal SC, Sporer S, Klopotowski M, et al. Brain edema

formation and neurological impairment after subarach-

noid hemorrhage in rats. Laboratory investigation.

J Neurosurg 2009; 111: 988–994.8. Gold EM, Su D, Lopez-Velazquez L, et al. Functional

assessment of long-term deficits in rodent models of trau-

matic brain injury. Regen Med 2013; 8: 483–516.9. Jeon H, Ai J, Sabri M, et al. Neurological and neurobe-

havioral assessment of experimental subarachnoid hem-

orrhage. BMC Neurosci 2009; 10: 103.10. Schallert T. Behavioral tests for preclinical intervention

assessment. NeuroRx 2006; 3: 497–504.11. Zorner B, Filli L, Starkey ML, et al. Profiling locomotor

recovery: comprehensive quantification of impairments

after CNS damage in rodents. Nat Meth 2010; 7: 701–708.


12. Morris R. Developments of a water-maze procedure for

studying spatial learning in the rat. J Neurosci Methods

1984; 11: 47–60.13. Paul CM, Magda G and Abel S. Spatial memory: theor-

etical basis and comparative review on experimen-

tal methods in rodents. Behav Brain Res 2009; 203:

151–164.

14. Langlois JA, Marr A, Mitchko J and Johnson RL.

Tracking the silent epidemic and educating the public:

CDC’s traumatic brain injury-associated activities under

the TBI Act of 1996 and the Children’s Health Act of

2000. J Head Trauma Rehabil 2005; 20: 196–204.15. McIntosh TK, Smith DH, Meaney DF, Kotapka MJ,

Gennarelli TA and Graham DI. Neuropathological

sequelae of traumatic brain injury: relationship to neuro-

chemical and biomechanical mechanisms. Lab Invest

1996; 74: 315–342.16. Morganti-Kossmann MC, Yan E and Bye N. Animal

models of traumatic brain injury: is there an optimal

model to reproduce human brain injury in the labora-

tory?. Injury 2010; 41(Suppl. 1): S10–S13.17. Dixon CE, Lyeth BG, Povlishock JT, et al. A fluid per-

cussion model of experimental brain injury in the rat.

J Neurosurg 1987; 67: 110–119.18. Carbonell WS, Maris DO, McCall T and Grady MS.

Adaptation of the fluid percussion injury model to the

mouse. J Neurotrauma 1998; 15: 217–229.

19. Dixon CE, Clifton GL, Lighthall JW, Yaghmai AA and

Hayes RL. A controlled cortical impact model of trau-

matic brain injury in the rat. J Neurosci Methods 1991;

39: 253–262.20. Hannay HJ, Feldman Z, Phan P, et al. Validation of a

controlled cortical impact model of head injury in mice.

J Neurotrauma 1999; 16: 1103–1114.21. Foda MA and Marmarou A. A new model of diffuse

brain injury in rats. Part II: Morphological characteriza-

tion. J Neurosurg 1994; 80: 301–313.

22. Marmarou A, Foda MA, van den Brink W, Campbell J,

Kita H and Demetriadou K. A new model of diffuse

brain injury in rats. Part I: Pathophysiology and bio-

mechanics. J Neurosurg 1994; 80: 291–300.23. Bederson JB, Levy AL, Ding WH, et al. Acute vasocon-

striction after subarachnoid hemorrhage. Neurosurgery

1998; 42: 352–360. discussion 60–62.

24. Broderick JP, Brott TG, Duldner JE, Tomsick T and

Leach A. Initial and recurrent bleeding are the major

causes of death following subarachnoid hemorrhage.

Stroke 1994; 25: 1342–1347.25. Macdonald RL. Delayed neurological deterioration after

subarachnoid haemorrhage. Nat Rev Neurol 2014; 10:

44–58.26. Fujii M, Yan J, Rolland WB, Soejima Y, Caner B and

Zhang JH. Early brain injury, an evolving frontier in

subarachnoid hemorrhage research. Transl Stroke Res

2013; 4: 432–446.27. Terpolilli NA, Brem C, Buhler D and Plesnila N. Are we

barking up the wrong vessels? Cerebral microcirculation

after subarachnoid hemorrhage. Stroke 2015; 46:

3014–3019.

28. Megyesi JF, Vollrath B, Cook DA and Findlay JM.

In vivo animal models of cerebral vasospasm: a review.

Neurosurgery 2000; 46: 448–460; discussion 60–61.29. Zoerle T, Ilodigwe DC, Wan H, et al. Pharmacologic

reduction of angiographic vasospasm in experimental

subarachnoid hemorrhage: systematic review and meta-

analysis. J Cereb Blood Flow Metab 2012; 32: 1645–1658.

30. Feiler S, Friedrich B, Scholler K, Thal SC and Plesnila N.

Standardized induction of subarachnoid hemorrhage in

mice by intracranial pressure monitoring. J Neurosci

Methods 2010; 190: 164–170.31. Kamii H, Kato I, Kinouchi H, et al. Amelioration of

vasospasm after subarachnoid hemorrhage in transgenic

mice overexpressing CuZn–superoxide dismutase. Stroke

1999; 30: 867–871; discussion 72.

32. Bederson JB, Germano IM and Guarino L. Cortical

blood flow and cerebral perfusion pressure in a new non-

craniotomy model of subarachnoid hemorrhage in the

rat. Stroke 1995; 26: 1086–1091; discussion 91–92.33. Solomon RA, Antunes JL, Chen RY, Bland L and Chien

S. Decrease in cerebral blood flow in rats after experi-

mental subarachnoid hemorrhage: a new animal model.

Stroke 1985; 16: 58–64.34. Kamp MA, Dibue M, Sommer C, Steiger HJ, Schneider

T and Hanggi D. Evaluation of a murine single-blood-

injection SAH model. PLoS One 2014; 9: e114946.35. Prunell GF, Mathiesen T and Svendgaard NA. A new

experimental model in rats for study of the pathophysi-

ology of subarachnoid hemorrhage. Neuroreport 2002;

13: 2553–2556.36. Friess SH, Lapidus JB and Brody DL. Decompressive

craniectomy reduces white matter injury after controlled

cortical impact in mice. J Neurotrauma 2015; 32: 791–800.37. Clifton GL, Allen S, Barrodale P, et al. A phase II study

of moderate hypothermia in severe brain injury.

J Neurotrauma 1993; 10: 263–271. discussion 73.

38. Torok E, Klopotowski M, Trabold R, Thal SC, Plesnila

N and Scholler K. Mild hypothermia (33 degrees C)

reduces intracranial hypertension and improves func-

tional outcome after subarachnoid hemorrhage in rats.

Neurosurgery 2009; 65: 352–359. discussion 9.39. Donnerer J and Lembeck F. Different control of the

adrenocorticotropin–corticosterone response and of pro-

lactin secretion during cold stress, anesthesia, surgery,

and nicotine injection in the rat: involvement of capsai-

cin-sensitive sensory neurons. Endocrinology 1990; 126:

921–926.

40. Xiong Y, Mahmood A and Chopp M. Animal models of

traumatic brain injury. Nat Rev Neurosci 2013; 14:

128–142.41. National Competent Authorities for the implementation

of Directive 2010/63/EU on the protection of animals

used for scientific purposes. Working document on a

severity assessment framework, http://ec.europa.eu/envir-

onment/chemicals/lab_animals/pdf/Endorsed_Severity_

Assessment.pdf (2012, accessed 19 October 2016).42. Jennings M, Batchelor GR, Brain PF, et al. Refining

rodent husbandry: the mouse. Report of the Rodent

Refinement Working Party. Lab Anim 1998; 32: 233–259.


43. Spani D, Arras M, Konig B and Rulicke T. Higher heartrate of laboratory mice housed individually vs in pairs.Lab Anim 2003; 37: 54–62.

44. Wallace DL, Han MH, Graham DL, et al. CREB regu-lation of nucleus accumbens excitability mediates socialisolation-induced behavioral deficits. Nat Neurosci 2009;12: 200–209.

45. Xu L, Nguyen JV, Lehar M, et al. Repetitive mild trau-matic brain injury with impact acceleration in the mouse:multifocal axonopathy, neuroinflammation, and neuro-

degeneration in the visual system. Exp Neurol 2016;275: 436–449.

46. Lee JY, Sagher O, Keep R, Hua Y and Xi G.

Comparison of experimental rat models of early braininjury after subarachnoid hemorrhage. Neurosurgery2009; 65: 331–343. discussion 43.

47. Park IS, Meno JR, Witt CE, et al. Subarachnoid hemor-rhage model in the rat: modification of the endovascularfilament model. J Neurosci Methods 2008; 172: 195–200.

48. Greenhalgh AD, Rothwell NJ and Allan SM. An endo-

vascular perforation model of subarachnoid haemor-rhage in rat produces heterogeneous infarcts thatincrease with blood load. Transl Stroke Res 2012; 3:

164–172.49. Hollig A, Weinandy A, Nolte K, Clusmann H, Rossaint

R and Coburn M. Experimental subarachnoid hemor-

rhage in rats: comparison of two endovascular perfor-ation techniques with respect to success rate,confounding pathologies and early hippocampal tissuelesion pattern. PLoS One 2015; 10: e0123398.

50. Kandasamy R, Calsbeek JJ and Morgan MM. Homecage wheel running is an objective and clinically relevantmethod to assess inflammatory pain in male and female

rats. J Neurosci Methods 2016; 263: 115–122.51. Kilkenny C, Browne WJ, Cuthill IC, Emerson M and

Altman DG. Improving bioscience research reporting:

the ARRIVE guidelines for reporting animal research.PLoS Biol 2010; 8: e1000412.

52. Rowe RK, Harrison JL, Thomas TC, Pauly JR, Adelson

PD and Lifshitz J. Using anesthetics and analgesics inexperimental traumatic brain injury. Lab Anim (NY)2013; 42: 286–291.

53. Hockel K, Trabold R, Scholler K, Torok E and Plesnila

N. Impact of anesthesia on pathophysiology and mortal-ity following subarachnoid hemorrhage in rats. ExpTransl Stroke Med 2012; 4: 5.

54. Flierl MA, Stahel PF, Beauchamp KM, Morgan SJ,Smith WR and Shohami E. Mouse closed head injurymodel induced by a weight-drop device. Nat Protoc

2009; 4: 1328–1337.55. Kooijman E, Nijboer CH, van Velthoven CT, et al. Long-

term functional consequences and ongoing cerebralinflammation after subarachnoid hemorrhage in the rat.

PLoS One 2014; 9: e90584.

56. Albert-Weissenberger C, Varrallyay C, Raslan F,Kleinschnitz C and Siren AL. An experimental protocolfor mimicking pathomechanisms of traumatic brain

injury in mice. Exp Transl Stroke Med 2012; 4: 1.57. Dirnagl U and members of the MCAO–SOP Group.

Standard operating procedures (SOP) in experimentalstroke research: SOP for middle cerebral artery occlusion

in the mouse. Nature Precedings 2009: http://dx.doi.org/10.1038/npre.2012.3492.3.

58. Milner E, Holtzman JC, Friess S, et al. Endovascular

perforation subarachnoid hemorrhage fails to causeMorris water maze deficits in the mouse. J Cereb BloodFlow Metab 2014; 3410.1038/jcbfm.2014.108.

59. Balcombe JP, Barnard ND and Sandusky C. Laboratoryroutines cause animal stress. Contemp Top Lab Anim Sci2004; 43: 42–51.

60. Dobrakovova M, Kvetnansky R, Oprsalova Z andJezova D. Specificity of the effect of repeated handlingon sympathetic–adrenomedullary and pituitary–adreno-cortical activity in rats. Psychoneuroendocrinology 1993;

18: 163–174.61. Meijer MK, Sommer R, Spruijt BM, van Zutphen LF

and Baumans V. Influence of environmental enrichment

and handling on the acute stress response in individuallyhoused mice. Lab Anim 2007; 41: 161–173.

62. Sotocinal SG, Sorge RE, Zaloum A, et al. The rat grim-

ace scale: a partially automated method for quantifyingpain in the laboratory rat via facial expressions. Mol Pain2011; 7: 55.

63. Langford DJ, Bailey AL, Chanda ML, et al. Coding of

facial expressions of pain in the laboratory mouse. NatMethods 2010; 7: 447–449.

64. Miller AL and Leach MC. The mouse grimace scale: a

clinically useful tool?. PLoS One 2015; 10: e0136000.65. Seggie JA and Brown GM. Stress response patterns of

plasma corticosterone, prolactin, and growth hormone

in the rat, following handling or exposure to novel envir-onment. Can J Physiol Pharmacol 1975; 53: 629–637.

66. Kant GJ, Eggleston T, Landman-Roberts L, Kenion CC,

Driver GC and Meyerhoff JL. Habituation to repeatedstress is stressor specific. Pharmacol Biochem Behav 1985;22: 631–634.

67. Ward A, Blanchard R, Bolivar V, et al. Recognition and

alleviation of distress in laboratory animals. Washington,DC: The National Academies Press, 2008, p.132.

68. Sanchez C. Stress-induced vocalisation in adult animals.

A valid model of anxiety? Eur J Pharmacol 2003; 463:133–143.

69. Wallace VC, Norbury TA and Rice AS. Ultrasound

vocalisation by rodents does not correlate with behav-ioural measures of persistent pain. Eur J Pain 2005; 9:445–452.


ab oratory

l i m i t e d

lan imals

Special Article

Morbidity scoring after abdominal surgery

Rolf Graf1,2, Paolo Cinelli3 and Margarete Arras2

AbstractPostoperative monitoring of pain and distress in small rodents is not standardized, and widely accepted scoresheets are not available. Here we describe a score sheet used in abdominal surgery of rodents, with particularreference to procedures involving the liver.

Keywordslaboratory animal welfare, pain, analgesia, pain assessment, animal model

The worldwide use of animals for medical research hasresulted in the development of an enormous number ofdifferent ‘disease models’ in a few select species, primar-ily rodents. The focus of these animal models is on thesimulation of a particular disease, either through gen-etic manipulation, application of drugs or surgicalinterventions. These experimentations can be a poten-tial source of pain and suffering. The protection of ani-mals used for scientific purposes is regulated by specificnational laws, but the way these laws are implementedshows considerable variation among countries.

Despite the lack of standardized animal protectionlaws, the development and use of animal models shouldnot only focus on the simulation of disease but alsoinclude an evaluation of proper pain management andsurveillance. Much effort has been invested in the pastin developing reliable protocols for the assessment ofpain and analgesic treatments of laboratory animals.

Nevertheless, due mainly to uncertainty regardingpain and stress detection, and inconsistencies in thedata obtained, no final protocols have been developedyet for appropriate and practicable monitoring andmanagement of pain and suffering.

Current efforts into improvements in behavioralmonitoring and pain assessment clearly indicate aneed for action. As an example, the use of the bupre-norphine opioid for postoperative analgesia is widelyaccepted.1 Although its half-life is longer than inother opioid–analgesics, it is less than 4 h, requiring atight and repetitive application every 4–6 h by injection.It is obvious that the investigator would not like tofollow such a schedule as it would involve applyingthe analgesia twice during the night. Also, repeated

injections at short intervals may be stressful for theanimals.2–4 One alternative is to provide the drugthrough drinking water which requires a 10-foldhigher amount of buprenorphine, due to the level ofconsumption and to the first-pass hepatic extractionof orally-applied buprenorphine. Furthermore, thereis also the possibility of spillage of the drinking waterwith most of the current caging systems. In addition, itremains unclear, whether individual animals voluntar-ily consume enough of the medicated water to ensurepermanent pain alleviation, especially during thedaylight phase, when mice do not drink regularly.Nevertheless, providing analgesia with drinking waterhas become popular, particularly for short half-life opi-oids (e.g. tramadol) or non-steroidal anti-inflammatorydrugs (NSAIDs, e.g. acetaminophen and paraceta-mol).5 Thus, long-lasting forms of analgesia are badlyneeded. However, sustained–release formulations ofpainkillers are not commercially available to date inEurope, although these drugs have been developed forrodents, particularly buprenorphine which provides 1–3days of continuous pain relief.3,6,7

NSAIDs (e.g. meloxicam and carprofen) are oftenadministered in mice; some of which are injected only

1Department of Visceral & Transplantation Surgery2Division of Surgical Research3Department of Trauma Surgery, University Hospital Zurich,Zurich, Switzerland

Corresponding author:Prof. Dr R Graf, University Hospital Zurich, Ramistrasse 100,CH-8091, Zurich, Switzerland.Email: [email protected]

Laboratory Animals

2016, Vol. 50(6) 453–458



sagepub.co.uk/


DOI: 10.1177/0023677216675188

la.sagepub.com

1–2 times per day based on the assumption that theyprovide pain relief for up to one day.1 However currentpublications have raised doubts about the analgesic effi-cacy of some NSAIDs. In particular the efficacy ofcommonly used dosages and the suggested length ofaction in each case remain unclear.8–10 One of the prob-lems arising from provision of pain-relief drugs toexperimental animals is that these can interfere withexperimental results. For example, in studies evaluatingnociception or the effect of pain on a disease model,treatment with an analgesic may counter the purposeof the experiment. Similarly, when studying inflamma-tory diseases, medication including analgesics mayaffect the inflammatory process itself and hence maypartially interfere with the experimental goal.8 Theseexamples indicate that an evaluation of efficient painmanagement is difficult and may require certaincompromises.

New indicators of pain are necessary to improve thequality of pain assessment. This is not simple toachieve; and probably only the identification of behav-ioral traits can provide sufficient information during thesimple cage-side monitoring, as is performed whenscore sheets are used. For example, audible vocaliza-tions cannot be unambiguously interpreted: are theydue to real pain or distress? Recent studies, aimed atthe identification of objective parameters for the assess-ment of pain, analyzed the link between behavioralchanges and alterations in heart rate and heart ratevariability using sensors implanted to continuouslymonitor physiological parameters.11,12 These experi-ments show a good correlation between cardiovascularreactions and signs of pain and distress. Monitoringlong-term physiological parameters and associatedbehaviors by continuous telemetric or video recordings(lasting for hours or days) can only provide evidence ofpain retrospectively. From a practical point of viewsuch sensors cannot be implanted into each and everyexperimental animal, first because this would involveadditional unnecessary stress and would undoubtedlybe painful for some time after surgical implantationof the transmitter, but also because not all experimentalsituations would allow such a sophisticated operation,e.g. in very young animals. Additionally, it has recentlybeen established that when alterations of specific spon-taneous home-cage behavior such as burrowing andnest-building are observed, these are not only signs ofdisturbed well-being13 or indications of the progressionof neurodegenerative diseases,14–16 but may also besigns of postoperative pain. However, performingthese observations for the detection of postoperativepain may be time-consuming.16,17 Thus short cage-side observations taken immediately after a painfulinsult have been used for many years, and allow theimmediate treatment for pain. In practical terms,

individual animals are monitored for approximately10min either in the home cage or at an observationarea for very short-term aberrations of bodily appear-ance. These aberrations, such as press, stagger, stretchand twitch, are known to be indicative of specific typesof pain, and are thought to be typical signs of abdom-inal pain in mice.18 Such very specific symptoms areoften combined into composite scorings, summarizingthe most frequent and indicative alterations.19

However, a well-known drawback of such cage-sideobservations is that the occurrence of aberrationsdoes provide some evidence of pain, but absence ofthese during the short observation period does notnecessarily prove that animals are pain-free.

Another approach to assess pain is the grimace scaleas facial changes and contortions may indicate pain.20

This approach, although plausible, requires videotap-ing of individual animals, and and its use during rou-tine monitoring is not yet validated. Such assessmentrequires adaptation to the experimental condition,animal strain, sex, etc. and the observer needs trainingand experience in order to recognize and categorizefacial features.21

Thus, methods that can reliably indicate pain arequite complicated, and need intensive analysis of indi-vidual animals and additional technical devices.Behavioral science has implemented quantitativeapproaches, based on statistical analysis of animalcohorts. Assessment of individual animals, however,is more subjective and may be biased due to the wishof investigators to continue without interfering with theexperiment. Hence, implementation of robust surveil-lance protocols, which are quantitative, reproducibleand allow longitudinal follow-up, are needed. Manyinvestigators have established their own ‘score’ sheetbased on the requirements of their local animal ethicscommittee. These score sheets are usually not includedor detailed in publications. Here we propose a scoresheet for postoperative surveillance that includes aquantitative approach with cut-offs defining termin-ation of the experiment and euthanasia of the animalin case of unacceptable suffering. We are wellaware that this score sheet is not exclusively based onscientific evidence but also on experience with severaldecades of animal experiments. The score sheet hasbeen developed along with animal welfare officers andauthorities in charge of regulations for animalexperimentation.

Animal models in abdominal surgery

Surgical modifications of organs in the abdominalcavity include operations on the liver, e.g. partialliver resection,22 liver transplantation,23 and bile ductligation.24 Partial or full splenectomies, kidney


transplantation,25 induction of strictures in the intes-tine, pancreatectomies,26 pancreatic duct ligation,27

etc. are other examples of intra-abdominal interven-tions. All these models have one thing in common,namely access to the abdominal cavity is via a laparot-omy (incision through the abdomen). The postopera-tive time frame of these experiments can be between afew hours and several months, whereby pain is experi-enced within the first 2–3 days.

The most common procedure (e.g. partial hepatect-omy) usually consists of a preoperative intraperitonealinjection of buprenorphine (0.1mg/kg body weight),approximately 30min before the operation.Subsequently, the animal is shaved and immobilizedunder inhalation anesthesia using isoflurane. Theanimal is fixed by tape on a stainless steel plate in asupine position. The plate is warmed with a heatingpad, maintaining a 37–40�C temperature. Eye ointmentis applied and the abdomen is opened by a sagittal inci-sion after disinfection with alcohol. To access the liver,the sternum is pulled upwards and fixed with a 6-0suture. The liver ligaments are then freed followed byindividual ligation of vessels to liver lobes, which canthen be dissected. After completion, the abdomen isthen closed with a running suture of the peritonealmuscle/fascia followed by closure of the skin. Duringrecuperation, the animal is placed in a warming cabinet(28–30�C) for a few hours or overnight, prior to itsreturn to the regular cage.

As in human surgery, the impact of a specific type ofintervention (e.g. partial hepatectomy) may be moder-ate to severe; and thus access to the abdomen, i.e. thelaparotomy itself, may have a variable impact onthe overall procedure. Furthermore, many proceduresin human patients can be performed by laparoscopy(minimally invasive), which strongly reduces postopera-tive morbidity. Although minimally invasiveapproaches have been tested in rodents, routine use isnot feasible.

In the current report, we focus predominantly onexperiments concerning liver surgery. A classic exampleis a 70% resection of liver lobes.22 This procedure isfairly standardized and benefits from the anatomy ofthe mouse or rat liver being quite lobulated. Therefore,parenchymal transections are not required because thelobes can be individually removed by ligation of thevessels. This procedure is well tolerated. The liver usu-ally regains its original size within a week. In addition,extended hepatectomies (e.g. removal of 86% and 91%of liver tissue) have been established to explore theeffects of insufficient liver volume.28 This situation canlead to a so-called small-for-size syndrome, the smallliver is unable to regenerate, most probably due to themetabolic overload and concurrent accumulation oftoxic products.29

The small-for-size syndrome, particularly with a fullblown liver failure, results in reduced activity, apathyand encephalopathy (‘hepatic coma’). Pain assessmentof these animals has been debated, as reports frompatients with a similar syndrome do not complain ofpain. In addition, analgesia may further impact on hep-atic function, and is given with caution.


Usually, pain is assessed by an investigator whoobserves the animals for a short period of time, lookingpredominantly for active behavior, ruffling of the furand hunchback position. In a cage with several animalsan individual repeated assessment can be inaccurateover time. Once an animal is identified that is clearlyin pain, it should properly be treated with analgesics.This leads to the following questions: Should all ani-mals in the experiment receive the same treatment toensure consistency? Would it be better to treat all ani-mals from the start with analgesics to avoid having tochange the experimental protocol based on an individ-ual animal? Alternatively, if a single animal showsabnormal behavior and pain, would it be better to elim-inate the animal from the experiment? These are ques-tions that should be clarified before beginning anexperiment, particularly if the animals are valuabledue to complex genetics or require special treatment,e.g. diabetic mice receiving insulin.

Below we describe a score sheet used for monitoringanimals after abdominal operations, particularly on theliver (see the score sheet, which is available online withthis article LAN.sagepub.com).

Postoperative score sheet

For every animal undergoing abdominal surgery, a scoresheet is prepared which shows animal ID, protocol ID,date and prospective severity degree (CH 0–III). The typeof intervention should also be indicated. This basic infor-mation will allow a replacement investigator to take oversurveillance, particularly if an animal caretaker or veter-inarian has reported that an animal has taken ill.

Depending on the protocol, the severity or expectedhealth issue, the investigator may be asked to supervisethe animals very closely, i.e. twice a day. For properobservation, it is advised to take each animal out of itscage and place it on top of the grid or lid, as is deemedbest for observation of the animal. If there are only oneor two animals per cage, observation may be performedwithout touching the animals, however, they must befreely visible. Overall, a solid knowledge of the normalbehavior and appearance of a mouse of the respectivestrain, sex and age is a prerequisite for propermonitoring.

Graf et al. 455

Activity

In the following, scored behaviors and signs of distressare described. Activity may be judged taking intoaccount the time of day, which affects the diurnal activ-ity pattern. Absence of activity is certainly an indicatorof impaired condition and health. However, normal oreven increased activity is not necessarily an indicator ofthe absence of pain or suffering, and it is mandatory toobserve other parameters or symptoms to assess thewhole picture. Aggressive animals may be observed asbeing very active, and aggression is a potential indica-tor of pain. Reduced activity is given one point whileimmobility is a strong factor and should be given threepoints. Increased activity is very subjective and mightbe induced by buprenorphine treatment3 and thereforeis not given any points in our score sheet.

Breathing

Breathing rate and depth are not easy to assess in micewithout specialized equipment. However, animals withflat breathing but normal frequency are given onepoint; while shallow or rapid breathing is an indicatorfor more severe physical impairment (two points).

Fur

The appearance of fur (piloerection and ruffling) is per-haps the easiest parameter to judge. Ruffled, unclean ormatted fur indicates absence of normal groomingbehavior. This is either based on physical impairment(through the experiment) or a stressful situation in thecage, particularly, if several males share a cage. In thiscase, presence of bites, wounds or aggressive behaviormay indicate social imbalance. Such an imbalance canbe prompted if one or two animals of the cohort areoperated on and may be weakened, resulting in achange of the social hierarchy. It is therefore recom-mended that operated animals should not be mixedwith untreated, healthy littermates during the recoveryperiod, i.e. during the first one to maximally four post-operative days only operated animals should be kepttogether. Ruffled fur receives one point.

Posture

Another parameter which is quite easy to gauge is pos-ture, such as hunchback or arching behavior. This isseen when the animals are in strong abdominal painand is found in combination with ruffled fur.Contradictory reports exist on the interpretation ofhunched posture in mice,18,30–33 and it was recentlydescribed as an unreliable indicator of pain.18,33

Based on our previous results on laparotomy11 we

have assigned one point for hunchback or archingbehavior.

Jaundice

The presence of jaundice indicates loss of liver function,and this may be due to a direct interference with theliver during surgery or pharmacological treatments tar-geting the liver. It appears as a yellowing of the skin.For mice with black or dark fur, the eyes or the foot-pads easily allow an assessment. If jaundice is part ofthe pathophysiology of the model, bilirubin should bemeasured regularly by blood analysis. If another organis targeted, blood parameters can be assessed, creatin-ine for kidney, amylase for pancreas, T3/4 for thyroid,etc. This parameter receives one point.

Operative site: assessment of the wound

Wound infections are rare in rodents. In contrast tohumans and larger animal species, particularly pets,the wound cannot be covered by a wound dressing. Itis therefore crucial that the wound be properly closed toavoid entry of foreign material, or worse, extrusion ofan intestinal loop. For proper postoperative examin-ation, particularly during the first two days, micehave to be removed from the cage and individuallyinspected.

Mice tend to bite and gnaw at their wounds particu-larly when pain medication is insufficient, furtherenhancing chances for infection. However, overdosingwith buprenorphine might also lead to aberrant behav-ior where the animals mutilate the wound (e.g. removalof sutures).

An open or infected wound receives two points.Additional criteria specific to the experiment can beintroduced and scored.

Overall score

At conclusion of an assessment, points should be addedup. A mouse having two or more points should receivepain treatment. Where it is given five points, a criticalphysical impairment has been reached which shouldreceive particular attention. Adequate pain manage-ment should be provided in both cases, i.e. a protocol(giving detailed advice on drug, dosage, administrationroute and interval) must be in place, which has beendeveloped and tailored for the model (e.g. interferencewith experimental goals) and the type of pain.Typically, we administer a bolus injection of buprenor-phine (0.1mg/kg subcutaneously) two or three times aday. If no improvement is observed within the next24 h, the animal should be euthanized. Below is a


summary of the criteria leading to termination. Someare independent criteria, e.g. cachexia (short- and long-term weight loss) or if the animal is self-mutilated.

Summary of termination criteria:

. The animal does not recuperate from the interven-tion (immobile, unresponsive to stimuli 2 h after theoperation).

. Score� 6 points.

. Score¼ 5 points: plus no improvement within thenext 24 h.

. The animal loses >15% of body weight within 12 hafter operation (short-term weight loss).

. The animal loses >20% of body weight comparedwith the start of experiment (long-term weight loss).

. Animal is self-mutilated.

. Animal exhibits signs of pain despite buprenorphinetreatment for three days.

. Animal does not respond to pain treatment within24 h.


An animal is transiently incapacitated and may loseits social status following an invasive procedure ofthe abdominal cavity. It is imperative that the animalhas opportunities to create a refuge in the form of nestbuilding or burrowing material or shelters. Particularlywith male animals, short-term isolation during inter-vention and physical impairment may severely disturbthe social structure. It should be evaluated whether asingle individual animal should be placed back in thecage with its healthy littermates. It is reasonable to pre-sume that if all animals in a cage are operated on at thesame time, the individual animal will suffer less fromthe possibly disrupted social structure, or may evenprofit from the social support.3


Historically, surgeons have preferred to operate onmale mice. However with their tendency to aggressivebehavior, increasing with age, female mice might bemore amenable. In particular, studies addressing age-related topics would greatly benefit from the inclusionof female mice. The social structure of the rat is differ-ent and males are less aggressive. Here, the need tofocus on one sex is not as crucial.

In both species, it might be advisable to use bothgenders in future experiments. First, to exclude conclu-sions that may or may not apply to both sexes, andsecond, the use of all animals reduces the overallnumber of animals to be produced. Currently, a vast

number of animals are still killed since the demand forone sex is much greater for certain animal models.

Conclusions

Proper animal surveillance in routine settings is onlypossible if the process is easy and quick. Otherwisethe likelihood is that investigators will not assess thewell-being of the animals thoroughly. This is to the det-riment of the animals. It may ultimately affect theexperimental outcome negatively and prevent relevantconclusions. The lack of diligence may also be a resultof mice or rats being observed simply as ‘happily’ andactively running around the cage, which may imply thatthese animals are not in pain. This misconception is stillcommonly made by many investigators and should berectified through education and training.


The author(s) declared no potential conflicts of interest with

respect to the research, authorship, and/or publication of thisarticle.

Funding


References

1. Stokes EL, Flecknell PA and Richardson CA. Reportedanalgesic and anaesthetic administration to rodents

undergoing experimental surgical procedures. Lab Anim2009; 43: 149–154.

2. Cinelli P, Rettich A, Seifert B, Burki K and Arras M.Comparative analysis and physiological impact of differ-

ent tissue biopsy methodologies used for the genotyping oflaboratory mice. Lab Anim 2007; 41: 174–184.

3. Jirkof P, Tourvieille A, Cinelli P and Arras M.

Buprenorphine for pain relief in mice: repeated injectionsvs sustained–release depot formulation. Lab Anim 2015;

49: 177–187.4. Meijer MK, Kramer K, Remie R, Spruijt BM, van

Zutphen LFM and Baumans V. The effect of routineexperimental procedures on physiological parameters in

mice kept under different husbandry conditions. AnimWelfare 2006; 15: 31–38.

5. Christy AC, Byrnes KR and Settle TL. Evaluation of

medicated gel as a supplement to providing acetamino-phen in the drinking water of C57BL/6 mice after surgery.

J Am Assoc Lab Anim Sci 2014; 53: 180–184.6. Carbone ET, Lindstrom KE, Diep S and Carbone L.

Duration of action of sustained–release buprenorphine in2 strains of mice. J Am Assoc Lab Anim Sci 2012; 51:

815–819.7. Healy JR, Tonkin JL, Kamarec SR, et al. Evaluation of an

improved sustained–release buprenorphine formulation

for use in mice. Am J Vet Res 2014; 75: 619–625.

Graf et al. 457

8. Roughan JV, Bertrand HG and Isles HM. Meloxicamprevents COX-2-mediated post-surgical inflammationbut not pain following laparotomy in mice. Eur J Pain

2016; 20: 231–240.9. Miller AL, Wright-Williams SL, Flecknell PA and

Roughan JV. A comparison of abdominal and scrotalapproach methods of vasectomy and the influence of

analgesic treatment in laboratory mice. Lab Anim 2012;46: 304–310.

10. Matsumiya LC, Sorge RE, Sotocinal SG, et al. Using the

mouse grimace scale to reevaluate the efficacy of post-operative analgesics in laboratory mice. J Am AssocLab Anim Sci 2012; 51: 42–49.

11. Arras M, Rettich A, Cinelli P, Kasermann HP and BurkiK. Assessment of post-laparotomy pain in laboratorymice by telemetric recording of heart rate and heart

rate variability. BMC Vet Res 2007; 3: 16.12. Cesarovic N, Arras M and Jirkof P. Impact of inhalation

anaesthesia, surgery and analgesic treatment on homecage behaviour in laboratory mice. Appl Anim Behav

Sci 2014; 157: 137–145.13. Jirkof P. Burrowing and nest building behavior as indi-

cators of well-being in mice. J Neurosci Methods 2014;

234: 139–146.14. Deacon RM. Assessing nest building in mice. Nat Protoc

2006; 1: 1117–1119.

15. Jirkof P, Cesarovic N, Rettich A, Nicholls F, Seifert Band Arras M. Burrowing behavior as an indicator ofpost-laparotomy pain in mice. Front Behav Neurosci2010; 4: 165.

16. Rock ML, Karas AZ, Rodriguez KB, et al. The time-to-integrate-to-nest test as an indicator of wellbeing in labora-tory mice. J Am Assoc Lab Anim Sci 2014; 53: 24–28.

17. Jirkof P, Fleischmann T, Cesarovic N, Rettich A, Vogel Jand Arras M. Assessment of postsurgical distress andpain in laboratory mice by nest complexity scoring. Lab

Anim 2013; 47: 153–161.18. Wright-Williams SL, Courade JP, Richardson CA,

Roughan JV and Flecknell PA. Effects of vasectomy sur-

gery and meloxicam treatment on faecal corticosteronelevels and behaviour in two strains of laboratorymouse. Pain 2007; 130: 108–118.

19. Dickinson AL, Leach MC and Flecknell PA. The anal-

gesic effects of oral paracetamol in two strains of miceundergoing vasectomy. Lab Anim 2009; 43: 357–361.

20. Langford DJ, Bailey AL, Chanda ML, et al. Coding

of facial expressions of pain in the laboratory mouse.Nat Methods 2010; 7: 447–449.

21. Miller AL and Leach MC. The mouse grimace scale: aclinically useful tool? PLoS One 2015; 10: e0136000.

22. Michalopoulos GK and DeFrances MC. Liver regener-

ation. Science 1997; 276: 60–66.23. Tian Y, Graf R, Jochum W and Clavien PA. Arterialized

partial orthotopic liver transplantation in the mouse: anew model and evaluation of the critical liver mass.

Liver Transpl 2003; 9: 789–795.24. Georgiev P, Navarini AA, Eloranta JJ, et al. Cholestasis

protects the liver from ischaemic injury and post-ischae-

mic inflammation in the mouse. Gut 2007; 56: 121–128.25. Tian Y, Chen J, Gaspert A, et al. Kidney transplantation

in mice using left and right kidney grafts. J Surg Res

2010; 163: e91–97.26. Bonner-Weir S, Trent DF and Weir GC. Partial pancrea-

tectomy in the rat and subsequent defect in glucose-

induced insulin release. J Clin Invest 1983; 71: 1544–1553.27. Wang RN, Kloppel G and Bouwens L. Duct- to islet-cell

differentiation and islet growth in the pancreas of duct-ligated adult rats. Diabetologia 1995; 38: 1405–1411.

28. Lehmann K, Tschuor C, Rickenbacher A, et al. Liverfailure after extended hepatectomy in mice is mediatedby a p21-dependent barrier to liver regeneration.

Gastroenterology 2012; 143: 1609–1619.e4.29. Dahm F, Georgiev P and Clavien PA. Small-for-size syn-

drome after partial liver transplantation: definition,

mechanisms of disease and clinical implications. Am J

Transplant 2005; 5: 2605–2610.30. Report of the Federation of European Laboratory

Animal Science Associations (FELASA). Working

Group on Pain and Distress accepted by the FELASABoard of Management November 1992. Pain and distressin laboratory rodents and lagomorphs. Lab Anim 1994;

28: 97–112.31. Burkholder T, Foltz C, Karlsson E, Linton CG and

Smith JM. Health evaluation of experimental laboratory

mice. Curr Protoc Mouse Biol 2012; 2: 145–165.32. Sevcik MA, Jonas BM, Lindsay TH, et al. Endogenous

opioids inhibit early-stage pancreatic pain in a mouse

model of pancreatic cancer. Gastroenterology 2006; 131:900–910.

33. Fentener van Vlissingen JM, Borrens M, Girod A,Lelovas P, Morrison F and Torres YS. The reporting of

clinical signs in laboratory animals: FELASA WorkingGroup Report. Lab Anim 2015; 49: 267–283.


ab oratory

l i m i t e d

lan imals

Special Article

Recommendation for severity assessmentfollowing liver resection and livertransplantation in rats: Part I

S Kanzler1,*, A Rix2,*, Z Czigany3, H Tanaka4, K Fukushima4,B Kogel4, K Pawlowsky4 and R H Tolba4

AbstractScore sheets were first introduced 30 years ago to assess pain, distress and suffering in animals. To date,however, there is still no general agreement on their use in research practice, and only a few publications canbe found on this topic. In the present work, we demonstrate the use of a special score sheet for severityassessment in the first three postoperative days in two showcased studies performed on Wistar and Lewisrats undergoing liver resection or orthotopic liver transplantation, respectively. Scoring of different criteriaand the total score were evaluated within each intervention. Additionally, both procedures were comparedregarding their degree of severity. Suitability of these score sheets was evaluated for assessing severity ofthe procedures and these showed a minor severity within each investigated study. A comparison of bothstudies showed slightly higher scores involving liver transplantation. In contradiction to the common classi-fication of these procedures as a moderate severity grade the score sheets applied here indicates a minorseverity grade within each investigated study. Also, limitations and possible improvements in the design of ourscore sheets for defined interventions are reconsidered.

Keywordsscore sheet, severity, liver, resection, transplantation

The principles of replacement, reduction and refine-ment (3Rs) in animal research were first described byRussell and Burch in 1959. The idea behind this con-cept is to ensure the most humane treatment for labora-tory animals as a prerequisite for successfulexperiments.1 With the implementation of the EUDirective 2010/63 on the protection of animals usedfor scientific purposes,2 the 3Rs are anchored withinEuropean law. One major point regarding the principleof refinement is the minimization of pain, suffering ordistress. Therefore, the recognition of pain and distressis a crucial criterion for assessing the level of discomfortin animals.

Several techniques have been developed for gradingthe level of pain in humans based on self-report meas-ures of patients such as visual rating scales or facialpain scales.3 These techniques represent the gold stand-ard as used in humans for over 30 years, but a transla-tion to animals is impossible due to the fact thatanimals cannot report verbally. This necessitates a

method to reliably recognize pain and distress basedon observations by the person conducting animalexperiments. Morton and Griffiths were the first whoselected criteria describing deviations from the normalbehavior of animals, which can be a sign of pain ordistress.4 This concept was taken up by variousgroups of scientists and adapted to their particularapplication, such as the assessment of discomfort in

1Institute for Pharmacology and Toxicology2Institute for Experimental Molecular Imaging3Department for General, Visceral, and Transplantation Surgery4Institute for Laboratory Animal Science and ExperimentalSurgery, RWTH Aachen University Hospital, Aachen, Germany

Corresponding author:R H Tolba, MD, PhD, Institute for Laboratory Animal Science andExperimental Surgery, RWTH Aachen University Hospital,Pauwelsstraße 30, 52074 Aachen, Germany.Email: [email protected]

*These authors contributed equally to this paper.

Laboratory Animals

2016, Vol. 50(6) 459–467



sagepub.co.uk/


DOI: 10.1177/0023677216678018

la.sagepub.com

hepatomegaly in mice and rats,5 the application ofhumane endpoints,6 or the assessment of pain afterlaparotomy in rats.7,8

Up to today, scoring has rarely been reported as arefinement measure in procedures involving laparot-omy in rats, only scarce evidence can be found onthis topic. Roughan and Flecknell have compared theeffects of different analgesic agents on the postoperativebehavior of rats.9,10 They have improved their scoringsystem regarding behavioral peculiarities found in thesestudies, and used it to also test different analgesicprotocols after laparotomy.8 Another study performedby Sotocinal et al. in 2011 have introduced the rat grim-ace scale as a method for quantifying pain followinglaparotomy.11 Nevertheless, according to the best ofour knowledge, there are no studies available on ascore sheet-based severity assessment after laparotomyin general or liver surgery specifically.

Therefore, the aim of the present paper is to reportthe results of severity assessment after liver resectionand liver transplantation in rats using a score sheetand to give recommendations for the prospective useof this tool. In brief, we scored rats undergoing liverresection or orthotopic liver transplantation for threepostoperative days, implementing a general score sheetbased on the original reports.4 A further aim of thisarticle is to analyze the feasibility of using a scoresheet, and to publicize its advantages and limitationsregarding both surgical procedures.

Experimental procedure

Experimental animals and statisticalanalysis

All animal experiments were performed in accordancewith the German Animal Welfare Law and the EUDirective 2010/63. The experimental protocol wasapproved by the governmental animal care and usecommittee (Landesamt fur Natur, Umwelt undVerbraucherschutz (LANUV) Nordrhein-Westfalen,Recklinghausen, Germany). The rats were housed ingroups of five per cage under specific pathogen-freeconditions with a 12 h light and dark cycle in a tem-perature- and humidity-controlled environment accord-ing to the Federation of European Laboratory AnimalScience Associations (FELASA, www.felasa.eu) recom-mendations. Water and standard diet for laboratoryrodents (ssniff GmbH, Soest, Germany) were offeredad libitum. In accordance with the German Society ofLaboratory Animal Science (GV–SOLAS) recommen-dation, ‘Pain management for laboratory animals2015’, all the rats were treated with 0.1mg/kg of bupre-norphine subcutaneously as preoperative analgesia aswell as postoperatively every 12 h for three days.12

The number of animals per group was calculated viaa power calculation using G*Power, version 3.1.9.2(Freeware, Kiel University, Kiel, Germany). All theanimals were allocated randomly to their groups, andthey all completed the study (i.e. no dropouts). Dataare expressed as mean� standard error of the mean(SEM). Statistical analyses were performed usingGraphPad Prism version 5.00 (GraphPad Software,San Diego, CA, USA). Data from the liver resectionor the liver transplantation were analyzed by one-wayanalysis of variance (ANOVA) with Tukey’s post hoctest. For the comparison of both protocols repeatedmeasures ANOVA were performed. Differencesbetween groups were considered significant whenP< 0.05.

Liver resection

For the liver resection experiments, 120 male Wistarrats (CS 4105; Janvier Labs, Le Genest-Saint-Isle,France) with a weight of between 290 and 400 g andaged 8–12 weeks were used. They were randomly allo-cated to 12 experimental groups (experimental unit)and liver resection was performed to investigate theefficacy of a tissue glue for bleeding control from theresection surface compared with the clinical gold stand-ard (fibrin glue). The non-anatomic hepatic resection ofthe left lateral lobe was performed as described byAysan et al.13 and Henderson et al.14 in 2010(Figure 1). On days 14, 21 and 90 after intervention,animals were sacrificed to investigate the liver resectionarea as well as the histological, hematological and bio-chemical parameters. Because all the animals werescored for three days after intervention using identicalscore sheets, animals from all the experimental groupscould be included in our study.

Liver transplantation

Whole-graft orthotopic liver transplantation (OLT)was performed on 40 male Lewis rats (LEW/OrlRj;Janvier Labs) with a body weight of between 260 and320 g and aged 8–10 weeks, according to our protocolpublished previously.15,16 Briefly, total hepatectomywas performed on the donor animals and their liverswere stored in a cold ischemic time for 8 h at 4�C in anorgan preservation solution (Figure 2). Recipient ani-mals were randomly assigned to the following experi-mental groups: (i) control group without remoteischemic conditioning, (ii) remote ischemic condition-ing before liver excision and recipient hepatectomy(RIC1), and (iii) remote ischemic conditioning aftergraft reperfusion (RIC2). Remote ischemic condition-ing treatment of the recipient animals was performed asdescribed previously.17 Cold ischemic time


(478.2� 9.0min) and anhepatic time (18.6� 0.7min)were comparable for all groups. One-week survivalrate was 100% in each group.

Score sheet

All animals were scored throughout the three con-secutive postoperative days. Therefore, alterations in

different parameters (body weight, general state, spon-taneous behavior and clinical result) were documentedand assessed by attributing scores in the range from 0to 20 points. 0 points indicated no alteration or physio-logical state, whereas �20 points represented the high-est severity (body weight loss of more than 10%, severechanges in general state, spontaneous behavior and aclinical state of more than 10 points or a combination

Figure 2. The timeline of the liver transplantation experiment.

Figure 1. Liver resection technique: A) After median laparotomy, the liver is positioned for resection B) The left laterallobe is resected with a scissor C) The liver resection surface is sealed with a glue.

Kanzler et al. 461

of all). In such cases, immediate euthanasia was recom-mended (humane endpoint). All animals were scored bya single senior technician to minimize inter-observerobjectivity. This person was blinded to the experimentalconditions and scoring was performed at the same timeof the day. The points of all parameters were summedup and the animals were classified into four ‘degree ofstrain (DS)’ groups depending on the total score(Table 1), in which DS0 represents no strain, andDS1, DS2 and DS3 correspond to the categories ofmild, moderate and severe, respectively.

Results

Liver resection

In Figure 3, the parameters of body weight, generalstate, spontaneous behavior and DS are shown. In

general, the condition of all the rats was good and atotal score of 6 points was not exceeded. No significantdifferences were detected between the rats treated withthe test glue, gold standard or saline (Figures 3a–c).The score for body weight of all groups increasedslightly on day 2, and decreased again on postoperativeday 3. The scores for the parameters of general stateand spontaneous behavior were constant, over theobservation time, of around 1. All the animals’ scoresranged between 1 and 9 points (DS1), resulting in aclassification of minor according to the score sheetused (Figure 3d).

Liver transplantation

The changes in body weights of rats that underwentliver transplantation were scored as between 1 and 5points over the examination period of three days. The

Table 1. Design of the score sheet used.

Parameter Score

Body weight

Unaffected 0

Variation <5% 1

Weight reduction 5–10% 5


Weight reduction >20% 20

General state

Fur smooth and shiny, stoma clean, eyes clear and shiny 0

Fur faulty (reduced/excessive body hygiene) 1

Fur lusterless and disheveled, scruffy stoma, eyes turbid, increased muscle tone 5

Dirty fur, sticky or moist stoma, abnormal posture, eyes turbid, increased muscle tone 10

Cramps, paralysis (trunk muscle, limbs), breath sounds, cold temperature 20

Spontaneous behavior

Normal behavior (sleep, reaction to blow and touch, curiosity, social contacts) 0

Slight variation to normal behavior 1

Unusual behavior, limited motor function or hyperkinetic 5

Self-isolating, lethargy, distinct hyperkinetic, stereotypy of behavior, coordination disorder 10

Pain noises by grabbing, self-amputation (automutilation) 20

Clinical results

Temperature, pulse and respiration normal, limbs warm, mucosa well supplied with blood 0

Slight variation to the normal situation 1

Variation of temperature 1–2�C, respiration and pulse �30% 5

Variation of temperature >2�C, respiration and pulse �50% 20

Degree of strain (DS)

DS0¼no strain 0

DS1¼minor strain: careful observation necessary 1–9

DS2¼moderate strain: veterinary support, additional analgesia 10–20

DS3¼high-grade strain: consult animal protection commissary, veterinary support necessary, ter-mination of experiment, euthanasia

>20


parameters of general state, spontaneous behavior andclinical signs (data not shown) were slightly increased(1 point) on postoperative day 1 but were alreadydecreased again on day 3 (Figures 2a–c). All animalsthat underwent liver transplantation were assignedDS1, which was defined as a minor strain accordingto the score sheet used (Figure 2d). No significant dif-ferences were found between treatment groups.

Comparison of liver resection and livertransplantation

Comparing both protocols (Figure 4), rats that under-went liver transplantation showed a slightly, but notsignificantly, higher score when the DS was analyzed(mean transplantation control: postoperative day[POd]1: 6.9� 0.7 SEM, POd2: 7.1� 0.6 SEM, POd3:

7.1� 1.3 SEM; mean resection gold standard: POd1:3.9� 0.9 SEM, POd2: 4.4� 0.8 SEM, POd3: 4.5� 0.9SEM). the scoring for the body weight of both inter-ventions showed only slight but not significant differ-ences (mean transplantation control: POd1: 3.9� 0.7SEM, POd2: 4.4� 0.6 SEM, POd3: 5.1� 0.9 SEM;mean resection gold standard: POd1: 2.9� 0.6 SEM,POd2: 3.9� 0.7 SEM, POd3: 3.6� 0.7).

Discussion

The use of score sheets for assessing pain and distresswas first suggested by Morton and Griffiths in 1985.4

They introduced general parameters that can be alteredin line with pain or distress in animals; and deviationsof the normal behavior were categorized into fourgroups: no obvious deviation from normal, a minor

Figure 3. Scoring of rats that underwent liver resection for three postoperative days. The parameters body weight (a),general state (b), spontaneous behavior (c) and the degree of strain (d) of animals treated with saline (n¼ 12), clinical goldstandard (n¼ 14) or a test glue (n¼ 25) are depicted. The degree of strain (DS) is defined as DS0 (no strain), DS1 (minorstrain) and DS2 (moderate strain). More than 20 score points (DS3) which is associated with severe strain, were notreached in our experiments. Values are depicted as mean�SEM.

Kanzler et al. 463

change, a definite change, and a gross change fromnormal. Depending on the respective group, scores of0 to 3 were assigned, with 0 indicating no deviationfrom the normal range. The scores from all groupswere then summed up to a total score, and treatmentrecommendations were then prescribed according tothis final value. Nevertheless, the authors emphasizedthat with increasing knowledge regarding special pro-cedures, these parameters should be modified accord-ingly. To date only a few applications of score sheetscan be found in the literature,5,6 all of which are modi-fications of the initial guidelines by Morton andGriffiths. Most studies from the pioneering era ofanimal welfare have focused on the effects of analgesiaon animal behavior following procedures involvinglaparotomy but not on the effects of the protocolitself.7 The present work implements the use of scoresheets to estimate the degree of pain and distress ofanimals after liver resection or orthotopic liver trans-plantation, and also to retrospectively assess severity,focusing on the surgical procedure itself, while usingappropriate analgesic treatment as a basic prerequisitefor these major surgical interventions (see methods). Asthere are no recommendations for the design of scoresheets for these specific procedures, the initial layout byMorton and Griffith was used. An advantage of thisdesign is the good understandability of the describedparameters and their alterations and, therefore, theeasy assignment of scores to those. This is a crucialpoint in designing score sheets, as all personnelinvolved in the experiments should be able to clearlyunderstand the score sheet.18

When more than one person is involved in theassessment, between-assessor variations have to betaken into account, because such scoring is partlybased on a subjective evaluation. In the describedstudy, we reduced inter-observer objectivity withineach protocol by appointing one senior technician forthe scoring of all rats on all postoperative days.Nevertheless, a limitation is the fact that betweenboth protocols (resection and transplantation), differ-ent personnel were responsible for the scoring of therats. One possible way of avoiding personal bias is tohave several observers, and to calculate average scoresfor each animal.5 Furthermore, to evaluate subjectiveparameters, such as general state or spontaneousbehavior, extensive knowledge of normal, healthybehavior of the animals is absolutely necessary. Inorder to improve personnel skills in better discriminat-ing normal from abnormal behavior, Roughan andFlecknell have suggested using video-based training8

to recognize relevant signs. As it is essential for theirsurvival in a natural environment, rats often mask pain-related behavior; video analysis of undisturbed animalscan also be used to observe alterations in spontaneousbehavior. In a study by Roughan and Flecknell it waspossible to identify uncommon criteria related to painafter laparotomy, such as back-arching, staggering orwrithing.7 The video-analysis technique could also beused to monitor the animals during the night, as ratsare nocturnal and changes in behavior or activity couldbe observed more easily during the active time of theanimal.19 Another possibility for assessing maskedbehavioral changes in rats is to measure their motor

Figure 4. Comparison of both liver interventions. The parameters body weight (a) and degree of strain (b) of both rats thatunderwent liver resection (n¼ 14) or liver transplantation (n¼ 8) are shown. Rats showed a slightly higher but not sig-nificant degree of strain after liver transplantation compared with liver resection. This difference was not caused by theirbody weight alterations, which did not differ in score after both interventions. Values are shown as mean�SEM.


coordination, i.e. the use of a Rotarod or parallel bars,where the animals have to solve a specified task, andscores could then be assigned to the time needed tofulfill that task.20 A relatively new approach in detect-ing pain is the rat grimace scale, where facial expres-sions of the animal are detected using photographsfrom video-material and assigned to different states ofpain.11 This method is translated from the facial actioncoding system which is used for the detection of pain innon-verbal human patients.21 The aforementionedtechniques can be applied to minimize the subjectivityin assessing the spontaneous behavior of animals.

Besides the subjectively evaluated parameters (gen-eral state and spontaneous behavior), we also usedobjective parameters, such as body weight and clinicalresults. With liver resection and transplantation, bodyweight is not always a reliable parameter because theoriginal body weight might change as a result of theintervention itself (resection of the liver, slight differ-ences between recipient liver and the graft, weightgain from fluid resuscitation, etc.). This could beavoided by weighing all animals immediately after sur-gery to register the post-intervention body weight, anddefined as the new baseline. Furthermore, analgesictreatment with buprenorphine should also be takeninto account, because as an opioid it could possiblylead to strong obstipation, among other effects.Nevertheless, body weight can be evaluated as anobjective parameter, although it must be interpretedvery carefully after liver resection and transplantationas well as following any other major surgical interven-tions. Thus, it may be reasonable to supplement thescoring system with a body condition score. Usingthese scores, the amount of muscle and soft-tissue cov-ering bony parts of the body, mostly the dorsal part ofthe hip, can be examined and estimated. These valuesshould show a better correlation with the amount ofsubcutaneous fat stored.22

Another limitation of using our score sheet to deter-mine severity after liver resection or transplantation isthat acute complications, such as shock or internalbleeding, cannot be detected with the present design.Due to the aforementioned deficiencies of the design,it is possible that animals scored with no or only aminor DS could die suddenly as result of a severeacute complication. Here we would like to emphasizethe importance of performing thorough autopsies in allcases of sudden unexpected death in every research set-ting to underline that the scoring did not fail. Thisproblem again highlights the necessity of adapting ageneral score sheet to an experiment. In doing so, per-forming a pilot study could be beneficial. Within theframework of a pilot project, animals could be observedintensively using continuous video recordings. The rec-orded materials could then be used to retrospectively

evaluate the observed, maybe procedure-specific, alter-ations, which are not included on a general score sheet.For the main study a score sheet, preferably containingthe identified important parameters, could be createdbased on the observations of the pilot study. Thiswould result in an easily understandable scoringsystem, which could be used to observe large numbersof animals. The establishment of a web-based databaseof a pool of specific alterations observed after certainprocedures might help researchers to select the mostrelevant parameters from this pool, and export themto an adapted score sheet which best fit their studydesign (measures for refinement and reduction – inline with the 3Rs).

In the studies described, we observed a slightlyhigher DS in rats that underwent liver transplantationcompared with animals that underwent liver resection,however this was not statistically significant. This dif-ference in DS was not caused by body weight changes,as it was comparable in both interventions. However,as two independent protocols were compared in thepresent study, it has to be mentioned that the scoringwas performed by different observers. Therefore itcannot be excluded if a real difference in severity ofboth procedures was detected, or if the differenceresulted from inter-observer deviation.

All animals in the studies described were assigned toa DS of 1, which was defined as a minor strain in ourscore sheet. This is in contrast to the severity classifica-tion in EU Directive 2010/63/EU Appendix VIII,2

where laparotomy and organ transplantation are bothassigned to the moderate severity category. Thisstrongly indicates the need for further evaluation ofour scoring system with special attention to the sensi-tivity and comparison with other scoring systems. Theconsequent use of precisely evaluated score sheets afterall interventions, such as liver surgeries, can lead to abetter and more precise evaluation of severity.Therefore, the use of score sheets should be obligatoryin order to obtain detailed information about the sever-ity of different interventions based on empirical data.By this means the score sheet could also be used as arefinement tool to improve the health and well-being ofanimals after interventions such as liver resection ortransplantation, respectively. Using analgesia as oneexample, the type and dosage of analgesics used canpossibly lead to severe postoperative complicationsdue to numerous side-effects.23 It would be beneficialto adapt the analgesic dosage to each animal dependingon its individual condition, which would be possible byusing the precisely-evaluated score sheet to detect theactual condition. In the present study a standard anal-gesic therapy was applied according to the GV–SOLASrecommendations for pain management of laboratoryanimals, meaning that the dosage of analgesia was

Kanzler et al. 465

not tailored individually to each animal. Nevertheless,the animals did not show any strong alterations in theirspontaneous behavior, leading to the assumptionthat no or only few side-effects, as a result of an over-dosage or signs of pain due to an inadequate dosage,occurred.

However, keeping all points in mind, more scoresheet data have to be collected from different protocolsin the field of liver surgery to ‘fine-tune’ severity assess-ment after these interventions in order to establish arobust basis for welfare assessment. This reliabledata-collection would also allow the scientific commu-nity to perform and publish preclinical meta-analysesfor severity assessments in specific procedures, whichagain contributes to the possibility of reducing thenumbers of animals used which is in line with the prin-ciple of the 3Rs.1

Conclusion

Within this article we were able to show that use of ascore sheet is a simple and effective way of assessingseverity following major liver surgery. The general con-dition of the animals was recorded, and slight but notsignificant differences were shown between two differ-ent interventions. Although the initial idea of a scoresheet was first implemented in 1985, there are stillongoing discussions within the scientific communityon the reliability of these score sheets. Based on thepresent work, we can conclude that the use of scoresheet is in general a suitable tool for evaluating severityafter liver resection and transplantation. However, limi-tations are identifiable in our score sheet, and severalimprovements might be needed in the future to assessthe condition of the rats in an adequate way.Additional measures such as pilot operations, videorecordings, thorough autopsies in all cases of suddendeath of animals and modifications in the scoringsystem to meet the special needs of certain procedures,might increase the overall sensitivity and specificity indetecting complications. In general, due to the lack ofverbal communication of animals, we as humans haveto be able to recognize and evaluate signs of pain anddistress, and the method presented is a good way offulfilling this task, at least in part.

Acknowledgement

The authors would like to thank Pascal Paschenda for hisexcellent technical assistance.



article.

Funding

The author(s) received no financial support for the research,

authorship, and/or publication of this article.

References

1. Russell W and Burch R. The Principles of Humane

Experimental Technique. London: Methuen, 1959.2. Directive 2010/63/EU of the European Parliament and of

the Council of 22 September 2010 on the protection of

animals used for scientific purposes. Official Journal of

the European Union L276. 2010; 20.10.2010: 33–79.

3. Dansie EJ and Turk DC. Assessment of patients with

chronic pain. Br J Anaesth 2013; 111: 19–25.4. Morton D and Griffiths P. Guidelines on the recognition

of pain, distress and discomfort in experimental animals

and an hypothesis for assessment. Vet Rec 1985; 116:

431–436.5. Beynen AC, Baumans V, Bertens AP, Havenaar R, Hesp

AP and Van Zutphen LF. Assessment of discomfort in

gallstone-bearing mice: a practical example of the prob-

lems encountered in an attempt to recognize discomfort

in laboratory animals. Lab Anim 1987; 21: 35–42.

6. Lloyd M and Wolfensohn S. Practical use of distress

scoring systems in the application of humane endpoints.

In: Hendriksen CFM and Morton DB (eds) Humane end-

points in animal experiments for biomedical research.

Proceedings of the International Conference 22–25

November 1998, Zeist, The Netherlands. London:

Royal Society of Medicine Press, 1999, pp.48–53.7. Roughan JV and Flecknell PA. Behaviour-based

assessment of the duration of laparotomy-induced

abdominal pain and the analgesic effects of carprofen

and buprenorphine in rats. Behav Pharmacol 2004; 15:

461–472.8. Roughan JV and Flecknell PA. Evaluation of a short

duration behaviour-based post-operative pain scoring

system in rats. Eur J Pain 2003; 7: 397–406.

9. Roughan JV and Flecknell PA. Behavioural effects of

laparotomy and analgesic effects of ketoprofen and car-

profen in rats. Pain 2001; 90: 65–74.10. Roughan JV and Flecknell PA. Effects of surgery and

analgesic administration on spontaneous behaviour in

singly housed rats. Res Vet Sci 2000; 69: 283–288.11. Sotocinal SG, Sorge RE, Zaloum A, et al. The rat grim-

ace scale: a partially automated method for quantifying

pain in the laboratory rat via facial expressions. Mol Pain

2011; 7: 55.

12. Committee on Anaesthesia of GV–SOLAS. Pain manage-

ment for laboratory animals. 2015.13. Aysan E, Bektas H, Ersoz F, Sari S, Kaygusuz A and

Huq GE. Ability of the Ankaferd Blood Stopper� to pre-

vent parenchymal bleeding in an experimental hepatic

trauma model. Int J Clin Exp Med 2010; 3: 186–191.14. Henderson PW, Kadouch DJ, Singh SP, et al. A rapidly

resorbable hemostatic biomaterial based on dihydroxy-

acetone. J Biomed Mater Res A 2010; 93: 776–782.

15. Nagai K, Yagi S, Uemoto S and Tolba RH. Surgical

procedures for a rat model of partial orthotopic liver


transplantation with hepatic arterial reconstruction. J VisExp 2013; 73): e4376.

16. Czigany Z, Iwasaki J, Yagi S, et al. Improving research

practice in rat orthotopic and partial orthotopic livertransplantation: a review, recommendation, and publica-tion guide. Eur Surg Res 2015; 55: 119–138.

17. Czigany Z, Turoczi Z, Onody P, et al. Remote ischemic

perconditioning protects the liver from ischemia–reperfu-sion injury. J Surg Res 2013; 185: 605–613.

18. Hawkins P. Recognizing and assessing pain, suffering

and distress in laboratory animals: a survey of currentpractice in the UK with recommendations. Lab Anim2002; 36: 378–395.

19. Jablonski P, Howden BO and Baxter K. Influence ofbuprenorphine analgesia on post-operative recovery intwo strains of rats. Lab Anim 2001; 35: 213–222.

20. Deacon RM. Measuring motor coordination in mice.J Vis Exp 2013; (75): e2609. doi: 10.3791/2609.

21. Ekman PFW. Facial action coding system. Palo Alto, CA:Consulting Psychologists Press, 1978.

22. Ullman-Cullere MH and Foltz CJ. Body condition scor-ing: a rapid and accurate method for assessing healthstatus in mice. Lab Anim Sci 1999; 49: 319–323.

23. Camprodon R and Bowles M. Perioperative analgesia inexperimental small bowel transplantation. TransplantProc 2006; 38: 1857–1858.

Kanzler et al. 467

ab oratory

l i m i t e d

lan imals

Special Article

Severity assessment in rabbits afterpartial hepatectomy: Part II

N Drude1, K Pawlowsky2, H Tanaka2, K Fukushima2,B Kogel2 and R H Tolba2

AbstractAlthough the recognition of pain, distress and discomfort has already been described in 1985 by Morton andGriffiths there is still very little known about the establishment of score sheets especially, regarding post-surgical pain and severity assessment for laboratory animals such as rabbits. In this paper we describe theestimation of severity and recovery status of 36 female New Zealand White rabbits (NZW) in a standardizedliver resection model using two different adhesive treatments and one control group. Welfare was assessed at3–4 consecutive days after surgery using a scoring system which included the following criteria: body weight,general state, clinical results, spontaneous behavior and clinical examination. Values could range from 0 to 20where increasing values indicated increasing severity with a predefined humane endpoint for a score �20points. Documented score points were almost exclusively a result of body weight loss, whereas clinical signsand general health status had no influence on the overall sum of points scored. Behavioral variation wassolely observed postoperatively, within the first 24 h, with an average score �1. In contrast to the classificationof a laparotomy as a moderate procedure in the EU Directive 2010/63 (annex VIII) the assessment hereinpresented showed a mild burden in all groups according to the scoring system used. The partial hepatectomyitself, as well as the adhesive treatment using either synthetic glue VIVO-107 or fibrin glue, were welltolerated.

Keywordsbehavior, distress, partial hepatectomy, rabbits, score sheet, severity

Historical background

Partial hepatectomy in rabbits has been practiced sincethe early 19th century; whereas the first operation, eventhough performed initially in the 1830s by Cruveilhierand Andral, was only described in 1879 by Tillmanset al. as a proof of liver regeneration.1,2 Surgical pro-cedures in (laboratory) animals are well known andhave been established for centuries, while animal wel-fare as a concept has only recently been implemented innational and international law. In 1959 Russell andBurch postulated the concept of replacement, reductionand refinement (the 3Rs) that has underpinned the eth-ical framework for animal experimentation ever since.3

Today the 3R concept is incorporated in national andinternational law as well as in published guidelines onanimal experimentation.4,5 Severity assessment as anaspect of refinement is emphasized in the EUDirective 2010/63, where refinement is described as

methods that avoid, alleviate or minimize the potentialpain, distress or other adverse effects suffered by theanimals involved, or which enhance animal well-being.A surgical procedure involving laparotomy such as thepartial hepatectomy discussed in this paper under gen-eral anesthesia and appropriate analgesia, associatedwith post-surgical pain, suffering or impairment ofgeneral condition is classified as a moderate inter-vention (annex VIII of EU Directive 2010/63).

1Department for Nuclear Medicine2Institute for Laboratory Animal Science and ExperimentalSurgery, RWTH Aachen University Hospital, Aachen, Germany

Corresponding author:R H Tolba, MD, PhD, Institute for Laboratory Animal Science andExperimental Surgery, RWTH Aachen University Hospital,Pauwelsstraße 30, 52074 Aachen, Germany.Email: [email protected]

Laboratory Animals

2016, Vol. 50(6) 468–475



sagepub.co.uk/


DOI: 10.1177/0023677216677949

la.sagepub.com

A procedure is considered to be moderate if theanimals are likely to experience only short-term mod-erate, or long-lasting mild pain, distress or suffering,or if the animals’ well-being or general condition isimpaired.4

The first guidelines on the recognition of pain, dis-tress and discomfort in laboratory animals using ascore sheet system were published by Morton andGriffiths in 1985 who described the relationshipbetween appearance, body weight, clinical and behav-ioral signs and the degree of pain or distress. It is pos-sible with an overall assessment, if well established, tocategorize the severity of a given procedure.6.Ideally ascore sheet should be modified according to specificspecies, and optimized for each procedure and animalmodel. A score sheet should include general as well astailored parameters for any given protocol.7

However, there is an ongoing discussion regardingthe limitation of a score sheet estimation to classify theseverity of an experiment or to predict pain and distressfor the laboratory animal, and to clearly identify atwhat point human intervention is required.

In this study, a standardized liver resection model inrabbits was used to test and compare the sealing with anovel synthetically medical adhesive, VIVO-107, to theclinical gold standard of fibrin glue (Tissucol Duo S),and to a control group treated with 0.9% sodium chlor-ide (NaCl). All the groups were part of a severity assess-ment with a semi-quantitative score as a sum of changesin body weight, general health status, spontaneousbehavior and/or clinical signs. The aim of the studywas to assess and to analyze data for severity scoringafter liver resection in rabbits. We discussed and com-pared our results with published data on the recogni-tion of postoperative pain. Furthermore, wehighlighted differences in the scoring method as wellas limitations, and compared the results of oursuggested score sheet system with the classification ofcomparable procedures as stated in the EU Directive2010/63.

Animals and statistical analysis

In total 36 female, 12–16-week-old rabbits (NewZealand White; specific pathogen-free according toFELASA recommendations; Charles RiverLaboratories, Saint-Germain-Nuelles, France) with abody weight range of between 2800 and 4446 g wereused in this study. The rabbits were acclimatized forat least one week after shipment. They were housedindividually in type 4421X or 4541P cages(Tecniplast, Buguggiate, Italy) under standardized con-ditions with a 12 h/12 h light/dark cycle at 22� 2�C,a relative air humidity range of 30 to 70% and15 air changes per hour (air movement <0.2m/s).

The animals were bedded on Lignocel hygienic animalbedding, type 3/4-S (Rettenmaier & Sohne GmbHþCoKG, Rosenberg, Germany), and fed with a ssniff diet(K-H; ssniff, Soest, Germany). Water was provided viasupply bottles filled with sterile water which was ultra-violet-treated, ozoned and pH-reduced.

The number of animals per group was calculated viaa power calculation using G*Power, version 3.1.9.2(Freeware, Kiel University, Kiel, Germany). Resultsare expressed as mean� SD. All statistical calculationswere performed using GraphPad Prism version 5.00(GraphPad Software, San Diego, CA, USA) forWindows. One-way analysis of variance (ANOVA)with post hoc comparisons were performed withTukey’s test. The results were considered to be statis-tically significant if P� 0.05.

Ethical statement

The experiments were performed in accordance withthe German legislation governing animal studiesaccording to the ‘Guide for the care and use of labora-tory animals’ (NIH publication, 8th edition, 2011) andthe EU Directive 2010/63/EU on the protection of ani-mals used for scientific purposes (Official Journal of theEuropean Union, 2010). Official permission wasgranted by the governmental animal care and useoffice (Landesamt fur Natur, Umwelt undVerbraucherschutz (LANUV) Nordrhein-Westfalen,Recklinghausen, Germany).

Experimental procedure

Liver resection

The rabbits were randomized to receive either VIVO-107(Adhesys Medical GmbH, Aachen, Germany),Tissucol Duo S fibrin glue (Baxter, Unterschleißheim,Germany) as the clinical gold standard, or saline as thecontrol. The non-anatomic hepatic resection of the mainleft lateral lobe was performed according to the methoddescribed by Kroez et al.8 Buprenorphine 0.5mg/kg(Temgesic; Essex Pharma GmbH, Munich, Germany)was administered subcutaneously as preoperative anal-gesia. An indwelling cannula was placed into the mar-ginal ear vein for induction and to maintain anesthesia.Anesthesia was induced using a mixture of medetomi-dine (Domitor 0.1mg/kg; Zoetis GmbH, Berlin,Germany) and ketamine 10%, 0.2mg/kg (MedistarArzneimittelvertrieb GmbH, Ascheberg, Germany).

All preparations were conducted under aseptic con-ditions (see Figure 1). Animals were placed in a supineposition. A ventral midline incision of approximately20 cm long was made, and bleeding was controlled byelectrocautery. Gauze swabs (10� 10 cm) soaked with

Drude et al. 469

lactated Ringer’s solution were used to retract the liveror intestines, and to keep the organs moist. The liverwas exposed and standardized hepatic trauma (about3 cm) was induced by sharp resection with a preciseincision of the elevated left lobe of the liver. VIVO-107, Tissucol Duo S or saline (about 2mL) were imme-diately applied to the cut edge. The liver was repos-itioned into the abdomen and the abdomen irrigatedwith 20mL of sterile saline solution. The abdominalwall was closed with a continuous suture (EthiconVicryl 2-0, #V443). The skin was closed as a separatelayer with interrupted suture (Ethicon Prolene 2-0,# EH7038. Ethicon Johnson & Johnson MedicalGmbH, Norderstedt, Germany). The animals recov-ered in a special intensive care unit (Vetario; BrinseaProducts Ltd, North Somerset, UK) with warmed air(30–35�C) for at least 20min. The resected part of the

liver was weighed to calculate the percentage of theresected liver. The calculation of the relative weight ofthe resected liver was based on an average liver weightof 102.8� 13.5 g of 12–16-week-old rabbits with aweight range of 2356 to 3740 g (historical data of thebreeder Charles River). Buprenorphine 0.1mg/kg(Temgesic, Essex Pharma GmbH) was administeredsubcutaneously every 12 h for three days as postopera-tive analgesia.9

Severity assessment

A semi-quantitative score for degree of severity (DS)was used to evaluate the described procedure,6 i.e.DS¼ 0 (no strain), DS¼ 1 (mild), DS¼ 2 (moderate),DS¼ 3 (severe) (Table 1). The recovery status (estima-tion of severity) of the animals was assessed at three

Figure 1. Surgical technique of partial hepatectomy in rabbits. (a) Preparation of anesthetized rabbit. (b) Midline lapar-otomy and exposure of liver. (c) Induction of hepatic trauma. (d) Resected liver. (e) Application of adhesive glue or NaCl.(f) Repositioning of organs into the abdomen.


(n¼ 34) to four (n¼ 20) consecutive days at a definedtime each day after surgery using a score system whichincluded the following criteria: body weight, generalhealth status, spontaneous behavior and clinical exam-ination. All animals were scored by a single seniortechnician to minimize inter-observer objectivity. Thetechnician was blinded to the experimental conditions.Spontaneous behavior was usually assessed from a dis-tance before evaluation of the other three criteria.During the time of scoring all animals received identi-cal analgesic treatment. For each criterion score,values could range from 0 to 20 where increasingvalues indicated increasing severity and correlated toan impairment of the general condition. A humaneendpoint was defined if the score exceeded 20 points,with immediate euthanasia if the endpoint wasreached.

Results

The amount of liver resected was comparable in allgroups. The relative amount of resected liver on theday of surgery was 9.12� 2.45% in the VIVO-107group, 10.05� 2.34% in the Tissucol Duo S groupand 11.62� 2.63% in the 0.9% NaCl control group.The survival rate was 100% for seven days in allgroups after surgery. In the VIVO-107 group, 2 outof 16 rabbits died during surgery due to technical com-plications (air emboli via central liver veins due to ana-tomical variation). These animals were excluded astechnical failures.

The severity assessment did not reach the predefinedendpoint (score� 20) in any of the animals. The overallassessment and sum of score points of the severity scor-ing showed a burden, defined as mild in our score sheet,

Table 1. Full score sheet used for severity assessment after partial hepatectomy in rabbits.

Scorepoints

POday 1

POday 2

POday 3

I. Body weightUnaffected 0

Variation <5% 1



Weight reduction >20% 20

II. General state

Fur smooth, shiny; stoma clean; eyes clear and shiny 0

Fur faulty (reduced or excessive body hygiene) 1

Fur lustreless, dishevelled; scruffy stoma; eyes turbid; increased muscle tone 5

Dirty fur; sticky or moist stoma; abnormal posture; eyes turbid; increased muscle tone 10

Cramps; paralysis (trunk muscle, limbs); breath sounds; cold temperature 20

III. Spontaneous behavior

Normal behavior (sleep, reaction to blow and touch, curiosity, social contact) 0

Slight variation to normal behavior 1

Unusual behavior, limited motor function or hyper kinetic 5

Self-isolation, lethargy, distinct hyper kinetic, stereotypy of behavior, coordination disorder 10

Pain noises by grabbing, self-amputation (automutilation) 20

IV. Clinical result

Temperature, pulse and respiration normal; limbs warm,;mucosa well supplied with blood 0

Slight variation to normal situation 1

Variation of temperature 1–2�C; respiration and pulse þ30% 10

Variation of temperature >2�C; respiration and pulse �50% 20

V. Degree of strain (DS)

DS0¼no strain 0

DS1¼mild (careful observation necessary) 1–9

DS2¼moderate (veterinary support; analgesics) 10–20

DS3¼ severe (high strain) (consult animal protection commissary; veterinary supportmandatory; termination of experiment; euthanasia)

�20

PO: postoperative.

Drude et al. 471

in all groups with a maximum of 4 out of a maximumof 20 points (Figure 2). There were no significant dif-ferences within the groups (VIVO-107, fibrin glue orNaCl control). Nevertheless, a slight increase in theoverall score was observed in that for some animals(n¼ 20; random number out of each group: n¼ 10(VIVO-107); n¼ 4 (fibrin glue); n¼ 6 (NaCl control))the score sheet was, as a direct consequence of scoring,repeated on day 4 for further evaluation of the well-being of the animals. The results are presented inFigure 3 and showed a slight decrease on day 4 afterthe analgesic treatment was stopped.

The body weight changes were comparable for thegroups treated with VIVO-107 and Tissucol Duo S.

The decrease in body weight in the first three dayswas <10% in the adhesive-treated groups as well asin the NaCl control. Numerically, on average thisaccounted to 3.1 score points for the adhesive treatmentand 3.0 score points for the control group. The bodyweight was not corrected for the weight of the resectedliver. Over seven days the 0.9% NaCl group showed thesmallest change of 2% body weight compared with4.0 to 5.3% weight changes for the VIVO-107 andfibrin glue, respectively. However, these differenceswere not statistically significant. Mean absolute valuesfor the different groups are shown in Figure 4.

In addition to suffering loss of body weight the spon-taneous behavior of the animals was observed to bedifferent, except in the case of the control group.However, these behavioral changes were exclusivelyvisible postoperatively within the first 24 h,

Figure 3. Mean scores of overall sum (of all groups) ofscoring points with body weight, spontaneous behavior,general health status and clinical signs having a part in thedegree of strain (mean�SEM; [n¼ 34 (POd1–POd3)];[n¼ 20 (POd4)]).POd: postoperative day.

Figure 2. Overall score points at three days post partialhepatectomy are given as mean�SD and showed no sig-nificant difference between the adhesive treatment withVIVO-107 (n¼ 14) (black sqaures), clinical gold standard(n¼ 10) (open circles) or with the NaCl control (n¼ 10)(grey triangle) groups (one-way ANOVA, Tukey’s test).Degree of strain (DS)¼ 1 (minor).POd: postoperative day.

Figure 5. Score points of spontaneous behavior in thethree different animal groups at three consecutive dayspost-surgery. Score points are presented as meanvalues�SD.POd: postoperative day.

Figure 4. Illustration of the mean body weight (�SD) withequivalent weight loss (�SD) at day 7 post-surgery in thethree different groups.POd: postoperative day.


and furthermore exclusively scored as slight variation ifany to the normal behavior (score� 1) (Figure 5 andTable 1).

No changes were documented regarding either thegeneral health state of the animals or clinical resultssuch as body temperature, respiration and pulse viapulseoxymetry (Nellcor; Medtronic Corp, Fridley,MN, USA) (Table 1 and Figure 5).

Discussion

The aim of this study was to evaluate the severity of anestablished rabbit liver resection model with a survivaltime of seven days and a scoring system with dailyobservation on three to four days post-surgery. Twodifferent adhesive treatments as well as an NaCl controlgroup were compared. Analgesia was given pre-surgeryas well as for three days post-surgery. The severityassessment of the animals was performed using ascore system based on body weight, general healthstatus, spontaneous behavior, and clinical examination.The analgesic treatment for this first evaluation wasidentical for all the animals. However, in future studiessubsequent doses could be adjusted according to theresults of the score sheet. Increasing values indicatedincreasing severity and correlated to an impairment ofthe general condition. The data indicated that the par-tial hepatectomy itself as well as the products used (syn-thetic glue VIVO-107 and fibrin glue) were welltolerated. Within the different groups the syntheticglue VIVO-107 showed no significant differences withrespect to the score sheets when compared with eitherthe clinical gold standard of fibrin glue or the NaClcontrol group. The results of the severity assessmentwith our score sheet indicated a mild burden in allgroups without any score point deriving from the sec-tion on general health status or clinical estimation onthe score sheet. The data showed the liver resectionherein described (with analgesia given postoperativelyfor three days) to be a minor intervention, in contrast tothe EU Directive 2010/63 which states that such a pro-cedure, even a laparotomy itself, is moderate. Thisshould not be seen as an overestimation of the severityof the procedure in the EU Directive 2010/63, becauseit might even be an underestimation due to the non-optimized scoring in this context. This could be evalu-ated in more detail, e.g. by comparing different scoresheet systems as well as by taking more objective datavia telemetry, e.g. blood pressure, heart rate, etc.

The rabbits from all groups lost weight over the timecourse of the study. The weight loss documented duringthe first three days was not corrected for the resectedpart of the liver. At day 7 the change in body weightwas on average 4.0 to 5.3% in the adhesive-treatedgroups, and 2% in the NaCl group (for absolute

values, see Figure 2). A body weight change of up to5% is rated as a low burden for animals (and is thusscored with a 1 score point), according to the modifiedscore system of Morton DB et al., which is recognizedby the governmental animal care and use committee inthe state of North Rhine Westphalia in Germany and isused to evaluate the burden of animals in experimentalprojects.10

However, when considering body weight onehas to keep in mind that the change in weight maybe delayed, reducing its usefulness as a criterion fordetecting acute causes of pain/distress in an animal. Itmight also be misleading, as for example in the case oftumors which can even lead to an increase in bodyweight. Before drawing conclusions from a decreasein body weight, the water and food intake of the ani-mals could give meaningful information on their well-being.

Another point that could be optimized is the moni-toring of clinical results, even though no variation wasseen in the present study. In general, a pilot study couldhelp to estimate and identify objective clinical signs thatlead to severe pain/distress in an animal. Early clinicalsigns can thus be used and listed in the scoring proced-ure to predict later ones, and thereby prevent suffering.7

Overall blood loss as well as bleeding time during theprocedure could be included in the score sheet as aseparate section or as a clinical sign; thus for exampleif a critical loss of blood is reached during surgery theanimal can be euthanized immediately instead of beingawakened again.

Apart from the objective signs the most widely non-invasive method used to assess pain/distress in animalsis still their behavior.11 However, before any abnormalbehavior can be identified, knowledge of the typicalnormal behavior of the species is required.Observation by the animal caretaker is necessary todetermine whether behavior has been affected by theprocedure or whether it preexisted. Very little informa-tion is available with regard to postoperative scoringsystems and pain assessment after surgical proceduresin rabbits. The most recent sources are Leach et al. in200912 and Farnworth et al. in 201113 who discussedand evaluated behavioral responses to pain in rabbitsafter abdominal surgery. Results from Leach et al. wereevaluated postoperatively for four days, and suggestedthat pain after ovariohysterectomy induced changes inthe frequency and duration of some behavior. Besidesfocusing on oral administration of meloxicam, Leachet al. postulated a scoring system for pain assessmentbased on their results. The most significant behaviorwas inactive pain behavior, which occurred very fre-quently immediately and exclusively after surgery, andwhich was suggested to be a direct response to surgicalpain and distress. This inactive pain behavior decreased

Drude et al. 473

postoperatively over time.12 Nevertheless, as was alsomentioned by Farnworth et al., one has to be careful indeciding whether a behavioral change is really a resultof the procedure or is a result of anesthesia/analgesia.Control groups could clarify this issue. Farnworth et al.referenced the work by Leach et al. to validate andestablish postoperative behavioral responses in rabbits.Data collection was performed by video recording andanalyzed in real time using the Observer XT software(Noldus Information Technology, Wageningen, TheNetherlands). Behavior with most significant indica-tions of animal well-being is summarized in Table 2.Adding these behavioral changes to our scoringsystem might have increased its sensitivity.

Even though Farnworth et al. focused exclusively onthe first 0–6 h post-surgery, their conclusions aboutbehavioral changes are in good agreement with thepoint scoring of our study where behavioral changeswere seen within the first 24 h but disappeared after-wards. In line with the works discussed, the evaluatedscore sheet system shows the importance and the highvalue of behavioral assessment related to pain and dis-tress in animals. Score sheets should thus be preciseregarding the questions of which behavioral changesneed to be followed that really are related to pain,and how they should be. Video observation shouldensure that changes are assessed correctly.

In summary, a score sheet for the assessment of theseverity of a surgical procedure is most likely to fail inpredicting acute failure (e.g. re-bleeding), especiallywhere symptoms occur only at a very late stage andwhere the animal’s welfare is already compromised.But it is unlikely that each different variable will bescored wrongly, and so there is a certain stability inthis type of scoring system.

Conclusion

The recovery status (estimation of severity) of rabbits ina standardized well-established liver resection model

with two different adhesive treatments and one controlgroup was assessed over three consecutive days aftersurgery using a score system which included the criteria:body weight, general health status, spontaneous behav-ior and clinical examination. In contrast to the classifi-cation of a laparotomy as a moderate procedure in theEU Directive 2010/63, our study categorized the pro-cedure as a mild intervention in all groups. However,observations especially during the first 24 h are largelymissing in the score sheet used herein and, as alreadydiscussed, an expansion of the score system mightincrease its sensitivity and thus might change the finalclassification. It is also necessary to correlate the scoresheet data in the future with more objective data, e.g.blood pressure, heart rate, cortisol levels, etc. The pro-cedure (partial hepatectomy) itself, as well as the adhe-sive treatment using either synthetic glue VIVO-107 orfibrin glue, was well tolerated. Score points were almostsolely a result of body weight loss, whereas clinicalsigns and general health status had no influence onthe overall sum of score points. Behavioral variationwas observed exclusively within the first 24 h post-surgery, which is in line with results from othergroups where behavioral changes were most frequentlyobserved close to the surgical procedure. The directcomparison with other studies of postoperative painexpression in rabbits suggests that behavioral indica-tors of pain have to be evaluated very precisely andwith detailed knowledge of the undisturbed species-specific behavior.14

Within this study we were not able to introduce ageneralized scoring system but discussed limitationsand considered how to make improvements. The lessobjective the parameters scored and the less precisethese parameters are defined, the more the outcomeof a score sheet relies on the skill and experience ofeach observer and the more it might differ from others.

In summary, score sheets need to be re-evaluatedand tailored to each procedure and each speciesundergoing the procedure. Even with its limitations

Table 2. Content from Farnworth et al. 2011.13

Behavior Description

Full-body flex Body jerks upwards for no apparent reason

Stagger Partial loss of balance

Hind-leg shuffle Simultaneous forward movement of both forelimbs followed byslow and shuffling movement of hind legs, one at a time

Shuffle Walking at a very slow pace

Grooming Self-directed cleaning behavior

Lateral lie Lying on either left or right side of the body. Head may or may not be raised

Tight huddle Sitting with spine arched and fore and hind limbs drawn in tightly

Eyelid position – closed Eyes tightly closed


(e.g. an inability to predict acute failure) it remains ahelpful tool in assessing and safeguarding animal well-being, and is thus acknowledged as good scientific prac-tice, and a valuable refinement tool in laboratoryanimal science.

Acknowledgement

The authors would like to thank Pascal Paschenda for his

excellent technical assistance.



article.

Funding


References

1. Tillmanns H. Experimental und anatomischeUntersuchungen uber Wunden der Leber und Niere.Virchows Arch 1879; 78: 437–465.

2. Power C and Rasko JE. Whither Prometeus’ liver? Greekmyth and the science of regeneration. Ann Intern Med2008; 149: 421–426.

3. Russell WMS and Burch RL. The principles of humaneexperimental technique. London: Methuen, 1959.

4. Directive 2010/63/EU of the European Parliament and of

the Council of 22 September 2010 on the protection ofanimals used for scientific purposes. Official Journal ofthe European Union, 2010.

5. Tierschutzgesetz in der Fassung der Bekanntmachung

vom 18 Mai 2006 (BGBl. I S. 1206, 1313), das zuletzt

durch Artikel 8 Absatz 13 des Gesetzes vom 3.Dezember 2015 (BGBl. I S. 2178) geandert worden ist.In: Verbraucherschutz ESdBdJuf (ed). 2006.

6. Morton DB and Griffiths PH. Guidelines on the recog-nition of pain, distress and discomfort in experimentalanimals and an hypothesis for assessment. Vet Rec1985; 116: 431–436.

7. Morton DB. The importance of non-statistical design inrefining animal experiments. REDVET 2008; IX: 1–14.

8. Kroez M, Lang W and Dickneite G. Wound healing and

degradation of the fibrin sealant Beriplast following par-tial liver resection in rabbits. Wound Rep Reg 2005; 13:318–323.

9. Coulter CA, Flecknell PA and Richardson CA. Reportedanalgesic administration to rabbits, pigs, sheep, dogs andnon-human primates undergoing experimental surgical

procedures. Lab Anim 2009; 43: 232–238.10. Regierung von Unterfranken. Score sheets, abbruchkriter-

ien, versuchsbezogene belastungseinschatzung. Wurzburg,2015.

11. Weary DM, Niel L, Flower FC and Fraser D. Identifyingand preventing pain in animals. Appl Anim Behav Sci2006; 100: 64–76.

12. Leach MC, Allweiler S, Richardson C, Roughan JV,Narbe R and Flecknell PA. Behavioural effects of ovar-iohysterectomy and oral administration of meloxicam in

laboratory housed rabbits. Res Vet Sci 2009; 87: 336–347.13. Farnworth MJ, Walker J, Schweizer KA, et al. Potential

behavioural indicators of post-operative pain in malelaboratory rabbits following abdominal surgery. Anim

Welfare 2011; 20: 225–237.14. Kohn DF, Martin TE, Foley PL, et al. Guidelines for the

assessment and management of pain in rodents and rab-

bits. J Am Assoc Lab Anim Sci 2007; 46: 97–108.

Drude et al. 475

ab oratory

l i m i t e d

lan imals

Special Article

Anaesthesia and analgesia in laboratoryadult zebrafish: a question of refinement

Tania Martins1, Ana M. Valentim1,2,3, Nuno Pereira4,5,6,7,8 andLuis M. Antunes1,2,3

AbstractAnaesthesia is used daily in fish experimental procedures; however, the use of an inadequate anaestheticprotocol can compromise not only the animal’s welfare but also the reliability of results. The use of zebrafish(Danio rerio) in biomedical research has increased in the last decades, highlighting the importance ofappropriate anaesthetic regimens for this species. This article reviews the main anaesthetic agents andprotocols used in laboratory adult zebrafish, and some of the analgesic methods to be used in this speciesthat still need more research. In addition, a systematized observation of signs is proposed to evaluate adultzebrafish welfare to reduce pain and distress.

Keywordsanaesthesia, analgesia, refinement, welfare, zebrafish

The use of zebrafish (Danio rerio) in biomedicalresearch has increased. Indeed, the percentage of pub-lications using zebrafish has almost tripled in the lastdecade1,2 in several research areas. In biomedicalresearch, the behavioural or physiological changesthat occur when an animal is exposed to a stressful orpainful event can lead to unreliable results. The use ofanaesthetic, sedative, or analgesic drugs is essential forreducing stress and/or pain in fish since they are sen-tient animals capable of pain perception.3–5 Studieshave demonstrated that fish show changes in normalbehaviour and in physiological responses after beingsubjected to noxious stimuli, which may be amelioratedby the use of analgesics.6–8 Despite these evidences, thediscussion about pain perception in fish is still ongoingamong researchers.9 Nevertheless, there is an agree-ment regarding the importance of fish welfare andthat efforts should be made to minimize painful andstressful conditions.9,10

Although the use of anaesthetics is important toensure zebrafish welfare, these drugs can also haveside-effects;11 and it is essential to establish an anaes-thetic regimen (doses, combinations) that suits eachresearch procedure so as to minimize collateral effects.This review summarizes the anaesthetic and analgesicdrugs that are used in laboratory adult zebrafish, anaes-thesia protocols, anaesthesia depth, and recovery. Also,during all experiments it is of major importance to

monitor zebrafish welfare, and for this, we propose ascore sheet to monitor distress and pain.

Anaesthesia

Anaesthesia and sedation (Table 1) are routinely usedin husbandry, clinical procedures and research. Thus, itis important to understand the side-effects of thesedrugs and how they may affect project outcomes

1Centre for the Research and Technology of Agro-Environmetaland Biological Sciences (CITAB), University of Tras-os-Montesand Alto Douro (UTAD), Vila Real, Portugal2Instituto de Investigacao e Inovacao em Saude, Universidade doPorto, Porto, Portugal3Laboratory Animal Science Group, Instituto de Biologia Moleculare Celular (IBMC), Universidade do Porto, Porto, Portugal4Oceanario de Lisboa, Esplanada D Carlos I, Lisboa, Portugal5Instituto Gulbenkian de Ciencia (IGC), Oeiras, Portugal6ISPA – Instituto Universitario, Lisboa, Portugal7Chronic Diseases Research Center (CEDOC), Nova MedicalSchool, Lisboa, Portugal8Faculty of Veterinary Medicine, Lusofona University, Lisboa,Portugal

Corresponding author:L Antunes, Centre for the Research and Technology ofAgro-Environmetal and Biological Sciences (CITAB), University ofTras-os-Montes and Alto Douro (UTAD), 5001-801 Vila Real,Portugal.Email: [email protected]

Laboratory Animals

2016, Vol. 50(6) 476–488



sagepub.co.uk/


DOI: 10.1177/0023677216670686

la.sagepub.com

in research. Anaesthesia has been described to nega-tively affect cognitive functions or to cause neurotox-icity, however, these effects are dependent on individualanimals’ age, regimen of administration, dose, durationand type of anaesthesia.12 These factors combined witha yet scarce knowledge regarding anaesthesia in fish,make it difficult to choose an appropriate anaestheticprotocol suitable for each procedure.

Anaesthesia in fish may be achieved by diluting theanaesthetics in water (inhalation anaesthesia), whichinduces a general anaesthesia with the drug beingabsorbed mainly through the gills but also through theskin in a few species.3,13,14 In fish, anaesthesia effectsmay vary depending on the administration route, pH,temperature, salinity, oxygenation, nitrogenous com-pounds and other water conditions.3 Furthermore,

Table 1. Anaesthesia stages in fish.

Stage ofanaesthesia Description Physiological and behavioural signs Clinical interest

0 Normal Total equilibriumNormal muscle toneNormal reaction to visual and

tactile stimuliNormal respiratory rate

I Light sedation Slight loss of reaction to visualand tactile stimuli

Can reduce stress and physicaltrauma during transport

II Deep sedation Slight decrease in muscle toneNo reaction to visual and light

tactile stimuliSmall decrease in respiratory rate

Appropriate stage for close visualobservation and for minimalmanipulation, weighing andmeasuring

III Light narcosis/excitement phase

Partial loss of equilibrium/ weakresponses to postural changes

Decrease in muscle toneIncreased reaction to visual and

tactile stimuliRespiratory rate increased

and/or irregular

Higher risk of physical injury orescape/jump from container oraquarium

IV Deep narcosis Total loss of equilibrium/lack ofresponses to postural changes

No reaction to minor visualand tactile stimuli

Respiratory rate decreasingto almost normal

Good plane for external samplingand blood sampling. Avoid painfulprocedures/analgesia may not bepresent. Suitable for imagingtechniques

V Light anaesthesia Complete loss of muscle toneNo reaction to painful stimuliDecrease in respiratory rateDecrease in heart rate

Minor surgical procedures: finbiopsies and gill biopsies

VI Surgical anaesthesia Absence of reaction to massivestimulation

Respiratory rate very lowSlow heart rate

Major surgical procedures

VII Medullary collapse/overdose

Flaccid muscle toneApnoea – absence of respiratory rate,

which can be followed in severalminutes by cardiac arrest ifanaesthesia depth is not decreased

Eventual death

Appropriate for euthanasia

Adapted from Murray 200217 and Pereira 2016.18

Martins et al. 477

anaesthesia depth and recovery depend on its duration,anaesthetic concentration, animal’s body weight andmetabolism, gill surface, fish health status, strain, age,and on the different particularities of fish species.3,11,15

Thus, anaesthesia trials with small numbers of fish, i.e.pilot studies, must be performed to determine the opti-mal dosage and exposure time prior to the establishmentof protocols. In addition, proper training and supervi-sion of fish anaesthesia are essential to avoid complica-tions that can lead to death. Not only should anaesthesiabe carefully monitored but also complete fish recov-ery,2,16 which has been disregarded in the literature.

Anaesthetic agents used in fish

The ideal anaesthetic agent should (i) be easy to admin-ister and effective at low dose or exposure; (ii) be ableto induce sedation or anaesthesia in less than 3minwith a minimum of stress; (iii) provide immobilizationand effective analgesia throughout the procedure;(iv) induce a quick recovery from the anaesthetic stagewithin 5min; and (v) induce no or minimal changes inphysiology and behaviour during or after anaesthesia.15

Ideally, the anaesthetic should be affordable, easily avail-able, practical to use, and safe for the operator.

A recent international survey showed thataround 93% of the surveyed participants used tricainemethanesulfonate (MS-222) for zebrafish anaesthesia.The use of 2-phenoxyethanol, benzocaine, cloveoil, isoeugenol, etomidate, and lidocaine was alsoreferred.19 In the following, several anaesthetic agentsare discussed.

MS-222

MS-222, classified as a local anaesthetic, is the mostused inhalant anaesthetic in fish. MS-222 is highlyabsorbed through the gills and is administrated bybath, inducing general anaesthesia. Overall it is a safeanaesthetic for fish,20 although there are some concernsregarding risks of overdose in deeper stages of anaes-thesia and long duration procedures, mainly in smallanimals such as zebrafish.2,14,20,21 The anaesthetic solu-tion of MS-222 should be buffered before use due to itsacidic nature, which may cause aversion, epidermal andcorneal lesions, and physiological alterations in thefish.14,22,23 Also, MS-222 can be toxic to humans.14,24

Clove oil

Eugenol is the major constituent (70–90% by weight) ofclove oil extracted from the plant Syzygium aromaticum.Clove oil and eugenol are used as inhalant anaesthetics,and must be mixed with ethanol to be soluble in a waterbath. They show rapid induction times and consistent

anaesthesia, however, fish recovery takes longer thanwith MS-222.3 Clove oil is efficient at a range of tem-peratures, easily available, and relatively inexpensive.3

Aqui-S is a similar product available in the market con-stituted of isoeugenol, another compound of clove oil,which is soluble in water. Both have been suitable forharvesting and fish transportation.3,15 In general, thereare equivocal evidence of carcinogenic activity of euge-nol and isoeugenol, while methyleugenol, another cloveoil constituent, is carcinogenic to rodents.25

Metomidate and etomidate

Metomidate and etomidate are non-barbituratehypnotic drugs, and both are used as inhalant anaes-thetics in a water bath in fish. These agents inducequick anaesthesia induction and recovery but theyshould only be used for minor procedures, as they donot induce a surgical anaesthetic stage or analgesia.3,15

Also, they alter fish physiology by suppressing cortisolproduction.11,15

Lidocaine

Lidocaine hydrochloride, a local anaesthetic and anal-gesic,15 is used as an immersion anaesthetic in fish. Itinduces anaesthesia within one minute and the recoveryis also rapid, taking about three to four times the induc-tion period.15 A high dose of lidocaine seems promisingas an anaesthetic agent for surgical procedures but hasa low margin of safety in zebrafish.20 Thus, the additionof another agent, such as propofol, may potentiate thiseffect and reduce the dosage.2 Moreover, perioperativeanalgesia with lidocaine seem to improve zebrafishwelfare.26

Propofol

Propofol is a sedative-hypnotic anaesthetic drug usedfor the induction and maintenance of general anaesthe-sia.27 Propofol can either be injected or used in ananaesthetic bath. It is rapidly metabolized, thus lackingany cumulative effects. Although propofol is highlylipophilic, it can induce anaesthesia in a rapid andsmooth way. Moreover, the recovery from anaesthesiais also quick and complete.2,28 Although the prelimin-ary results seem promising2 it may not be fully solublein the water and more research is needed.

Ketamine

Ketamine is an injectable agent often used in mammalsand induces a dissociative anaesthesia with some anal-gesia.29,30 In fish, it can either be injected or used in ananaesthetic bath. The use of ketamine in fish depends


Table

2.Anaestheticandanalgesicagents

testedin

laboratory

adultzebrafish

.

Species

Anaesthetic

or

analgesic

agent

Anaesthetic

stage/

analgesia

Dose

Obse

rvationsreported/

Comments

Reference

Zebrafish

(Danio

rerio)

MS-222

StageV

100,120,140,160,

180and200mg/L

Immersion

Theach

ievementoflightanaesthesia,induce

dbyalldose

s,rangedfrom

ameantimeof

1min

to2.75min

inaninverserelationsh

ipresp

onse

todose

.Theach

ievementofsu

rgicalanaesthesia,

induce

dbyalldose

s,rangedfrom

amean

timeof3min

to14min

inaninverse

relationsh

ipto

dose

.Inter-individualvariabilityto

ach

ieve

anaesthesiaprese

ntatalldose

s.Thetotalreco

very

ofequilibrium,andreaction

toexternalstim

uliin

thereco

very

phase

of

anaesthesiaoccurredatapproximately

3min

regardless

ofthedose

.

Grush

etal.

200435

StagesIII–IV

100and120mg/L

Immersion

Dose

swere

unable

toinduce

lightanaesthesia

inallsu

bjects.

Huangetal.

201021

StageV

140,160,180

and200mg/L

Immersion

Dose

sinduce

dlightanaesthesiawithin

2min.

Alldose

s(100–200mg/L)

Immersion

Tim

erequiredto

ach

ieve

resp

iratory

arrest

rangedfrom

ameantimeof3min

to12min


ipresp

onse

todose

.Subjectstookover2.2min

toreco

verafter

showingsignofresp

iratory

arrest.

N/A

100mg/L

Immersion

Inductionofbehaviouralaversion.

Readmanetal.

201322

StageVI

150mg/L

Immersion

Surgicalanaesthesiawasach

ievedwithin

ameantimeof4.2min

andreco

very

2.3min

post-anaesthesia.

Nomortality

occurredduringanaesthesiaor

post-anaestheticperiod.

Nodistress

fulbehaviours

were

obse

rvedduring

inductionorreco

very.

Collym

ore

etal.

201420

(continued)

Martins et al. 479

Table

2.Continued

Species

Anaesthetic

or

analgesic

agent

Anaesthetic

stage/

analgesia

Dose

Obse

rvationsreported/

Comments

Reference

N/A

168mg/L

Immersion

Oneexp

osu

rehadnoeffect

onse

veralbehav-

iouralparameters

ofanxiety

andactivity

immediately,5min,30min

and1hafter

reco

very.

Nordgreenetal.

201416

N/A

150mg/L

Immersion

Oneexp

osu

reca

use

dco

nditionedplace

aver-

sionand53%

ofthesu

bjectsco

mpletely

rejectedthepreviouspreferredsideof

thetank.

Wongetal.

201436

StageV

75mg/L

Immersion

Lightanaesthesiawasach

ievedin

less

than

2min.

Tim

eto

reco

verwasless

than1min.

Chambel

etal.201537

StageV

100mg/L

Immersion


ievedin

less

than

2min.

Tim

eto

reco

verwasless

than1min.

StageV

125mg/L

Immersion


ievedwithin

1min.

Reco

very

occurredwithin

1min.

StageV

150mg/L

Immersion


ievedwithin

1min.

Reco

very

occurredwithin

1min.

Exp

osu

refor30min

atthis

dose

induce

d50%

of

mortality.

StageV

200mg/L

Immersion


ievedin

less

than

1min.

Reco

very

occurredwithin

1min.

Exp

osu

refor30min

atthis

dose

induce

d100%

ofmortality.

StageV

250mg/L

Immersion


ievedin

less

than

1min.

Reco

very

occurredwithin

1min.

Exp

osu

refor30min

atthis

dose

induce

d100%

ofmortality.

StageVI

100mg/L

Immersion

Nomortality



Valentim

etal.

20162 (continued)


Table

2.Continued

Species

Anaesthetic

or

analgesic

agent

Anaesthetic

stage/

analgesia

Dose

Obse

rvationsreported/

Comments

Reference

Zebrafish

(Danio

rerio)

Clove

oil

N/A

55mg/L

Immersion

Oneexp

osu

reca

use

dso

meco

nditionedplace

aversionandonly

19%

ofthesu

bjectsco

m-

pletely

rejectedthepreviouspreferredside.

Wongetal.

201436

StageV

60,80,100,

120and

140mg/L

Immersion

Theach

ievementoflightanaesthesia,induce

dbyalldose

s,rangedfrom

ameantimeof

0.5min

to1min


ipresp

onse

todose

.Theach

ievementofsu

rgicalanaesthesia,

induce

dbyalldose

s,rangedfrom

amean

timeof2.5min

to8.5min

inanegative

exp

o-

nentialresp

onse

todose

.Thetotalreco

very

ofequilibrium

afteranaes-

thesiaoccurredatapproximately

5min

regardless

ofeugenoldose

.Thereactionto

externalstim

uliin

thereco

very

phase

ofanaesthesiaoccurredatapproxi-

mately

8min,regardless

ofeugenoldose

.

Grush

etal.

200435

Zebrafish

(Danio

rerio)

Metomidate

hydroch

loride

StageI

2and4mg/L

Immersion

Thetimeto

reco

verfrom

anaesthesiaranged

from

2.8min

to5.2min.

Collym

ore

etal.201420

StageIV

6,8and10mg/L

Immersion

Lightanaesthesiaoccurredbetw

een2.3min

and

4.3min.

Nosu

rgicalanaestheticplanewasach

ieved

duringthe10min

ofexp

osu

re.

Thetimeto

reco

very

from

anaesthesiaranged

from

5.6min

to10.4min

regardless

ofthe

dose

.Nodistress

fulbehaviours

obse

rvedduring

anaesthesia.

Nomortality

duringorafteranaesthesia.

N/A

13.5mg/L

Immersion

Oneexp

osu

reca

use

dso

meco

nditionedplace

aversionandonly

11%

ofthesu

bjectsco

m-

pletely

rejectedthepreviouspreferredside.

Wongetal.

201436

(continued)

Martins et al. 481

Table

2.Continued

Species

Anaesthetic

or

analgesic

agent

Anaesthetic

stage/

analgesia

Dose

Obse

rvationsreported/

Comments

Reference

Zebrafish

(Danio

rerio)

Etomidate

StageIV

4to

10mg/L

Immersion

Inductionoflightanaesthesiawasach

ievedfor

dose

sof4mg/L

andhigher,

andoccurredin

less

than1min.

Thetimeto

reco

very

variedfrom

around25min

for4mg/L

to60min

forthehighest

dose

.

Amend

etal.198238

StageIV

8to

10mg/L

Immersion

Dose

shigherthan8mg/L

induce

dmortality.

N/A

3mg/L

Immersion

Exp

osu

reformore

than10min

and20min

showed5%

and20%

ofmortality,

resp

ectively.

N/A

9mg/L

Immersion

Exp

osu

refor120sonthreeco

nse

cutive

days

induce

d30%

ofmortality.

N/A

15mg/L

Immersion

Exp

osu

refor120sonthreeco

nse

cutive

days

induce

d95%

ofmortality.

N/A

2mg/L

Immersion

Nobehaviouralevidence

ofaversionwas

obse

rvedin

thesu

bjectsexp

ose

d.

Readmanetal.

201322

Zebrafish

(Danio

rerio)

Lidoca

ine

hydroch

loride

N/A

100mg/L

Immersion

Inductionofbehaviouralaversion.

Readmanetal.

201322

StageI

300mg/L

Immersion

Inductionoflightse

dationthatoccurredwithin

ameantimeof5.4min

withagreatvariability.

Thereco

very

from

anaesthesiaoccurredata

meantimeof3.4min


Nomortality



Collym

ore

etal.201420

StageVI

325mg/L

Immersion

Inductionofsu

rgicalanaesthesiaatamean

timeof0.85min


Thereco

very

from


meantimeof5.8min


Nomortality



StageVI/VII

350mg/L

Immersion

Inductionofsu

rgicalanaesthesiaatamean

timeof1.7min,andso

mesu

bjectsexp

eri-

ence

d

(continued)


Table

2.Continued

Species

Anaesthetic

or

analgesic

agent

Anaesthetic

stage/

analgesia

Dose

Obse

rvationsreported/

Comments

Reference

StageVII–medullary

collapse

/overdose

.Thereco

very

from


meantimeof1.9min


30%

mortality

obse

rvedduringanaesthesia.

Analgesic

0.1–2mg/kg

(IM)

Analgesicdrug.

Noside-effectsobse

rved.

Very

efficientat1mg/kg.

Sneddon201211

Analgesic

2–5mg/L*

Immersion

Pre-andpost-surgicaladministrationfortailfin

clipping.

Reductionofpain-relatedbehaviours.

Sch

roeder

201626

Zebrafish

(Danio

rerio)

Propofol

StageV

2.5mg/L

Immersion

Thedose

induce

dlightanaesthesiain

93.4%

of

thesu

bjectsbutwasunable

toinduce

anal-

gesiain

40%

ofthesu

bjects.

This

dose

showedanoccurrence

of33%

mortality.

Valentim

etal.20162

StageV

5and7.5mg/L

Immersion

Both

dose

sinduce

dlightanaesthesiabutwere

unable

toinduce

analgesiain

36%

and18%

of

thesu

bjects,

resp

ectively.

Both

dose

ssh

owedanoccurrence

of9%

mortality.

Zebrafish

(Danio

rerio)

Propofol(P)

combinedwith

lidoca

ine(L)

StageVI

2.5mg/L

(P)þ

50mg/L

(L)

Immersion

Nomortality

occurredduringanaesthesia

orpost-anaestheticperiod.

Valentim

etal.

20162

StageVI

2.5mg/L

(P)þ

100mg/L

(L)

Immersion

Occurrence

of23%

mortality.

StageVI

2.5mg/L

(P)þ

150mg/L

(L)

Immersion

Occurrence

of9%

mortality.

(continued)

Martins et al. 483

Table

2.Continued

Species

Anaesthetic

or

analgesic

agent

Anaesthetic

stage/

analgesia

Dose

Obse

rvationsreported/

Comments

Reference

Zebrafish

(Danio

rerio)

Ketamine

N/A

2000mg/L

(acu

teexp

osu

reduring5min

and

chronic

exp

osu

refor5co

nse

cutive

days)

Immersion

Subanaestheticdose

produce

dbehavioural

abnorm

alitiessu

chasincrease

dcircling

behaviour,

decrease

dgillmovement(venti-

latory

resp

onse

tohypoxia),anddecrease

dstress

resp

onse

tohypoxia(bodypulses).

Therepeatedadministrationofketaminedid

not

cause

tolerance

orse

nsitizationto

specific

drugeffects.

Zakhary

etal.201139

StageIV

(atleast)

8000mg/L

Immersion

Physiologicalanaestheticdose

inducingadeep

levelofunco

nsciousn

ess

.

Zebrafish

(Danio

rerio)

Isoflurane

StageIII

0.5mL/L

Immersion

Distress

fulbehaviours

were

obse

rved

(twitch

ing,erraticsw

imming,andpiping).

Signsofdisorientation,difficu

ltymaintaining

buoyancy,rolling,a

ndsw

immingupsidedown

alsooccurred.

Reco

very

from

anaesthesiawithin

ameantime

of4.2min.

30%

mortality

duringanaesthesiaand1day

post-anaesthesia.

Collym

ore

etal.

201420

Zebrafish

(Danio

rerio)

Isoflurane(I)

combinedwith

MS-222(M

)

StageIII–IV;

StageV

(10–20%)

50mg/L

(I)þ

50mg/L

(M)

Immersion

Lightanaesthesiawasinduce

dbythis

dose

inonly

10%

–20%

ofthesu

bjects.

Huangetal.

201021

StageIV

60mg/L

(I)þ

60mg/L

(M);

65mg/L

(I)þ

65mg/L

(M);

70mg/L

(I)þ

70mg/L

(M);

80mg/L

(I)

80mg/L

(M)

Immersion

Theco

mbinationdose

sbetw

een60þ60mg/L

and80þ80mg/L

induce

dlightanaesthesia

within

1.5min.

Alldose

s

(continued)


largely on the species, since it could cause incompleteanaesthesia, apnoea, prolonged recovery and excite-ment in salmonid species.3 Ketamine has beenrevealed to be neurotoxic to zebrafish larvae,31,32

and to interfere with the development of embryos.33

Isoflurane

Isoflurane is a hypnotic volatile drug routinely usedin mammal anaesthesia. Studies evaluating volatiledrugs in fish are scarce, since the anaesthetic depthis difficult to control, causing an overdose.3,20

Furthermore, anaesthetic preparation and anaesthe-sia should be conducted in a chemical fume hood forscavenging waste gas, and reducing the risk to theoperator. These characteristics and the observed clin-ical effects render volatile anaesthetics less practic-able in fish anaesthesia.3,20

Anaesthesia in laboratory adult zebrafish

Laboratory zebrafish has emerged as a powerful ver-tebrate model system in research.1,34 Despite thisinterest, research on welfare and refinement of pro-cedures in this species are scarce. Pain and unex-pected mortality due to incorrect anaesthesia inzebrafish can constitute a serious animal welfareissue, which increases data variability, imposing animportant scientific and economical cost on research.

Table 2 provides a summary of the main anaes-thetic agents used in adult zebrafish, and their effects.Immersion is the most common method used, espe-cially in small fish, such as zebrafish, where otherinvasive routes are impractical. Protocols for fishanaesthesia usually include only one anaestheticagent instead of a combination; however, some stu-dies have demonstrated that a mixture of two typesof anaesthetics can result in a safer and more effectiveanaesthesia.2,21

Monitoring zebrafish welfare

In addition to the importance of a good anaestheticprotocol to ensure fish welfare during experimentalprocedures, animals should also be monitored fordistress, discomfort and pain throughout the experi-mental protocol and maintenance. Fish have demon-strated to react consistently to noxious chemicalstimuli and present reliable phenotypes of stress,fear, and anxiety.40,41 Thus, the use of analgesicdrugs in fish during and after painful experimentalprocedures should be considered, especially due tothe fact that not all anaesthetic agents available forfish have proved to have adequate analgesic proper-ties.3 The information concerning the use ofT

able

2.Continued

Species

Anaesthetic

or

analgesic

agent

Anaesthetic

stage/

analgesia

Dose

Obse

rvationsreported/

Comments

Reference

Regardingalldose

s,timerequiredto

ach

ieve

resp

iratory

arrest

rangedfrom

ameantime

of6min

to55min

inanegative

exp

onential

resp

onse

todose

.Foralltheco

mbineddose

sthesu

bjects

reco

veredwithin

2min

aftersh

owingsigns

ofresp

iratory

arrest.

N/A:notapplica

ble/notevaluated.All

anaestheticagents

were

administeredbyim

mersion,withexceptions:

IM:intramuscularinjection.*P

aulSch

roeder,

personalco

mmunication,

19Ju

ly2016.

Martins et al. 485

analgesics in zebrafish is extremely scarce (Table 2).Furthermore, the available analgesics tested in fish areadministered intramuscularly and/or by the intraperito-neal route,11 which make it difficult to perform in smallfish, such as zebrafish, or in a large number of ani-mals.26 Lidocaine has recently been described as a pro-mising analgesic for zebrafish,26 which is an importantstep in the development of analgesic protocols to beadministered in a water bath.

In Table 3 we suggest a score sheet to monitor thesigns of distress and pain in adult zebrafish. This tableis an initial proposal of a score sheet to monitor

zebrafish welfare, and, as such, it should be furtherinvestigated for each individual protocol and geneticbackground.

Concluding remarks

The use of a suitable anaesthetic protocol able to pro-duce anaesthesia with effective analgesia is an import-ant refinement for painful procedures. Studiesaddressing the effects of anaesthetics in zebrafish arevariable and lack important information such as thetime during which the anaesthetic stage can be

Table 3. Proposal of a pain and distress score sheet for laboratory adult zebrafish.

Score

Physical appearance*

Normal 0

Missing operculum, and missing fins repeatedly, indicating possibleantagonist behaviours; darkening/inflammation of fin

1

Mild scoliosis/lordosis, tegument lesions, mucus production, over- orunder-conditioned (obese or thin), abrupt colour change, especiallyblanching

2

General emaciation (low body to head ratio), general body deformities,missing or protuberant scales

3

Food consumption

Normal 0

Unresponsive to food during 1 day 1

Unresponsive to food during 2 days, not even live food 2

Unresponsive to food, more than 3 days, not even live food as artemiaor rotifers (**)

3

Respiratory pattern

Normal 0

Piping or extremely low rate, almost no opercular movement 3

Swimming behaviour (**)

Swimming through the water column 0

Difficulties to control buoyancy and/or to maintain equilibrium 2

Systematic swimming on the surface or in the bottom of the tank 3

Activity (**)

Normal 0

Hyperactive (erratic swimming) or hypoactive 2

Lethargic, no reaction to external stimuli 3

Social behaviour (**)

Normal. Shoaling 0

Individual often isolated when group-housed 1

Individual always chasing or being chased by conspecifics 2

Individual does not respond to conspecific behaviours towards itself 3

TOTAL

Judgement: 0–1: normal; 2–8: monitor carefully. Consider veterinary treatment including analgesics and consider also to analyse waterquality; 9–12: suffering, provide relief, consult the specialized veterinarian, consider euthanasia; 13–18: severe status, euthanasia,rethink experimental procedure. Euthanasia may be considered if a score of 3 is observed in any of the categories, except for beha-viour-related categories (**), in which scores of 3 suggests repeated and close observation for a final decision regarding euthanasia.*Take in consideration animals’ age.


maintained without secondary effects, a clear descrip-tion of the anaesthesia stage achieved with a certaindose, and data about recovery. This lack of informationcan have a negative impact on zebrafish welfare wheninvestigators are not experienced in the use of zebrafishanaesthesia. In addition, analgesia in zebrafish isanother topic that needs further investigation in orderto refine analgesic protocols during experimental pro-cedures and for postoperative pain management, thusensuring zebrafish welfare and reliable data collection.

In general, MS-222 is a good anaesthetic, justifyingits wide use in zebrafish,19 but further refinement couldbe valuable when long duration procedures arerequired.11,21 Furthermore, two studies have describedzebrafish aversion towards this anaesthetic.22,36 Thus,finding other anaesthetic protocols are advisable, forexample, the use of anaesthetic combinations todecrease individual concentrations and the risk ofunwanted side-effects. Also, lidocaine seems to be themost promising analgesic to be used in zebrafish, how-ever, its efficacy needs to be tested in different painfulprocedures and experimental situations.

Acknowledgements

The authors would like to thank Hugo Santos (BOGA,CIIMAR – Interdisciplinary Centre of Marine andEnvironmental Research), Ana Cristina Borges, Maysa

Franco e Liliana Vale (IGC – Instituto Gulbenkian deCiencia), Petra Pintado e Fabio Valerio (CEDOC – NOVAMedical School), and Sandra Martins (Champalimaud

Foundation) for their critical comments on the scoring sheettable.



article.

Funding

The author(s) disclosed receipt of the following financial sup-

port for the research, authorship, and/or publication of thisarticle: This work was supported by the PostdoctoralFellowship SFRH/BPD/103006/2014 funded by FCT, and by

the Fellowship BI/CITAB/UTAD/VET/2015 supported by:European Investment Funds by FEDER/COMPETE/POCI –Operacional Competitiveness and Internacionalization

Programme, under Project POCI-01-0145-FEDER-006958 andNational Funds by FCT – Portuguese Foundation for Scienceand Technology, under Project UID/AGR/04033/2013.

References

1. Kalueff AV, Echevarria DJ and Stewart AM. Gaining

translational momentum: more zebrafish models for

neuroscience research. Prog Neuropsychopharmacol Biol

Psychiatry 2014; 55: 1–6.

2. Valentim AM, Felix LM, Carvalho L, Diniz E and

Antunes LM. A new anaesthetic protocol for adult zebra-

fish (Danio rerio): propofol combined with lidocaine.

PLoS One 2016; 11: e0147747.3. Neiffer DL and Stamper MA. Fish sedation, analgesia,

anesthesia, and euthanasia: considerations, methods, and

types of drugs. ILAR J 2009; 50: 343–360.

4. Sneddon LU. Pain in aquatic animals. J Exp Biol 2015;

218: 967–976.

5. Jones RC. Fish sentience and the precautionary principle.

Anim Sent 2016; 1.3: 10.

6. Sneddon LU, Braithwaite VA and Gentle MJ. Do fishes

have nociceptors? Evidence for the evolution of a verte-

brate sensory system. Proc Biol Sci 2003; 270: 1115–1121.7. Sneddon LU, Braithwaite VA and Gentle MJ. Novel

object test: examining nociception and fear in the rain-

bow trout. J Pain 2003; 4: 431–440.8. Mettam JJ, Oulton LJ, McCrohan CR and Sneddon LU.

The efficacy of three types of analgesic drugs in reducing

pain in the rainbow trout, Oncorhynchus mykiss. Appl

Anim Behav Sci 2011; 133: 265–274.

9. Rose JD, Arlinghaus R, Cooke SJ, et al. Can fish really

feel pain? Fish Fisheries 2014; 15: 97–133.

10. Iwama GK. The welfare of fish. Dis Aquat Organ 2007;

75: 155–158.

11. Sneddon LU. Clinical anesthesia and analgesia in fish.

J Exot Pet Med 2012; 21: 32–43.

12. Jevtovic-Todorovic V, Absalom AR, Blomgren K, et al.

Anaesthetic neurotoxicity and neuroplasticity: an expert

group report and statement based on the BJA Salzburg

Seminar. Br J Anaesth 2013; 111: 143–151.13. Ferreira JT, Schoonbee HJ and Smit GL. The uptake of

the anaesthetic benzocaine hydrochloride by the gills and

the skin of three freshwater fish species. J Fish Biol 1984;

25: 35–41.14. Carter KM, Woodley CM and Brown RS. A review of

tricaine methanesulfonate for anesthesia of fish. Rev Fish

Biol Fisheries 2011; 21: 51–59.15. Ross LG and Ross B. Anaesthetic and sedative tech-

niques for aquatic animals. Oxford: Blackwell

Publishing, 2008.16. Nordgreen J, Tahamtani FM, Janczak AM and Horsberg

TE. Behavioural effects of the commonly used fish anaes-

thetic tricaine methanesulfonate (MS-222) on zebrafish

(Danio rerio) and its relevance for the acetic acid pain

test. PLoS One 2014; 9: e92116.17. Murray MJ. Fish surgery. Semin Avian Exot Pet Med

2002; 11: 246–257.18. Pereira N. Introduction to anaesthesia and surgery in

fish. In: Oliveira M, Bernardo F and Robalo JI (eds)

Practical notions on fish health and production. Sharjah:

Bentham Science Publishers, 2016, pp.127–182.19. Lidster K, Readman GD, Prescott MJ and Owen SF.

International survey on the use of zebrafish in research.

(Under submission).20. Collymore C, Tolwani A, Lieggi C and Rasmussen S.

Efficacy and safety of 5 anesthetics in adult zebrafish

(Danio rerio). J Am Assoc Lab Anim Sci 2014; 53:

198–203.

Martins et al. 487

21. Huang WC, Hsieh YS, Chen IH, et al. Combined use ofMS-222 (tricaine) and isoflurane extends anesthesia timeand minimizes cardiac rhythm side effects in adult zebra-

fish. Zebrafish 2010; 7: 297–304.22. Readman GD, Owen SF, Murrell JC and Knowles TG.

Do fish perceive anaesthetics as aversive? PLoS One 2013;8: e73773.

23. Davis MW, Stephenson J and Noga EJ. The effect oftricaine on use of the fluorescein test for detecting skinand corneal ulcers in fish. J Aquat Anim Health 2008; 20:

86–95.24. Berstein PS, Digre KB and Creel DJ. Retinal toxicity

associated with occupational exposure to the fish anes-

thetic MS-222. Am J Ophthalmol 1997; 124: 843–844.25. Center for Veterinary Medicine (CVM). Concerns related

to the use of clove oil as an anesthetic for fish. Rockville:

CVM, 2007.26. Schroeder P. Exploring suitable analgesics in zebrafish –

a combined approach. In: 13th FELASA Congress.Brussels, Belgium, 13–16 June 2016, p. 32.

27. Steinbacher DM. Propofol: a sedative-hypnotic anes-thetic agent for use in ambulatory procedures. AnesthProg 2001; 48: 66–71.

28. GholipourKanani H and Ahadizadeh S. Use of propofolas an anesthetic and its efficacy on some hematologicalvalues of ornamental fish Carassius auratus. Springerplus

2013; 2: 76.29. Karmarkar SW, Bottum KM and Tischkau SA.

Considerations for the use of anesthetics in neurotoxicitystudies. Comp Med 2010; 60: 256–262.

30. Kohrs R and Durieux ME. Ketamine: teaching an olddrug new tricks. Anesth Analg 1998; 87: 1186–1193.

31. Kanungo J, Cuevas E, Ali SF and Paule MG. Ketamine

induces motor neuron toxicity and alters neurogenic and

proneural gene expression in zebrafish. J Appl Toxicol2013; 33: 410–417.

32. Burgess HA and Granato M. Sensorimotor gating in

larval zebrafish. J Neurosci 2007; 27: 4984–4994.33. Felix LM, Antunes LM and Coimbra AM. Ketamine

NMDA receptor-independent toxicity during zebrafish(Danio rerio) embryonic development. Neurotoxicol

Teratol 2014; 41: 27–34.34. Oliveira RF. Mind the fish: zebrafish as a model in cog-

nitive social neuroscience. Front Neural Circuits 2013; 7:

131.35. Grush J, Noakes DL and Moccia RD. The efficacy of

clove oil as an anesthetic for the zebrafish, Danio rerio

(Hamilton). Zebrafish 2004; 1: 46–53.36. Wong D, von Keyserlingk MA, Richards JG and Weary

DM. Conditioned place avoidance of zebrafish (Danio

rerio) to three chemicals used for euthanasia and anaes-thesia. PLoS One 2014; 9: e88030.

37. Chambel J, Pinho R, Sousa R, et al. The efficacy of MS-222 as anaesthetic agent in four freshwater aquarium fish

species. Aquacult Res 2015; 46: 1582–1589.38. Amend DF, Goven BA and Elliot DG. Etomidate: effect-

ive dosages for a new fish anesthetic. Trans Am Fish Soc

1982; 111: 337–341.39. Zakhary SM, Ayubcha D, Ansari F, et al. A behavioral

and molecular analysis of ketamine in zebrafish. Synapse

2011; 65: 160–167.40. Maximino C. Modulation of nociceptive-like behavior in

zebrafish (Danio rerio) by environmental stressors.Psychol Neurosci 2011; 4: 149–155.

41. Roques JA, Abbink W, Geurds F, van de Vis H and FlikG. Tailfin clipping, a painful procedure: studies on Niletilapia and common carp. Physiol Behav 2010; 101:

533–540.


www.sdsdiets.comthe essential resource for quality research diets

Special Diets Services

Special Diets ServicesPO Box 705, Witham, Essex, England CM8 3AD

Telephone: +44 (0) 1376 511260Fax: +44 (0) 1376 511247Email: [email protected]

Special Diets Services is the largest supplier of Laboratory Animal diets in Europe and the only dedicated manufacturer in the UK.

Special Diets Services has a global reputation for the quality of its diets and manufacturing and storage facilities.

Expert in the world of research diets

ZOONLAB GmbH

Hermannstraße 6 | 44579 Castrop-Rauxel | Germany

www.zoonlab.de

The only thing that really counts isquality. Our safe ergonomic ZOONLAB cages and enclosure systems

are aimed at species-appropriate accommodation for laboratoryanimals – owing to the fact that we are widely experiencedspecialists in metal processing. The perfect combination of expertise and technology to achieve outstanding quality – made in Germany and certified in accordance with current directives.

To ensure the smooth and efficient operation of your facility weguarantee rapid and professional maintenance and repair work.

Learn more about our experience in this field – we look forwardto your enquiry: [email protected] or +49 / 23 05 / 97 30 40

SOFTWARE FOR THE MANAGEMENT OF ANIMAL FACILITIES

ANY TYPE OF ANIMAL FACILITY SECURITY, TRACEABILITYSUPPORT IN 9 LANGUAGES

ANIMAL MANAGEMENTBreedingTransgenic linesPedigree

NorayBio Software for BiosciencesT +34 94 403 69 98 +33 04 26 73 68 [email protected] www.noraybio.com

COMMUNICATION WITH RESEARCHERSNotesServices

ETHICAL COMMITTEE

YOUR SOFTWARE

ASK FOR A DEMO - [email protected]

NEW FEATURES

NorayDocs Project Evaluation Software

AniCloud

AniXplore Get the most out of your data

Productivity statisticsORDER MANAGEMENT

ANIMAL FACILITY MANAGEMENTUser managementStock controlEquipmentFinancial management

STATISTICS AND REPORTSEU legislationAnimals entries/ exits record

DRUG REPORT MODULEData records and managementEntry/ exit follow-upReport extractionDrugs administration control

Supplying the Innovations of

Modern Research

IPS – working with leading industry brands and products to supply innovative research facilities worldwide

TestDiet®

Custom made Speciality Diets

LabDiet®

Leader in Research Animal Nutrition

Bedding & Nesting

Creating Natural Environments

Environmental Enrichment

Caring for your Laboratory Animals

BenchGuard®

& Tray Liners

Protective &Absorbent Papers

Logistics & Storage

A comprehensive range of services

Irradiation Services

Tailored to your specific needs

Customer Service

+44 (0)870 [email protected]

www.ipsltd.biz

L a b o r a t o r y A n i m a lS c i e n c e A s s o c i a t i o n

Providing leadership in laboratory animal science and welfare in support of ethical and effective animal research

JOIN TODAY!Benefits of being a member:

• Membership of a highly regarded and influential scientific association• Access to support for career development at all levels • The opportunity to serve on high level Sections and Task Forces charged with delivering scientific guidance and policy proposals• Preferential registration for the Annual Conference and other events• Access to Association bursaries, travel awards and CPD opportunities• Regular newsletter: ‘LASA Forum’

The professional society for those with an interest in...

Animal use • Care and WelfareEducation and training • 3Rs • EthicsAnimal research regulation and policy

www.lasa.co.uk/join_us [email protected]

��

��

For further information,please contact:

AddressBell Isolation Systems LtdUnit 12aOakbank Park WayLivingston EH53 0THScotland UK

Telephone+44 (0) 1506 442916

Fax+44 (0) 1506 440780

[email protected]

Websitewww.bell-isolation-systems.com

Institute of Animal Technology

The Career Pathway for Animal Technologists

www.iat.org.uk Advancing and promoting excellence in

Facility Manager / Specialist

Core Career Pathway including

Continuous Professional Development

Senior Animal Technologist

Animal Technologist

Animal Technician

Trainee Animal Technician

Level 6 LAS&T Diploma or 1st Degree

Entry to Master’s Degree

Level 5 LAS&T Diploma




Level 2 Lab Animal Husbandry Diploma

Local Induction, Health & Safety, etc

Alignment to EducationFramework

Alignment to ProfessionalRecognition

EU Modules3.2, 7, 8, 9 & 23requirementfor ICARE

EU FunctionGroup C(animal carers)Modules1, 2, 3.1, 4, 5,6.1 & 3.2

Named Animal Care and

Fellow of the Institute ofAnimal Technology (FIAT)

Registered Animal Technologist (RAnTech)

Member of the Instituteof Animal Technology (MIAT)

8years+

5years

3years

1year

6months

Alignmentto Directive 2010/ 63/EU

Appointment

TypicalIndicativeTimescales

Please see the IAT website www.iat.org.uk for further details

the care and welfare of animals in science www.iat.org.uk

IAT Education ProgrammeThe IAT is an Ofqual ofqual.gov.uk/ Awarding Organisation and offers qualifications at Further (Levels 2 and 3) and Higher (Levels 4 - 6) Education, providing internationally recognised qualifications that ensure animal technologists are prepared to meet their legal and ethical responsibilities, high standards of welfare for laboratory animals, good science and career advancement.

The Qualification and Credit Framework (QCF) is the framework for creating and accrediting learning in the United Kingdom and is recognised by the European Qualifications Framework (EQF) ec.europa.eu/ploteus/search/site which ensures that QCF qualifications are nationally and internationally recognised enabling employers to value the level of achievement allowing units gained by learners to be transferred to other relevant qualifications.

An indication of the educational levels: Level 3 is recognised as suitable for the UK Named Animal Care and Welfare Officer (NACWO), equivalent to the EU Institutional Care and Animal Welfare Responsible Person (ICARE); whereas Level 6 allows entry into Masters Degrees.

The IAT is an Ofqual accredited Awarding Organisation.

Further Education Qualifications Awarded by IAT IAT Level 2 Diploma In Work Based Animal Technology (Apprenticeship) IAT Level 2 Diploma in Laboratory Animal Husbandry IAT Level 2 Diploma in Laboratory Animal Science and Technology IAT Level 3 Diploma in Laboratory Animal Science and Technology (MIAT)*

Higher Education Qualifications Awarded by IAT IAT Level 4 Diploma in Laboratory Animal Science & Technology

Communication skills, Supervisory management, Biological science, Control of disease IAT Level 5 Diploma in Laboratory Animal Science & Technology

Animal law and welfare, Experimental design, Toxicology, Reproduction and Genetic Alteration IAT Level 6 Diploma in Laboratory Animal Science & Technology (FIAT)*

Project planning, Project, Building design and management, Stress and pain *Subject to IAT membership criteria.

“This Laboratory Animal Science & Technology Career Pathway will importantly enable animal facility staff, both in industry and academia, to achieve high standards of competence and professionalism within Home Office licensed establishments.

The Animals (Scientific Procedures) Act, 1986 and the EU Directive 2010/63 both identify the need for laboratory animal care personnel to be formally recognised as being professionally qualified, including Higher Education Qualifications at levels 4 to 6. All these qualifications will therefore build on that need. They will deliver the knowledge to enable individuals to achieve a full and sustainable career pathway. For some, this may include the entry into Master’s Degree programmes and to aspire to ever more senior positions.

I am confident that graduates from these courses will be able to make a very significant contribution in influencing the application of high standards of animal welfare, care and use in the UK Life Sciences. This will importantly include promoting and implementing the 3Rs (Replacement, Reduction and Refinement of animal use) in research facilities throughout the UK – something which the Government strongly advocates.”

Dr Judy MacArthur Clark CBE MRCVS Head, Animals in Science Regulation Unit

ab oratory

l i m i t e d

lan imals

News

Iniciativa espanola para mejorar latransparencia en el uso de animales deexperimentacion

Javier Guillen

El pasado 20 de Septiembre se presento el Acuerdo detransparencia sobre el uso de animales en experimentacioncientıfica (http://www.cosce.org/pdf/Acuerdo_Transparencia_COSCE_2016.pdf), un Acuerdo que pre-tende potenciar la apertura y comunicacion a la sociedadde la experimentacion que involucra el uso de animales.La Sociedad Espanola para las Ciencias del Animal deLaboratorio (SECAL), como miembro de laConfederacion de Sociedades Cientıficas de Espana(COSCE) promovio en 2013 la creacion de un grupode trabajo COSCE sobre experimentacion animal. Elprimer producto de este grupo fue la publicacion en2015 del Documento COSCE sobre el Uso deAnimales en Investigacion Cientıfica (http://www.cosce.org/pdf/documento_cosce_comision_animal_research.pdf), que trataba de explicar la todavıa necesaria aporta-cion de los animales al avance cientıfico para el beneficiode personas, animales y medio ambiente. Ahora se halanzado este nuevo Acuerdo con la adhesion de 90 insti-tuciones espanolas tanto publicas como privadas,incluyendo centros usuarios (universidades, institutosde investigacion, empresas), asociaciones cientıficas yasociaciones de pacientes. La adhesion de organiza-ciones sigue abierta, en un proceso que se mantendrade modo permanente tras esta presentacion. LaCOSCE ha contado con la muy valiosa colaboracionde la European Animal Research Association (EARA:www.eara.eu), una asociacion que promueve la comuni-cacion de la investigacion con animales en toda Europa.EARA se encargara de la coordinacion de las adhesionesde nuevas instituciones a traves de su web. Este Acuerdocopia la iniciativa surgida en el Reino Unido a finales de2012 que cristalizo en Mayo de 2014 con el lanzamientodel Concordato sobre Transparencia en InvestigacionAnimal (http://www.understandinganimalresearch.org.uk/policy/concordat-openness-animal-research/), mante-niendo los mismos compromisos.

Los compromisos que las instituciones adquieren aladherirse al Acuerdo de Transparencia son:

1. Hablar con claridad sobre cuando, como y por quese usan animales en investigacion.

2. Proporcionar informacion adecuada a los medios decomunicacion y al publico en general sobre las con-diciones en las que se realiza la investigacion querequiere el uso de modelos animales y los resultadosque de ella se obtienen.

3. Promover iniciativas que generen un mayor conoci-miento y comprension en la sociedad sobre el uso deanimales en investigacion cientıfica.

4. Informar anualmente sobre el progreso y compartirexperiencias.

Las primeras obligaciones tangibles para las institu-ciones seran en primer lugar, publicar una declaracionen la pagina web de cada organizacion firmante dentrodel primer ano de la firma de este acuerdo, donde seexplique la polıtica de la misma sobre el uso de ani-males en investigacion. De esta manera se proporcio-nara informacion clara sobre la naturaleza de suparticipacion en los experimentos con animales en elcontexto de la investigacion que se realiza en la institu-cion. Y en segundo lugar, informar al final del primerano de las actuaciones llevadas a cabo. Estas activi-dades pueden incluir la referencia al papel de los ani-males en los descubrimientos cientıficos, la promocionde una detallada y correcta descripcion de los experi-mentos con animales segun recomendaciones interna-cionales a la hora de publicar resultados en revistasespecializadas, la informacion de las mejoras en laimplementacion de las 3Rs, la inclusion de informacionsobre experimentacion animal en eventos publicos, o laapertura de animalarios (de acuerdo a limitacionessanitarias y de seguridad) a visitas interesadas, porejemplo de colegios o periodistas.

SECAL, Madrid, Spain

Corresponding author:Javier Guillen, C/ Maestro Ripoll 8, 28006 Madrid, Spain.Email: [email protected]

Contributions to the News section are not subject to peer reviewand reflect the opinion of the subscribing society.

Laboratory Animals

2016, Vol. 50(6) 489–491



sagepub.co.uk/


DOI: 10.1177/0023677216676736

la.sagepub.com

La SECAL anima a otras sociedades en otros paısesa promover similares procesos que pueden ayudar amejorar la percepcion que la sociedad tiene de este tra-bajo de tanta importancia e interes social.

Spanish initiative to improve transparencyon the use of experimental animals

The Spanish Concordat on openness on the use of ani-mals in scientific research (http://www.cosce.org/pdf/Acuerdo_Transparencia_COSCE_2016.pdf) waslaunched past September 20th, aiming to promoteopenness and communication to society on the scien-tific progress involving the use of animals. The SpanishAssociation for Laboratory Animal Science (SECAL),as member of the Spanish Confederation of ScientificAssociations (COSCE), promoted in 2013 the creationof a working group on animal research within COSCE.The first outcome of this working group was the releasein 2015 of the COSCE Document on the use of animalsin scientific research (http://www.cosce.org/pdf/docu-mento_cosce_comision_animal_research.pdf), thatexplained the still needed contribution of animals toscientific progress for the benefit of people, animalsand environment. Now, this new Agreement has beenlaunched with the signature of 90 Spanish institutionsfrom both public and private sectors including userestablishments (academic, research institutes, and com-panies), scientific associations and patient groups. Thesigning process is open and will be kept permanentlyafter the launching of the Concordat. COSCE hascounted with the very valuable cooperation of theEuropean Animal Research Association (EARA:www.eara.eu), an association that promotes the com-munication of animal research across Europe. EARAwill be responsible through its website for the

coordination of the signing process for new institutions.This Concordat is copying the initiative established inthe United Kingdom (UK) in 2012 that ended with thelaunching of the Concordat on openness on animalresearch in the UK (http://www.understandinganimal-research.org.uk/policy/concordat-openness-animal-research/) in May 2014, and maintains the samecommitments.

The commitments that the organisations undertakewhen signing the Spanish Concordat are:

1. Speak out clearly on when, how and why animals areused in research.

2. Provide adequate information to the media and tothe public in general on the conditions in whichresearch involving the use of animals is performedand on the results obtained from it.

3. Promote initiatives to generate a better societalknowledge and understanding of the use of animalsin scientific research.

4. Inform annually on the progress and exchangeexperiences.

The initial obligations that signing institutions willhave to implement in practical terms will be: firstly tomake public a statement in the institution’s websitewithin the first year after the signing of thisConcordat, with a description of the institutionalpolicy on the use of animals in research. This wayclear information on the nature of its involvement inanimal research will be provided, in the context of theresearch being performed. And secondly, inform by theend of the first year on the activities performed relatingto the Concordat commitments. These activities caninclude: references to the role of animals in scientificachievements; promotion of detailed and appropriate


description of the experiments according toInternational recommendations when publishing in sci-entific journals; informing on the implementation of3Rs; including information on animal research inpublic events; or accepting visits to animal facilities(according to sanitary and safety limitations) by inter-ested parties such as schools of journalists.

SECAL encourages other associations in other coun-tries to promote similar initiatives the may help improv-ing the societal perception on this field of work withsuch important social benefit.

Guillen 491

Calendar of events

Meetings of interest to laboratory animals scientists and technicians: references to Laboratory Animals are for furtherdetails. Items for inclusion should be sent to Notes and Comments Editor, LAL, PO Box 373, Eye, Suffolk, IP22 9BS, UK.Email to [email protected]. The deadlines for inclusion of material are: February issue, 10 November; April issue,10 January; June issue, 10 March; August issue, 10 May; October issue, 10 July; December issue, 10 September.

20166 December Scottish TSE network annual symposium, ‘‘Neurodegeneration: Models, Mechanisms

& Resources’’, Roslin Institute, Edinburgh, Scotland. For further information visit http://www.stn.ed.ac.uk/stn/seminars.html

8 December RSPCA Lay Members’ Forum, London, England. For further information visit https://goo.gl/forms/Nch88tFyw3qMGvzD3

13–15 December British Pharmacological Society annual meeting 2016, London, England. For further infor-mation visit British Pharmacology Society website.

201712 January AFSTAL/ComTech 15th symposium, Cite Internationale Universitaire de Paris, Paris,

France. For further information visit http://www.alphavisa.com/comtech/2017/17–19 January Introduction to Mathematical and Computational Modelling: from mouse to computer and

back, Eindhoven University of Technology, Eindhoven, The Netherlands. For further infor-mation visit https://docs.google.com/forms/d/e/1FAIpQLSeD4GWCCLfyikMx3vwCKLjxSrrPrgIFbpKsCt3jr5jtsWS_nw/viewform?c=0&w=1

26 January LAVA IAT LASA Named Persons’ Workshop, Central UK. For further details visit www.iat.org.uk

8–10 March Organizing and Operating Activities in a Laboratory Animal Facility, FGB , Milan, Italy. Forfurther details visit www.fondazioneguidobernadini.org

21–24 March Institute of Animal Technology Congress, West UK. For further details visit http://www.iat.org.uk/congress

22–24 March Charles River European Short Course. ‘‘Empower your Professional Self’’. Berlin, Germany.For further details visit http://www.iat.org.uk/congress

30 March LASA and Fish Vets’ Society – Fish Anaesthesia, Edinburgh, Scotland. Further informationto follow.

30 May–2 June Scand-LAS Annual Conference. Do we need animal experimentation? Copenhagen,Denmark. For further information visit http://eushortcourse.criver.com/

11–13 June 6th Infusion Organisation Conference, Cologne, Germany. For further information visithttps://www.ito-org.eu/index.php?id=109

11–13 September GV-SOLAS National Meeting, Cologne, Germany. For further information visit www.gv-solas2017.de

15–19 October AALAS National Meeting, Austin, Texas. For further information visit https://www.aalas.org/national-meeting/general-information/future-meetings#.WA4P8OArK1s

28–30 November LASA Annual Conference, Birmingham, England. Further information to follow.

Index to Advertisers December 2016

Altromin International OBCAnLab Ltd xix

Bell Isolation Systems Ltd xxiv

Charles River Laboratories IFC

Datesand Ltd xiDatesand Ltd xviiiDustcontrol AB xix

Edstrom Inc xEnvigo viii

Fine Science Tools GmbH iv

Granja San Bernardo SL xxvGVG Diagnostics ix

Institute of Animal Technology xxvi, xxviiIPS Product Supplies Ltd xxITO Infusion Technology Organisation xxi

Laboratory of Pharmacology and ToxicologyGmbH & Co KG vii

LASA xxiiLBS xxv

Marshall BioResources xii

NorayBio xvii

PFI Systems Ltd xiv

R C Hartelust bv xxiii

SAFE xiiiScionics Computer Innovation GmbH xxiiiSpecial Diets Services xvSurrey Diagnostics xxviii

Tecniplast SpA iiiTecniplast SpA IBC

ZOONLAB xvi

xxviii...............................................................................................................................................................

Volume 50 Number 6 December 2016 ISSN 0023-6772 ......la.sagepub.com Published on behalf of...

Documents

Transcript of Volume 50 Number 6 December 2016 ISSN 0023-6772 ......la.sagepub.com Published on behalf of...