Official Journal of AFSTAL, ECLAM, ESLAV, FELASA, GV-SOLAS, ILAF, LASA, NVP, SECAL, SGV, SPCAL

Volume 53 Number 2 April 2019 ISSN 0023-6772

Laboratory Animals THE INTERNATIONAL JOURNAL OF LABORATORY ANIMAL SCIENCE, MEDICINE, TECHNOLOGY AND WELFARE

Published on behalf of Laboratory Animals Ltd. journals.sagepub.com/home/lan by SAGE Publications Ltd. Does your facility have all the right people with all the right skills?

Your animal facilities can get the training, consulting, FDQKHOSƬOOWKHJDSVLQ\RXUSURJUDPPH:KHWKHU\RX DQGVWDƯQJVROXWLRQVWKH\QHHGZLWKInsourcing DUHORRNLQJIRUSHUPDQHQWRUWHPSRUDU\VWDƪZHFDQ SolutionsSM)URPUHJXODWRU\DQG&RQWLQXLQJ3URIHVVLRQDO TXLFNO\SURYLGHTXDOLƬHGSHUVRQQHOLQFOXGLQJEDFNJURXQG 'HYHORSPHQWWUDLQLQJWRSURYLGLQJIXOO\TXDOLƬHGDQLPDO FKHFNV7UDLQLQJDQGIXUWKHUHGXFDWLRQLVDYDLODEOHYLDRXU WHFKQLFLDQVYHWHULQDULDQVDQG3K'OHYHOVFLHQWLVWVZH classroom and eLearning solutions.

Visit us at stand A15 at FELASA or at www.criver.com/insourcing Volume 53 Number 2 April 2019 Contents

Review Article Guidelines for porcine models of human bacterial infections 125 LK Jensen, NL Henriksen and HE Jensen

Working Party Report FELASA accreditation of education and training courses in laboratory animal science according to the Directive 2010/63/EU 137 M Gyger, M Berdoy, I Dontas, M Kolf-Clauw, AI Santos and M Sjo¨quist

Original Articles Improved timed-mating, non-invasive method using fewer unproven female rats with pregnancy validation via early body mass increases 148 AK Stramek, ML Johnson and VJ Taylor

A rat model of nerve stimulator-guided brachial plexus blockade 160 Y Zhang, B Cui, C Gong, Y Tang, J Zhou, Y He, J Liu and J Yang

4D cardiac magnetic resonance imaging, 4D and 2D transthoracic echocardiography: a comparison of in-vivo assessment of ventricular function in rats 169 H Stegmann, T Ba¨uerle, K Kienle, S Dittrich and M Alkassar

The effects of sucrose on urine collection in metabolic cages 180 A Kovalcˇ´ıkova´, M Gyura´szova´, R Gardlı´k, M Borisˇ, P Celec and Lˇ To´thova´

Impact assessment of tail-vein injection in mice using a modified anaesthesia induction chamber versus a common restrainer without anaesthesia 190 M Resch, T Neels, A Tichy, R Palme and T Ru¨licke

News Don’t forget to register for the FELASA 14th Congress 205 J-P Mocho

Expert information from the Working Group on Hygiene: Harmonisation of Health Monitoring Reports 208 W Nicklas and K Seidel

Calendar of events/Index to advertisers 212

Serving e cient care in research.

From genotyping to colony management, help eliminate human error and promote e cient care of animals in research at your institution.

Learn how you can create e ciencies with animal models through trusted, outsourced genetic services at Transnetyx.com. CLASSIC CRAFTSMANSHIP

Scissors • Retractors • Magnifiers • Probes & Hooks • Bone Instruments • Animal Identification Hemostats • Forceps • Surgical & Laboratory Equipment • Feeding Needles • Spatulae & Spoons Wound Closure • Surgical Plates • Instrument Care & Sterilization • Rongeurs • Scalpels & Knives Clamps • Pins & Holders • Needles & Needle Holders • Student Quality Instruments & Much More

FINE SURGICAL INSTRUMENTS FOR RESEARCHTM Visit us at finescience.de or call +49 (0) 6221 905050 ...... 113 Laboratory Animals Subscription information Annual subscription (2019) including postage: AALAS Member, (print and electronic) [US$120]. Combined Institutional Rate (print and electronic) [£304/US$562]. Electronic only and print Editorial Board only subscriptions are available for institutions at a discounted Editor-in-Chief B Riederer rate. Note VAT is applicable at the appropriate local rate. Deputy Editors G Jarvis, P Jirkof Visit sagepublishing.com for more subscription details. To activate your subscription (institutions only) visit Section Section Editors online.sagepub.com. Abstracts, tables of contents and contents alerts are available on this site free of charge Anaesthesia, Analgesia, P Flecknell, M Leach, P Foley, for all. Student discounts and single issue rates are available from Pain & Stress P Hedenqvist SAGE Publications Ltd, 1 Oliver’s Yard, 55 City Road, London EC1Y 1SP, UK, tel. þ44 (0)20 7324 8500, email subscriptions@sa- Anatomy and Neuroscience B Riederer, S Wells (neuro) gepub.co.uk and in North America, SAGE Publications Inc, Aquatic Organisms K Finger-Baier (fish), PO Box 5096, Thousand Oaks, CA 91320, USA. JP Mocho, M Crim Behaviour D Preissmann, M Gyger, Advertisement Managers L Lewejohann Biostatisics & Experimental R-D Gosselin, H Wu¨ rbel PRC Associates Ltd, 1st Floor Offices, 115 Roebuck Road, Design Chessington, Surrey KT9 1JZ, UK; Tel: +44 (0) 20 8337 3749; Fax: Education P Vergara, C Tho¨ne-Reineke +44 (0) 20 8337 7346; Email: [email protected] Imaging Techniques L van der Weerd, J Tremoleda Laboratory Animals Ltd Large Animal Models M Jensen-Waern, D Anderson, Laboratory Animals Ltd is a company limited by guarantee and has T Morris no share capital. The Memorandum of Association obliges the Management of Animal Facilities J-B Prins, M Dennis company to apply all its resources to the advancement of public Molecular & Genetic Engineering T Ruelicke, P Cinelli Nutrition and Diets G Tobin, T Nortey education in laboratory animal science, technology and welfare. It is a registered charity (Registered Charity Number 261047) and Pathology & Microbiology P Clements, D Salvatori, none of its directors may receive any fee or remuneration. A Bleich Physiology & Clinical Chemistry M Sommers, T Hough Registered Office: Primates G Rainer, P Honess Laboratory Animals Ltd, 44 Springfield Road, Horsham, West 3Rs & Ethics G Griffin, A Olsson Sussex, RH12 2PD, UK Reproductive Biology H Hedrich, B Pintado, C Gilbert Council of Management Small Animal Models M Berard, J-B Prins (temporary), S Wells Chairman J-B Prins Surgical Procedures D Bouard, R Tolba Secretary E Weir Systematic Review M Ritskes-Hoitinga, Treasurer J Gregory BS Kousholt L Antunes P Nowlan Toxicology F Rutten K Applebee J Orellana Veterinary Medicine E Rivera, J Sanchez-Morgado, V Baumans B Riederer L Whitfield, M Berard A Ritchie N Kostomitsopoulos N Ezov M Ritkes-Hoitinga Special Issue Microbiota guest editor: C Gilbert A Shortland Axel Kornerup-Hansen and J Guillen S Wells Craig Franklin J Helppi M Wilkinson C Johner Comment and correspondence relating to editorial matters may be sent to the Chairman of the Editorial Board by email: laeditorial@- ß 2019 Laboratory Animals Ltd. Apart from any fair dealing for the sagepub.co.uk; or post: LAL, PO Box 373, Eye, Suffolk, IP22 9BS, purposes of research or private study, or criticism or review, as permitted UK. under the UK Copyright, Designs and Patents Act, 1988, no part of this See also http://www.lal.org.uk publication may be reproduced, stored, or transmitted, in any form or by Laboratory Animals, (ISSN 0023-6772) is published and distributed any means, without the prior permission of the publishers, or in the case bimonthly (February, April, June, August, October, December) in of reprographic reproduction in accordance with the terms of licences both print and electronic form by SAGE Publications Ltd, issued by the Copyright Licensing Agency in the UK, or in accordance 1 Oliver’s Yard, 55 City Road, London EC1Y 1SP, UK. with the terms of licences issued by the appropriate Reproduction Rights Organization outside the UK. Enquiries concerning reproduction outside All manuscripts submitted for publication should the terms stated here should be sent to SAGE at the address. be prepared in accordance with the Guidelines Whilst every effort is made to ensure that no inaccurate or misleading for Authors which can be found online at data, opinion or statement appears in the journal, Laboratory Animals Ltd journals.sagepub.com/home/lan. Please submit your paper wish to make it clear that the data and opinions appearing in the articles online at mc.manuscriptcentral.com/la, and advertisements herein are the responsibility of the contributor or contributions for news items can also be made via advertiser concerned. Accordingly, Laboratory Animals Ltd and their offi- manuscript central. cers and agents accept no liability whatsoever for the consequences of any such inaccurate or misleading data, opinion or statement. Printed in Great Britain ...... 114 AFSTAL Secretary President Nicole Linklater Teresa Rodrigo Association Franc¸aise des Sciences et Faculty of Biology Techniques de I’Animal de Laboratoire Philipps University Vice President President Karl-von-Frisch Str. 8 Isabel Blanco Sebastian Paturance 35043 Marburg Vice President Germany Secretary Elodie Bouchoux (www.gv-solas.de) Clara Mun˜oz

Secretariat: 28, rue Saint Dominique, 75007, Treasurer Paris, France David Munõz (www.afstal.com) ILAF Israeli Laboratory Animal Forum Secretariat: c/Maestro Ripoll, 8, President 28006 Madrid, Amir Rosner Spain (www.secal.es) ECLAM Secretary European College of Laboratory Animal David Castel Medicine Neufeld Cardiac Research President Institute SGV Rony Kalman Sheba Medical Center Schweizerische Gesellschaft fu¨r Tel Hashomer 52621 Secretariat: PO Box 9101NL-6500 Versuchstierkunde Israel HB Nijmegen, The Netherlands Socie´te´ Suisse pour la Science des Animaux de (www.ilaf.org.il) (http://eslav-eclam.org/eclam) Laboratoire (Swiss Laboratory Animal Science Association) President ESLAV LASA Dr. Birgit Ledermann European Society of Laboratory Animal Laboratory Animal Science Association Secretary Veterinarians President Dr. med. vet. Andrina Zbinden President David Anderson Faculty of Science and Medicine Aurora Bronstad University of Fribourg Secretary General Ch. du Museé8 Honorary Secretary Miles Maxwell CH-1700 Fribourg, Switzerland Massimiliano Bardotti (www.naturalsciences.ch/organisations/sgv) PO Box 524, Hull, Secretariat HU9 9HE, UK c/o PO Box 24, (www.lasa.co.uk) Hull, HU12 8YJ, UK SPCAL Sociedade Portuguesa de Cieˆ ncias NVP em Animais de Laborato´ rio (Portuguese Society for Laboratory Animal FELASA Nederlandse Vereniging voor Science) Federation of European Proefdierkunde Laboratory Animal Science Associations (Dutch Association for Laboratory Animal President President Science) Isabel Vito´ria Figueiredo Hanna-Marja Voipio Vice-President President President-elect Ricardo Afonso Ana Santos Martje Fentener van Vlissingen Secretary Past President Secretary Catarina Pinto Reis Heinz Brandstetter Jan Langemans Secretariat: Laborato´rio de BPRC Hon. Secretary Farmacologia Lange Kleiweg 139 Jean-Philippe Mocho Faculdade de Farmaćia 2288 GJ Rijswik Largo de D. Dinis The Netherlands Secretariat: PO Box 372, 3000 Coimbra (www.proefdierkunde.nl) Eye, IP22 9BR, UK Portugal (www.felasa.eu) (www.spcal.pt)

SECAL GV-SOLAS Sociedad Espanˇ ola para las Ciencias del Gesellschaft fu¨r Versuchstierkunde Animal de Laboratorio (Society for Laboratory Animal Science) (Spanish Society for Laboratory Animal Science) President Bettina Kraenzlin

Innovative solutions for laboratory animal containment

BioFlex® Isolators > High Quality > Expert Knowledge > Experience > Innovation > Latest Technology > Reliability > Design Flexibility > Specialise in Bespoke Designs > Flexible Film, Semi-Rigid or Fully-Rigid Designs > Large Animal Isolator Designs

Bell Isolation Systems lead the way in the design and manufacture of high quality flexible film and rigid isolators using the latest materials and technology, coupled with over forty years of innovation developing standard, customised and bespoke products.

Contact for further information: Bell Isolation Systems Ltd Telephone: +44 (0) 1506 442916 Unit 12a, Oakbank Park Way Fax: +44 (0) 1506 440780 Livingston EH53 0TH Email: [email protected] Scotland, UK Website: www.bell-isolation-systems.com In 2019 Marshall celebrates its 80th anniversary...

Marshall Beagles® are now produced both in France and the United Kingdom!

...with the opening of new possibilities for researchers in Europe.

www.marshallbio.com Delivering higher standards Environment Enrichment

Enhance your research with stimulating Enrichment products from LBS Biotech... Choose from our extensive product range so your research animals can enjoy an active, stimulating and comfortable environment. DesResTM Rodent Houses, Fun Tunnels, Toys, Balls, Chews, Treats, Bedding, Foraging - just some of our quality assured products, suitable for use in biotechnology conditions.

Contact the experts: Tel: +44 (0)1293 827940 Email: [email protected] www.lbs-biotech.com

ssniff

Spezialdiäten

“The Best or Nothing”

Standard feeds – cereal based (chow) for all animal species Special / Purified diets Medicated feed

Perfect Animal – Diet models Purified diets • for every research purpose (e.g. diabetes) Obesity, metabolic syndrome, NIDDM, atherosclerosis, (high fat and/or cholesterol)

Hypertension and chronic kidney disease (± adenine) Chronic liver diseases (steatohepatosis, ASH, NASH) Diets with deficits or excesses of certain nutrients

Test compound diets with

Doxycycline Tamoxifen (flavoured diets)

Customers’ statins

Customers’ compounds for toxicity or PK studies ssniff has been the test manufacturer for several CROs Diets with hormone supplementation (17-βββ-estradiol, DHT)

Spezialdiäten GmbH Ferdinand-Gabriel-Weg 16 batch sizes as small as 1 kg DE-59494 Soest - Germany +49-(0)2921-9658-0 [email protected] / www.ssniff.com ssniff – dedicated to r esearch for decades

conventional systems • air ﬂ ow systems • enrichment • enclosure systems • transport systems • cleaning systems • conventional systems

High quality work with attention to detail – and a passion for animal welfare

ZOONLAB is for more than 70 years one of the leading manufacturers and suppliers of solutions for scientiﬁ c animal husbandries, for small and large animals. Not only the vast product range but also decades of experience, ﬁ rst-class quality and innovative and individual product solutions are reasons why customers from all over the world rely on ZOONLAB.

For example: Enclosure systems for keeping marmosets usually consist of several single cage modules which are mounted together and can be modiﬁ ed into larger units with very little effort – the modular arrangement system by ZOONLAB.

Learn more about us – we are looking forward to your enquiry: [email protected] or +49 / 23 05 / 97 30 40

ZOONLAB GmbH Hermannstraße 6 | 44579 Castrop-Rauxel | Germany www.zoonlab.de ﬂ exible ﬁ lm and rigid isolation solutions for all your containment needs

● Holding Isolators

● Surgical Isolators

● Hypoxic Chambers

● Wall Mounted Transfer Hatches

● Desiccator Cabinets

Holding Isolators Standard range or custom designed

44 Potters Lane Milton Keynes MK11 3HQ tel: 01908 305 725 fax 01908 305 729 email: sales@pﬁ systems.co.uk www.pﬁ systems.co.uk

Review Article Laboratory Animals 2019, Vol. 53(2) 125–136 ! The Author(s) 2018 Guidelines for porcine models of Article reuse guidelines: sagepub.com/journals- human bacterial infections permissions DOI: 10.1177/0023677218789444 journals.sagepub.com/home/lan Louise K Jensen , Nicole L Henriksen and Henrik E Jensen

Abstract During the last 10 years the number of porcine models for human bacterial infectious diseases has increased. In the future, this tendency is expected to continue and, therefore, the aim of the present review is to describe guidelines for the development and reporting of these models. The guidelines are based on a review of 122 publications of porcine models for different bacterial infectious diseases in humans. The review demonstrates a substantial lack of information in most papers which hampers reproducibility and continuation of the work that was established in the models. The guidelines describe overall principles related to the inoculum, the animal, the infected animal and the post-mortem characterization that are of crucial importance when porcine models of infectious diseases are developed, validated and reported.

Keywords infectious diseases, pigs, animal models

Date received: 19 March 2018; accepted: 27 June 2018

Introduction preferred as experimental animals is due to their size, anatomy and physiology which are comparable to The use of pigs as experimental animals is increasing.1 humans.4,6 Recently, the complete genome of the In the EU 239,089 pigs have been used for experimental domesticated pig was sequenced8 and more laboratory purposes during the last 5 years (Figure 1(a)).2 The use reagents specific for porcine tissue have become avail- of pigs as experimental animals has primarily been in able.9 However, two additional points make pigs espe- the areas of toxicology, metabolism, cancer and xeno- cially favourable for modelling of infectious diseases in transplantation.1 However, during the last 10 years humans. Firstly, the immune response of pigs is com- the number of porcine models for human bacterial parable to that of humans as the porcine and human infectious diseases has increased, and it is reasonable genome show similarity, both structurally and function- to assume that many more models will follow in ally, among immune-related proteins.10 Furthermore, the future (Figure 1(b)).3 Pigs have been used to as in humans, pigs have a large percentage of neutro- model the different human bacterial infectious dis- philic granulocytes in the peripheral blood.11 Second, eases listed in Table 1. The aims of the different porcine models are advantageous due to spontaneous models were to study either the efficacy of novel and development (naturally-occurring infections) of the well-known antimicrobials, prevention and treatment same infectious diseases as in humans.3 As an example, of infection, specific virulence mechanisms of the pathogenesis and lesions of spontaneously occur- pathogens or the process of inflammation and disease ring chronic pressure wounds, endocarditis, pyeloneph- development. The increased popularity of pigs as ritis and hematogenous osteomyelitis are comparable to experimental animals is also reflected in several reviews about the advantages of porcine models in biomedical Department of Veterinary and Animal Science, University of research.1,4–6 Copenhagen, Copenhagen, Denmark Animal models of infections are important tools to Corresponding author: study different aspects of infectious diseases in humans Louise K Jensen, Faculty of Health Sciences, University of because they provide all factors involved in a patho- Copenhagen, Ridebanevej 3, 2200 Copenhagen, Denmark. 7 gen–host interaction. The reason why pigs are Email: [email protected] 126 Laboratory Animals 53(2)

Figure 1. (a) Pigs used for biomedical research from 2012–2016 in the EU. Pigs are frequently used as experimental animals in most of the European countries.2 (b) Number of publications where pigs have been used to model human bacterial infectious diseases. The colours indicate the target organ system infected in the models. References for all the models can be found in the supplementary material.

Table 1. Human bacterial infectious diseases modelled those seen in humans and are also often caused by the in pigs. same bacteria.3 Homologous models (same aetiology, pathogenesis and symptoms) are described as the most Main organ system 12 of the infection Disease ideal for capturing the complex nature of diseases. Beside all the advantages of using porcine models in Respiratory Pneumonia translational research, it is also important to consider Pneumonia in cystic fibrosis the differences between pigs and humans. As an exam- Ventilator associated pneumonia ple, there are some major anatomical differences in the Tuberculosis gastrointestinal tract. In pigs the caecum, which has a Whooping cough (acute respiratory remarkably larger size, and colon are arranged in a infection in infants) series of centrifugal and centripetal coils differing 4 Genital Chancroid (sexually from those in humans. In a review by Swindle et al. transmitted genital ulcer) the anatomical similarities and dissimilarities between Chlamydia in females pigs and humans are described in detail for all organ 4 Urinary Pyelonephritis systems. Another difference, which is very important for bacterial challenge trials is the body temperature. In Reflux nephropathy normal pigs the rectal temperature can vary between Vascular Vascular catheter infection 38–40C which can make it difficult to define the occur- Aortic graft infection rence of pyrexia.13 Prosthetic vascular graft infection For scientific, ethical and economic reasons, animal Metallic stent placement infection experiments should be appropriately designed, cor- Eye Endopthalmitis rectly analysed and clearly described. However, it has Bone Hematogenous osteomyelitis been documented recently that there is a lack of detail Implant associated osteomyelitis in the way research using animal models is reported.14 Skin Wound infections Furthermore, several authors have stated the need for Heart Endocarditis more efficiency in translational medicine and called for Gastrointestinal Gastroenteritis international guidelines to set higher standards for the 15 Colitis validity of animal models in different disciplines. Therefore, the aim of this review was to investigate Enteritis the details of the reported data from existing porcine Enterocolitis models of bacterial infectious diseases in humans. Typhoid fever Furthermore, the aim was also to describe guidelines Systemic Acute respiratory distress syndrome of overall principles that are of crucial importance Sepsis when porcine models are developed, validated and Pyemia reported. Jensen et al. 127

Material and methods without self-citation) in Scopus. The extracted data Data collection and all references can be found as supplementary material. Papers were collected by a comprehensive search of Medline, Web of Science and Google Scholar. The Results and discussion aim of the search was to find papers in which pigs were used to model bacterial infections in humans. In total, 122 different papers were collected for the pre- Prior to the search, it was decided to allocate all sent review (Figure 1(b)). All the models followed the papers into groups based on the target organ-system same structure: a pathogen was introduced into an of the infection. The following target-organ systems animal, which then became infected for a fixed period were defined: systemic (sepsis models); respiratory; of time followed by post-mortem analyses. Therefore, urinary; brain; gastrointestinal; genital; eye; heart; the results were organized in four sections dealing with skin; bone; vascular (graft/catheter). Thereafter, the inoculum, the pig, the infected pig and post-mortem Group A search words were combined with Group B characterization (Figure 2). search words in the databases: Group A search words: The inoculum . Porcine model OR pig model OR swine model NOT The inoculum consists of an exact dose of a specific guinea, human OR man; bacterial strain suspended in a fixed volume . Infection OR infectious OR bacteria OR micro- (Figure 2).16–20 The different bacterial strains used for organism OR pathogen OR disease NOT virus. inoculation of pigs for human bacterial infections are shown in Figure 3(a). The most commonly used gram Group B search words: positive and gram negative bacterial species, regardless of target organ system, were Staphylococcus aureus and . Lung OR pneumonia OR respiratory OR airway; Escherichia coli, respectively. Staphylococcus aureus . Kidney OR nephritis OR urinary OR cystitis; was primarily used for establishing catheter,21 bone,22 . Brain OR CNS OR neuro OR encephalitis OR heart,23 skin24 and systemic infections25 and E. coli for meningitis; induction of urinary,17 respiratory,26 gastrointestinal27 . Gastro OR intestine OR enteritis OR colitis; and systemic infections.28 The bacterial strains most . Genital OR reproductive; often originated from humans (48%) or pigs . Sepsis OR pyemia OR bacteraemia OR bacteremia; (35%).20,29 For E. coli, there was no difference in inocu- . Graft OR catheter; lation dose among human and porcine strains. For . Eye OR orbital OR endophthal OR retin OR uve S. aureus there was a tendency towards a lower inocu- OR kerat; lation dose when the strain originated from pigs.23,30 . Bone OR bone infection OR osteomyelitis; An optimal inoculum should result in the develop- . Heart OR endocarditis; ment of infection in all inoculated animals. However, . Skin OR wound. this was not obtained in all studies.3 In order to find the optimal inoculum dose, a dose-response study can be The following data were extracted from all included performed.31 In 80% of the papers reviewed, the inocu- publications: bacterial strain used for inoculation lation dose was between 106 and 109 CFU (colony (including origin); inoculation dose; inoculation forming units) (Figure 3(b)). However, for the bone, volume; animal data (age, breed, sex, immune status, eye, respiratory system and grafts the most commonly number); housing data (acclimatization period, diet, used doses (median) were lower than 106 CFU individual/group housing, isolation stable, specification (Figure 3(b)). Very high bacterial doses might not of housing condition); inoculation route; anaesthesia mimic the clinical situation of, for example, surgical during inoculation; surgery during inoculation; meth- contamination of bone implants and colonization of ods used to monitor local and systemic infection in catheters with skin commensals.32 Most often the alive animals; pain management; time frame; full- inoculation dose was reported as CFU ml1 followed body necropsy; occurrence of secondary systemic bac- by the number of ml injected or as CFU in a fixed terial (inoculum) spread; methods used to characterize volume. However, in many of the models the inocula- local and systemic inflammation post mortem; methods tion volume was not registered (Table 2). In a study used to identify the inoculated bacteria post mortem. performed by Soerensen et al. in 2012 it was demon- Finally, the impact of each study was evaluated by strated that the volume – that is, the same dose of registration of the number of citations (with and a specific bacterial strain suspended in two different 128 Laboratory Animals 53(2)

Figure 2. Guidelines for development and reporting of porcine models for human bacterial infectious diseases. The inoculum, the pig, the infected pig and the post-mortem characterization must be well considered and reported to allow reproduction. The bullets listed in the figure should be mandatory in all papers reporting porcine models of human bacterial infections. volumes – had a major impact on the inflammatory SPF pigs was reported in 23 of the reviewed publica- response seen in a sepsis model.20 tions. Besides SPF, other descriptions of immune status included the use of gnobiotic pigs and colostrum- The pig deprived pigs, which were used to model gastrointestinal infections. Usually, one to 20 pigs were used in the Different breeds of farm pigs raised for slaughter and reported studies (Figure 3(c)). Only one study reports mini-pigs were used in 63% and 7.3% of the reviewed the use of a power analysis to calculate the sample papers, respectively. In the rest of the papers the breed size.33 Most of the animals were below 12 weeks of of the pigs was not reported (Table 2). The differences age (Figure 3(d)), despite the fact that most of the mod- between conventional farm pigs and mini-pigs are pri- elled infectious diseases are seen in adult patients. marily related to the growth rate and size at sexual However, in some studies pigs of 0–2 weeks of age maturity. The growth rate of conventional pigs is were used to replicate infectious diseases of the respira- high (the body weight of an adult is >200 kg) compared tory, intestinal and urinary systems which only occur in to mini-pigs (the body weight of an adult is 40–80 kg), infants.34–38 Unfortunately, the age of the pigs was making mini-pigs a more favourable model for adults.4 often not reported (Table 2); the age of the pig might There are various breeds of both farm pigs and mini- influence the immune response towards an infection. In pigs available, and there is a wide variation in their a porcine model of implant-associated osteomyelitis no genotype and phenotype which impacts the growth difference was observed between the use of 12 and rate, size and body weight. Therefore, it is always 32 week old pigs, respectively.39 However, in a porcine important to report the used breed as it may have an model of pneumonia the possibility of inducing infec- influence on, for example, the amounts of drug needed tion with Bordetella pertussis was significantly lower in for testing.4 The health status of domestic farm breed 4–5 week old pigs compared to newborns.40 In the pre- pigs is variable depending on country, region and sent review, 38% of the pigs were females, 16% were farmer management.6 Specific pathogen free (SPF) males and in 46% the sex of the pigs was not registered. status is a specific term used in swine management Due to the physiological differences between female ensuring that the animals are free of specific infectious and male pigs, it is relevant to report the sex in order diseases. However, they can still have had previous to secure that a study is 100% reproducible. The opti- exposure to a broad range of pathogens. The use of mal inoculation route (getting the inoculum into the Jensen et al. 129

Figure 3. (a) Bacterial species used in porcine models of human infections. (b) Inoculation doses (colony forming units; CFU) used in porcine models of human bacterial infections with respect to target organ system for the infection. Doses reported as an interval – for example, 102–4 – have been reported as 102,103 and 104. In studies where multiple doses of the same 10 factor was used, only one dose was reported. (c) Animal number, (d) age and (e) time-frame reported in porcine models of human bacterial infectious diseases. The highest inoculation time for each study was chosen if multiple time points of euthanasia were described. The colours indicate the target organ system in the models. pig) should replicate the pathogenesis of the infection bacterial infections are anaesthetized or sedated seen in humans. Therefore, the inoculation routes often during the inoculation procedure. Furthermore, in reflect the main target organ (Table 3). However, dif- 50% of the models some kind of surgery (defined by ferent anatomical inoculation points for the same cutting) was performed prior to inoculation. pathogenesis can result in diverse results; for example, Reporting of housing details was in general found to in a model of female genital chlamydia infection a be very limited. Only 22% of the studies report a period longer infection period was seen for intrauterine inocu- of acclimatization, even though a pre-study stabiliza- lation compared to vaginal inoculation.41 The intraven- tion period is recommended for experimental pigs to ous inoculation route has been widely used to model be used in survival studies.6 In 23 of the 122 papers it many different kinds of infectious diseases; for example, was mentioned if the animals were housed individually graft infections,42 pneumonia,43 osteomyelitis,18 endo- or in groups. Of these 23 studies, group housing was carditis23 and sepsis.20 However, it might be difficult to reported in 12 papers and the number of pigs housed control and contain an infection using the intravenous together ranged from two to six. Swine are social ani- inoculation route.18 The present review demonstrated mals and to reduce stress, it is recommended that they that two-thirds of all pigs used to model human have the opportunity to interact with or see other 130 Laboratory Animals 53(2)

Table 2. Overview of ‘not reported data’ in 122 publications describing porcine models of human bacterial infectious diseases.

Percent (rounded up) of publications without Category Data extracted information on the data

The inoculum Origin of the bacterial strain 17% (21/122) Inoculation dose 11% (13/122) Inoculation volume 32% (40/122) The pig Breed of the used pigs 29% (36/122) Number of pigs 5% (6/122) Age of the used pigs 38% (46/122) Sex of the used pigs 29% (36/122) The infected pig Pain management 65% (80/122) Local infection monitoring 17% (21/122) Systemic infection monitoring 18% (22/122) Time from inoculation to euthanization 2% (3/122) Post mortem Characterization of local pathology 21% (19/90) (articles for sepsis not included) Characterization of systemic pathology 51% (63/122) Full necrospy 60% (72/122) Identification of inoculated bacteria 31% (38/122)

members of their species.6 The use of an isolation stable sequester formation have been established despite daily was reported in 13% of the papers included in the pre- oral NSAID treatment.39 Beside analgesic treatment, sent review. Diet details were reported in 25% of the diseased pigs should have an increased number of included papers and housing conditions – for example, daily inspections and special requirements to minimize pen size, flooring, bedding specification, humidity or their discomfort, such as fluid therapy or oxygen temperature – were only reported in 11% of the papers. supply, should be established if possible.25 In the 122 papers collected for the present review, the develop- The infected pig ment of a local infection in the main target organ system was monitored by either clinical signs, imaging It is necessary to consider how to monitor and handle techniques (e.g. CT and MR scan), fluid and tissue sam- the infected pig with regards to disease development. As ples and samples taken for microbiological examin- the models replicate infectious diseases in humans, dis- ation. The systemic response to the local infection, or comfort and signs of pain can be expected. In 34% of a systemic response due to spread of infection from the the papers reviewed analgesic treatment was reported main target organ system, was most often monitored by during the study period. Analgesic treatment was clinical examinations, blood analysis including micro- primarily used in studies of sepsis, bone infections, biological culture and urinary analysis. However, the infections of the respiratory system and catheter/ systemic response was only seldom considered in the graft associated infections.20,39,44,45 Both opioids and models (Table 2). Optimally, the infection time or NSAIDs have been used as analgesic treatments in time-frame of the experiment (time from inoculation porcine models of human bacterial infections.20,45,46 to euthanization) should reflect the disease stage of A drawback of NSAIDs is their anti-inflammatory interest (e.g. acute or chronic). For all organ systems, effect, however recently opioids have also been shown except for sepsis, eye, urinary and genital, porcine to have anti-inflammatory activity.47 In a porcine models with a time frame of more than four weeks model of osteomyelitis, macroscopic bone lesions with have been developed.43,48–51 However, in 69% of all Jensen et al. 131

Table 3. Inoculation routes used in order to establish mortem.16,17,21,23,27,39,41,43,52,53 Local inflammatory porcine models of human bacterial infectious diseases. It is lesions will occur in the main target organ and have registered if the inoculation routes demand anaesthesia been characterized by several methods (Figure 4(a)). and surgery (defined by cutting). However, most often macroscopic pathology and hist- Main organ Reported ology were applied. The inflammatory lesions should be system for inoculation described and quantified by semi-quantitative scoring 17,39 the infection routes Anaesthesia Surgery systems or objective measurements. It is relevant, in all porcine models of bacterial infectious diseases, to Respiratory Intra-pulmonary yes yes/no include a definition of infection. Not all inoculated ani- Oropharyngeal yes yes mals may develop an infection, or the same grade of Transthoracic yes no infection. Therefore, it is important to have criteria that Tonsil deposition yes no define if an animal was infected or not.22,39,54 From the Intravenous yes no main target organ the local infection can spread either Intra-tracheal yes no locally into the adjacent tissue or systemically. A local Genital Transcervical yes yes spread will be registered in the description of the lesions Intra-uterine yes yes pattern of distribution. However, an infection in the Intra-vaginal yes no main target organ may result in an infective embolus (bacteria and particular material in the blood) which Urinary Directly in the yes yes renal pelvis can settle in other organ systems and initiate secondary lesions. This will result in signs of infection, which can Intra-vesical yes yes interfere with the signs caused by the primary infection Vascular Infected graft/ yes yes of interest.18,30,52 Therefore, it is always important to catheter look for and describe the occurrence of secondary, Intravenous yes yes embolic infectious lesions. Macroscopic pathology Intra-arterial yes yes and histology of internal organs (i.e. lungs, kidneys, Eye Intra-viterous yes yes spleen and liver) were the predominantly used methods Bone Intravenous yes yes/no to exclude secondary foci of infection (Fig. Intra-arterial yes yes 4(a)).18,19,39,55,56 However, in half of the porcine Traumatic yes yes models reviewed, occurrence of systemic inflammation Skin Topical yes yes was not described, and a full necropsy was only Heart Intravenous yes/no yes/no reported in 40% of the papers (Table 2). Subcutaneous no no Microbiologically, it is important to look for the inocu- Gastrointestinal Peroral no no lated bacteria both locally within lesions and on 57 Intra-gastrical no no inserted implants/devices and systemically. In 28% of the publications (sepsis models excluded) the inocu- Intestinal yes yes lated bacteria were recovered from blood or other Intra-nasal no no organs than the local organ in question, or animals Systemic Intra-bronchial yes no were reported dead due to sepsis. The techniques Intra-pulmonary yes yes listed in Figure 4(b) have been used to localize and Intravenous yes/no yes/no re-isolate the inoculum post mortem. The lungs of Intra-nasal no no pigs trap blood-borne pathogens due to the presence Intra-abdominal yes yes of specialized pulmonary intravascular macrophages.20 Intra-peritoneal yes yes Therefore, lung samples for quantitative microbiology Intestinal yes yes were commonly used to look for a hematogenous spread of the inoculated bacteria.18,31,56 Culture (by the use of swabs) was frequently used, however, the the models included in this review the time-frame was technique will not identify bacteria located inside the below two weeks (Figure 3(e)). tissue.58 Clinically, it has been demonstrated that swabs are less effective compared to biopsies for the diagnosis 58 Post-mortem characterization of infections. The occurrence of bacterial biofilm formation should be considered in future porcine models To characterize a porcine bacterial infectious dis- of bacterial infections.59 Biofilm can be formed both ease model, it is important to focus on both inside tissue and on inserted implants and devices, local and systemic signs of infection post and biofilm is an important factor in sustaining 132 Laboratory Animals 53(2)

Figure 4. (a) Common methods used to characterize the local (bottom) and systemic (top) inflammatory response post mortem in porcine models of bacterial infectious diseases in humans. (b) Common methods used to isolate the inoculum post mortem in porcine models of bacterial infectious diseases in humans. infections in humans.59 During the last decades, it has been realized that bacterial biofilm formation is involved in chronic infections.59 Chronic infections are of increasing concern due to their high healthcare burden and, recently, diagnostic and therapeutic guidelines for biofilm infections were developed by ESCMID (European Society of Clinical Microbiology and Infectious Diseases).59 The guidelines raised the following question: what research is urgently needed to improve diagnosis and treatment of bacterial biofilm infections? Among others, one of the answers was the development of better animal models that realistically reflect infectious diseases in humans. Biofilm formation can be confirmed by different types of microscopy and 3 sonication of implants. Figure 5. The 20 most cited porcine models of bacterial infectious diseases in humans. The models have been Citation impact regularly cited since publication. The last name of the first author and the year of publication is mentioned on the The analysed porcine models of bacterial infectious dis- x-axis. Blue, pyelonephritis model; green, sepsis models; eases in humans showed citation impact. In total, 88 of pink, gastroenteritis models; grey, graft model; red, the 122 papers included in the present review were pneumonia models; yellow, wound model. registered in Scopus. The mean number of citations for the 88 papers were 29.6 SEM 4.48 with self- pneumonia, wounds, graft infections, pyelonephritis citation and 22.73 SEM 3.617 without self-citation. and gastroenteritis (Figure 5). These most cited If the porcine models were grouped by organ system models were developed between 1975 and 2013 and the number of citations, without self-citations, from they were cited regularly since publication – that is, the highest to the lowest was as follows: urinary the mean number of citations per year since 85.5 SEM 83.5; respiratory 39.7 SEM 15.9; gastro- publication ¼ 7 SEM 2.6 (6 SEM 2.3 without self- intestinal 31.6 SEM 10.72; skin 21.4 SEM 5.9; citation). The most cited porcine model for human bac- sepsis 20.24 SEM 4.723; vascular 17.1 SEM 4.478; terial infections was published in Nature in 201360 and genital 11 SEM 6; heart 6.25 SEM 1.1 and bone demonstrated pathophysiological aspects of cystic 4.6 SEM 1.7. The 20 most cited porcine models of fibrosis and pneumonia: since publication this model human bacterial infections were models of sepsis, has been cited 250 times by other authors. Jensen et al. 133

Guidelines for reporting of porcine models caused by the local infection and not by a systemic of bacterial infectious diseases in humans spread of the inoculum. The reported guidelines The inoculum should hopefully help authors to include all essential information in future publications as this will encour- The inoculum must be strictly reported and consist of age reproducible, clear, comprehensive, accurate, con- the bacterial strain, the dose and the volume. The bac- cise and well-written manuscripts of porcine models for terial strain defines the virulence of the inoculum. human bacterial infections. The guidelines will also Therefore, it is recommended to report as many details help journal editors to facilitate improvement in the as possible for the used bacterial strain; for example, quality and validity of porcine bacterial infectious dis- human or porcine origin, diseases/outbreaks caused by ease models. Recently, the ARRIVE (Animal Research: the strain, severity of disease development, expression Reporting of In-Vivo Experiments) guidelines were of known virulence factors including antimicrobial developed to improve general design, analysis and resistance. reporting of research using animals. These guidelines are now endorsed by many scientific journals. The pig However, the ARRIVE guidelines do not fulfil all the special requirements needed for optimal development The breed, number, age and sex of the pigs should and reporting of animal models for bacterial infectious always be reported. Consider using mini-pigs if the diseases in humans. Application of the present guide- growth of the pigs is a major drawback for the study lines for porcine models will build on the ARRIVE or if mature pigs are essential. Report housing condi- guidelines and comply with the principles of the 3Rs tions. The inoculation route/procedure should reflect (replacement, reduction, refinement), as maximum the natural pathogenesis which might demand anaes- information published for each model will decrease thesia and often also surgery. the number of unnecessary animal studies (reduction). Furthermore, reproducible, well-described and charac- The infected pig terized animal models will have the least possible negative impact on animal welfare (refinement). Clinical signs of both local and systemic infection/ Porcine models of bacterial infections must be dis- response should be monitored, and it is important to criminative for the disease seen in humans. The optimal report how the diseased animal is handled; for example, discriminative model reflects the aetiology, pathogen- analgesic treatment, humane endpoints, any special esis, distribution and appearance of lesions. Pathology requirements relevant for the disease. The time frame (i.e. the macroscopic and histologic examination of of the infection should reflect that of human infections. lesions) is fundamental for evaluation of whether the developed lesions are discriminative or not. Therefore, Post-mortem characterization pathology should always be used to characterize porcine models of human bacterial infections. A complete necropsy should be performed in order to disclose both local and systemic infectious lesions. Use Declaration of Conflicting Interests blinded semi-quantitative or objective methods to evalu- The author(s) declared no potential conflicts of interest with ate the lesions. Include a definition of infection. Identify respect to the research, authorship, and/or publication of this the inoculum and consider the formation of biofilm. article.

Conclusion Funding We have demonstrated an increase in the number of The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this porcine models for bacterial infections in humans. article: This work was supported by a grant from the Danish Furthermore, we have also demonstrated that the Research Council (grant number 4005-00035B). models are regularly cited – that is, they have scientific impact. However, there is a substantial lack of information in most of the papers describing the models ORCID iD and, therefore, it was highly important to develop the Louise K Jensen http://orcid.org/0000-0002-0372-4923 present guidelines. It is important to emphasize that reporting of methods and their negative results such References as ‘no signs of systemic inflammation were found post 1. Gutierrez K, Dicks N, Glanzner WG, et al. Efficacy of the mortem’ is also valuable. Only then can one rely on, for porcine species in biomedical research. Front Genet 2015; example, that a high C-reactive protein response is 293: 1–9. 134 Laboratory Animals 53(2)

2. European Commission. Animals used for scientific pur- 18. Jensen HE, Nielsen OL, Agerholm JS, et al. A non-trau- poses, http://ec.europa.eu/environment/chemicals/lab_ matic Staphylococcus aureus osteomyelitis model in pigs. animals/member_states_stats_reports_en.htm (2016, In Vivo 2010; 24: 257–264. accessed 15 September 2017). 19. Leifsson PS, Iburg T, Jensen HE, et al. Intravenous 3. Jensen LK, Johansen ASB and Jensen HE. Porcine inoculation of Staphylococcus aureus in pigs induces models of biofilm infections with focus on pathomor- severe sepsis as indicated by increased hypercoagulability phology. Front Microbiol 2017; 8: 1–13. and hepatic dysfunction. FEMS Microbiol Lett 2010; 309: 4. Swindle MM, Makin A, Herron AJ, et al. Swine as 208–216. models in biomedical research and toxicology testing. 20. Soerensen KE, Skovgaard K, Heegaard PM, et al. The Vet Pathol 2012; 2: 344–356. impact of Staphylococcus aureus concentration on devel- 5. Schomberg DT, Tellez A, Meudt JJ, et al. Miniature opment of pulmonary lesions and cytokine expression swine for preclinical modeling of complexities of human after intravenous inoculation of pigs. Vet Pathol 2012; disease for translational scientific discovery and acceler- 49: 950–962. ated development of therapies and medical devices. 21. Hernandez-Richter T, Schardey HM, Lo¨ hlein F, et al. Toxicol Pathol 2016; 3: 299–314. The prevention and treatment of vascular graft infection 6. Swindle MM, Smith AC, Laber-Laird K, et al. Swine in with triclosan (Irgasan)-bonded Dacron graft: an experi- biomedical research: management and models. ILAR J mental study in the pig. Eur J Vasc Endovasc Surg 2000; 1994; 1: 1–5. 20: 413–418. 7. Rouse MS and Wilson WR. General methodologies for 22. Nielsen OL, Afzelius P, Bender D, et al. Comparison of animal models. In: Zak O and Sande MA (eds) Handbook autologous 111In-leukocytes, 18F-FDG, 11C-methio- of Animal Models of Infection, 1st ed. London: Academic nine, 11C-PK11195 and 68GA-citrate for diagnostic Press, 1999, pp.9–12. nuclear imaging in a juvenile porcine haematogenous 8. Swindle MM, Smith AC, Laber-Laird K, et al. Swine in Staphylococcus aureus osteomyelitis model. Am J Nucl biomedical research: management and models. ILAR J Med Mol Imaging 2015; 2: 169–182. 1994; 1: 1–5. 23. Christiansen JG, Jensen HE, Johansen LK, et al. Porcine 9. Meurens F, Summerfield A, Nauwynck H, et al. The pig: models of non-bacterial thrombotic endocarditis (NBTE) a model for human infectious diseases. Trends Microbiol and infective endocarditis (IE) caused by Staphylococcus 2012; 1: 50–57. aureus: a preliminary study. J Heart Valve Dis 2003; 22: 10. Dawson HD, Loveland JE, Pascal G, et al. Structural and 368–376. functional annotaion of the porcine immunome. BMC 24. Nusbaum AG, Gil J, Rippy MK, et al. Effective method Genomics 2013; 332: 1–16. to remove wound bacteria: comparison of various 11. Fairbairn L, Kapetanovic R, Sester DP, et al. The mono- debridement modalities in an in vivo porcine model. nuclear phagocyte system of the pig as a model for under- J Surg Res 2012; 176: 701–707. standing human innate immunity and disease. J Leukoc 25. Soerensen KE, Nielsen OL, Birck MM, et al. The use of Biol 2011; 6: 855–871. sequential organ failure assessment parameters in an 12. Hau J and Schapiro SJ. Handbook of Laboratory Animal awake porcine model of severe Staphylococcus aureus Science, Volume II, 3rd ed. Florida: CRC Press, 2011, sepsis. APMIS 2012; 120: 909–921. pp.1–11. 26. Elman M, Goldstein I, Marquette CH, et al. Influence 13. Pullar EM. The rectal temperature in normal and infected of lung aeration on pulmonary concentrations of pigs. Br Vet J 1949; 105: 437–453. nebulized and intravenous amikacin in ventilated piglets 14. Kilkenny C, Parsons N, Kadyszewski E, et al. Survey of with severe bronchopneumonia. Anesthesiology 2002; 97: the quality of experimental designs: Statistical analysis 199–206. and reporting of research using animals. PLoS One 27. Moxley RA, Francis DH, Tamura M, et al. Efficacy of 2009; 4: 1–11. Urtoxazumab (TMA-15 Humanized monoclonal anti- 15. Reitmaier S, Graichen F, Shirazi-Adl A, et al. Separate body specific for shiga toxin 2) against post-diarrheal the sheep from the goats. Use and limitations of large neurological sequelae caused by Escherichia coli animal models in intervertebral disc research. J Bone O157:H7 infection in the neonatal gnotobiotic piglet Joint Surg 2017; 99: 1–11. model. Toxins 2017; 49: 1–18. 16. Erneholm K, Lorenzen E, Bøje S, et al. Genital tract 28. Goldfarb RD, Marton A, Szabo´ E, et al. Protectice lesions in sexually mature Go¨ ttingen minipigs during effect of a novel potent inhibitor of poly(adenosine the initial stages of experimental vaginal infection with 5’-diphosphate-ribose) synthetase in a porcine model of Chlamydia trachomatis serovar D. BMC Vet Res 2016; severe bacterial sepsis. Crit Care Med 2002; 5: 974–980. 200: 1–11. 29. Skorup P, Maudsdotter L, Lipcsey M, et al. Beneficial 17. Isling LK, Aalbæk B, Birck MM, et al. Host response to antimicrobial effect of the addition of an aminoglycoside porcine strains of Escherichia coli in a novel pyeloneph- to a b-lactam antibiotic in an E.coli porcine intensive care ritis model. J Comp Pathol 2011; 144: 257–268. severe sepsis model. PloS One 2014; 9: 1–10. Jensen et al. 135

30. Johansen LK, Koch J, Frees D, et al. Pathology and 45. Aboshady I, Raad I, Shah AS, et al. A pilot study of a biofilm formation in a porcine model of Staphylococcal triple antimicrobial-bonded dacron graft for the preven- osteomyelitis. J Comp Pathol 2012; 147: 343–353. tion of aortic graft infection. J Vasc Surg 2012; 3: 31. Johansen LK, Frees D, Aalbæk B, et al. A porcine model 794–801. of acute, haematogenous, localized osteomyelitis due to 46. Johansen LK, Svalastoga EL, Frees D, et al. A new tech- Staphylococcus aureus: a pathomorphological study. nique for modeling of hematogenous osteomyelitis in APMIS 2010; 119: 111–118. pigs: inoculation into femoral artery. J Invest Surg 32. Harrasser N, Gorkotte J, Obermeier A, et al. 2012; 3: 149–153. A new model of implant-related osteomyeliyis in the 47. Slater D, Kunnathil S, McBride J, et al. Pharmacology of metaphysis of rat tibiae. BMC Musculoskelet Disord nonsteroidal antiinflammatory drugs and opioids. Semin 2016; 152: 1–11. Intervent Radiol 2010; 27: 400–411. 33. Gao H, Sandermann J, Prag J, et al. Prevention of pri- 48. Gao H, Lund L, Prag J, et al. Laparoscopic diagnosis and mary vascular graft infection with silver-coated polyester treatment of aortic vascular prosthetic graft infections in a graft in a porcine model. Eur J Vasc Surgy 2010; 39: porcine model. Eur J Vasc Endovasc Surg 2008; 35: 41–45. 472–477. 49. Wood RL, Cutlip RC and Shuman RD. Osteomyelitis 34. Elahi S, Buchanan RM, Babiuk LA, et al. The host and arthritis induced in swine by Lancefield’s group A defense peptide beta-defensin 1 confers protection against Streptococci (Streptococcus pyogenes). Cornell Vet 1970; Bordetella pertussis in newborn piglets. Infect Immun 61: 457–470. 2006; 5: 2619–2627. 50. Dewar HA, Jones MR, Griffin SG, et al. A study of 35. Elahi S, Buchanan RM, Attah-Poku S, et al. Maternal experimental endocarditis in pigs. J Comp Pathol 1987; immunity provides protection against Pertussis in new- 97: 567–574. born piglets. Infect Immun 2006; 4: 2338–2352. 51. Breuing K, Kaplan S, Liu P, et al. Wound fluid bacterial 36. Van Kam AH, Lutter R, Lachmann RA, et al. Effect levels exceed tissue bacterial counts in controlled porcine of ventilation strategy and surfactant on inflammation partial thickness burn infections. Plast Reconstr Surg 2003; 2: 781–788. in experimental pneumonia. Eur Respir 2005; 26: 52. Thibodeaux LC, James KV, Lohr JM, et al. Infection of 112–117. endovascular stents in a swine model. Am J Surg 1996; 37. Cilieborg MS, Thymann T, Siggers R, et al. The incidence 172: 151–154. of necrotizing enterocolitis is increased following pro- 53. Chevaleyre C, Riou M, Bre´ a D, et al. The pig: a relevant biotic administration to preterm pig. J Nutr 2010; 141: model for evaluating the neutrophil serine protease activ- 223–230. ities during acute Pseudomonas aeruginosa lung infection. 38. Coulthard MG, Flecknell P, Orr H, et al. Renal scarring PloS One 2016; 11: 1–17. caused by vesicureteric reflux and urinary infection: a 54. Hodson CJ, Maling TM and McManamom PJ. The patho- study in pigs. Pediatr Nephrol 2002; 17: 481–484. genesis of reflux nephropathy (Chronic atrophic pyeloneph- 39. Jensen LK, Koch J, Dich-Jorgensen K, et al. Novel por- ritis). Br J Radiol 1975; (Suppl 13): 1–23. cine model of implant-associated osteomyelitis: a compre- 55. Nielsen OL, Iburg T, Aalbaek B, et al. A pig model of hensive analysis of local, regional and systemic response. acute Staphylococcus aureus induced pyemia. Acta Vet J Orthop Res 2017; 35: 2211–2221. Scand 2009; 14: 1–8. 40. Elahi S, Brownlie R, Korzeniowski J, et al. Infection of 56. Soerensen KE, Olsen HG, Skovgaard K, et al. newborn piglets with Bordetella pertussis: a new model Disseminated intravascular coagulation in a novel porcine for pertussis. Infec Immun 2005; 6: 3636–3645. model of severe Staphylococcus aureus sepsis fulfills human 41. Lorenzen E, Follmann F, Secher JO, et al. Intrauterine clinical criteria. J Comp Pathol 2013; 149: 463–474. inoculation of minipigs with Chlamydia trachomatis 57. McDermid KP, Morck DW, Olson ME, et al. A porcine during diestrus establishes a longer lasting infection com- model of Staphylococcus epidermidis catheter-associated pared to vaginal inoculation during estrus. Microbes infection. J Infect Dis 1993; 168: 897–903. Infect 2017; 19: 334–342. 58. Aggarwal VK, Higuera C, Deirmengian G, et al. Swabs 42. Ricci MA, Mehran RJ, Petsikas D, et al. Species differ- cultures are not as effective as tissue cultures for diagnosis ences in the infectability of vascular grafts. J Invest Surg of periprosthetic joint infections. Clin Orthop Relat Res 1991; 4: 45–52. 2013; 471: 3196–3203. 43. Bolin CA, Whipple DL, Khanna KV, et al. Infection of 59. Høiby N, Bjarnsholt T, Moser C, et al. ESCMID guide- swine with Mycobacyerium bovis as a model of human lines for the diagnosis and treatment of biofilm infections. tuberculosis. J Infect Dis 1997; 176: 1559–1566. Clin Microbiol Infect 2015; 21: S1–S25. 44. Goldstein I, Bughalo MT, Marquette CH, et al. 60. Pezzulo AA, Tang XX, Hoegger MJ, et al. Reduced Mechanical ventilation-induced air-space enlargement airway surface pH impairs bacterial killing in the porcine during experimental pneumonia in piglet. Am J Respir cystic fibrosis lung. Nature 2013; 487: 109–113. Crit Care Med 2001; 163: 958–964. 136 Laboratory Animals 53(2)

Re´sume´ Au cours des 10 dernie`res anneés, le nombre de mode`les animaux pour les maladies infectieuses bacte´r- iennes a augmente´. Cette tendance devrait se poursuivre a` l’avenir, aussi la pre´sente e´tude a-t-elle pour objet de dećrire des lignes directrices concernant le de´veloppement de ces mode`les et l’e´laboration des rapports y affe´rents. Les lignes directrices sont fondeés sur un examen de 122 publications de mode`les porcins pour diffe´rentes maladies infectieuses bacte´riennes chez les humains. L’examen de´montre un manque substantiel d’informations dans la plupart des publications, ce qui entrave la reproductibilite´ et la continuite´ du travail e´tabli dans les mode`les. Les lignes directrices dećrivent les principes geńe´raux relatifs a` l’inoculum, a` l’animal, a` l’animal infecte´ et a` la caracte´risation post-mortem, qui sont d’une importance cruciale lors de l’e´laboration et de la validation des mode`les porcins de maladies infectieuses ainsi que pour la re´daction des rapports y affe´rents.

Abstract In den letzten 10 Jahren hat die Zahl der Schweinemodelle fu¨r bakterielle Infektionskrankheiten des Menschen zugenommen. Es wird erwartet, dass sich diese Tendenz in Zukunft fortsetzen wird, und es ist deshalb Ziel des vorliegenden Berichts, Leitlinien fu¨r die Entwicklung und Berichterstattung dieser Modelle zu beschreiben. Die Leitlinien basieren auf einer Durchsicht von 122 Vero¨ffentlichungen u¨ber Schweinemodelle fu¨r verschiedene bakterielle Infektionskrankheiten beim Menschen. Der Bericht zeigt einen erheblichen Mangel an Informationen in den meisten Beitra¨gen auf, wodurch die Reproduzierbarkeit und Fortsetzung der Arbeit, die bei den Modellen aufgebaut wurde, behindert wird. Die Leitlinien beschreiben die allgemeinen Prinzipien des Inokulums, des Tieres, des infizierten Tieres und der Post-mortem- Charakterisierung, die fu¨r die Entwicklung, Validierung und Berichterstattung von Schweinemodellen fu¨r Infektionskrankheiten von entscheidender Bedeutung sind.

Resumen Durante los u´ltimos 10 anõs ha aumentado el nu´mero de modelos porcinos para enfermedades infecciosas bacterianas en humanos. En el futuro, se espera que esta tendencia continué y, por lo tanto, el objetivo de la presente revisioń es describir las pautas para el desarrollo y la presentacioń de informes de estos modelos. Las directrices se basan en una revisioń de 122 publicaciones de modelos porcinos para diferentes enfermedades infecciosas bacterianas en humanos. La revisioń demuestra una falta sustancial de informaciońen la mayorıá de los documentos que obstaculiza la reproducibilidad y la continuacioń del trabajo que se establecio´ en los modelos. Las directrices describen principios generales relacionados con el inoćulo, el animal, el animal infectado y la caracterizacioń post mortem que son de crucial importancia cuando se desarrollan, validan y registran modelos porcinos de enfermedades infecciosas. Working Party Report Laboratory Animals 2018, Vol. 53(2) 137–147 FELASA accreditation of education ! The Author(s) 2018 Article reuse guidelines: and training courses in laboratory sagepub.com/journals- permissions animal science according to the DOI: 10.1177/0023677218788105 Directive 2010/63/EU journals.sagepub.com/home/lan

Marcel Gyger1 , Manuel Berdoy2, Ismene Dontas3, Martine Kolf-Clauw4, Ana Isabel Santos5 and Mats Sjo¨quist6

Abstract This document describes how the Federation of European Laboratory Animal Science Associations (FELASA) accreditation addresses both the Directive 2010/63/EU and the related European Commission guidance document. The four EU Functions and beyond: FELASA accredits courses that fulfil the requirements of Functions A, B, C and D as defined by EU Directive, Article 23, as well as for designated veterinarians and specialists in laboratory animal science. Modularity and mobility: Cohesive courses for Functions and for very specific topics are accredited, but flexibility and mobility are possible: a researcher can start his/her training with one FELASA accredited course and complete other modules with another. A course organizer will deliver a FELASA certificate relating to the successfully completed modules. Accreditation process: The process consists of two major steps: (1) a review of full course documentation provided by the applicant will lead, if successful, to FELASA accreditation. The course is posted on the FELASA website as ‘FELASA accredited’ and the course provider can deliver FELASA certificates upon successful completion of the course; (2) successful accreditation is followed by an on-site course audit. In the case of a negative outcome of the audit, FELASA accreditation is withdrawn, the course is deleted from the list of FELASA accredited courses and FELASA certificates cannot be issued. To ensure that quality is maintained, continuation of accreditation requires regular revalidation.

Keywords education, training, Directive 2010/63/EU, accreditation, mobility

Date received: 19 February 2018; accepted: 7 May 2018 1Centre of PhenoGenomics, School of Life Sciences, Swiss Federal Polytechnic School of Lausanne, Switzerland 2BMS, University of Oxford, UK 3Laboratory for Research of the Musculoskeletal System, School of Medicine, National & Kapodistrian University of Athens, KAT Background of this document Hospital, Greece 4Toxicology, Toulouse National Veterinary School, CREFRE, Article 23 of the Directive on the Protection of Animals University of Toulouse, ENVT, France Used for Scientiﬁc Purposes 2010/63/EU (http://ec. 5Physiology, NOVA Medical School, Universidade Nova de Lisboa, europa.eu/environment/chemicals/lab_animals/pubs_ Portugal 6Swedish Centre for Animal Welfare, Swedish University of guidance_en.htm) recognizes the importance of educa- Agricultural Sciences, Uppsala, Sweden tion and training of all persons involved with the breeding, supplying and use of laboratory animals. Annex V All authors contributed equally to this article. of the Directive identiﬁes a list of topics to be included in education and training and the National Competent Corresponding author: Marcel Gyger, Centre of PhenoGenomics, School of Life Sciences, Authorities endorsed a European Union (EU) working Swiss Federal Polytechnic School of Lausanne, Station 19, 1015 document proposing a common education and training Lausanne, Switzerland. 1 framework. Email: [email protected] 138 Laboratory Animals 53(2)

Article 23 of the Directive recognizes competence of personnel required to carry out four Functions, designated as A (carrying out procedures on animals), B (designing procedures and projects), C (taking care of animals) and D (killing animals). These four Functions, however, differ from the Categories used previously by the Council of Europe and by the Federation of European Laboratory Animal Science Associations (FELASA) since the 1990s (see Appendix 1). Thus, the FELASA accreditation scheme was adapted to address Functions A to D as defined by 2010/63/EU (Article 23, paragraph 2) and their associated learning outcomes. In addition to these four Functions, FELASA accreditation is also available to the educa- Figure 1. The steps of the EU education process that tion and training of what we define as specialists in FELASA accredits (adapted from the European Commission laboratory animal science (Specialist in LAS) – a guidance document, 2014). person who may be involved in tasks described in Articles 24, 25 and 26 (fulfilling specific requirements demonstrated. Guidelines for supervision requirements for the welfare and care of animals, for the education are available.2 and training of personnel or any other requirement for These requirements are reflected in different the education and training of the designated approaches to the assessment of satisfactory training veterinarian). and attainment of competence. The FELASA accreditation scheme addresses the first steps of the education The current European education and process – acquisition of basic knowledge and skills, training process which are concluded by an examination (Figure 1) – and ensures their international recognition as a bench- The Expert Working Group established by the mark of high quality. Commission recommended a modular approach to the development of competencies, with defined learning outcomes1 (see Appendix 2). It acknowledged that the The FELASA accreditation scheme objective of initial training is to instil basic knowledge Functions and modules to match specific and/or understanding and is only the first step of the needs of personnel learning process. It involves a programme of work/ study leading to specific learning outcomes, which pro- FELASA accreditation is available for courses that vide basic understanding and skills appropriate to the develop initial training leading to competency in each Function. of the four Functions identified in the Directive. Satisfactory completion of this initial training is fol- A course is defined here as a training programme con- lowed by working with animals under supervision, lead- sisting of one or more modules designed to provide the ing to deeper understanding (Figure 1). This second education and training needs of personnel dedicated to level of training promotes the necessary in vivo com- a specific Function. It may or may not have been petences for caring for and working with experimental endorsed/accredited by a National or Regional animals in a fully responsible way and in accord- Authority or a Laboratory Animal Science Association. ance with the ‘3Rs’. This may be developed fur- For a few exceptions, please see the section entitled ther with a programme of Continuous Professional ‘Accreditation of training for Function and for stand- Development. alone modules’ below. Moreover, the scheme is also The EU Expert Working Group recommended that available for designated veterinarians and specialists persons performing one of the Functions A, B, C or D in laboratory animal science (EU Directive, Articles during which there is a likelihood of pain, suffering, 24, 25 and 26). distress or lasting harm should have completed relevant Because of its modular structure, the education training prior to working under supervision.1 In other scheme recommended by the EU allows the develop- cases, the trainee could begin working under supervi- ment of competence by following courses tailored to sion before the relevant modules have been satisfactor- the needs of the participants. A brief description of ily completed.1 Responsibility for correct performance the modules can be found in Appendix 2. of tasks always remains with the supervisor until train- Table 1 gives a summary of the different types ing has been completed and the requisite competence of modules that participants must or can attend Gyger et al. 139

Table 1. Mapping of modules to Functions and specific tasks.

EU Function Species EU ID Module description ABCDa specific

1 National legislation CCCC 2 Ethics, animal welfare and the 3Rs (level 1) CCCC 3.1 Basic and appropriate biology C C C C Yes 4 Animal care, health and management C C C C Yes 5 Recognition of pain, suffering and distress C C C C Yes 6.1 Humane methods of killing C C C C Yes 3.2 Basic and appropriate biology – skills F F F Yes 7 Minimally invasive procedures without anaesthesia FF Yes 8 Minimally invasive procedures without anaesthesia – skills F Yes 9 Ethics, animal welfare and the 3Rs (level 2) F 10 Design of procedures and projects (level 1) TF 11 Design of procedures and projects (level 2) F 6.2 Humane methods of killing – skills T T F Yes 20 Anaesthesia for minor procedures TT 21 Advanced anaesthesia for surgical and prolonged procedures T T 22 Principles of surgery TT 23 Advanced animal husbandry, care and enrichment practices T

EU Function A: carrying out procedures on animals; EU Function B: designing procedures and projects; EU Function C: taking care of animals; EU Function D: killing animals. Level 1: At this level the trainee should describe and explain the subjects taught; level 2: at this level the trainee should show detailed understanding and should be able to critically evaluate the subjects taught. aModule 6.3 is a stand-alone module for EU Function D (not mentioned here). C: core modules; modules that are required for all Functions; F: function-specific (prerequisite) modules; T: task-specific modules: modules that are relevant to specific tasks within a Function. for EU Functions A to D defined under Article 23 students carrying out research under supervision of and illustrates how courses can be organized to the principal investigator. provide an educational programme that covers A course for EU Function B could be built as a several Functions at once. It is important to stand-alone course that is entirely theoretical and remember that the course should be relevant to the spe- does not include skills training. However, it is easy to cies chosen. conceive, as a way of example, that a course organizer A course provider can therefore build up a course by may propose a programme combining both the combining modules. The simplest course structure is required elements for EU Functions A (persons carry- the EU Function C for person taking care of animals. ing out animal experiments) and B (persons responsible It comprises six core modules and one Function-specific for designing experiments). Thus, a person aiming for module (Module 3.2). EU Function B could decide to attend the additional The addition of the two modules on minimally inva- elements required for EU Function A. The benefit of a sive procedures without anaesthesia (Modules 7 and 8: combined training is that such a person would be better theory and skills) leads to a course covering require- informed regarding the practicalities of performing pro- ments for EU Function A (carrying out procedures cedures and the necessity of adopting best practices in on animals) in addition to EU Function C. the design of projects than a person with only EU The EU Directive and particularly the European Function B education. A course based on such a unified Commission guidance document1 allow tailoring of programme would deliver FELASA accredited certifi- training required by specific tasks within a Function. cates with relevant modules for EU Functions A or B This is the role of the additional task-specific modules for persons who have successfully completed the rele- (Table 1). Note that Module 10 (on experimental vant parts of the programme, or a certificate for EU design) is a Function-specific module of EU Function Functions A and B for those who have completed both. B (i.e. compulsory) but can be included in an EU If only EU Function B courses are delivered, it is Function A curriculum (i.e. task specific); this choice possible to cover the requirements of EU Function C could, for example, be specifically delivered to PhD by adding Module 3.2 to the EU Function B course. 140 Laboratory Animals 53(2)

Humane killing of animals is recognized as a separate Veterinarian. In addition to Module 24 it includes Function under the EU Directive (Function D). other modules as described below: A stand-alone module (6.3) can fulfil the training requirement for those who carry out euthanasia only. In practice, however, this Function is often carried Module 1 ‘National legislation’ with additional out alongside others such as caring for animals (EU Learning Outcomes as described in Function C) or carrying out procedures (EU Function Module 24 (24.1–24.5); A). In these cases, it is recommended to add Module 6.2 Module 9 ‘Ethics, animal welfare and the 3Rs’ to those of EU Function A or C to fulfil the training (level 2) with additional Learning requirements of EU Function D as well (Table 1). Outcomes as described in Module 24 (24.6–24.12); Accreditation of training for Function Module 10 ‘Design of procedures and projects’ (level 1); and for stand-alone modules Module 50 ‘Introduction to the local environment’. The FELASA accreditation scheme is based on the modules set out in the document endorsed by the National Competent Authorities (European 1 Commission guidance document; see Appendix 2) Mobility of students and FELASA accredited but for these to be accredited they are expected to be course attendance part of a broader programme fulfilling the requirements of a specific Function. Thus, a FELASA accredited The FELASA scheme accredits courses or training pro- programme will be expected to consist of at least: a) grammes, not persons. The rules governing the the ‘Core Modules’, b) the modules that are prerequis- accreditation of individuals are, of course, the respon- ite to that Function (see ‘Function-specific modules sibility of relevant Competent Authorities, not of (prerequisite)’ – Table 1/Appendix 2), and c) cover at FELASA. To facilitate access to training as well as least one species or a group of species. These constitute mobility of personnel, persons undertaking training the minimum requirements that are expected to be cov- for a specific Function will be able to combine ered before a course application can be accepted for FELASA accredited courses in different institutions FELASA accreditation. to gain qualification. In this scenario, students can However, the FELASA Accreditation Board for start training at one institution and complete their Education and Training (FELASA E&T Board) will training by taking part in another FELASA accredited accept applications for a few stand-alone modules. course at a different institution. Organizers of FELASA These include: accredited courses will be able to deliver a FELASA certificate for the specific module(s) of the course com- . Function B modules 9, 10 or 11. pleted by the student. . Task-specific modules 20, 21 and 22, or 23. . Other additional modules 50 and 51. Entry qualifications . Stand-alone module 6.3. Although there are no specific entry qualifications for See Table 1/Appendix 2 for a description of each three of the four Functions, the EU Directive indicates module. that a person with a Function B responsibility ‘...shall The FELASA E&T Board will not apply the stand- have received instruction in a scientific discipline rele- alone module accreditation policy to any core module. vant to the work being undertaken and shall have species-specific knowledge...’ (Article 23, 2). More specifically the EU Commission guidance document Roles other than the EU Functions A, B, 1 C and D described in the Directive (p. 7), recommends that the person designing procedures and projects should have received a university In addition to the four Functions mentioned above, the degree or an equivalent degree in the relevant scientific FELASA accreditation covers courses for other roles; discipline. the Designated Veterinarian and the Specialist in The FELASA E&T Board regards a bachelor’s LAS are already part of its portfolio. Courses for degree or equivalent as the minimum entry qualification Project Evaluators will be potential candidates for a Function B course. As a consequence, when a for accreditation. course covers several Functions, including among The European Commission guidance document1 them Function B, entry qualification criteria for that gives the list of learning outcomes for a Designated Function must be satisfied. Gyger et al. 141

If there is no legally defined entry qualification, animals under specific conditions that are described the course organizer has to ensure that, for each module below: taught, all students should attain the minimal level of understanding of the learning outcomes at the end of 1. The applicant must provide a proof that the manual the course, irrespective of their prior level of knowledge. skills training has been submitted to an ethical evaluation process and has been accepted by the compe- Manual skills training using live animals tent authorities. 2. As recommended by the European Commission The issue of the use of live animals in laboratory animal guidance paper,1 the applicant should demonstrate science education and training courses has been debated that he/she applies a tiered approach from non- for many years. When launching its accreditation scheme animal alternatives to cadavers and live animals, in 2003, FELASA issued the following statement: ‘It can depending on the procedure. The severity of the be concluded that the use of live animals in courses for procedures should be non-recovery or mild. The scientists can contribute both to the development of skills approach should demonstrate the effective imple- and enhancing attitude, when the teaching is of a high mentation of the 3Rs. standard and stresses the moral dimension...’.3 We 3. The training of manual skills should address housing believe that this statement remains valid today. and basic care of animals, include handling and The EU commission has made it clear that the restraint of conscious animals and some minimally Directive allows the use of live animals for education invasive procedures on anesthetized animals. and training and describes under which conditions the 4. The animals to be used for skill training should come use of live animals is justified (EU Commission guid- whenever possible from surplus stocks or from the re- ance document, pp.31–321). Such justification is more- use of animals according to the EU Directive regula- over substantiated by past experiences. The FELASA tions. Their fate at the end of the training should be E&T Board has accumulated information with the indicated. The room where animals are housed accreditation of courses under the previous FELASA should be clearly separated from the training rooms category scheme. Review of past annual reports and in order to minimize the stress to the animals. audits indicates that students’ assessments of courses Euthanasia should be performed in a separate room. place the manual skills training using live animals among the preferred parts of the course; very often, The teaching environment and guidance of students students ask for more practical sessions. Moreover, should not be neglected. Rooms and space should be the skill acquisition process is performed in a pedagogic appropriate for skills training. A detailed guide should and educational environment where theoretical con- be provided beforehand to students. The ratio between cepts taught during the lectures are directly translated number of tutors and number of students should not in best practices during skills training; this might not be exceed one tutor for five trainees. Waste and personal the case in an on-the-job set-up carried out in the protection utilities and equipment should be available research laboratory (see also Carlsson et al.4). The and clearly marked. Protection against laboratory FELASA E&T Board’s view is that the provision of animal allergies should be included. contextual understanding in educational courses gives a solid basis for building up skills for future research Certificate format and content with live animals. This results in a positive effect on scientific integrity and prevents avoidable compromises The certificate should include sufficient details of what to animal welfare. has been learnt in order to promote transparency, which Not all courses use live animals for skills training. will facilitate movement from one institution to another However, the FELASA E&T Board considers that at and/or acceptance by a different competent authority. least handling and restraint of live animals (EU Module Thus, the certificate should include the FELASA logo, 3.2) should be mandatory. In order to be able to guide course ID, participant’s details, modules completed and students to treat animals humanely and to assess atti- a course email address (in case complementary informa- tude, no alternative to working with live animals exists tion is requested). Animals that have been handled as today. Prior to learning how to perform procedures on part of the practical parts of the course should be listed animals, certain manual skills could be taught non- by the level of the species (e.g. rat, mouse, zebrafish, invasively, such as handling needles, syringes, or sutur- medaka) rather than as a group of species (e.g. rodents, ing and others. Further, in addition to such training fish) whenever possible. The teaching of the theoretical using alternatives, the FELASA E&T Board recom- part, however, can cover either a single species or a mends that, whenever possible, training in minimally group of species at the discretion of the course organizer invasive procedures (Module 8) is carried out on live as long as this is listed in the certificate. Thus, it is 142 Laboratory Animals 53(2) possible to cover a group of related species as part of the This form seeks information about the applicant and theory, and focus the practical training on the relevant the institution where the course is held. It asks for con- species. In the case of a future change of animal model firmation that the course has been run in its present from one rodent species to another, for example, the form at least once; the frequency with which it is researcher can receive specific hands-on training on the held; the number of students and admission qualifica- new rodent species in the institution without needing to tions when appropriate (see above for EU Function B); attend additional lectures. whether the use of living animals has been authorized by competent authorities and whether the course is The FELASA accreditation process recognized as part of a formal educational programme. Applicants must provide information about course In order to combine flexibility of training and speed of structure and topics, practical classes and how the process, the FELASA accreditation scheme is based on learning outcomes are covered, the way in which know- a two-stage approach based on verification of course ledge and competency are assessed and the results of quality (Figure 2). those assessments. The credentials of teaching staff must also be provided along with details of how stu- Written submission dents are able to comment on the quality of the learning experience. Course organizers seeking accreditation are required to On the basis of this written submission, the complete an application form obtainable from the FELASA E&T Board verifies that the training deliv- FELASA Secretariat. ered meets the requirements of the relevant EU Function,1 reviews the mix of teaching methodology and the use made of supporting materials, such as course notes and recommended reading, and ensures that the objectives of the course conform with relevant national laws, recommendations and guidelines with official supporting documents where appropriate. Consideration is also given to the method of student assessment, the range and depth of evaluation of learning outcomes and the way in which practical skills are assessed. Course organizers must have implemented a mechanism by which students can reflect and report on their learning experience, and must provide examples of course evaluations to show that the course is balanced, effective and worthwhile. Copies of all examination papers, all student evaluation results and all teaching materials are not requested at this stage. If the information provided is judged satisfactory, the course will be given accreditation. Following the decision, an accreditation contract is to be signed between FELASA and the Institution that will issue the training certificates. The FELASA certificate is to be distributed to every participant who ful- filled the requirements to enter and attend the course and who passed the examination. There is only one kind of FELASA accredited certificate delivered. The course is posted on the FELASA website as ‘FELASA accredited’.

Audit Since the FELASA E&T Board implemented a sys- Figure 2. The FELASA Education and Training tematic programme of course audits in 20065 it has Accreditation Process. become apparent that assessment of course quality FELASA: Federation of European Laboratory Animal based only on written evidence may lack robustness Science Associations. E&T Board: FELASA Accreditation and can on some occasions be misleading. The Board for Education and Training. Gyger et al. 143

FELASA E&T Board therefore audits all courses communication with the applicant goes through the whilst they are running. This enables the gathering of FELASA Secretariat. independent first-hand information about the quality of the learning environment. Requirements for maintaining FELASA The audit consists of a visit lasting usually one or accreditation two days by two FELASA auditors. The board requests documentary materials before the visit and a Accreditation for programmes of training delivered list of materials that will be examined during the audit. over a period lasting less than one year will normally The latter includes examination papers and written stu- be valid for five years. Courses for Functions A, B, C dent course evaluations, as well as notes of teachers’ and D fall into this category. For courses lasting over a meetings, records of course attendance, copies of year, for example, courses for laboratory animal spe- course materials provided to the students, checklists, cialists, accreditation will be valid for 10 years. including teaching aids and those used for practical Organizers of FELASA accredited courses must training, and a list of certificates issued. In addition submit an annual report, confirming adherence to con- to reviewing these documents, the auditors attend lec- ditions specified on the application or providing explan- tures and practical sessions included in the accredited ations for changes. Further details and a report programme. The auditors interview students who have template are available from the FELASA Secretariat. already completed the course and supervisors, as well as Significant changes of content from the programme students and teachers of the current course and others approved (and not pre-advised and approved by the responsible for course organization and delivery. E&T Board), under-staffing for course delivery, At the end of the audit, the auditors submit a inappropriate content or teaching methodology, con- report to the FELASA E&T Board. This leads to the cerns about animal welfare, occupational health issues decision on whether the course maintains accreditation etc. could be sufficient cause for the withdrawal of with or without changes, or if accreditation is FELASA accreditation. required to be suspended until crucial changes are It is expected that course organizers will continually incorporated. evaluate the content format and teaching methodology The audit is a crucial part of the accreditation. of the programme and, where appropriate, introduce The FELASA E&T Board regards it as the most modifications to improve the quality of the learning reliable method of establishing the quality of the experience in light of comments from participants. In student learning experience. It is inevitably a formal many cases, new ideas should be evaluated by introduc- process, but it is also an opportunity for an open and tion on a trial basis, assessed by teachers and partici- constructive discussion on LAS training. Course organ- pants and reported in the next annual report. izers often report that they have found the audit If it is proposed to introduce major changes to a inevitably formal but surprisingly enjoyable and an FELASA accredited course, a detailed proposal opportunity to talk about issues with like-minded should be submitted to the FELASA secretariat people. before these are implemented, so that the FELASA The FELASA E&T Board reserves the right to con- E&T Board can give them prior consideration and, if duct audits, other than those involved in the accredit- appropriate, approval. This requirement refers to ation process, subject to the provision of at least 10 changes of course structure and/or delivery process days’ advance notice. and is independent of the obligation to submit an The FELASA E&T Board publishes a list of courses annual progress report. that have been FELASA accredited on the FELASA web page (www.felasa.eu). Renewal

Appeal procedure FELASA accreditation will lapse at the end of each accreditation period. If continued accreditation is An applicant has the right to appeal against a decision required, course organizers must submit a new applica- made by the FELASA E&T Board by writing to the tion in accordance with the requirements for accredit- FELASA President within 15 days of receiving written ation at least six months before the existing rejection. accreditation expires. During the renewal process the course will be entitled to retain its status of Confidentiality accreditation. If the first or the renewal audit reveals major prob- The FELASA accreditation process is conducted in lems leading potentially to the cancelling of the strict confidentiality by all those involved. All accreditation, the FELASA E&T Board will decide 144 Laboratory Animals 53(2) whether the application is rejected or, if satisfactory ORCID iD changes have been implemented, that accreditation Marcel Gyger http://orcid.org/0000-0001-6374-681X may be continued. A second audit will then confirm accreditation of the course if the changes have References addressed the shortcomings (Figure 2). In the case of 1. European Commission guidance document. National a second audit being necessary, the course organizer Competent Authorities for the implementation of Directive will have to pay the costs of the auditors. 2010/63/EU on the protection of animals used for scientific purposes. A working document on the development of a Fee policy common education and training framework to fulfil the requirements under the Directive. Brussels, http://ec.europa. The FELASA accreditation programme is a non-profit eu/environment/chemicals/lab_animals/pdf/guidance/educa- making venture, but is required to be financially tion_training/en.pdf (2014, accessed 4 January 2018). self-supporting. Applications for accreditation will be 2. Jennings M and Berdoy M (eds). LASA 2016 guiding prin- processed only after an application fee has been received. ciples for supervision and assessment of competence as The application fee is independent of the number of required under EU and UK legislation. 2nd ed. A report by the LASA Education, Training and Ethics Section, courses given annually and is non-refundable, irrespect- http://www.lasa.co.uk/wp-content/uploads/2016/09/LASA_ ive of the outcome of the accreditation application. The supervision_and_competence_2016.pdf (2016, accessed 4 FELASA E&T Board will propose a scale of fees annu- January 2018). ally to the FELASA Board of Management. 3. Ritskes M and Hau J. FELASA statement on the use of live For courses given in Europe, the fees for the dur- animals in teaching and training, http://www.felasa.eu/? ation of the accreditation period include the cost of the ACT¼43&file_id¼PMh7rM%2BTYA5nKny4HER14 audit. In the case of an audit revealing major failures in UJGh2%2F%2FyBpeRdWZqt7i9sbjNTwpHEYFfLvgP the course quality and a second audit being required, RCP6mPNxKrtdwH3zyaSQiNl3zG48Q%3D%3D&acce the costs of the second audit are covered by the course ss¼YPCn9I2eJRBxlXNvCmqNBnXwgTDnfCvMV% organizer. 2BtAgOMTqdGtDgHVEh4FIk5nhhpthPuG5PJFVH4vpL- For courses offered in countries outside Europe, the lMx6FgPxxSGQ%3D%3D (2003, accessed 4 January 2018). 4. Carlsson H-K, Hagelin J, Hoglund AU, et al. fee for accreditation or renewal is identical to that for ¨ Undergraduate and postgraduate students’ responses to European courses, but does not include the cost of the mandatory courses (FELASA category C) in laboratory audit, which will be levied as an additional cost. animal science. Lab Anim 2001; 35: 188–193. An annual fee plus a small fee for each student cer- 5. Nevalainen T, Blom HJ, Guaitani A, et al. FELASA recom- tificate issued during that year for the accredited course mendations for the accreditation of laboratory animal sci- is payable for the duration of the accreditation. This ence education and training. Lab Anim 2002; 36: 373–377. annual fee is payable even if the course has not been 6. Council of Europe. European Convention for the held during that year. The current fees are listed on the Protection of the Vertebrate Animals used for FELASA website. Experimental and other Scientific Purposes. Article 26, In the event of a course wishing to opt out of Strasbourg, http://conventions.coe.int/Treaty/en/Treaties/ accreditation status, the remaining annual fees for the Word/123.doc (1986, accessed 4 January 2018). 7. Council of Europe. Resolution on Education and Training of duration of the accreditation will still be payable. Persons Working with Laboratory Animals adopted by the Multilateral Consultation on 3 December 1993, Strasbourg, Acknowledgements http://www.coe.int/t/e/legal_affairs/legal_co-operation/bio- The authors thank Bryan Howard, Hans Peter Ka¨sermann, logical_safety_and_use_of_animals/laboratory_animals/ Timo Nevalainen and Patri Vergara for their contributions to Res%20training.asp (1993, accessed 4 January 2018). the FELASA accreditation scheme and Nicole Linklater for 8. FELASA recommendations on the education and training comments on this manuscript. of persons working with laboratory animals: Categories A and C. Reports of the Federation of European Laboratory Declaration of Conflicting Interests Animal Science Associations Working Group on Education accepted by the FELASA Board of The author(s) declared no potential conflicts of interest with Management. Lab Anim 1995; 29: 121–131. respect to the research, authorship, and/or publication of this 9. Nevalainen T, Dontas I, Forslid A, et al. FELASA recom- article. mendations for the education and training of persons carrying out animal experiments (Category B). Report of the Funding Federation of European Laboratory Animal Science The author(s) disclosed receipt of the following financial sup- Associations Working Group on Education of Persons port for the research, authorship, and/or publication of this Carrying out Animal Experiments (Category B) accepted article: The authors received financial support for the open by the FELASA Board of Management. Lab Anim 2000; access publication ofthis article from FELASA. 34: 229–235. Gyger et al. 145

10. Weiss J, Bukelskiene V, Chambrier P, et al. FELASA Appendix 2. Modules considered as recommendations for the education and training of minimum training necessary before laboratory animal technicians: Category A. Report of the Federation of European Laboratory Animal persons are allowed to carry out a Science Associations Working Group on Education of Function (A to D) Animal Technicians (Category A) accepted by the These modules have been endorsed by the National FELASA Board of Management. Lab Anim 2010; 44: 163–169. Competent Authorities for the implementation of 11. Nevalainen T, Berge E, Gallix P, et al. FELASA guide- Directive 2010/63/EU on the protection of animals lines for education of specialists in laboratory animal sci- used for scientiﬁc purposes (see European 1 ence (Category D). Report of the Federation of European Commission guidance document for explanation and Laboratory Animal Science Associations Working Group details of learning outcomes). on Education of Specialists (Category D) accepted by the FELASA Board of Management. Lab Anim 1999; 33: 1–15. Core modules – Functions A, B, C and D

Appendix 1. The FELASA Categories and 1 National legislation; Directive Functions: An historical reminder 2 Ethics, animal welfare and the 3Rs (level 1); 3.1 Basic and appropriate biology – species specific Three out of the four former FELASA Categories6–10 (theory); correspond roughly to three of the Functions defined in 4 Animal care, health and management – species the Directive 2010/63/EU (Table 2) although some specific (theory); additional adjustments are clearly needed, for example, 5 Recognition of pain, suffering and distress – spe- learning outcomes instead of syllabus, recommenda- cies specific; tions for FELASA Category A8,10 do not match the 6.1 Humane methods of killing (theory). current Directive proposals for Function C. FELASA Category D11 does not map directly into one of the four Directive Functions. However, it may be applicable partly to ‘persons overseeing the welfare Function-specific (prerequisite) modules – and care of the animals, ensuring adequate information Function A: Carrying out procedures specific to the species housed available to the staff, and on animals ensuring that the staff has the required competences’ (Article 24), to ‘designated veterinarians with expertise 3.2 Basic and appropriate biology – species specific in laboratory animals medicine’ (Article 25) and to (practical); ‘persons responsible for the welfare and care of the 7 Minimally invasive procedures without anaesth- animals’ (Article 26). esia – species specific (theory);

Table 2. Comparison of the Federation of European Laboratory Animal Science Associations (FELASA) Categories and the Functions identified in Directive 2010/63/EU.

FELASA Category Role Directive EU Function

A Persons taking care of animals C (may include D) B Persons carrying out animal experiments A (may include C/D) C Persons responsible for directing animal experiments B (may include A/C/D) D Specialists in LAS Refer to Articles 24, 25, 26 – Persons killing animals D

LAS: laboratory animal science. 146 Laboratory Animals 53(2)

8 Minimally invasive procedures without anaesth- 6.2 Humane methods of killing (skills). esia – species speciﬁc (skills). Alternatively:

Function-specific (prerequisite) modules – 6.3 Stand-alone module for Function D (only). Function B: Designing procedures and projects Additional task-specific modules 7 Minimally invasive procedures without anaesthesia – species speciﬁc (theory); 20 Anaesthesia for minor procedures; 9 Ethics, animal welfare and the 3Rs (level 2); 21 Advanced anaesthesia for surgical or prolonged 10 Design of procedures and project (level 1); procedures; 11 Design of procedures and project (level 2). 22 Principles of surgery; 23 Advanced animal husbandry, care and enrichment practices; Function-specific (prerequisite) modules – 24 Designated Veterinarian; Function C: Taking care of animals 25 Project Evaluator.

3.2 Basic and appropriate biology – species speciﬁc (practical). Other additional modules

50 Introduction to the local environment (establish- Function-specific (prerequisite) modules – ment) for persons taking speciﬁc roles under the Function D: Killing animals Directive; 51 Information provision and retrieval. 3.2 Basic and appropriate biology – species speciﬁc (practical);

Re´sume´ Ce document dećrit la façon dont l’accre´ditation de la Fe´de´ration des Associations europeénnes pour la science des animaux de laboratoire (FELASA) re´pond a` la fois aux exigences de la Directive 2010/63/UE et a` celles du document d’orientation connexe de la Commission europeénne. Les quatre fonctions de´finies par la Directive et au-dela` : La FELASA accre´dite les formations qui re´pondent aux exigences des fonctions A, B, C et D, telles que de´finies par l’article 23 de la Directive de l’UE, ainsi que des formations destineés aux ve´te´rinaires de´signe´s et des spećialistes de la science des animaux de laboratoire. Modularite´ et mobilite´ : Les formations cohe´sives concernant des fonctions et celles tre`s spećifiques sont accre´diteés, mais la flexibilite´ et la mobilite´ sont possibles : un chercheur peut ainsi de´buter sa formation avec une formation accre´diteé par la FELASA et comple´ter lesautres modules dans le cadre d’une autre formation accre´diteé. Un organisateur de formation offrira un certificat FELASA relatif aux modules termine´s avec succe`s. Processus d’accre´ditation : Le processus comprend deux grandes e´tapes : (1) un examen de la documentation comple`te du cours fourni par le demandeur, et, si satisfaisant, l’accre´ditation par la FELASA. Le cours est publie´ sur le site web de la FELASA comme formation « accre´diteé par la FELASA » et le fournisseur du cours peut fournir des certificats FELASA lorsque la formation est acheveé avec succe`s ; (2) l’accre´ditation reússie est suivie par un audit du cours sur place dans l’anneé qui suit l’accre´ditation. Dans le cas ou` cette ve´rification ne s’ave´rait pas satisfaisante, l’accre´ditation FELASA est retireé, le cours est supprime´ de la liste des cours accre´dite´s par la FELASA et les certificats ne peuvent plus eˆtre de´livre´s par le fournisseur de la formation. Afin d’assurer une qualite´ continue, l’accre´ditation doit faire l’objet d’une revalidation re´gulie`re.

Abstract Dieses Dokument erla¨utert, inwiefern die EU-Richtlinie 2010/63 sowie das dazugeho¨rige Leitliniendokument der Europa¨ischen Kommission beim Akkreditierungsprozess der Federation of European Laboratory Animal Gyger et al. 147

Science Associations (FELASA) Beru¨cksichtigung finden. Die vier Funktionen (Ta¨tigkeiten) der EU und daru¨ber hinaus: FELASA akkreditiert Kurse, die die Anforderungen der Ta¨tigkeiten A, B, C und D im Sinne der EU- Richtlinie, Artikel 23, erfu¨llen, sowie Kurse fu¨r Tiera¨rzte und Spezialisten fu¨r Versuchstierkunde. Modularita¨t und Mobilita¨t: Es werden feste Kurse fu¨r Funktionen und fu¨r sehr spezifische Themen akkreditiert, wobei entsprechende Flexibilita¨t und Mobilita¨t gewa¨hrt werden: So ko¨nnen Forscher ihre Ausbildung mit einem FELASA-akkreditierten Kurs beginnen und andere Module im Rahmen eines anderen Kurses absolvieren. Ein Kursveranstalter stellt ein FELASA-Zertifikat u¨ber die erfolgreich absolvierten Module aus. Akkreditierungsprozess: Der Prozess besteht aus zwei Hauptschritten: (1) Eine U¨berpru¨fung der vollsta¨ndi- gen Kursunterlagen durch den Antragsteller fu¨hrt bei Erfolg zur FELASA-Akkreditierung. Der Kurs wird auf der FELASA-Website als ‘‘FELASA-akkreditiert’’ vero¨ffentlicht und der Kursanbieter kann nach erfolgreichem Abschluss des Kurses FELASA-Zertifikate ausstellen; (2) die erfolgreiche Akkreditierung wird von einem Kursaudit vor Ort begleitet. Fa¨llt der Audit negativ aus, wird die FELASA-Akkreditierung zuru¨ckgezogen, der Kurs wird aus der Liste der FELASA-akkreditierten Kurse gestrichen und FELASA-Zertifikate du¨rfen nicht ausgestellt werden. Zur Wahrung der Qualita¨t muss die Akkreditierung regelma¨ßig verla¨ngert werden.

Resumen Este documento describe co´mo la acreditacioń de la Federacioń Europea de las Asociaciones de Ciencias de Animales de Laboratorio (FELASA) trata la Directiva 2010/63/UE y el documento de guıá relacionado de la Comisioń Europea. Las cuatro funciones de la UE y ma´s alla´: FELASA acredita cursos que cumplen con los requisitos de las funciones A, B, C y D tal y como se define en la Directiva de la UE, artıćulo 23, adema´s de ir dirigidos a veterinarios expertos y especialistas en las ciencias de animales de laboratorio. Modularidad y movilidad: Los cursos cohesivos para Funciones y para temas muy concretos estań acre- ditados, pero hay flexibilidad y movilidad: un investigador puede empezar su formacioń con un curso acreditado de FELASA y completar otros mo´dulos con otros. Un organizador de un curso facilitara´ un certificado FELASA para los mo´dulos completados satisfactoriamente. Proceso de acreditacioń: El proceso consiste en dos pasos principales: (1) una revisioń de toda la documen- tacioń del curso facilitada por el solicitante llevara´, de completarse satisfactoriamente, a una acreditacioń FELASA. El curso se publica en la pa´gina web de FELASA como ‘‘Acreditado por FELASA’’ y el proveedor del curso puede otorgar certificados FELASA tras la finalizacioń satisfactoria del curso; (2) a la acreditacioń satisfactoria le sigue una auditorıá del curso in situ. En el caso de un resultado negativo de. Original Article Laboratory Animals 2019, Vol. 53(2) 148–159 ! The Author(s) 2018 Improved timed-mating, non-invasive Article reuse guidelines: sagepub.com/journals- method using fewer unproven female rats permissions DOI: 10.1177/0023677218774076 with pregnancy validation via early body journals.sagepub.com/home/lan mass increases

Agata K Stramek, Michelle L Johnson and Victoria J Taylor

Abstract For studies requiring accurate conception-timing, reliable, efficient methods of detecting oestrus reduce time and costs, whilst improving welfare. Standard methods use vaginal cytology to stage cycle, and breeders are paired-up using approximately five proven females with proven males to achieve at least one conception on a specific day. We describe an alternative, fast, consistent, non-invasive method of timed-mating using detection of lordosis behaviour in Wistar and Lister-Hooded rats that used unproven females with high success rates. Rats under reverse lighting had body masses recorded pre-mating, day (d) 3–4, d8, d10 and d18 of pregnancy. Using only the presence of the oestrus dance to time-mate females for 24 hours, 89% of Wistar and 88% of Lister-Hooded rats successfully conceived. We did not observe behavioural oestrus in Sprague- Dawleys without males being present. Significant body mass increases following mating distinguished pregnant from non-pregnant rats, as early as d4 of pregnancy (10% 1.0 increase cf. 3% 1.2). The pattern of increases throughout gestation was similar for all pregnant rats until late pregnancy, when there were smaller increases for primi- and multiparous rats (32% 2.5; 25% 2.4), whereas nulliparous rats had highest gains (38% 1.5). This method demonstrated a distinct refinement of the previous timed-mating common practice used, as disturbance of females was minimised. Only the number required of nulli-, primi- or multiparous rats were mated, and body mass increases validated pregnancy status. This new breeding management method is now established practice for two strains of rat and has resulted in a reduction in animal use.

Keywords Gestation, oestrus dance, lordosis, pregnancy-prediction, reduction

Date received: 2 January 2018; accepted: 9 April 2018

Introduction or requiring monitoring of the additional females for possible pregnancy. Our overall study aim was to Rats are model organisms with a short oestrous cycle improve on current methods and replace them with a that are used in a multitude of research areas. For some simpler method of selecting single females in oestrus for studies, accurate timing of pregnancy or stage of off- time-mating and then accurately predicting their preg- spring development is critical, and it is common prac- nancy status, as early as possible. tice for multiple dams to be time-mated for each study dam required, which allows for the probability of most dams not conceiving during the required time window. School of Life, Health and Chemical Sciences, The Open Current practices of non-invasive time-mating involves University, UK pairing approximately five proven females with proven Corresponding author: males to achieve at least one conception per group of Vicky J Taylor, School of Life, Health and Chemical Sciences, five females mated. This is both time-consuming and an The Open University, Walton Hall, Milton Keynes, MK7 6AA, UK. inefficient use of animals, either generating excess litters Email: [email protected] Stramek et al. 149

The rat oestrous cycle is 4 to 5 days long and consists lactation. In rats, maternal body mass increases of clearly identifiable stages of proestrus, oestrus, throughout pregnancy, starting from as early as day 3 metoestrus and dioestrus.1,2 The cycle is divided into post-conception,6 partly resulting from increased accu- several phases of different length. Proestrus lasts mulation of adipose tissue,7,8 ready to support lacta- approximately 12 hours and corresponds to the onset tion. Body mass increases in pregnant dams follows of mating behaviour, known as behavioural oestrus. increased food intake.9 Hyperphagia during pregnancy The female ovulates at the end of proestrus and in rodents has been reported to increase food intake by enters the oestrus phase, which lasts approximately 26 20% during the first week of pregnancy,10,11 peaking at hours. Metoestrus follows oestrus for approximately between 50%10 and 60%6 by the end of pregnancy 6–8 hours, and then dioestrus ensues, a period of compared to nulliparous controls. Maternal leptin 50 þ hours that is subdivided into dioestrus I (first resistance also occurs during pregnancy; leptin 24 þ hours) and dioestrus II (second 24 þ hours). increases in proportion to the adipose reserves accumu- After dioestrus, the rat returns to proestrus and starts lated and develops from mid-gestation to support a new cycle, if she has not conceived. Rats do not form hyperphagia and allow further adipose accumulation. a fully functioning corpus luteum unless they are coi- Previous work with Wistars has established body tally stimulated, so infertile mating can lead to a state mass increases from day 14 (d14) to day 19 (d19) of of pseudopregnancy that can persist for approximately pregnancy as a reliable method for recognising preg- 10 days before the cycle resumes. The gold standard of nancy and determining the exact day of gestation,12 cycle stage determination in rodents is vaginal cytology, but this is obviously a retrospective and later stage with preference towards vaginal lavage rather than a method. A further aim of this study was to see if we cotton swab due to collection of an increased number could track body mass increases from mating to con- of cells with undisturbed morphology.3 However, it is ception and during pregnancy, to establish the earliest possible to accidently induce pseudopregnancy when time point following conception, to confidently predict conducting vaginal cytology, particularly amongst pregnancy status, rather than having to wait until preg- untrained users. nancy was physically very obvious. Additionally, we Female rats display behavioural signs of oestrus. aimed to establish if these methods could be used When receptive to males during the peri-ovulatory with unproven females who had not previously phase, females have increased running activity, hop- conceived. ping, darting, freezing and ear quivering.1 This leads to male mounting, which in turns solicits lordosis from the female due to pelvic stimulation. Lordosis is Animals, materials and methods a distinctive mating behaviour of the female rat that Ethical approval involves dorsiflexion of the vertebral column.4 Our main aim was to determine if we could make use of This work took place in the Biomedical Research Unit this observed behaviour prior to oestrus to predict and was licensed under the Home Office Animals when females were likely to be receptive to the male to (Scientific Procedures) Act 1986 and had approval increase the individual chances of conception when pair- from The Open University Animal Welfare and ing them for timed-mating. We also kept the animals Ethical Review Body (AWERB). under reverse lighting as this is more appropriate for 5 studying natural behaviour in nocturnal animals, so Animals and housing observations were made during their active phase. Pregnancy confirmation in laboratory rats in Female rats were housed in groups of four in wiretop modern research facilities can involve a range of differ- ventilated cages (NKP, UK) within larger cabinets ent methods, some of which may be invasive, such as (Scantainer, Scanbur, Denmark) and maintained taking blood samples to check for raised progesterone under a 12-hour reverse light cycle (lights off between concentrations. Other methods involve animal separ- 11:00 and 23:00 in Wistar room and 8:00 and 20:00 in ation and handling, such as ultrasound scanning for Lister-Hooded room, for practical reasons) with free conceptus visualisation, or checking for physical access to standard RM3 breeding diet (SDS Diets, changes including enlarged abdomen and eventually http://www.sdsdiets.com/pdfs/RM3-E-FG.pdf) and noticeable body mass increases, detectable by weighing. water. Animals were given time to adjust to the reverse Another study aim was to confidently predict preg- light cycle prior to the study, if needed. Intact stud male nancy as soon as possible following conception. rats for breeding were housed singly. All cages of repro- Pregnancy is a time of extensive maternal adaptation ductively cycling females of the same strain were kept in to facilitate foetal growth and development and also to the same Scantainer along with one cage of male rats to build energy reserves in preparation for the demands of keep females cycling normally. 150 Laboratory Animals 53(2)

Female nulliparous Wistars for ovarian cycle stage viable and the dam had not lost the conceptus and determination were obtained from Harlan, UK (n ¼ 53) returned to ovarian cyclicity and proestrus behaviour. and an in-house colony was subsequently bred from The body mass data for one Lister-Hooded rat are more purchased animals. Wistar rats that were also missing. used in another pregnancy and lactation study as well Whilst handling the Wistar animals to reduce stress as this one were bred in-house (two to three females per prior to the commencement of oestrous monitoring cage, n ¼ 54; n ¼ 3 males). To obtain initial litters from (details below), a distinct behaviour was observed, the colony, females were introduced to a male (one to which was identified as an ‘oestrus dance’ (Figure 2). two females per male) in his home cage for a period of 5 When in proestrus (confirmed by vaginal cytology and days and litters (un-manipulated) were kept to maturity uterine masses (Figure 3) for Wistar rats subgroup/in for future pregnancy and lactation time points. Nulli, pilot studies), gentle physical manipulation with stimu- primi and multiparous Lister-Hooded (breeding stock lation of the back of the rat near the hind legs would derived from Harlan, UK, n ¼ 91) and Sprague-Dawley result in initiation and presentation of lordosis with ear rats (in-house strain, n ¼ 15) were included in the study. wiggling, hopping, darting and freezing, without a male Rats were regularly handled to reduce stress during rat being present or even visible, as described by Feder.1 regular weighing, and for the subset of Wistars that This behavioural stimulation was repeated in Lister- underwent daily oestrous monitoring, vaginal lavage Hooded rats, without vaginal cytology; the check for (detailed below) was conducted at 24-hourly intervals the ‘dance’ was carried out between 11:00 and 12:00 between 11:00 (lights off) and 13:00, to observe and and took approximately 1 min per animal (see supple- determine cycle stages (see below and Figure 1) and mentary video, which is available online with this art- therefore confirm normal cyclicity prior to studies.All icle). The oestrus dance was then used on its own, handling and data collection activities were carried out without cytology, for accurately time-mating Wistar during the dark phase, so they were obtained when females for the pregnancy time points of days 4 most physiologically relevant for natural activity and (n ¼ 7), 12 (n ¼ 7) and 18 (n ¼ 7) of pregnancy, then in reproductive behaviour. Body masses were recorded a much larger group of Lister-Hooded rats for work at approximately the same time each day at the follow- requiring specific embryonic time points and early post- ing time points: pre-mating, day 3–4 pregnant and days natal studies. 8, 10 and 18 pregnant. Rats were checked for lordosis behaviour at pre-mating and then again at each weigh- Oestrous cycle monitoring ing time point; checking at the later time points was included to confirm that the behaviour seen was Rats have an oestrous cycle of approximately 5 days. mating-specific, also that the pregnancy was still As guided by the literature,3,13,14 proestrus (Figure 1(a))

(a) (b) Proestrus Oestrus

Corniﬁed Nucleated ‘spindle- cells shaped’ cells

Figure 1. Representative images for each stage of the Wistar rat oestrous cycle (100 magnification). (a) Proestrus; (b) oestrus; (c) metoestrus; and (d) dioestrus. Stramek et al. 151

Figure 2. The oestrus dance in Wistar females. (a and b) The oestrus dance occurred in proestrus during the dark phase of the lighting cycle (lights turned on for photography only). See Video 1 for Lister-Hooded rat showing behavioural oestrus – note the hopping and ear twitching.

(a) (b)

Representave proestrus Representave dioestrus uterus uterus ** (c) *

Proestrus Oestrus Metoestrus Dioestrus

Figure 3. Uteri (not to scale) became fluid filled at proestrus (a), but not at other cycle stages (b). (c) Intact uterus mass corrected for individual body mass (proestrus, n ¼ 12; oestrus, n ¼ 11; metoestrus, n ¼ 9; dioestrus, n ¼ 11; *P < 0.05; **P < 0.01). Uteri were significantly heavier at proestrus in Wistar rats. was identified by the presence of a large number of by an abundance of leucocytes with a small number of spherical nucleated cells, which resulted in a very cornified, particularly spindle-shaped, cells present. cloudy cell suspension. Leucocytes were often present During dioestrus (Figure 1(d)), few cells of any cell but in very small numbers, with cornified (keratinised, type were present, which was characteristic of the non-nucleated) cells being rare. A predominance of cor- stage in itself. A dioestrus lavage would typically con- nified cells confirmed oestrus (Figure 1(b)), with a tain a lot of mucus, sticking leucocytes together in lines. number of these cells becoming spindle-shaped in No vaginal cytology was undertaken in the Lister- appearance. Metoestrus (Figure 1(c)) was characterised Hooded rats. 152 Laboratory Animals 53(2)

Additionally, in 57 Wistar rats that were culled for a more rat strains: Lister-Hooded and Sprague-Dawley. separate experimental purpose, the opportunity to All rats were adjusted to reverse lighting conditions, remove their uteruses was taken and their post- then only checked to confirm the presence of an oestrus mortem uterine mass increases confirmed proestrus dance before being paired with a male if they did exhi- and the other ovarian stages identified by the vaginal bit this proestrus behaviour, by taking body mass meas- lavage method used (Figure 3). These Wistar females urements at the designated time points. Results are not were fully anaesthetised (isoflurane; IsoFlo, Abbott) presented for Sprague-Dawleys as this method was not and decapitated to collect trunk blood and tissue sam- successful; females only displayed the oestrus dance ples for other studies.15,16 once they were placed with a male in his home cage. Wistar rats have a gestation period of approximately 21 days, which is slightly shorter than that of Lister- Behavioural observations: ‘oestrus dance’ Hooded rats (23 days). Wistar rats were checked regularly close to their expected delivery date (d21 post The majority of females presented an obvious oestrus mating), and cages were carefully checked for pups dance during handling in proestrus (except for Sprague- and counted, without disturbing the dams. Signs of Dawleys, as mentioned above), every 4–5 days, so this blood were also checked for, in case of postnatal distinctive behaviour was used as confirmation of readi- death. Pups were sexed at weaning at approximately ness to mate/conceive. Additionally, animals were 25 days old. Lister-Hooded rats that showed the oestrus observed to have an agitated demeanour and be more dance and were paired with a male but failed to pro- physically active during proestrus. For the breeding duce a litter (or any signs of giving birth, e.g. blood) colony of Wistars used in this study, the females were classed as not pregnant and used as a non- demonstrated this sexual behaviour shortly after lights pregnant body mass comparison group. Three out and the Lister-Hoodeds, within the first 6 hours Lister-Hooded rats were suspected of pseudopregnancy following lights out. and placed in another group for data analysis purposes. Confirmation of proestrus cycle stage Statistics in Wistars Values represent mean standard error of the mean. The Wistar rats in the initial study were monitored for Statistical analysis was carried out using a one-way two (and up to four) complete ovarian cycles to ensure analysis of variance (ANOVA; uterine masses, uterine they were cycling normally. After they were culled at masses/body mass) and repeated measures ANOVA the different cycle stages, increased fluid in the uterus (body mass percentage increases following mating for was used as an additional confirmation of successful nulliparous, primi/multiparous and not pregnant/pseu- proestrus identification by vaginal lavage. Intact dopregnant rat groups, with age as covariate) with uterus mass was found to be significantly highest at Tukey post hoc tests on normally distributed data. proestrus (F(3, 39) ¼ 7.998, P < 0.001), when the uterine The ANOVA F test results are reported in the cavity became fluid filled (Figure 3(a)). When corrected format, e.g. F(2,27) ¼ 18.06; P < 0.01; where the for individual rat body mass, uteri remained signifi- bracketed numbers are the degrees of freedom between cantly heavier at proestrus (F(3, 39) ¼ 8.035, groups and within groups. Pearson correlations were P < 0.001) than at all other cycle stages. used to determine any relationships of body mass and body mass percentage increases following mating with Refinement of time-mating procedure age at mating. All statistical tests were performed using IBM SPSS Statistics 21. P < 0.05 was considered stat- Using oestrus dance observation alone to identify oes- istically significant. trus females that were then time-mated with proven males for 24 hours, 48/54 (89%) Wistar and 80/91 Results (88%) Lister-Hooded rats successfully conceived and littered down. Observations of reproductive behaviour and body mass data were obtained from three different strains of rat: Wistars. Body masses were recorded pre-mating and Wistar, Lister-Hooded and Sprague-Dawley. The ini- during early pregnancy: d3P and d4P (but these d4 tial work was carried out in Wistar rats, validated measurements were taken in rats fasted overnight). with cycle monitoring, which progressed to checking For seven Wistar rats that were 212 þ/– 2.63 days old only for the oestrus dance, once the method was (range 205–218), their mean pre-mating body mass was refined. This simpler method was then used in Wistars 254 þ/ 6.4 g (range 223–276) and by d3P was 267 þ/ for pregnancy and lactation time points and in two 7.6 g (232–294), with no lordosis behaviour observed at Stramek et al. 153

Table 1. Changes in body mass after mating in Lister-Hooded rats. Mean body masses (g) before (d0) and after mating (d1–18) in nulli/primi/multiparous Lister-Hooded rats and those that did not conceive/have a successful pregnancy.

Parity d0 d1 d4 d8 d10 d18

Nulliparous (n ¼ 67) 205 3.0 214 3.0 227 2.9 238 3.1 242 2.6 284 3.3 Primiparousa (n ¼ 10) 240 6.4 250 5.9 263 6.3 278 5.5 281 7.3 316 8.0 Multiparousa (n ¼ 2) 246 20.3 253 23.3 267 20.1 279 17.4 281 15.0 307 19.6 Combined primi/multi (n ¼ 12) 241 5.9 250 5.7 263 5.8 278 5.1 281 6.3 314 7.1 Not pregnant/unsuccessful 223 9.9 229 9.0 235 7.6 234 6.8 238 7.0 241 7.1 pregnancy (n ¼ 11) Not pregnant without possible 237 6.0 241 5.7 240 4.3 240 4.5 245 3.9 246 5.9 pseudopregnants (n ¼ 8) Pseudopregnantsb (n ¼ 3) 186 22.7 198 22.9 214 39.4 216 21.1 221 20.9 227 21.5 aPrimi- and multiparous are shown separately (in italics), then as combined as there were only 2 multiparous and their values were similar. bPossible pseudopregnants were separated from the not pregnant group.

Table 2. Percentage changes in body mass after mating in Lister-Hooded rats. Mean body mass differences (%) after mating in nulli/primi/multiparous Lister-Hooded rats and those that did not conceive (within columns, a>b, P<0.001).

Parity d1 d4 d8 d10 d18

Nulliparous (n ¼ 67) 4% 0.3a 10% 0.5a 15% 0.7a 19% 0.7a 38% 1.5a Primiparousa (n ¼ 10) 4% 1.2 10% 1.4 14% 1.9 17% 1.8 32% 2.5 Multiparousa (n ¼ 2) 3% 1.0 9% 0.9 14% 2.4 15% 3.4 25% 2.4 Combined primi/multi (n ¼ 12) 4% 1.0a 10% 1.1a 14% 1.a 17% 1.6a 31% 2.3a Not pregnant/unsuccessful 3% 0.9 3% 1.2 6% 2.8 8% 3.1 9% 3.3 a pregnancy (n ¼ 11) Not pregnant without possible 2% 0.7b 2% 1.1b 2% 1.5 b 3% 1.9b 4% 2.7b pseudopregnants (n ¼ 8) Pseudopregnantsb (n ¼ 3) 7% 1.5 8% 2.2 17% 5.7 19% 4.3 23% 3.2 aPrimi and multiparous are shown separately (in italics), then as combined as there were only 2 multiparous and their values were similar. bPossible pseudopregnants were separated from the not pregnant group. this latter time point.Their body masses had increased body masses, Table 2 shows the pregnancy-related by 5.1% þ/ 0.54 by d3P (3–7%); the d4P body masses body mass increases as a percentage of initial body are not shown as the rats were fasted as a requirement mass at mating/conception (d0). The 11 rats that did for a diﬀerent study. not conceive or with unsuccessful pregnancies had much smaller body mass changes. Lister-Hooded. There were 67 nulliparous (130 þ/ The body mass of all pregnant rats increased 6.7 days old at pre-mating weigh point, range 50– between d1P and d4P (with no further lordosis behav- 254), 10 primiparous (193 þ/ 10.8 days, range iour observed) and between all later time points 146234) and 2 multiparous rats (236 þ/ 6.0 days, (Tables 1 and 2; also see Figure 4). The day following range 230–242). Body masses were recorded pre-mating mating, all rats showed modest increases of 4%, then (d0), then throughout gestation at d4P, d8P, d10P and by day 4 of pregnancy (when no physical signs are vis- d18P. The mean values are shown in Table 1 subdivided ible), consistent body mass percentage increases of 10% into groupings based on parity: nulliparous for no pre- were recorded, followed by 15–19% by days 8–10, vious births, primiparous for one previous litter and then up to 38% by day 18 (Table 2). multiparous for two previous litters. Multi- and prim- Primi- and multiparous rats were heavier prior iparous rats were older and slightly heavier than nul- to mating than nulliparous rats (d0, Table 1) as liparous rats both at mating and throughout gestation, they were slightly older). However, the pattern of as expected. To account for variation in individual body mass increase (% increase of d0, Table 2) 154 Laboratory Animals 53(2)

145

140 nulliparous (n=67) 135 primi/mulparous (n=12)

130 not pregnant (n=8)

125

120 Body mass (% of d0) 115 * * 110 *

105 * * 100 d1 d4 d8 d10 d18 Day post mang

Figure 4. Percentage body mass increases throughout pregnancy from mating, for nulliparous and combined primi/ multiparous pregnant and not pregnant (excluding three pseudopregnant). was similar for all pregnant rats until d18, when it was Larger body mass increases in nulliparous reduced for the rats that had already had prior litters rats in late gestation (31% cf. 38%, Table 2). Nulliparous rats had the highest body mass percentage gains, especially by d18 Body mass increases by the end of gestation were larger of pregnancy, as they were younger and still physic- in younger nulliparous rats (Table 2) so we explored ally growing (Table 2; re: age, see next section and these data further. Figure 5 shows the range of ages Figure 5). of the nulliparous rats against their body masses at The not pregnant rat group, not including three sus- mating (d0). There was a positive correlation for nul- pected pseudopregnant animals, showed a very slight liparous rats only (r ¼ 0.739, n ¼ 67, P < 0.001 cf. primi/ body mass gain (up to 4%) from d0 through to d18 multiparous r ¼ 0.244, n ¼ 12, P ¼ 0.445, ns), suggesting post mating (Figure 5) but values were significantly that they were still growing physically/maturing to lower than all the pregnant rat time points adult body sizes despite being reproductively (F(2,67) ¼ 22.563, P < 0.001; with age as covariate). competent. The small pseudopregnant group had similar body Likewise, for nulliparous rats only, there was a nega- mass gains to pregnant rats, although the increase by tive correlation between age at mating and body mass d18 was reduced compared to the younger rats with increases throughout pregnancy: the youngest pregnant successful pregnancies (23% cf. 38% nulliparous cf. rats had the highest body mass percentage increases by 32% primiparous, Table 2). Although the numbers d18, with the oldest having the smallest increases are small (n ¼ 3) and age range was wide (53–230 (d1%–d18%: r ¼0.320 to 0.648, n ¼ 57–67, days), the pseudopregnant rats had a similar final P ¼ 0.008 to < 0.001; Figure 6). There were no signifi- body mass increase to multiparous rats (23% cf. 25%). cant correlations for primi- or multiparous rats (d1%– The body mass percentage increases of the pregnant d18%: r ¼0.274 to –0.501, n ¼ 1112, P ¼ 0.097– rats were all significantly higher than the not pregnant 0.389, ns). rat time points (F(2,67) ¼ 22.563, P < 0.001) (*P < 0.001, Table 2, Figure 4) from day 1 following Discussion mating. Table 3 shows the similar litter sizes per parity of the Lister-Hooded dams and sex ratio of the pups We have described an alternative, non-invasive and born in each litter. consistent timed-mating method that uses gentle Stramek et al. 155

276.8

247.9

244.5

225.5

219.0

212.8

209.3

205.9

202.1

198.4

196.0

Body mass at mating d0 (g) 193.0

190.9

187.7

177.8

168.8

159.8

25423422920419718317516916415014714514113812511910910194898482787671686150

Age at mating (days)

Figure 5. Body mass at mating (d0) by age (in days) in nulliparous Lister-Hooded rats. There was a positive correlation for nulliparous rats only (r ¼ 0.739, n ¼ 67, P < 0.001). Body masses at mating (d0) of nulliparous Lister-Hooded rats positively correlated with their age.

Table 3. Pup totals and sex produced by nulliparous, new and accurate method has important welfare impli- primiparous and multiparous Lister-Hooded rats. Data cations as it greatly reduced the number of animals shown standard error of the mean. needed and produced a pregnancy rate approaching Male Female 90% with only a single, unproven female required per Parity Total pups pups pups time point. This represents a further refinement as proven breeders are nearly always used with the exist- Nulliparous (n ¼ 67) 11 0.3 6 5 ing method. Additionally, the disturbance of females Primiparousa (n ¼ 10) 11 1.1 6 5 with either oestrous monitoring and/or to check for Multiparousa (n ¼ 2) 8 3.5 4 4 the appearance of a vaginal plug to confirm mating Combined primi/multi (n ¼ 12) 10 1.1 6 5 was reduced, and then removed entirely. Females were subject to less stressful situations, as they were aPrimi and multiparous are shown separately (in italics), then as combined as there were only 2 multiparous and their values were only paired with males when sexually receptive. similar. In the current study, pregnancy in Wistar and Lister- Hooded rats could be confirmed non-invasively, with physical manipulation during handling to instigate lor- confidence as early as d4 (when visualisation of any dosis behaviour in female Wistar and Lister-Hooded conceptus is not yet possible) using body mass increases rats. Implementing the oestrus dance to time-mate the of 10% from prior to mating body masses, compared to dams was very successful and a distinct refinement of the not pregnant control group. This was much earlier the commonly used time-mating procedure (over 5 days and an improvement on previous studies that used with male using five females/time point required). This body mass increases from mating to d14–d19 of 156 Laboratory Animals 53(2)

159.7

150.9

147.9

145.7

144.4

140.6

139.3

136.2

135.2

134.0

130.6

128.4

127.0 Body mass increase (%) from mating d0 to d18 pregnant 123.3

114.1

119108948782776450 137 146141 168150 175 204194 254233

Age at mating (days)

Figure 6. Relationship between age at mating (days) and percentage body mass increases from mating to day 18 of pregnancy in nulliparous Lister-Hooded rats (r ¼0.608, n ¼ 74, P < 0.001). Age of nulliparous Lister-Hooded rats negatively correlated with body mass increases during gestation. pregnancy.11 Furthermore, confirmation of pregnancy dams had smaller litter sizes than the nulli- and prim- in unproven, nulliparous dams was very reliable as they iparous groups. had the highest body mass gains towards the end of Non-pregnant animals showed small body mass pregnancy. This is a potential improvement in study increases, which might be expected to be greater in design for future studies, which would not need to use younger, growing animals, but the much larger proven breeders. Study animals could be selected increases in mass due to pregnancy in the other rat with less physiological variation: those within a reduced groups became apparent as early as the day following age and body mass range, and that were also mating and had more than doubled by day 4 of preg- nulliparous. nancy, continuing to increase further thereafter. Based This work has also revealed a further consideration on our findings for Lister-Hooded and Wistar rats that for studies using unproven nulliparous rats. As they displayed the oestrus dance, we would predict that were still growing, this resulted in the youngest increases of greater than 3% by day 1 following rats achieving the highest body mass gains from mating indicate conception and over 8% by day 4 con- mating to d18 of pregnancy. Thus, we can further pre- firm a successful pregnancy. dict from this work that older nulliparous rats, over However, we offer one potential caveat with this approximately 20 weeks (5 months) old, are likely body mass increase method, relevant to studies using to have smaller body mass increases to d18P, closer or requiring multiparous dams. Based on our findings, to the values achieved by primi- and multiparous albeit with very small numbers, there could be the pos- dams. These slight body mass differences may sibility of a false positive pregnancy diagnosis in a mul- also reflect reducing litter sizes, as the multiparous tiparous dam when a pseudopregnancy had actually Stramek et al. 157 occurred (n ¼ 3), as their body mass increases mirrored findings it should also be expected that the youngest the values of the multiparous dams (n ¼ 2) for the whole nulliparous females will gain more body mass in the study period. This was surprising as pseudopregnancy later stages of pregnancy, which highlights the import- is expected to resolve by day 10, so it is possible that ance of them having access to enough additional food these pseudopregnant animals did conceive but their at this stage of pregnancy, compared with primi- and pregnancies were resorbed at a late stage, which could multiparous dams. explain the late pregnancy body mass gains. Further Setting up a study that relies on early embryonic studies are needed using larger sample sizes of pseudo- stages requires practice of cytological methods to con- pregnant animals compared with multiparous dams to firm reproductive cycle staging, but once reproductive confirm this observation. cyclicity has been established in the breeding stock, the In our study, the rats were housed under reverse use of behavioural signs of oestrus, induced by gentle lighting conditions, so they were observed and weighed handling, can replace the need to undertake the previ- during their active phase. This was simple to instigate in ously labour-intensive and potentially disruptive pro- separate rooms within the facility, as they each had cedures in the females. This refined breeding individual light cycle monitors. After being housed in management procedure has been adopted as a new effi- rooms with lighting schemes that suit human workers, cient practice in the Biomedical Research Unit (BRU) brought-in rats easily adapt to reverse lighting within at The Open University for two strains of rat, Wistars about 2 weeks. Rats born under reverse lighting did not and Lister-Hooded, housed under reverse lighting, require any adjustment time. Reverse lighting rooms which is effective for nulliparous (unproven) as well were fitted with red overhead lights, and workers also as proven females. This has resulted in a reduction in used red headtorches to allow cleaning, feeding and the number of rats required to provide new rat litters or other day-to-day tasks without disrupting the lighting females at specified time points of pregnancy or conditions. lactation. We were unable to observe the behavioural signs of oestrus in Sprague-Dawley rats, which were also stu- Acknowledgements died under reverse lighting, before they were placed in The authors thank Steve Walters, Karen Evans, Sophie the home cages of male rats, so our new methods were Brooks and Gareth Williams for outstanding technical sup- not replicated in all strains tested. Thus, more animal port and other life, health and chemical sciences (LHCS) and labour-intensive studies are required to achieve the researchers for the opportunity to monitor their breeding same outcomes for timed-mating in this rat strain when stock. We are also grateful to anonymous reviewers whose housed under similar conditions, and this may also comments on an earlier version have helped us improve the apply to other rat strains not studied here. It remains current manuscript. The School of Life, Health and Chemical unknown why some laboratory rat strains show overt Sciences (LHCS), The Open University, provided the funding behavioural signs of oestrus on handling whilst others for this work. do not unless they are in close proximity to a sexually mature male rat, which may indicate that the role of Declaration of Conflicting Interests pheromonal or other intersexual signals varies depend- The author(s) declared no potential conflicts of interest with ing on genetic background. It was noted that cyto- respect to the research, authorship, and/or publication of this logical changes in Sprague-Dawleys were much harder article. to follow, thus their ovarian cycles may have been irregular and less predictable. Sprague-Dawleys have Funding demonstrated more anxiety-like behaviours and The author(s) received no financial support for the research, reduced motor function compared with other rat authorship, and/or publication of this article. strains,17,18 so their natural behaviours may be more inhibited during handling. Thus, testing of other rat strains to find out which ones can be gently stimulated ORCID iD to show lordosis behaviour will also be required before ML Johnson http://orcid.org/0000-0002-6252-610X attempting to fully adopt this new method. We have established a new robust method that com- References bines a behavioural method to increase mating success 1. Feder HH. Estrous cyclicity in mammals. In: Adler NT rates and additionally validated the use of body mass (ed.) Neuroendocrinology of reproduction: physiology and increases (as % body mass since mating) across a range behavior. Boston, MA: Springer US, 1981, pp.279–348. of ages and parities to confidently predict whether a 2. Westwood FR. The female rat reproductive cycle: a prac- nulli-, primi- or multiparous dam is, and/or remains, tical histological guide to staging. Toxicol Pathol 2008; 36: pregnant from day 4 of pregnancy. Based on our 375–384. 158 Laboratory Animals 53(2)

3. Becker JB, Arnold AP, Berkley KJ, et al. Strategies and Wistar rats housed in individually-ventilated cages. Lab methods for research on sex differences in brain and Anim 2015; 49: 188–195. behavior. Endocrinology 2005; 146: 1650–1673. 13. Goldman JM, Murr AS and Cooper RL. The rodent 4. Barnett SW. Manual of animal technology. Oxford, UK: estrous cycle: characterization of vaginal cytology and Blackwell Publishing, 2007. its utility in toxicological studies. Birth Defects Res B 5. Prager EM, Bergstrom HC, Grunberg NE, et al. The Dev Reprod Toxicol 2007; 80: 84–97. importance of reporting housing and husbandry in rat 14. Marcondes FK, Bianchi FJ and Tanno AP. research. Front Behav Neurosci 2011; 5: 38. Determination of the estrous cycle phases of rats: some 6. Cripps AW and Williams VJ. The effect of pregnancy and helpful considerations. Braz J Biol 2002; 62: 609–614. lactation on food intake, gastrointestinal anatomy and 15. Johnson ML, Saffrey MJ and Taylor VJ. Plasma ghrelin the absorptive capacity of the small intestine in the concentrations were altered with oestrous cycle stage and albino rat. Br J Nutr 1975; 33: 17–32. increasing age in reproductively competent Wistar 7. Lopez-Luna P, Maier I and Herrera E. Carcass and tissue females. PLoS One 2016; 11: e0166229. fat content in the pregnant rat. Biol Neonate 1991; 60: 16. Johnson ML, Saffrey MJ and Taylor VJ. Glucagon-like 29–38. peptide-1 (GLP-1) increases in plasma and colon tissue 8. Pujol E, Proenza AM, Roca P, et al. Changes in mam- prior to estrus and circulating levels change with increas- mary fat pad composition and lipolytic capacity through- ing age in reproductively competent Wistar rats. Peptides out pregnancy. Cell Tissue Res 2006; 323: 505–511. 2017; 90: 55–62. 9. Gualillo O, Caminos JE, Nogueiras R, et al. Effect of 17. Ferguson SA and Cada AM. Spatial learning/memory food restriction on ghrelin in normal-cycling female rats and social and nonsocial behaviors in the spontaneously and in pregnancy. Obes Res 2002; 10: 682–687. hypertensive, Wistar-Kyoto and Sprague-Dawley rat 10. Crean GP and Rumsey RD. Hyperplasia of the gastric strains. Pharmacol Biochem Behav 2004; 77: 583–594. mucosa during pregnancy and lactation in the rat. 18. Turner KM and Burne TH. Comprehensive behavioural J Physiol 1971; 215: 181–197. analysis of Long Evans and Sprague-Dawley rats 11. Trujillo ML, Spuch C, Carro E, et al. Hyperphagia and reveals differential effects of housing conditions on tests central mechanisms for leptin resistance during preg- relevant to neuropsychiatric disorders. PLoS One 2014; 9: nancy. Endocrinology 2011; 152: 1355–1365. e93411. 12. Paronis E, Samara A, Polyzos A, et al. Maternal weight as an alternative determinant of the gestational day of

Re´sume´ Pour les e´tudes nećessitant un moment de conception prećis, des me´thodes de de´tection de l’ovulation fiables et efficaces permettent de re´duire les de´lais et les couˆts, tout en ame´liorant le bien-eˆtre animal. Les me´thodes standard reposent sur la cytologie vaginale pour de´terminer l’e´tape du cycle, et les animaux sont accouple´s en utilisant environ cinq femelles dont l’ovulation est confirmeéa` des maˆles reproducteurs afin de parvenir a` au moins une conception a` un jour donne´. Nous dećrivons une me´thode non invasive alternative, rapide et uniforme d’accouplement temporise´ reposant sur la de´tection du comportement de lordose chez les rats Lister-Hooded et Wistar, utiliseé avec succe`s pour se reproduire avec des femelles dont l’ovulation n’est pas confirmeé. La masse corporelle de rats sous lumie`re inverseéae´te´ enregistreé avant l’accouplement, et aux jours (d) 3-4, d8, d10 et d18 de la gestation. En utilisant uniquement la pre´sence de la danse de l’ovulation pour accoupler de manie`re temporiseé les rats sur 24 h, 89% et 88 % des rates Wistar et Lister-Hooded avaient conçuavec succe`s. Nous n’avons pas observe´ de comportement lie´ a` l’ovulation chez les rates Sprague-Dawleys sans maˆles pre´sents. D’importantes augmentations de masse corporelle apre`s l’accouplement permettaient de distinguer les rates gestantes des rates non gestantes, de`s le 4 e jour de gestation (10 % 1,0 % cf d’augmentation 3 % 1,2). Le mode`le d’augmentations tout au long de la gestation e´tait similaire pour toutes les rates gestantes jusqu’a` la fin de la gestation, moment ou` de plus petites augmentations pouvaient eˆtre observeés chez les rates primi- et multipares (32 % 2,5 ; 25 % 2,4), tandis que les gains les plus e´leve´s concernaient les rates nullipares (38 % 1,5). Cette me´thode a de´montre´ une nette ame´lioration par rapport a` la me´thode d’accouplement preće´dente couramment utiliseé, la perturbation des femelles e´tant re´duite au minimum. Seul le nombre requis de rates nulli-, primi- ou multipares a e´te´ accouple´, et l’augmentation de la masse corporelle permettait de confirmer la gestation. Cette nouvelle me´thode de gestion de l’e´levage est maintenant la pratique e´tablie pour deux souches de rat et a permis de re´duire le nombre d’animaux utilise´s. Stramek et al. 159

Abstract Fu¨r Studien, die eine genaue Empfa¨ngnisplanung erfordern, reduzieren zuverla¨ssige und effiziente Methoden zur O¨struserkennung Zeit und Kosten bei gleichzeitiger Verbesserung des Tierschutzes. Standardmethoden stu¨tzen sich auf die Vaginalzytologie fu¨r den Zyklus, wobei Zu¨chter ca. fu¨nf getestete Weibchen mit getesteten Ma¨nnchen paaren, um mindestens eine Empfa¨ngnis an einem bestimmten Tag zu erreichen. Wir beschreiben eine alternative, schnelle, konsistente, nicht-invasive Methode der zeitlichen Paarung durch die Erkennung von Lordoseverhalten bei Wistar- und Lister-Hooded-Ratten, bei der ungetestete Weibchen mit hohen Erfolgsraten eingesetzt wurden. Bei Ratten unter Gegenlicht wurden die Ko¨rpermassen vor der Paarung und an Tra¨chtigkeitstagen (d) 3-4, d8, d10 und d18 aufgezeichnet. Allein unter Ausnutzung des Brunsttanzes zur zeitlich gesteuerten Paarung von Weibchen fu¨r 24 Stunden wurden 89 % der Wistar- und 88% Lister-Hooded-Ratten erfolgreich gedeckt. Wir haben bei Sprague-Dawleys keinen Verhaltens-O¨strus beobachtet, wenn keine Ma¨nnchen pra¨sent waren. Signifikante Zunahme der Ko¨rpermasse nach der Paarung unterschied tra¨chtige von nicht tra¨chtigen Ratten bereits ab d4 der Tra¨chtigkeit (10 % 1,0 % Zunahme cf 3 % 1,2). Das Muster der Zunahme wa¨hrend der Tra¨chtigkeit war fu¨r alle tra¨chtigen Ratten bis zur spa¨ten Tra¨chtigkeit a¨hnlich, ab dann sahen wir geringere Zunahmen fu¨r primi- und multipare Ratten (32 % 2,5; 25 % 2,4), wa¨hrend nullipare Ratten die ho¨chsten Zunahmen verzeichneten (38 % 1,5). Diese Methode zeigte eine deutliche Verbesserung der bisher u¨blichen terminlichen Verpaarung, da Sto¨rungen der Weibchen minimiert wurden. Es wurde nur die erforderliche Anzahl der nulli-, primi- oder multiparen Ratten verpaart, und die Ko¨rpermassenzunahme validierte den Tra¨chtigkeitsstatus. Diese neue Zuchtmanagementmethode hat sich inzwischen fu¨r zwei Rattensta¨mme etabliert und hat zu einer Reduzierung der Anzahl beno¨tigter Versuchstiere gefu¨hrt.

Resumen Para estudios que requieren una concepcioń programada precisa, los me´todos eficientes y fiables de detec- cioń del celo reducen el tiempo y los costes adema´s de mejorar el bienestar. Los me´todos estańdar utilizan citologıá vaginal para organizar los ciclos, y los reproductores se emparejan utilizando aproximadamente cinco hembras probadas con cinco machos probados a fin de conseguir una concepciońenundıá en concreto. Describimos un me´todo alternativo, ra´pido, consistente y no invasivo de crıá programada utilizando la detec- cioń de conductas de lordosis en ratas Wistar y Lister-Hooded donde se habıá utilizado hembras sin probar con un gran porcentaje de e´xito. Las ratas bajo iluminacioń revertida se sometieron a un registro de masa corporal anterior a la crıá en los dıás 3-4, 8, 10 y 18 del embarazo. Utilizando so´lo la danza del celo para programar la crıá de las hembras durante 24 horas, el 89 % de las ratas Wistar y el 88 % de las ratas Lister- Hooded concibieron fructı´feramente. No observamos ninguń comportamiento de celo en las ratas Sprague- Dawleys sin la presencia de machos. Una masa corporal significativamente superior tras la crıá distinguio´ las ratas embarazadas de las no embarazadas desde el dıá 4 del embarazo (10 % 1,0 aumento del 3 % 1,2). El patroń de aumento durante la gestacioń fue similar en todas las ratas embarazadas hasta la etapa final del embarazo, cuando se registraron unos aumentos inferiores para las ratas primı´paras y multı´paras (32 % 2,5; 25 % 2,4), mientras que las ratas nulı´paras registraron los mayores aumentos (38 % 1,5). Este me´todo demostro´ un refinamiento distintivo de la praćtica de crıá programada utilizada anteriormente, ya que la perturbacioń de las ratas hembras se redujo. Uńicamente el nu´mero requerido de ratas nulı´paras, primı´paras y multı´paras se sometieron a crıá, y los aumentos de la masa corporal validaron su embarazo. Este nuevo me´todo de gestioń de la crıá ahora es una praćtica comuń para dos cepas de ratas y permitio´ una reduccioń del nu´mero de animales utilizado. Original Article Laboratory Animals 2019, Vol. 53(2) 160–168 ! The Author(s) 2018 A rat model of nerve stimulator-guided Article reuse guidelines: sagepub.com/journals- brachial plexus blockade permissions DOI: 10.1177/0023677218779608 journals.sagepub.com/home/lan Yanzi Zhang1, Bo Cui1, Chunyu Gong2, Yidan Tang1, Jianxiong Zhou1,YiHe1, Jin Liu1,3 and Jing Yang1,3

Abstract It is important to develop a feasible animal model of regional anesthesia other than sciatic nerve blockade for pharmacokinetic investigations of new local anesthetics or analgesia in upper extremity surgery. Herein, we explored a nerve stimulator (NS)-guided brachial plexus block (BPB) in a rat model. The anatomy of the brachial plexus in rats was delineated in cadavers, and various BPBs were examined. The puncture point was located 0.5–1.0 cm below the lateral one-third of the clavicle. The efficacy and safety of the NS-guided BPB were evaluated using an injection of 2% lidocaine or 0.5% bupivacaine in 16 live animals; saline injection was used as a control. Both sides of the brachial plexus were located successfully using the NS-guided technique. Sensory blockade (nociception assessment) and motor blockade (grasping and straightening tests) appeared after application of the two classical local anesthetics, but not normal saline. The motor and sensory blockade induced by bupivacaine exhibited a longer duration than that induced by lidocaine (p < 0.05). All rats recovered uneventfully from general anesthesia and BPB. No abnormal results were found in pathological studies or behavioral observations. Thus, a rat model of NS-guided BPB was established, and BPB induced an overall reversible sensory and motor blockade in the thoracic limbs. Evaluation of the efficacy and safety demonstrated that this rat BPB model was feasible, reproducible, and safe.

Keywords brachial plexus block, bupivacaine, lidocaine, nerve stimulator, rat

Date received: 31 January 2018; accepted: 4 May 2018

Introduction nerve block (ventral limb anesthetization of the hind limb). Brachial plexus block (BPB) is a commonly Regional anesthesia is widely used as a powerful anes- used peripheral nerve block for the upper extremities thetic method in clinical surgery.1,2 Furthermore, that provides complete anesthetization of the entire there is growing research interest in investigating the mechanism, safety, duration, and distinct sensory and 1Department of Anesthesiology, West China Hospital, Sichuan motor blockades of new local anesthetics and additives University, China for inducing regional anesthesia. Thus, a feasible and 2Department of Surgery, West China Teaching Hospital, Sichuan safe animal model of regional anesthesia is essential for University, China such studies. 3Translational Neuroscience Center, West China Hospital, Sichuan Some regional anesthesia models have been devel- University, China 3–7 oped in rats, cats, dogs and sheep. Sciatic nerve Corresponding authors: blockade models in rats are the most widely used Jing Yang, Department of Anesthesiology, West China Hospital, animal models in experimental studies because of Sichuan University. No.38, Guoxue Ave, Wuhou, Chengdu, their simplicity and reproducibility. However, a sciatic Sichuan, 610041, China. nerve block only provides dorsal limb anesthetization Email: [email protected] Chunyu Gong, Department of Surgery, No. 4 West China Teaching of the hind limb, whereas it cannot provide complete Hospital, Sichuan University. No. 18, South Renmin Road, Wuhou, anesthetization of the hind limb (i.e., the dorsal and Chengdu, Sichuan, 610041, China. ventral limb) without including a femoral or saphenous Email: [email protected] Zhang et al. 161 upper limb. The anatomical characteristics of the rat China, approved all protocols (Ethics No.: brachial plexus allow for the development of a rat 2015014A). Thirty fresh cadavers of adult Sprague- BPB model.8 Dawley rats (weighing 200–300 g) that died or were Therefore, we developed a rat BPB model. We used euthanized for reasons unrelated to this study were rat cadavers to investigate the BPB and verified the allocated to this study. effectiveness and safety of the BPB in living rats. Dissection of the brachial plexus. The brachial plexus (BP) morphology of 20 fresh cadavers of adult rats Materials and Methods (euthanized with propofol) was investigated. Rats Cadaver Experiment were positioned in a dorsal recumbent position with the blocked limb extended straight. The skin, fascia, Animals. The Committee of Scientific Research and and muscle were removed throughout the thoracic Institutional Animal Experimental Ethics Committee, and forelimb areas. The BP on the left and right side West China Hospital, Sichuan University, Chengdu, of each cadaver was dissected from the spinal cord to the thoracic limb.

Results of BP dissection. We identified the roots, trunks, divisions, cords, and terminal branches of the BP (Figures 1 and 2). The BP emerged at five different levels, C5, C6, C7, C8, and T1, similar to humans (Figure 1). The C5 and C6 branches merged to form the upper trunk, the C7 branch separately and continu- ously formed the middle trunk, and the C8 and T1 branches passed over the first rib and merged to form the lower trunk (Figure 2). The trunks divided into the anterior and posterior divisions. The cords continued and became the five major nerves of the upper extremity: the musculocutaneous nerve, the axillary nerve, the radial nerve, the median nerve, and the ulnar nerve. Figure 2 shows a schematic depiction of these nerves. The cutaneous branch of the five major sensory nerves innervated the area shown in Figure 3(a) and (b). Figure 1. Anatomy of the position of the BP. A lamina ven- The beginning of the trunks was located posterior to tral, B the first rib, C the cords of BP, D clavicle, E deltoids. the external jugular vein, which ran diagonal from the BP:Brachialplexus;Cv:cervicalvertebra. lateral neck, crossing to the medial clavicle (Figure 4).

Figure 2. Schematic drawing of the roots, trunks, divisions, cords, and terminal branches of the BP in our study. n.: nerve. 162 Laboratory Animals 53(2)

Figure 4. Schematic drawing of the position of the BP. The dashed line indicates the line passing the point of the lateral one-third of the clavicle parallel to the mid-sternal line of the rat. The position at the three circles indicates the blockade target position. The gray rectangle indicates the safe areas for puncture. Figure 3. The skin area for evaluating mechanical noci- n.: nerve; a: artery; v: vein. ceptors. (a) The ventral thoracic limb. (b) The dorsal thor- A acic limb. The median nerve, the skin overlying the and accuracy of this method. The rats were positioned palmaromedial aspect of digit 2. B The ulnar nerve, the in a dorsal recumbent position with the limb to fourth digital pad. C The radial nerve, the skin area below the dorsal aspect of the antebrachiocarpal joint. D The be blocked extended straight, and the landmarks musculocutaneous nerves, the skin area below the ventral (0.5–1.0 cm below the lateral one-third of the clavicle) aspect of the antebrachiocarpal joint. E The axial nerve, the were determined (Figure 5(a)). A syringe needle dorsal and ventral aspects of the deltoid muscles. (0.8 mm 38 mm, Becton Dickinson SA, Mequinenza, Fraga, Spain) was used to create a puncture, and a non- The divisions (anterior and posterior) and cords (lat- traumatic peripheral nerve block (PNB)-stimulating eral, medial and posterior cords) of the BP were con- needle (Uniplex Nanoline, 22 G 50 mm, Pajunk centrated and situated 0.5–1.0 cm below the middle and GmbH Medizintechnologie, Geisingen, Germany) was lateral clavicle. There are no major vessels at this pos- used to inject 0.2 ml of methylene blue diluted in saline ition, except the axillary artery or vein, in addition to at 1:25 with the needle inserted vertically to a depth of the lung. The branches spread out from this position. approximately 0.7 cm. Dissection was immediately per- The axillary vessels ran along and beneath the cords but formed to expose the BP. were not surrounded by the three cords like human vessels. Staining results. Twenty stained positions were rec- This dissection revealed that the space from the orded; 14 BP cords were stained, and six BPs lateral one-third of the clavicle to the axillary vessels were not stained (Figure 5(b)). Two of the BPs, which (distance of approximately 0.5–1.0 cm) was a safe punc- were located in the pectoralis minor muscle, lacked blue ture area (Figure 4). The anterior and posterior areas of dye due to too shallow insertion of the needle. The this space were composed of muscles and fascia. other four BPs, which were located in the posterior of Therefore, we deﬁned the puncture point for rat BPB the BP cords, lacked blue dye due to too deep insertion at 0.5–1.0 cm below the lateral one-third of the clavicle of the needle. No vessel ruptures or intrathoracic punc- (Figure 4). The depth of the cords of the BP to the skin tures occurred during the 20 injections. was approximately 0.7 cm. Experiments in live rats Staining Studies Animals. This study was performed in strict accordance A percutaneous, perineural injection technique was with the guidelines for the care and use of laboratory developed for the divisions and cords of the BP using animals,9 and the Committee of Scientiﬁc Research and the injection method described in the results of BP dis- Institutional Animal Experimental Ethics Committee, section. Ten additional adult rat cadavers (weighing West China Hospital, Sichuan University, Chengdu, 200–300 g) were obtained to determine the feasibility China, approved the study (Ethics No.: 2015014A). Zhang et al. 163

distal thoracic limb were marked using picric acid. A cotton swab was dipped into a 3% picric acid solution and applied to the skin (Figure 3). Rats were randomly divided into two groups (eight per group) based on a computer-generated randomized list. Rats received 2% lidocaine/saline (Group Lido) or 0.5% bupivacaine/saline (Group Bupi). One forelimb was randomly selected for saline or local anesthesia in each rat. Randomization of limbs was performed using a coin toss. The same investigator who was blinded to the drugs injected performed all injections. The rats were positioned in a dorsal recumbent position with the limb blocked, and the landmark of the puncture point (0.5–1.0 cm below the lateral one-third of the clavicle) was determined (Figure 5(a)). A puncture was created at this point using a syringe needle (0.8 mm 38 mm, Becton Dickinson SA, Mequinenza, Fraga, Spain) under surgical aseptic conditions. A non- traumatic PNB-stimulating needle (Uniplex Nanoline, Figure 5. Rats were placed in a dorsal recumbent position 22 G 50 mm; Pajunk GmbH Medizintechnologie, with the limb to be blocked extended straight. (a) The body Geisingen, Germany) connected to a nerve stimulator landmarks for BPB in rats. Arrows indicate the point for (Stimuplex HNS 12, B Braun Melsungen AG, puncture at 0.5–1.0 cm below the lateral one-third of the clavicle. (b) The stained and unstained (deep injections) Melsungen, Germany) was inserted vertically through BPs in cadaver rats. the hole in the skin. Stimulation was performed at a frequency of 1 Hz for 0.1 ms beginning at a current of 0.5 mA. The needle was slowly advanced until twitches Sixteen healthy Sprague-Dawley rats (8 males and 8 of the forelimb and forepaw were observed, and the females) aged 3–4 months with a mean body weight of current was then decreased. Twitches should be main- 238 g (range 220–260 g) were provided by Dashuo tained at 0.2 mA but abolished at lower stimulus inten- Laboratory Animal Co., Ltd (Chengdu, China). Rats sities to avoid intraneural injection.12 Drugs (0.2 ml) were free of Hantavirus, Coronavirus, Dacryoadenitis were injected when no blood or air appeared after virus, Mycoplasma spp., respiratory and enteric bac- gently withdrawing the syringe plunger. The twitches teria, ectoparasites, and endoparasites. The animals were immediately abolished following the drug injec- were housed in polycarbonate cages (five rats per cage) tion.12 BPB for the other forelimb was performed at an ambient temperature of 26C under a 12-h light/ using the same protocol. Inhalational anesthesia was dark cycle. Rats were provided with nest-building mater- stopped immediately after surgery of the other forelimb ial and allowed free access to a commercial pellet diet was finished. and pure water. Dashuo Laboratory Animal Co., Ltd (Chengdu, China) provided the diet, which conformed Blockade evaluation. The motor and sensory func- with the Chinese laboratory animal nutrient guidelines tions of the blocked forelimb were assessed qualita- (GB 14924.3-2010) and hygienic standards (GB 14924.2- tively immediately after recovery of the righting reflex 2001).10,11 Rats were handled daily for 3 days to become (i.e., defined as T0). Assessments were performed at 15- familiarized with the environment and to observe their min intervals until normal motor function was restored. physiological functions and limb activity. The same investigator who was blinded to the injection process performed all assessments. Nerve stimulator-guided BPB. The results of the staining studies revealed six failed blocks (6/20) using Motor blockade. Motor block was assessed using digit blind injection. Therefore, we tested the nerve stimula- grasping and the straightening test. A three-point sub- tor (NS)-guided BPB in 16 live rats. General anesthesia jective rating scale was used: 1 ¼ normal, normal ability was induced with 5% isoflurane in 100% oxygen fol- to grasp a grid mesh with a weight of 200 g or straighten lowed by a combination of 1.5–2% isoflurane in 50% the digits; 2 ¼ partial blockade; and 3 ¼ complete block- oxygen for maintenance. Respiratory and pulse rates ade, no active digit flexion or straightening.13 were closely observed. The fur of the rats was removed by shaving the chest and forelimbs. The skin areas Sensory blockade. Mechanical hyperalgesia in the innervated by the five major sensory nerves of the dermatomes innervated by the five major sensory 164 Laboratory Animals 53(2) nerves of the thoracic limb (Figure 3) was evaluated were cross and coronally sliced, and hematoxylin and using the same method for assessing sensory blockade eosin staining was used to determine nerve injury. A in cats.6 The evaluation was always performed in the blind histological examination was performed. The per- same sequence: median, ulnar, radial, musculocuta- centage of the unstained area (%UA), in which the neous, and axially nerves. myelin was retreated, destroyed or dissolved, was used An alligator clip was used to elicit nociception. Only to evaluate the histological damage of nerve fibers. The the first two jaws of the clip were used to clamp. The jaws formula was %UA ¼ unstained area/entire area of the were continually closed for a maximum time of 5 s until a nerve fiber 100.14 positive response was elicited, and the following three- point rating scale was used: 1 ¼ normal, normal response Statistical analysis to the stimulus, with rapid withdrawal of the limb, weight shifting and/or head movement toward the testing site Statistical analyses were performed using Prism 6 and/or vocalization; 2 ¼ partial blockade, delayed or (GraphPad, San Diego, CA, USA). Statistical tests were attenuated response to the stimulus; and 3 ¼ complete two-tailed, and data are presented as the means SD. blockade, absence of a response to the stimulus. The onset and offset times of the motor and sensory The following times (min) were also recorded in this blocks were analyzed using the Mann-Whitney test and study: onset time of blockade ¼ time from the drug two-way analysis of variance (ANOVA). The %UA was injection to complete blockade (score of 3); offset analyzed by using one-way ANOVA. Differences were time of blockade (total duration time of block- considered significant at p < 0.05. ade) ¼ time interval after complete blockade (score of 3) was fully achieved and gradually reduced to partial Results blockade (score of 2). All 32 BPB in the 16 rats were accomplished in 2–5 min. Post-treatment evaluation and pathological study. Four All rats recovered after isoflurane delivery was stopped, rats (two from each group) were observed for 14 days and the righting reflex was restored within 2 min of after BPB to determine possible complications. The isoflurane cessation. Complete motor and sensory remaining rats (six rats per time-point) were euthanized blockades were observed in all forepaws after 2% lido- using 2 ml propofol at 3 and 8 days after drug injection, caine and 0.5% bupivacaine injection at T0. The onset and the BP nerves (12 per time-point) were retrieved. No times were 5.50 2.07 min after 2% lidocaine injection injury of the surrounding tissues was observed in the BP and 4.89 1.73 min after 0.5% bupivacaine injection nerves. The BP nerves were evaluated pathologically for (p ¼ 0.578). mechanical injury and potential drug influence. The The offset times of the motor and sensory blockades cords of the BP (the drug injection site) were removed induced by 2% lidocaine (p < 0.0001, Figure 6(a)) and and fixed in 10% formaldehyde at 4C overnight. Cords 0.5% bupivacaine (p < 0.0001, Figure 6(b)) were

Figure 6. The offset time of BPB in rats (each group, n ¼ 8). (a) Comparison of offset times in Group Lido (2% lidocaine vs. saline, *p < 0.001). (b) Comparison of offset times in Group Bupi (0.5% bupivacaine vs. saline, *p < 0.001). n.: nerve; grasping: the grasping test; straightening: the straightening test. BPB Brachial plexus block. Zhang et al. 165

Table 1. Offset time of brachial plexus nerve block with 2% lidocaine or 0.5% bupivacaine in rat.

Sensory blockade Motor blockade

Grasping Straightening Axillary n. Radial n. musculocutaneous n. Median n. Ulnar n. test test

Lidocaine 50.6 19.5 61.9 28.3 76.9 21.9 71.3 27.5 75.0 22.7 73.1 27.1 67.5 33.1 Bupivacaine 103.1 38.8 116.3 43.0 120.0 43.9 121.9 39.6 116.3 43.0 112.5 34.9 106.9 33.5 p value 0.017 0.004 0.011 0.018 0.023 0.019 0.006

Data were analyzed using two-way ANOVA. p value (column factor) < 0.0001. p value (row factor) was 0.293. Differences were considered a significant at p < 0.05. n.: nerve. significantly different compared with saline. The offset All NS-guided BPBs in our study were accomplished times of the motor (grasping, straightening) and sensory in 2–5 min, which indicates that this method is simple (axillary, radial, musculocutaneous, median, and ulnar and easy to perform. No skin injury, hematoma, nerves) blocks in the 0.5% bupivacaine group were signifi- pneumothorax, intrathecal block or nerve damage cantly longer than those in the 2% lidocaine group was observed in our model. Pathological studies (p < 0.05, Table 1). Table 1 shows the total duration of revealed no mechanical injury or potential drug influ- blockades induced by 2% lidocaine and 0.5% bupivacaine. ence. This model differs from previously reported BPB No obvious differences were found between the offset models that did not discuss complications or reported times of the sensory and motor blockades induced by relatively high complication rates.5,15 The possible rea- 2% lidocaine or 0.5% bupivacaine (Table 1). sons for these differences included that the anatomical Two saline-injected limbs in the lidocaine group characteristics of the rat BP and the NS-guided tech- exhibited complete and partial motor and/or sensory nique improved the safety of the rat BPB model. The blockade for 30 min (Figure 6(a)). There were no sig- rat model of NS-guided BPB is a feasible, safe, effective nificant differences in % UA among the lidocaine, bupi- and reproducible model. vacaine, and saline injections. No abnormal behavior, Autopsy was performed to examine easily identifiable color change, ulcers, or necrosis of the thoracic limbs surface landmarks and to define an effective entry were observed. approach for developing a rat BPB model. Our cadaver study revealed widespread morphological homology Discussion between the rat and human BP (Figure 1). However, some slight differences were observed between the rat The present study developed and investigated a rat BP and the human BP. The axillary vessels ran along model of BPB. Completely reversible BPBs were and beneath the cords but were not surrounded by the observed in all limbs that received 2% lidocaine or three cords. There were no clear medial, posterior, or 0.5% bupivacaine but not in the limbs that received lateral cords in the rat BP, which was also reported in a normal saline. These results demonstrate high success previous study.8 The trunks of the human BP pass and reliability rates in the rat BPB model. The anatom- between the anterior and middle scalene muscles at the ical and BPB results revealed that the rat BP innervated level of the supraclavicular space, over the first rib, and the cutaneous and muscular structures of the entire continue beneath the clavicle. The sternum and clavicle in thoracic limb, and the NS-guided BPB provided com- rats are at a higher position than in humans. The supras- plete anesthesia (motor and sensory block). The motor ternal border in rats is located at the 6th cervical vertebra and sensory blockade induced by bupivacaine exhibited (Figure 1) but is in front of the 1st thoracic vertebra in a longer duration than that induced by lidocaine. humans. The divisions and cords of the rat BP are located This characteristic is consistent with the pharmacody- beneath the clavicle. namic function of these two drugs, as was demon- The results of the BP dissection in rats revealed that strated in other models.14 Therefore, this BPB the space from the lateral one-third of the clavicle to the model is effective and reproducible and is suitable for axillary vessels (distance approximately 0.5–1.0 cm) was evaluating the efficacy, safety, and pharmacokinetics of a safe puncture area. The anterior and posterior areas new local anesthetics and additives. This model will aid of this space were composed of muscles and fascia. Our in the investigation of the chemical and pharmaco- staining study demonstrated that 6 of the 20 BPs from logical injurious effects of substances injected onto per- the cadavers were not stained with methylene blue, ipheral nerves and the effects of nerve block on specific which indicates inaccurate blind injections. Therefore, outcomes. we used the nerve stimulator-guided method to improve 166 Laboratory Animals 53(2) the success rate of BPB. All BPBs were successful with elucidated. There was no statistically significant differ- the guidance of the nerve stimulator in the live rats, and ence between the onset times of lidocaine and bupiva- no severe injuries or side effects were observed. caine. Motor and sensory function was blocked prior to The duration of bupivacaine on rat BPB in this study recovery from anesthesia. The actual onset time was was 103.1–121.9 min (Table 1). This duration of shorter than the recorded onset time. The recorded action was shorter than that of bupivacaine described onset time was 5.50 2.07 min after 2% lidocaine injec- in rat anesthesia and analgesia formulary.16 However, tion and 4.89 1.73 min after 0.5% bupivacaine injec- one reason for the shorter duration in this study was tion. BPB block without general anesthesia could be the difference in the definition of block duration. useful in the determination of the onset times of local We defined the onset time as the time from drug injec- anesthetics. The high concentration of local anesthetic tion to complete blockade (score of 3) and the offset time (2%, lidocaine) may also contribute to this limitation. as the time interval after achieving complete blockade A lower concentration of local anesthetic may be bene- and to reversion to a partial blockade (score of 2), but ficial for observing the exact onset time and recovery not to normal function (score of 1). Brummett et al. used sequence of motor and sensory blocks. A previous 0.5% bupivacaine in 0.2 ml for rat sciatic nerve block.17 study demonstrated a motor deficit anterior to nocicep- The duration of complete blockade (101.3 20.7 min) tion block in rats after local anesthetic injection was similar to that reported in our study, but the time (0.05-1%, lidocaine).22 A graded series of doses and to restoration of normal function (223.1 22.4 min) was concentrations of local anesthetics should be used to similar to the duration described in the formulary. further evaluate the model. Sensory or motor blockade after placebo injection was observed in 2 of the 16 rats in the present study Conclusions (Figure 6(a)), but this block was shorter than the blocks from lidocaine and bupivacaine (Table 1). The unex- We developed a feasible and promising BPB model in pected blockade was not associated with permanent rats using cadavers and live rats that can be used for neurological injury or functional impairment in the analgesia in upper limb surgery in rodents to evaluate hind limbs, and pathological study of the BP nerves the efficacy, safety and pharmacokinetics of new local revealed no demyelination or Wallerian degeneration. anesthetics and additives. The possible reasons behind the saline-induced block should be further investigated. This phenomenon is Acknowledgments sometimes referred to as a possible physiological We thank colleagues in the laboratory of Translational block, which is unrelated to pharmacological effects. Neuroscience Centre, West China Hospital, Sichuan A psychological reason may also be considered because University, for their assistance with the study. pain may be transferred via olfaction in co-housed mice.18 Whether there is a social transfer of neural Declaration of Conflicting Interests blockade in rodents is not known. The author(s) declared no potential conflicts of interest with Ultrasound (US)-guided BPB is prevalent in upper respect to the research, authorship, and/or publication of this extremity surgery. However, NSs are cheaper and easier article. to obtain and operate than US systems. A randomized controlled trial indicated no difference in the efficacy of Funding axillary BPB performed by novices when using ultrasound 19 The author(s) disclosed receipt of the following financial sup- guidance vs. the NS technique. Axillary blocks with US port for the research, authorship, and/or publication of this guidance exhibited similar success rates and complication article: This work was supported by the National Research 20 ratesasNSguidance. Our study demonstrated the feasi- Foundation of Nature Sciences, Beijing, P.R. China, (grant bility of a BPB rat model and confirmed the efficacy and number 81271201). safety of BPB under NS-guided conditions. Further studies should be performed to investigate the US-guided ORCID iD BPBmodelinratsandtocompareUSandNSguided- Jing Yang http://orcid.org/0000-0002-0971-4566. BPB in rats for the development of a safer and more 21 convenient BPB method. Vuckovic et al. demonstrated References that measuring the injection pressure was an effective 1. Brattwall M, Jildenstal P, Warren Stomberg M, et al. Upper method for evaluating the safety of BPB in the axilla of extremity nerve block: how can benefit, duration, and safety rats. This technique would be a good supplement for NS be improved? An update. F1000Res 2016; 5: 1–9. or US-guided BPB to improve injection safety. 2. Knezevic NN, Anantamongkol U and Candido KD. One limitation of the current approach is that the Perineural dexamethasone added to local anesthesia for onset times of the motor and sensory blocks were not brachial plexus block improves pain but delays block Zhang et al. 167

onset and motor blockade recovery. Pain Physician 2015; 13. Bertelli JA and Mira JC. The grasping test: a simple 18: 1–14. behavioral method for objective quantitative assessment 3. Buvanendran A, Kroin JS, Li J, et al. Relative contribu- of peripheral nerve regeneration in the rat. J Neurosci tion of adjuvants to local anesthetic for prolonging the Methods 1995; 58: 151–155. duration of peripheral nerve blocks in rats. Reg Anesth 14. Li Z, Chai Y, Gong C, et al. Evaluation of the antinoci- Pain Med 2016; 41: 589–592. ceptive effects of lidocaine and bupivacaine on the tail 4. Memari E, Hosseinian MA, Mirkheshti A, et al. nerves of healthy rats. Basic Clin Pharmacol Toxicol Comparison of histopathological effects of perineural 2013; 113: 3–36. administration of bupivacaine and bupivacaine-dexmede- 15. Rioja E, Sinclair M, Chalmers H, et al. Comparison of tomidine in rat sciatic nerve. Exp Toxicol Pathol 2016; 68: three techniques for paravertebral brachial plexus block- 559–564. ade in dogs. Vet Anaesth Analg 2012; 39: 190–200. 5. Ricco C, Shih A, Killos M, et al. Different volumes of 16. The Institutional Animal Care and Use Committee. injectate using electrostimulator and blinded techniques Anesthesia and analgesia in laboratory animals. Rat for- for brachial plexus block in dogs. Vet Rec 2013; 173: 608. mulary, http://www.iacuc.ucsf.edu/Proc/awRatFrm.asp 6. Anson A, Laredo FG, Gil F, et al. Evaluation of an (2017, accessed 7 April 2018). ultrasound-guided technique for axillary brachial plexus 17. Brummett CM, Norat MA, Palmisano JM, et al. blockade in cats. J Feline Med Surg 2017; 19: 146–152. Perineural administration of dexmedetomidine in com- 7. Estebe JP, Le Corre P, Chevanne F, et al. Motor block- bination with bupivacaine enhances sensory and motor ade by brachial plexus block in the sheep. Anesthesiology blockade in sciatic nerve block without inducing neuro- 2000; 93: 291–293. toxicity in rat. Anesthesiology 2008; 109: 502–511. 8. Angelica-Almeida M, Casal D, Mafra M, et al. Brachial 18. Smith ML, Hostetler CM, Heinricher MM, et al. Social plexus morphology and vascular supply in the wistar rat. transfer of pain in mice. Sci Adv 2016; 2: e1600855. Acta Med Port 2013; 26: 243–250. 19. Barrington MJ, Gledhill SR, Kluger R, et al. A rando- 9. National Research Council. Guide for the Care and Use of mized controlled trial of ultrasound versus nerve stimu- Laboratory Animals, 8th ed. Washington, DC: The lator guidance for axillary brachial plexus block. Reg National Academies Press, 2011. Anesth Pain Med 2016; 41: 671–677. 10. National Standard. Laboratory animals. Nutrients for for- 20. Kumar A, Sharma D, Sibi ME, et al. Comparison of mula feeds, https://www.chinesestandard.net/Default. peripheral nerve stimulator versus ultrasonography aspx?PDF-English-ID¼GB%2014924.3-2010 (2010, guided axillary block using multiple injection technique. accessed 7 April 2018). Indian J Anaesth 2014; 58: 700–704. 11. National Standard. Laboratory animals. Hygienic stand- 21. Vuckovic I, Hadzic A, Dilberovic F, et al. Detection of ard for formula feeds, https://www.chinesestandard.net/ neurovascular structures using injection pressure in Default.aspx?PDF-English-ID¼GB%2014924.2-2001 blockade of brachial plexus in rat. Bosn J Basic Med (2001, accessed 7 April 2018). Sci 2005; 5: 79–85. 12. Grant GJ, Vermeulen K, Zakowski MI, et al. A rat sciatic 22. Gokin AP, Philip B and Strichartz GR. Preferential block nerve model for independent assessment of sensory and of small myelinated sensory and motor fibers by lido- motor block induced by local anesthetics. Anesth Analg caine: in vivo electrophysiology in the rat sciatic nerve. 1992; 75: 889–894. Anesthesiology 2001; 95: 1441–1454.

Re´sume´ Il est important de de´velopper un mode`le animal faisable d’anesthe´sie re´gionale autres que le blocage du nerf sciatique pour des e´tudes pharmacocine´tiques portant sur de nouveaux anesthe´siques locaux ou analge´siques destine´sa` la chirurgie du membre supe´rieur. Nous avons e´tudie´ un bloc du plexus brachial (BPB) guide´ par stimulateur de nerf (SN) sur un mode`le de rat. L’anatomie du plexus brachial chez le rat a e´te´ de´limiteé sur des cadavres, et divers BPB ont e´te´ examine´s. Le point de ponction est situe´ a` 0.5-1.0 cm sous le tiers late´ral de la clavicule. L’efficacite´ et l’innocuite´ du BPB guide´ par SN ont e´te´ e´valueés en utilisant une injection de lidocaı¨ne 2% ou de bupivacaı¨ne 0.5% sur 16 animaux vivants; l’injection de solution saline servait de controˆle. Les deux coˆte´s du plexus brachial ont e´te´ localise´s en utilisant la technique du guidage par SN. Le bloc sensoriel de la nociception (e´valuation) et moteur (tests de prise et de redressement) sont apparus apre`s l’administration des deux anesthe´siques locaux, mais pas de la solution saline normale. Le bloc moteur et sensoriel induit par la bupivacaı¨ne durait plus longtemps que celui induit par la lidocaı¨ne (P < 0.05). Tous les rats ont rećupe´re´ sans proble`me apre`s l’anesthe´sie geńe´rale par BPB. Aucun re´sultat anormal n’a e´te´ observe´ dans les e´tudes pathologiques ni dans les observations du comportement. Ainsi, un mode`le de BPB guide´ par SM chez le rat a e´te´ e´tabli, et le BPB a induit un bloc moteur et sensoriel re´versible dans l’ensemble au niveau des membres thoraciques. L’e´valuation de l’efficacite´ et de la sećurite´ ont de´montre´ que ce mode`le de BPB chez le rat e´tait reálisable, reproductible et suˆr. 168 Laboratory Animals 53(2)

Abstract Fu¨r pharmakokinetische Untersuchungen neuer Lokalana¨sthetika oder Analgesie in der Chirurgie der oberen Extremita¨ten ist es wichtig, ein praktikables Tiermodell der Regionalana¨sthesie, abgesehen von einer Blockade des N. ischiadicus, zu entwickeln. Hier untersuchten wir eine Nervenstimulator (NS)-gesteuerte Plexus-brachialis-Blockade (BPB) in einem Rattenmodell. Die Anatomie des Plexus brachialis bei Ratten wurde fu¨r Kadaver beschrieben und verschiedene BPBs wurden untersucht. Der Punktionspunkt lag 0.5-1.0 cm unter dem seitlichen Drittel des Schlu¨sselbeins. Wirksamkeit und Sicherheit der NS-gesteuerten BPB wurden mit einer Injektion von 2% Lidocain oder 0.5% Bupivacain bei 16 lebenden Tieren evaluiert, Kochsalzlo¨sungs-Injektion wurde als Kontrolle verwendet. Beide Seiten des Plexus brachialis wurden mit der NS-gesteuerten Methode erfolgreich lokalisiert. Sensorische Blockade (Nozizeptions-Beurteilung) und motorische Blockade (Greif- und Begradigungstests) traten nach Anwendung der beiden klassischen Lokalana¨sthetika auf, jedoch nicht bei normaler Kochsalzlo¨sung. Die motorische und sensorische Blockade durch Bupivacain hielt la¨nger an als die durch Lidocain induzierte (P < 0.05). Alle Ratten erholten sich ereignislos von Vollnarkose und BPB. Es wurden keine anormalen Ergebnisse in pathologischen Untersuchungen oder Verhaltensbeobachtungen gefunden. Damit wurde ein Rattenmodell NS-gesteuerter BPB etabliert, und BPB induzierte eine insgesamt reversible sensorische und motorische Blockade in den thorakalen Extremita¨ten. Die Bewertung von Wirksamkeit und Sicherheit hat gezeigt, dass dieses Ratten- BPB-Modell praktikabel, reproduzierbar und sicher ist.

Resumen Es importante desarrollar un modelo animal factible de anestesia regional distinto al bloqueo del nervio cia´tico para las investigaciones farmacocine´ticas de nuevos aneste´sicos locales o analgesia en las opera- ciones de las extremidades superiores. Por tanto, exploramos un bloqueo de plexo braquial guiado por un estimulador nervioso (NS) (BPB) en un modelo de rata. La anatomıá del plexo braquial en ratas estaba delineada en cada´veres, y se examinaron varios BPB. El punto de puncioń se situo´ a 0.5-1.0 cm por debajo de un tercio lateral de la clavıćula. La eficacia y seguridad del BPB guiado por un NS se evaluaron utilizando una inyeccioń de un 2% de lidocaıńa o un 0.5% de bupivacaıńa en 16 animales vivos; se utilizo´ una inyecciońde salina como control. Ambos lados del plexo braquial fueron ubicados satisfactoriamente utilizando la tećnica del guiado NS. El bloqueo sensorial (evaluacioń de nocicepcioń) y el bloqueo motor (pruebas de estiramiento y agarre) aparecieron despue´s de la aplicacioń de los dos aneste´sicos cla´sicos, pero no la salina normal. El bloqueo motor y sensorial inducido por la bupivacaina mostro´ una mayor duracioń que la induccioń con lidocaina (P < 0.5). Todas las ratas se recuperaron sin novedades de la anestesia general y el BPB. No se encontraron resultados anormales en los estudios patolo´gicos u observaciones de comportamiento. Por tanto, se establecio´ un modelo de rata de BPB guiado por NS, y el BPB provoco´ un bloqueo general sensorial y motor en las extremidades toraćicas. La evaluacioń de la eficacia y la seguridad demostro´ que este modelo BPB de rata era factible, reproducible y seguro. Original Article Laboratory Animals 2019, Vol. 53(2) 169–179 ! The Author(s) 2018 4D cardiac magnetic resonance imaging, Article reuse guidelines: sagepub.com/journals- 4D and 2D transthoracic echocardiography: permissions DOI: 10.1177/0023677218789971 a comparison of in-vivo assessment of journals.sagepub.com/home/lan ventricular function in rats

Hedwig Stegmann1 , Tobias Ba¨uerle2, Katharina Kienle1, Sven Dittrich1 and Muhannad Alkassar1

Abstract Preclinical cardiovascular research is the foundation of our understanding and broad knowledge of heart function and cardiovascular disease. Reliable cardiac imaging modalities are the basis for applicable results. Four-dimensional cardiac magnetic resonance (4D CMR) has been set as the gold standard for in-vivo assessment of ventricular function in rodents. However, technical improvements in echocardiography now allow us to image the whole heart, which makes four-dimensional echocardiography (4DE) a possible alternative to 4D CMR. To date, no study has systematically assessed 4DE in comparison with 4D CMR in rats. In total we studied 26 juvenile Sprague-Dawley rats (Crl: CD (SD) IGS). Twenty rats underwent echocardiographic imaging (2D and 4D) and 4D CMR. Five of those rats underwent a ligation of the superior and inferior vena cava to reduce the cardiac inflow as a disease model. Six additional rats were used to assess reproducibility of echocardiography and underwent three echocardiographic examinations. 4D CMR was performed on a 7T scanner; 2D and 4D echocardiography was conducted using a 40 MHz transducer. Correlation between 4D CMR, 4DE and 2DE for left-ventricular ejection fraction (LVEF) was assessed. An excellent correlation was observed between 4DE and 4D CMR (r ¼ 0.95, p < 0.001). Correlation of 2DE and 4D CMR was weak (r ¼ 0.57, p < 0.01). 4DE provides results that are equally precise as 4D CMR and highly reproducible with less technical effort than 4D CMR.

Keywords echocardiography, cardiac magnetic resonance, left ventricular function, rat, four-dimensional cardiac imaging

Date received: 6 February 2018; accepted: 27 June 2018

Introduction (4D CMR). At this point in time, 4D CMR is the gold standard for assessing left ventricular (LV) func- Non-invasive cardiac imaging is an integral part of pre- tion in humans4 and in rodents5 due to its ability to clinical cardiovascular research.1 Rodents in particular obtain 3D images over time (with time being the are frequently used to evaluate cardiovascular dis- orders2 in order to understand physiological, molecular 1Department of Paediatric Cardiology, University Hospital and biochemical mechanisms of cardiovascular dis- 3 Erlangen, Friedrich-Alexander University Erlangen-Nuremberg, eases. This research requires a reliable imaging tech- Erlangen, Germany nique to gain valid data. Different imaging modalities 2Institute of Radiology, University Hospital Erlangen, Friedrich- are available to determine heart function, each having Alexander University Erlangen-Nuremberg, Erlangen, Germany their advantages and disadvantages concerning acqui- Corresponding author: sition time, availability and costs. Hedwig Stegmann, Uniklinik Erlangen Loschgestraße 15 Erlangen, Among these tools, the first to be mentioned is four- 91054 Germany. dimensional cardiac magnetic resonance imaging Email: [email protected] 170 Laboratory Animals 53(2) fourth dimension) with few geometric assumptions.6,7 often cut off due to the geometrical circumstances.15 The high spatial and moderate temporal resolution of To eliminate this error the VevoÕ 3100 imaging 4D CMR provides highly reproducible images with min- platform uses linear array transducers creating a imal inter- and intra-observer variations.8 However, cube-like model. most research institutions do not have small animal Accordingly, the objective of our study was to com- MR scanners available. As a result, echocardiography pare four-dimensional echocardiography (4DE) with is the method of choice for many researchers,9,10 as it the gold standard 4D CMR in order to test the produces images with high resolution at lower cost and hypothesis that 4DE is a valid method for cardiac shorter scan duration than 4D CMR. imaging. To the best of our knowledge, no evaluation Currently the determination of LV volumes from 2D of 4D CMR, 4DE and 2DE has been undertaken in images is a widely used approach in small animal echo- rats. Therefore, we sought to determine the accuracy, cardiography. Due to the circumstance that only one objectivity and reproducibility of 4DE in comparison slice is used to calculate volumetric heart function, two- with 4D CMR. dimensional echocardiography (2DE) is only accurate in normal functioning hearts and highly dependent on the experience of the operator.11 There are different Animals, material and methods approaches to determine LV function from transthor- Study protocol acic echocardiography (TTE), relying on either 1D M-mode evaluation, 2D short axis view, or 2D Animals were obtained and handled as approved by the long axis view. M-mode only provides information Franz-Penzoldt Centre of the Friedrich-Alexander about two opposing myocardial walls, e.g. the antero- University Erlangen-Nu¨rnberg in accordance with the septum and the posterior wall from parasternal views. EU Directive 2010/63/EU for animal experiments. The Calculation of left ventricular ejection fraction (LVEF) study was confirmed by the government according to from parasternal short axis or long axis view in B-Mode the German law for animal protection using LV-Trace can provide a more representative (Tierschutzgesetz Regierung Unterfranken). In total evaluation of LVEF, but it displays only one slice of we studied 26 juvenile Sprague-Dawleys rats (Crl: CD the heart not capturing the rest of the myocardium. (SD) IGS) weighing 75–100 g from our institutional In 2004 Dawson et al12 developed a method of three- breeding colony. The original animal supplier was dimensional echocardiography (3DE): 10 to 18 con- Charles River (Wilmington, MA, USA). We decided secutive short axis views were acquired by moving the to examine juvenile rats because those animals were mouse platform in 0.5 mm steps with the linear array needed for further experiments at a later age. Both probe being secured in one position. Respiratory gating male and female rats were used. Animals were kept in and ECG were used. Using this technique Dawson at al a 12 h:12 h light:dark cycle at room temperature measured no significant difference concerning LV mass, between 21–24C, food (V1534000, ssniff rat/mouse, volumes and EF in comparison with 4D CMR.12 Ssniff GmbH, Soest, Germany) and water were pro- Arias et al.13 compared 4D CMR with 2DE that used vided ad libitum. All rats were not handled before geometric models (hemisphere cylinders) to calculate LV their inclusion in the study except for routine cage volumes. In their publication they claimed matrix-array cleaning maintenance and had no history of health transducers and gated imaging need to be able to acquire abnormalities. Echocardiography and 4D CMR were 4D imaging by echocardiography. In 2016 a new method performed in 20 rats at four weeks age. To compare of echocardiography became available, one that pro- the method of assessment in healthy and diseased ani- duces 3D images of the beating heart. mals, five rats underwent ligation of the superior and Similar to 4D CMR many slices are acquired; they inferior vena cava to reduce cardiac inflow two weeks are then put together to create a 3D model of the beat- prior to imaging (method in publishing process). The ing heart (Figure 1). time interval between 4D CMR and echocardiography Compared with Dawson et al12 the number of was 0 1 d. The other 6 animals underwent three echo- slices that can be acquired are increased significantly cardiographic scans to assess reproducibility. Rats were by the technical progress of hardware and software. anaesthetized with 1.5% isoflurane for both imaging The probe is now connected to a motor that is able modalities.16 to move through the heart in steps of a few micro- metres. The acquired slices are automatically combined Disease model to a 3D model of the beating heart. Clinical 4D ultrasound uses convex probes14 that are only able to create Five animals aged 14 1 d underwent surgical cone-shaped 4D models. Unfortunately, important banding of the superior (SVC) and inferior (IVC) structures such as the apex and myocardial walls are vena cava. Anaesthesia was induced with 5% isoflurane Stegmann et al. 171

Figure 1. Principal geometric figures used to represent the left ventricle. (a) Algorithms for a full volume data set of 4D CMR and 4DE; the volume of each slice gets calculated separately and to get the full volume all slices are added. (b) Volume quantification for 2DE by ellipsoid formula using only one single slice. and maintained with 1.5% isoflurane. The animals Cardiac magnetic resonance imaging were intubated and buprenorphine (0.05–0.1 mg kg1 bodyweight) was given for analgesia. Chest hair was 4D CMR was performed with a 7T horizontal bore removed with a small animal clipper. animal MR scanner (Bruker BioSpin GmbH, The thoracic cavity was opened and the SVC and Ettlingen, Germany) using a birdcage coil with an IVC were carefully dissected. The diameters of SVC inner diameter of 72 mm. Animals were administered and IVC were reduced to a standardized 0.6 mm to with 1.5% isoflurane via a nose cone.17,18 Heart rate minimize blood flow to the right heart. and respiratory rate were monitored. All rats were After closing the thorax, isoflurane was switched off kept warm using heated coils. A three-plane localizer and a second dose of buprenorphine was given intra- was used to plan 4D CMR acquisition. The cine loop peritoneal when the animal was breathing sufficiently. images (15 frames per cardiac cycle) were obtained The fully awake animal was returned to its mother’s using a self-gated protocol with the following param- care and closely monitored for the next 24 h. Analgesia eters: 3D acquisition type with 86 encoding steps; was provided with buprenorphine 0.1 mg kg1 body- TR 5.3 ms; TE 2.52 ms; number of averages 3; matrix weight every 8 h for the next three days. 256 144; slice thickness 0.6 mm; pixel spacing 172 Laboratory Animals 53(2)

0.13 mm; flip angle 35. Typically, seven or eight slices current guidelines.11 Papillary muscle was excluded in were acquired in 15 phases per cardiac cycle. Scan time the LV volume calculation. was set to 40 min per rat (overview in Table 1). The total study duration including localizers was 60 min Echocardiography per animal. LV volumes were evaluated using Mimics Research19.0 (Figure 2 (c) and (d)). End-diastole was The images were acquired using a 3100 VevoÕ Imaging defined as the frame with the largest LV cavity, end- Platform (Fujifilm VisualSonics Inc, Toronto, Canada) systole as the smallest LV cavity according to the with a 40 MHz linear array transducer (MS 400). In 2016 the VevoÕ 3100 imaging platform integrated a motor into their system and developed software that can calculate a 3D model of the beating heart Table 1. Comparison of acquisition parameters of 4D CMR and 4DE. (time as fourth dimension, therefore also referred to as real-time 3DE) by combining the imaged slices. 4D CMR 4DE In addition, gated imaging allows for the heart cycle Õ (7T, Bruker (3100 Vevo to be split into different phases to differentiate between BioSpin GmbH) Imaging Platform) systole and diastole. For 4DE the probe is mounted Frames per cycle 15 200 into a bracket ensuring a horizontal movement of the Slice thickness 600 mm10mm probe without any vertical deviation. The rats were 16 Scan time 40 min 8 min under anaesthesia with 1.5% isoflurane via mask and placed in a supine position on a heat-regulated Resolution 130 mm 130 mm50mm 110 mm table. Chest hair was removed, using a small animal

Figure 2. Four-dimensional cardiac magnetic resonance (4D CMR). (a) Long axis in end-diastole. (b) Short axis in end- diastole. (c) Mimics Research 19.0, end-diastolic 4D CMR images including the calculated 3D model of the left and right ventricle (red) and the myocardium (yellow). (d) End-diastolic 3D model of the whole heart for volume measuring using Mimics Research 19.0 based on 4D CMR images. Dark red, left ventricle; light red, aortic arch; purple, right ventricle and pulmonary artery; blue, superior vena cava and left persistent superior vena cava. Stegmann et al. 173

Figure 3. Transthoracic echocardiography (TTE). (a) Photograph illustrating TTE set-up with 40 MHz transducer fixed in bracket that is connected to the motor. (b) Coordinate 4D view of the rat heart showing the left ventricular long axis (apex left, base right). (c) M-Mode sequence. (d) Representative 2D parasternal long axis in end-diastole.

clipper. The experimental set-up is shown in muscle was not included in the blood volume using either Figure 3(a). Heart rate and respiratory rate were software. monitored. A rectal probe registered core temperature, which was held stable at 37.0 0.5 C. 4DE Reproducibility (Figure 3(b)) and 2D parasternal long axis (Figure 3(d)) were performed. Parasternal long axis To assess the intra- and interobserver reproducibility of (PSLAX) was chosen as 2DE method, because this echocardiography six additional animals were examined was the starting position for 4D scanning and the twice by the same observer and underwent a further examination time was not prolonged by short axis echocardiographic scan by an independent observer. image acquisition. The investigator uses adjusting All three scans were performed in different periods of screws, moving the bracket with the probe to get into anaesthesia with an interval of at least two hours. the position of PSLAX. Once the probe is in the right position, the system is to scan through the whole heart Statistics systematically across the orientation of PSLAX. The total study duration including 2DE took between 15–20 min IBM SPSS Statistics 21 (IBM Corp. Released 2012. per animal. 4D scanning was performed at the beginning IBM SPSS Statistics for Windows, Version 21.0. of the examination and took 8 min per animal in the Armonk, NY) was used for statistical analyses. LVEF highest definition with a slice thickness of 10 mmand values were expressed as mean standard deviation 200 frames per cycle (overview in Table 1). The average (SD). Bland-Altman analysis was conducted with number of slices was 480. All data was evaluated offline mean bias, standard deviation and 95% limits of agree- using the Vevo LAB 3.0 software. Using this software the ment (mean 1.96 SD) calculated. Pearson’s correl- investigator had to mark the endocardium in an average ation analysis was performed between LVEF derived of six slices, in diastole and systole. The software was then by 4DE, 4D CMR and 2DE. Intra- and interobserver able to trace the endocardium in the other slices. Papillary reproducibility was tested by calculating the interclass 174 Laboratory Animals 53(2) correlation for two consecutive measurements of 4DE whereas 2DE (Figure 4(b) and (d)) has wider distribution and 2DE. Variabilities were described as relative and a significant lower LVEF with a mean difference of changes in percent between two datasets of different 8% in comparison with 4D CMR. Overall, 4DE and 4D recordings as a coefficient of variation (CV) of differ- CMR gave concordant measures of EDV, ESV and ences and were calculated as following for n animals: LVEF (Table 2).

Coefficent of variation ð%Þ !Intra-observer agreement jjObserver 1 Observer 2 The CV, mean differences in EF (EF) and interclass ¼ Observer 1þ Observer 2 100 2 correlation (ICC) for each animal study between the first and second analyses were evaluated for 4DE and X6 Observer 1 Observer 2 2DE (PSLAX) as shown in Table 3. For 4DE the mean Mean Difference ¼ i i 6 difference was 2.8 3.4% while it was appreciably i¼1 higher for 2DE with a mean difference of 6 6.7%. Intra-observer agreement was excellent for 4DE Results (ICC ¼ 0.94, p < 0.01), whereas 2DE derived EF was only moderate in agreement (ICC ¼ 0.68, p < 0.05). Imaging was successfully completed in all 26 animals. Bland-Altman analysis (Figure 5(a) and (b)) reiterated Two sets of 4D CMR images (control group) could not the high level of intra-observer agreement for 4DE. be included in the evaluation due to large artefacts in the area of the LV cavity. Mean heart rate was 353 Inter-observer agreement 45 beats per min. Respiratory rate remained stable between 40–60 breaths per min. LV volumes (end-dia- Mean differences in EF between two different obser- stolic volume (EDV) and end-systolic volume (ESV)) vers were lower for 4DE (2.8 4.5) than for 2DE and LVEF measured by 4DE, 4D CMR and 2DE (5.3 8.9). Inter-observer agreement was excellent for (PSLAX) are shown in Table 2. 4DE (ICC ¼ 0.89, p < 0.01) (Figure 5(c)) whereas 2DE derived EF was only moderate in agreement Correlation between modalities (ICC ¼ 0.71, p < 0.05) (Figure 5(d)). Bland-Altman analysis revealed a trend for one observer to systemat- Linear regression analysis (Figure 4(a)) and Bland- ically measure 4DE EF marginally lower (bias 2.5%). Altman analysis (Figure 4(b)) showed excellent correl- For 2DE there was no significant bias, however differ- ations and strong agreements between 4DE LVEF and ences were found between observers at either end of the 4D CMR LVEF. Furthermore, the correlation coeffi- EF spectrum. cient between 4DE LVEF and 4D CMR demonstrated better correlation than that of 2DE LVEF (2DE Discussion LVEF v. 4D CMR: r ¼ 0.577, p < 0.01; 4DE LVEF v. 4D CMR: r ¼ 0.952, p < 0.01; Figure 4(b) and (d)). The results of our study imply two main messages: (a) As seen in the Bland-Altman analysis (Figure 4(c)) 4DE 4D CMR and 4DE show excellent correlation in meas- and 4D CMR show strong concordance regarding LVEF, urements of volumetric heart function, and (b) 4DE is

Table 2. Comparison of 4D CMR, 4DE and 2DE (PSLAX) for EDV, ESV and LVEF in diseased and normal animals, separately and for the whole study sample. Values are expressed as mean standard deviation (SD).

EDV [ml] ESV [ml] LVEF [%]

4D CMR 4DE 2DE 4D CMR 4DE 2DE 4D CMR 4DE 2DE

Diseased 160 14 158 14 108 57 55 10 53 10 56 11 65 466 448 6 n ¼ 5 Control 154 23 153 24 126 37 47 10 45 10 47 18 70 571 563 8 n ¼ 13 Overall 156 21 154 21 121 43 49 11 47 10 50 16 68 569 561 8 n ¼ 18 Stegmann et al. 175

Figure 4. Correlation between imaging modalities. Dots are used for control animals (n ¼ 13), stars for diseased animals (n ¼ 5). (a), (b) Linear correlation charts with regression lines for each comparison. (c), (d) Corresponding Bland-Altman plots with mean bias and 95% limits of agreement (mean 1.96 SD) displayed.

Table 3. Coefficient of variation (CV), mean difference with very accurate and highly reproducible. This is the ﬁrst a confidence interval of 95%, and interclass correlation are time that echocardiography has produced results that displayed for intra- and inter-observer reproducibility of are comparable with measurements done by 4D CMR. 13 4DE and 2DE. Previously, HC 2D-echocardiographic models and measurements of LV cross-sectional area derived ejec- Mean Interclass tion fraction7 were compared with CMR-derived full- CV difference correlation volume LVEF. Geometric assumptions of HC models Intra-observer led to an overestimation of LV volumes.13 Another 4DE 3.8 2.3 2.8 3.4 0.94 (p < 0.01) method to calculate LV function is Simpson’s rule. It 2DE 8.6 5.7 6 6.7 0.68 (p < 0.05) uses biplane apical discs throughout the heart instead Inter-observer of just one single slice. Simpson’s rule produces accurate 4DE 3.8 3.0 2.8 4.5 0.89 (p < 0.01) results in normal functioning hearts and is very common in clinical echocardiography.11 However, it is inaccurate 2DE 7.6 6.6 5.3 8.9 0.71 (p < 0.05) when wall motion abnormalities are involved.19 176 Laboratory Animals 53(2)

Figure 5. Bland-Altman plots with mean bias, standard deviation and 95% limits of agreement displayed. (a) Intra- observer reproducibility of 4DE; (b) intra-observer reproducibility of 2DE; (c) inter-observer reproducibility of 4DE; (d) inter-observer reproducibility of 2DE.

Van de Weijer10 described 4D CMR as superior to probe isn’t as crucial as it is in 2DE, because the echocardiography because the volume measurement is whole heart is divided into many slices to calculate based on many acquired slices, therefore crucially mini- heart function. Therefore, the results of longitudinal malizing the geometric assumptions. 4DE is now pro- studies are more reliable when measured by 4DE in ducing images on the same technological basis as 4D comparison with other echocardiographic approaches, CMR by combining many acquired slices (Figure 1). even when the scans are carried out by independent Evaluating LVEF by 4DE with minimal geometric observers. assumptions leads to results that are consistent with As natural variation in EDV, ESV, and LVEF is 4D CMR. Another advantage of 4DE is that by fairly large20 (in this study both 4DE and 4D CMR displaying the whole heart it becomes very useful for show that LVEF ranges from 60–80%), this might pro- diagnostic imaging. Further studies are needed to con- duce undetectable variation at the group level, when the firm that it is able to detect wall motion abnormalities mean between groups is examined. In these cases, lon- or changes in structures of the myocardium. In add- gitudinal studies with 4DE offer the possibility to detect ition, we have shown that the intra- and inter-observer subtle changes in systolic function by individual follow- reproducibility of 4DE is significantly higher than up. This may allow researchers to use smaller groups in in 2DE. The impact of the initial position of the detecting changes in heart function. Stegmann et al. 177

A major advantage of 4DE is the high spatiotem- Declaration of Conflicting Interests poral resolution and the fast scan time compared with The author(s) declared no potential conflicts of interest with 4D CMR (Table 2). A high definition 4DE of the heart respect to the research, authorship, and/or publication of this takes 8 min, which is less than a quarter of the time article. needed for 4D CMR. As a result, stress on rodents is reduced since the animals only require approximately Funding 10 min of light anaesthesia.21 This time can be further The author(s) disclosed receipt of the following financial sup- reduced when the slice thickness (in our study 10 mm) is port for the research, authorship, and/or publication of this increased. article: this research was supported by the Klaus Tschira In our experience there were fewer problems Stiftung gGmbH. with gating for ultrasound than we would expect with 4D CMR, which is more susceptible to disruptions ORCID iD and disturbances concerning cardiac monitoring, because ECG-gating is strongly influenced by the mag- Hedwig Stegmann http://orcid.org/0000-0002-9269-8850 netic field.22 However ultrasound is more vulnerable to gas and bone artefacts, and extra care must be taken References when scar tissue is present. The Vevo 3100 imaging 1. Hasenfuss G. Animal models of human cardiovascular platform provides probes for rats of all ages, so we disease, heart failure and hypertrophy. Cardiovasc Res don’t expect other results when 4DE is performed in 1998; 39: 60–76. older animals. Further studies have to confirm this 2. Russell JC and Proctor SD. Small animal models of car- assumption. diovascular disease: tools for the study of the roles of metabolic syndrome, dyslipidemia, and atherosclerosis. Cardiovasc Pathol 2006; 15: 318–330. Conclusion 3. Patten RD and Hall-Porter MR. Small animal models of To the best of the authors’ knowledge, this is the first heart failure: development of novel therapies, past and present. Circ Heart Fail 2009; 2: 138–144. study to compare 4DE with 4D CMR in rats. Excellent 4. Greupner J, Zimmermann E, Grohmann A, et al. correlation of LV size and heart function was seen Head-to-head comparison of left ventricular function between the newly available method of 4DE and 4D assessment with 64-row computed tomography, biplane CMR in healthy and diseased animals, showing that left cineventriculography, and both 2- and 3-dimensional 4DE is an improvement compared with older echocar- transthoracic echocardiography: comparison with mag- diographic approaches. Availability, expertise and the netic resonance imaging as the reference standard. JAm aim of studies will determine which imaging modality Coll Cardiol 2012; 59: 1897–1907. will be used. Our results suggest that researchers can 5. Prunier F, Marescaux L, Franconi F, et al. Serial mag- now confidently rely on 4DE as the method of choice netic resonance imaging based assessment of the early for cardiac imaging. effects of an ACE inhibitor on postinfarction left ventricular remodeling in rats. Can J Physiol Pharmacol 2005; 83: 1109–1115. Limitations 6. Schneider JE, Cassidy PJ, Lygate C, et al. Fast, high- resolution in vivo cine magnetic resonance imaging in This study was done in young Sprague-Dawley rats; normal and failing mouse hearts on a vertical 11.7 T further studies for adult rats are needed. system. J Magn Reson Imaging 2003; 18: 691–701. Reproducibility analysis was only done for echocardi- 7. Stuckey DJ, Carr CA, Tyler DJ, et al. Cine-MRI versus ography not for 4D CMR. Therefore we could only two-dimensional echocardiography to measure in vivo show that measurements by 4DE are highly reprodu- left ventricular function in rat heart. NMR Biomed cible, but were not able to compare reproducibility of 2008; 21: 765–772. 4DE and 4D CMR. Simpson analysis was not included 8. Bunck AC, Engelen MA, Schnackenburg B, et al. in the study. Feasibility of functional cardiac MR imaging in mice using a clinical 3 Tesla whole body scanner. Invest Radiol 2009; 44: 749–756. Acknowledgements 9. Ram R, Mickelsen DM, Theodoropoulos C, et al. New We thank Dipl.-Phys. Jutta Jordan, Dr. rer. nat. Clarissa approaches in small animal echocardiography: imaging Gillmann and Dr. rer. nat. Marc Schwarz from the Institute the sounds of silence. Am J Physiol Heart Circ Physiol of Radiology of the UK Erlangen for their technical support. 2011; 301: H1765–780. The present work was performed in fulfilment of the require- 10. van de Weijer T, van Ewijk PA, Zandbergen HR, et al. ments for obtaining the degree Dr. med for Hedwig Geometrical models for cardiac MRI in rodents: com- Stegmann. parison of quantification of left ventricular volumes and function by various geometrical models with a 178 Laboratory Animals 53(2)

full-volume MRI data set in rodents. Am J Physiol Heart structure and function in rats. Basic Res Cardiol 2007; Circ Physiol 2012; 302: H709–715. 102: 28–41. 11. Lang RM, Badano LP, Mor-Avi V, et al. 17. Hanusch C, Hoeger S and Beck GC. Anaesthesia of small Recommendations for cardiac chamber quantification rodents during magnetic resonance imaging. Methods by echocardiography in adults: an update from the 2007; 43: 68–78. American Society of Echocardiography and the 18. Kober F, Iltis I, Cozzone PJ, et al. Cine-MRI assessment European Association of Cardiovascular Imaging. Eur of cardiac function in mice anesthetized with ketamine/ Heart J Cardiovasc Imaging 2015; 16: 233–270. xylazine and isoflurane. MAGMA 2004; 17: 157–161. 12. Dawson D, Lygate CA, Saunders J, et al. Quantitative 3- 19. Wandt B, Bojo L, Tolagen K, et al. Echocardiographic dimensional echocardiography for accurate and rapid assessment of ejection fraction in left ventricular hyper- cardiac phenotype characterization in mice. Circulation trophy. Heart 1999; 82: 192–198. 2004; 110: 1632–1637. 20. Hoit BD, Kiatchoosakun S, Restivo J, et al. Naturally 13. Arias T, Chen J, Fayad ZA, et al. Comparison of echo- occurring variation in cardiovascular traits among inbred cardiographic measurements of left ventricular volumes mouse strains. Genomics 2002; 79: 679–685. to full volume magnetic resonance imaging in normal and 21. Albrecht M, Henke J, Tacke S, et al. Influence of diseased rats. J Am Soc Echocardiogr 2013; 26: 910–918. repeated anaesthesia on physiological parameters in 14. Kwon SH and Gopal AS. 3D and 4D Ultrasound: male Wistar rats: a telemetric study about isoflurane, Current Progress and Future Perspectives. Curr ketamine-xylazine and a combination of medetomidine, Cardiovasc Imaging Rep 2017; 10: 43. midazolam and fentanyl. BMC Vet Res 2014; 10: 310. 15. Zeidan Z, Erbel R, Barkhausen J, et al. Analysis of global 22. Nacif MS, Zavodni A, Kawel N, et al. Cardiac magnetic systolic and diastolic left ventricular performance using resonance imaging and its electrocardiographs (ECG): volume–time curves by real-time three-dimensional echo- tips and tricks. Int J Cardiovasc Imaging 2012; 28: cardiography. J Am Soc Echocardiogr 2003; 16: 29–37. 1465–1475. 16. Stein AB, Tiwari S, Thomas P, et al. Effects of anesthesia on echocardiographic assessment of left ventricular

Re´sume´ La recherche cardiovasculaire prećlinique repre´sente le fondement de notre compre´hension et de notre connaissance geńe´rale de la fonction cardiaque et des maladies cardiovasculaires. Des modalite´s d’imagerie cardiaque fiables constituent la base nećessaire a` l’obtention de re´sultats applicables. L’imagerie cardiaque par re´sonance magne´tique a` quatre dimensions (IRMc 4D) a e´te´ de´finie comme la norme pour e´valuer la fonction ventriculaire in vivo chez les rongeurs. Cependant, les ame´liorations techniques survenues en ećhocardiographie nous permettent de´sormais d’obtenir des images du cur entier, ce qui fait de l’ećhocar- diographie en 4 dimensions (E4D) une alternative possible a` l’IRMc 4D. A` ce jour, aucune e´tude n’a e´value´ l’E4D de façon syste´matique par rapport a` l’IRMc 4D chez le rat. Au total, nous avons e´tudie´ 26 rats Sprague- Dawley (Crl : CD (SD) IGS). Les rats ont subi vingt examens d’imagerie ećhocardiographique (2D et 4D) et un IRMc 4D. Cinq de ces rats avaient subi une ligature de la veine cave infe´rieure et supe´rieure visant a` re´duire le de´bit cardiaque comme mode`le de maladie. Six autres rats ont e´te´ utilise´s pour e´valuer la reproductibilite´ de l’ećhocardiographie et ont subi trois examens ećhocardiographiques. L’IRMc 4D a e´te´ reáliseé sur un scanner 7T; les ećhocardiographies 2D et 4D ont quant a` elles e´te´ reáliseés avec un transducteur de 40 MHz. La corre´lation entre l’IRMc 4D, l’E4D et l’E2D concernant la fraction d’e´jection ventriculaire gauche (FEVG) a e´te´ e´valueé. Une excellente corre´lation a e´te´ observeé entre et l’IRMc 4D (r ¼ 0.95, p < 0.001). La corre´lation de l’IRMc 4D et l’E2D e´tait faible (r ¼ 0.57, p < 0.01). L’E4D fournit des re´sultats qui sont tout aussi prećis que l’IRMc 4D et tre`s reproductibles tout en impliquant moins d’efforts techniques qu’avec l’IRMc 4D.

Abstract Die pra¨klinische Herz-Kreislauf-Forschung ist die Grundlage unseres Versta¨ndnisses und unseres gesamten Wissens u¨ber Herzfunktion und Herz-Kreislauf-Erkrankungen. Zuverla¨ssige kardiale Bildgebungsverfahren sind die Basis fu¨r anwendbare Ergebnisse. Die vierdimensionale kardiale Magnetresonanz (4D CMR) gilt als Goldstandard fu¨r die In-vivo-Bewertung der ventrikula¨ren Funktion bei Nagetieren. Die technischen Fortschritte in der Echokardiographie erlauben es uns jedoch nun, das ganze Herz abzubilden, was die vierdimensionale Echokardiographie (4DE) zu einer mo¨glichen Alternative zu 4D CMR macht. Bisher gibt es keine systematischen Bewertungsstudien u¨ber einen Vergleich von 4DE und 4D CMR bei Ratten. Wir haben insgesamt 26 junge Sprague-Dawley Ratten (Crl: CD (SD) IGS) untersucht. Zwanzig Ratten wurden echokardiographischen Aufnahmen (2D und 4D) und 4D CMR unterzogen. Fu¨nf dieser Ratten wurden einer Ligatur der Stegmann et al. 179 oberen und unteren Hohlvene unterzogen, um den kardialen Zufluss als Krankheitsmodell zu reduzieren. Sechs weitere Ratten wurden zur Beurteilung der Reproduzierbarkeit der Echokardiographie eingesetzt und drei echokardiographischen Untersuchungen unterzogen. 4D CMR wurde mit einem 7T-Scanner durchge- fu¨hrt, 2D- und 4D-Echokardiographie mit einem 40 MHz-Wandler. Die Korrelation zwischen 4D CMR, 4DE und 2DE fu¨r die linksventrikula¨re Ejektionsfraktion (LVEF) wurde untersucht. Es wurde eine ausgezeichnete Korrelation zwischen 4DE und 4D CMR beobachtet (r ¼ 0.95, p < 0.001). Die Korrelation von 2DE und 4D CMR war schwach (r ¼ 0.57, p < 0.01). 4DE liefert Ergebnisse, die genauso pra¨zise wie 4D CMR sind und mit weniger technischem Aufwand reproduzierbar sind als 4D CMR.

Resumen La investigacioń cardiovascular preclıńica es la base de nuestra comprensioń y amplio conocimiento de la funcioń cardıáca y las enfermedades cardiovasculares. Las modalidades fiables de escańeres cardıácos son la base de los resultados aplicables. La resonancia magne´tica cardıáca tetradimensional (4D CMR) se ha establecido como el patroń de oro para la evaluacioń in vivo de la funcioń ventricular en roedores. Sin embargo, las mejoras tećnicas en la ecocardiografıá ahora nos permiten obtener ima´genes de todo el cor- azoń, lo que hace que la ecocardiografıá tetradimensional (4DE) sea una alternativa posible a 4D CMR. Hasta la fecha, ninguń estudio ha evaluado sistema´ticamente 4DE en comparacioń con 4D CMR en ratas. En total, se estudiaron 26 ratas Sprague-Dawley jo´venes (Crl: CD (SD) IGS). Veinte ratas se sometieron a escańeres ecocardiogra´ficos (2D y 4D) y 4D CMR. Cinco de esas ratas se sometieron a una ligadura de la vena cava superior e inferior para reducir el flujo cardıáco como un modelo de enfermedad. Seis ratas adicionales se utilizaron para evaluar la reproducibilidad de la ecocardiografıá y se sometieron a tres exa´menes ecocardiogra´ficos. Se realizo´ una 4D CMR en un escańer 7T; La ecocardiografıá 2D y 4D se realizo´ con un transductor de 40 MHz. Se evaluo´ la correlacioń entre 4D CMR, 4DE y 2DE para la fraccioń de eyeccioń del ventrıćulo izquierdo (FEVI). Se observo´ una excelente correlacioń entre 4DE y 4D CMR (r ¼ 0.95, p < 0.001). La correla- cioń de 2DE y 4D CMR fue de´bil (r ¼ 0.57, p < 0.01). 4DE proporciona resultados que son igualmente precisos como 4D CMR y altamente reproducibles con menos esfuerzo tećnico que 4D CMR. Original Article Laboratory Animals 2019, Vol. 53(2) 180–189 ! The Author(s) 2018 The effects of sucrose on urine Article reuse guidelines: sagepub.com/journals- collection in metabolic cages permissions DOI: 10.1177/0023677218781674 journals.sagepub.com/home/lan Alexandra Kovalcˇ´ıkova´1, Marianna Gyura´szova´1, Roman Gardlı´k1,2, Martin Borisˇ3, Peter Celec1,2,4 and Lˇubomı´ra To´thova´1,5

Abstract Representative urine collection that respects the standards of animal welfare is still an issue in experimental nephrology. The commonly used metabolic cages induce stress in rodents. In mice, the volume of collected urine is sometimes insufficient for further analysis. The aim of this experiment was to analyse the effects of time of day, temperature and 2%, 5% or 10% sucrose solutions on diuresis, weight change and liquid intake of adult mice placed in metabolic cages for urine collection. Mice were placed in metabolic cages for 12 h during the day or night at standard ambient (22C) and thermoneutral (28C) temperatures. To determine the effect of acclimatisation, mice were placed in metabolic cages for five consecutive days. Diuresis increased with concentrations of sucrose. Body weight reduction was most rapid in the group given tap water and decreased with increasing sucrose concentrations. A drastic drop in body weight was observed in mice placed in metabolic cages for four consecutive days with access to tap water and food, indicating that time spent in metabolic cages should be kept to a minimum, as prolonged confinement in metabolic cages can be harmful to mice. The administration of concentrated sucrose solutions can potentially aid in mouse urine collection by reducing the time spent in metabolic cages. Sucrose supplementation increased the albumin/creatinine ratio. However, without showing estimates of glomerular filtration rate, renal haemodynamics, plasma electrolytes and urinary electrolyte excretions, the results of this study do not provide any conclusion about the effect of sucrose on renal function.

Keywords metabolic cages, renal function, urine collection, sucrose, diuresis

Date received: 26 December 2017; accepted: 14 May 2018

Introduction 1Institute of Molecular Biomedicine, Faculty of Medicine, Comenius University, Bratislava, Slovakia In experimental nephrology, urine collection is a cru- 2Institute of Pathophysiology, Faculty of Medicine, Comenius cial and irreplaceable tool for assessing renal function. University, Bratislava, Slovakia Collecting spot urine is difficult, even impossible, in 3Institute of Electronics and Photonics, Faculty of Electrical some experimental settings, and/or insufficient in terms Engineering and Information Technology, Slovak University of of volume for further qualitative or quantitative ana- Technology, Bratislava, Slovakia 4Department of Molecular Biology, Faculty of Natural Sciences, lyses. Housing mice in metabolic cages is a widespread 1 Comenius University, Bratislava, Slovakia and commonly used method for urine collection. The 5Institute of Physiology, Faculty of Medicine, Comenius University, use of metabolic cages is, however, problematic and can Bratislava, Slovakia stress experimental animals. In mice, as well as in rats, solitary housing in a new environment introduces so- Corresponding author: Peter Celec, Institute of Molecular Biomedicine, Faculty of called isolation stress. These cages do not contain any Medicine, Comenius University, Sasinkova 4, 811 08 Bratislava, enrichment, and the mice must move on wire mesh with- Slovakia. out any nesting material. This discomfort might affect Email: [email protected] Kovalcˇ´ıkova´ et al. 181 physiological variables, such as glucose metabolism, sucrose solutions on diuresis, weight change and liquid reactivity to stressors, animal behaviour, as well as the intake of adult mice placed in metabolic cages for urine production of reactive oxygen species, leading to oxida- collection. Pilot experiments testing time of day and tive stress.2–6 ambient temperature were performed at the same time Ideally, the urine collected from experimental ani- to make the protocol more practical and animal-friendly. mals should be of sufficient volume, free of contamin- The purpose of these preliminary experiments was to ation, collected within the shortest time possible, and determine the effect of time of day and temperature on obtained without pain or distress. Several factors are urine output in order to minimise time spent in meta- known to influence urine production, including metabolic cages. bolic activity, liquid composition, temperature, and time spent in metabolic cages. Stechman et al. showed that weight, diuresis, food, and water intake were sta- Methods 7 bilised after spending 3–4 days in metabolic cages. On Animals the other hand, Kalliokoski et al. showed that mice do not adapt to metabolic cages.8 Indicators of distress The investigation comprised 20 male and 20 female such as activation of the hypothalamic–pituitary–adre- mice aged 15 months., A strain of mice that is com- nal axis, oxidative stress, and increased overall metab- monly used for the induction of kidney diseases and, olism persist even after 3 weeks in a metabolic cage.8 In thus, for monitoring metabolic and renal markers using light of these contradictory results, a better understand- urine collection in metabolic cages, 129Sv4 (Anlab, ing of murine response to housing in metabolic cages is Prague Czech Republic), was used.14,15 In addition, needed for optimisation of the protocol for urine col- this strain is typically used for the generation of gene- lection while limiting stress on the animal. targeted mice.16 The age of 15 months was chosen to Furthermore, metabolic activity can influence urine mimic the onset and pathology of human metabolic production. Metabolic activity is affected by internal disorders and chronic kidney disease, both of which as well as external factors, such as day/night cycle increase in frequency with age. The total number of and ambient temperature. However, these factors are animals required for the main experiment was deter- often not considered in studies where metabolic cages mined according to G power analysis using G power are used for urine collection and analysis. software (G-Power 3.1.9.2., Kiel, Germany) from the Several studies have shown that the administration of preliminary results of our previous experiments a 5% sucrose solution results in a multiple-fold increase (F3,32 ¼ 2.90, effect size f ¼ 0.75, number of groups ¼ 4, in urine output compared to water administration.9–11 In total samples size ¼ 36). Animals were housed in groups addition, in these studies, short-term sucrose administra- of five in standard type III cages (Trigon plus, Czech tion had no effect on body weight and serum glucose Republic) and were placed in one room. Mice had concentrations. Thus, replacing drinking water with a wood bedding and a 12/12 h light/dark cycle at an sucrose solution could solve some of the issues surround- ambient temperature of 22C and a humidity of 40– ing urine collection, such as obtaining sufficient amounts 50%. All animals had free access to tap water and a of urine and minimising the distress of the animals. commercial pellet diet (Standard 19% protein, 4% fat However, this might be associated with changes in rodent diet, Velaz, Czech Republic). All experiments urine content, including markers of renal function. were performed in the same room and by the same Arikawe et al. showed that rats fed a diet containing researchers in order to standardise the effects of sur- 25% or 50% sucrose for 12 weeks had increased rounding factors such as human contact, smell, and plasma and urinary creatinine and urea.12 On the other fluctuations of light, humidity and temperature. All hand, Kobayashi et al. showed that the administration but experimental conditions were kept equal for all of sucrose at a dose of 1.5 g/kg to healthy males did not groups of mice. The experiment was approved by the affect creatinine clearance.13 The same study showed ethical committee of the Institute of Molecular that sucrose enhances purine degradation, which results Biomedicine, Faculty of Medicine, Comenius in an increased plasma concentration of uric acid, while University, Slovakia and complies with the animal wel- urinary excretion remains unaffected.13 However, fare and ethical guidelines of the European Convention whether the short-term administration of sucrose solu- for the Protection of Vertebrate Animals used for tions instead of water in metabolic cages affects markers Experimental and Other Scientific Purposes. of renal function is unknown. To our knowledge, a systematic analysis of the effects Experimental design of sucrose on urinary output and proteinuria in mice housed in metabolic cages has not yet been published. The same 40 mice were used in all experiments. The The aim of this experiment was to analyse the effect of order of the experiments is as follows: (1) effect of 182 Laboratory Animals 53(2) light/dark cycle; (2) effect of temperature; (3) effect of after 12 h in the metabolic cages. Liquid intake and sucrose; and (4) effect of acclimatisation. The allocation diuresis were measured after 12 h in the metabolic of mice to specific groups was randomised, and a recov- cages, and urine samples were stored at 20C for sub- ery period of 7 days was used between experiments. sequent analysis. The first session included male mice only, and the last session included only half of the mice of both sexes. Effect of acclimatisation Thus, the total number of housings of a single mouse in a metabolic cage was 3 or 4 for female mice and 5 or Determination of the effect of acclimatisation and a 6 for male mice. After the completion of all planned reduced-stress environment on diuresis, water intake, experiments, the mice were anaesthetised using keta- and weight reduction was performed in accordance mine and xylazine (100 mg/kg and 10 mg/kg of the with Stechman et al.,7 with several minor differences bodyweight, in ratio 3:1) and then humanely killed by including the total time spent in the cages (5 instead cardiac puncture and exsanguination. Samples of liver, of 7 days), the age of mice (15 month instead of 6–8 kidney, brain and heart were taken, snap-frozen in month old mice), food availability (4 out of 5 days liquid nitrogen and kept for further analysis. instead of full), and cage enrichments (missing instead of present). Mice of both sexes were individually placed Effect of light/dark cycle in metabolic cages for 5 days (n ¼ 10 for each sex) after a 7 day recovery period from the previous experiment. Male mice (n ¼ 20) were individually placed in meta- Additionally, mice had free access to water during the bolic cages (Tecniplast, Buguggiate, Italy) with a floor entire experiment. Animal weight, diuresis and food area of 200 cm2. All animals had free access to tap and water intake were measured daily. To obtain water but no access to chow. The mice were housed urine without contamination, food was removed after in metabolic cages for 12 h during the day or night the fourth day. Urine collected during the fifth day was (n ¼ 10 for each group). Subsequently, the groups stored at 20C for analysis. were given a recovery period of 36 h and then swapped. Diuresis was measured as the volume of urine collected Urine sample analysis in the metabolic cages. The urine collected in the tubes of metabolic cages was measured by pipetting the urine Creatinine was measured using the Jaffe method.17 from the collection tubes using micropipettes. Body This method is based on the reaction of creatinine weight loss during the 12 h spent in the metabolic with picric acid under alkaline conditions, which results cages was also measured. in a yellow-orange complex. Absorbance was measured at 492 nm using the Olympus AU400 analyser Effect of temperature (Beckman Coulter, California, USA). The concentration of proteins in urine was measured using the pyrogallol Mice of both sexes (n ¼ 40) were individually placed in red-molybdate method, which is characterised by the the metabolic cages for 12 h during the day with free binding of proteins on pyrogallol red in the presence of access to tap water, no access to chow, and at a tem- molybdate.18 The coloured complex was read at 592 nm perature of either 22Cor28C. All mice were used in using COBAS 6000 (Roche, Basel, Switzerland). In both turn for both temperatures (n ¼ 10 for each turn) with a methods, 10 ml of the samples was used and diluted five- recovery period of 36 h between turns. Diuresis and fold with distilled water. The albumin/creatinine ratio body weight loss were assessed. expressed in mg/mmol was calculated from urinary protein and creatinine concentrations. Effect of sucrose Statistical analysis To analyse the effect of sucrose on diuresis, concentrations of 2%, 5% and 10% sucrose in tap water were GraphPad Prism 5.0 (GraphPad Software, San Diego, provided as a drinking liquid, with tap water used as a California, USA) was used for statistical analysis. To control. With a 7-day recovery period after the previous evaluate the effect of the dark/light cycle on diuresis experiment, mice of both sexes were individually placed and body weight, a paired t-test was used for compari- in metabolic cages for 12 h (5 male and 5 female mice son. To analyse the effect of temperature in both sexes for the water group, 5 male and 5 female mice for each on diuresis and body weight, two-way analysis of vari- of the three different sucrose concentration groups, ance (ANOVA) was used with temperature and sex as total n ¼ 40 mice, 20 male and 20 female mice). The the tested factors. To assess the effect of acclimatisation mice had free access to the corresponding liquid and and different concentrations of sucrose in drinking no access to chow. Animals were weighed before and water on diuresis, liquid intake, and other parameters, Kovalcˇ´ıkova´ et al. 183 repeated measures (RM) one-way ANOVA with sub- r2 ¼ 0.88, p < 0.001 for females; r ¼ 0.95, r2 ¼ 0.91, sequent Bonferroni post-hoc tests and Pearson’s correl- p < 0.001 for males), diuresis (r ¼ 0.95, r2 ¼ 0.90, ation analysis was performed. The data sets of urine p < 0.001 for females; r ¼ 0.78, r2 ¼ 0.61, p < 0.001 for output were log-transformed for the analysis. The males) and albumin/creatinine ratio (r ¼ 0.76, r2 ¼ 0.58, results are reported as the mean standard deviation; p < 0.001 for females; r ¼ 0.41, r2 ¼ 0.16, p < 0.05 for p-values below 0.05 were considered statistically males). These correlations were found in both sexes significant. (Figures 2 and 3). In the acclimatisation experiment, a significant Results increase in urine output was observed during the monitored 5 day period in males (F4,55 ¼ 8.61, p < 0.001, In a preliminary experiment, male mice were housed in Figure 4(d)) but not in females (F4,53 ¼ 1.07, p ¼ ns, metabolic cages for 12 h during the day or during the Figure 5(d)). Similarly, liquid intake was only night (Figure 1). The mean value of diuresis during the increased in male mice (in males F4,44 ¼ 34.4, day was 55 ml (SD ¼ 84), whereas the mean value of p < 0.001, Figure 4(c); in females F4,44 ¼ 0.72, p ¼ ns, diuresis during the night was 90 ml (SD ¼ 125). The dif- Figure 5(c)). Food intake was highest on the fourth ference was not found to be significant (t18 ¼ 1.19, day in both sexes (mean ¼ 3.19, SD ¼ 0.65 g per 24 h p ¼ 0.25). The mice lost 1.8 g (SD ¼ 0.27) of their in males, F3,33 ¼ 101, p < 0.001, Figure 4(b); mean ¼ body weight in the group that spent 12 h in metabolic 2.19, SD ¼ 0.75 g per 24 h in females, F3,33 ¼ 8.09, cages during the day and 2.6 g (SD ¼ 0.49) in the group p < 0.01, Figure 5(b)). Before the fourth day, a drastic that spent 12 h in the cages at night (t19 ¼ 6.35, loss in body weight was observed (in males F5,55 ¼ p < 0.001). Afterwards, mice of both sexes were 156.7, p < 0.001, Figure 4(a); in females F5,55 ¼ 97.0, housed in metabolic cages during daylight hours at a p < 0.001, Figure 5(a)). Body weight loss reached a standard ambient temperature of 22C or at a thermo- peak on the third day (mean ¼ 5.55, SD ¼ 1.25 g loss neutral temperature of 28C. No differences in body in males and mean ¼ 3.48, SD ¼ 0.92 g loss in females weight loss (F1,22 ¼ 3.70, p ¼ 0.08) or collected urine compared to baseline) and stabilised on the fourth day, volume (F1,22 ¼ 2.58, p ¼ 0.12) were observed between but it did not reach the baseline values. these temperatures. However, a significant gender effect of temperature on diuresis was found (F1,22 ¼ 17.37, Discussion p < 0.001). Thus, further experiments were conducted during the day and at the standard temperature This study focused on some of the shortcomings of the of 22C. widely established method of urine collection in mice In the experiments with sucrose solutions, a signifi- using metabolic cages, while also pointing to possibili- cant association between sucrose intake and measured ties for its optimisation. We found no effect of the day/ physiological parameters was found. Pearson’s correl- night cycle or increased ambient temperature to ther- ation analysis showed a negative correlation of sucrose moneutrality on diuresis. Studies suggest that there is a intake with body weight (r ¼0.42, r2 ¼ 0.18, p < 0.01 circadian rhythm of diuresis in rodents, with urine for females; r ¼0.56, r2 ¼ 0.32, p < 0.001 for males) output being increased during the active phase of the and a positive correlation with liquid intake (r ¼ 0.99, animal.19 The effect of the day/night cycle was tested on male mice only, as gender differences were not expected in this rhythm. On the other hand, the fact that only male mice were used in the first round might represent a limitation of the study, since the male mice were exposed to a higher number of experimental sessions in total. The apparent absence of a circadian rhythm could be attributed to the fact that the passive phase of laboratory rodents is sometimes disrupted (by the animal house staff, who can make noise in the corridors or enter their housing area briefly). This could lead to increased daytime activity and diuresis. Mice are nocturnal animals, and the thermoneutral temperature of 28C has been suggested as being the ideal temperature for testing metabolism in mice.20 A decreased loss of body weight was observed in mice Figure 1. Effect of day and night on diuresis. Using paired housed in metabolic cages during the day. This could be t-test no significant differences were found (p ¼ 0.25). explained by the fact that mice are active during the 184 Laboratory Animals 53(2)

Figure 2. Linear regression showing the correlation of sucrose and (a) bodyweight reduction, (b) liquid intake, (c) diuresis and (d) albumin/creatinine ratio (ACR) in males.

Figure 3. Linear regression showing the correlation of sucrose and (a) bodyweight reduction, (b) liquid intake, (c) diuresis and (d) albumin/creatinine ratio (ACR) in females. Kovalcˇ´ıkova´ et al. 185

Figure 4. Effect of acclimatisation on males: (a) bodyweight, (b) food intake, (c) water intake and (d) diuresis during 5 days in metabolic cages. All animals had free access to food for 4 days. On the fifth day food was removed to collect uncontaminated urine. Results are expressed as mean þ SD. *** denotes p < 0.001; ** denotes p < 0.01 in comparison to day 0 or 1, respectively. night and probably burn fewer calories during the day diuresis were lowest in the group that received only while sleeping.21 water and gradually and significantly increased with Several studies show that mice are under constant the increasing concentrations of sucrose. Body weight cold stress when housed at 22C and that this tempera- reduction was the most rapid in the group with tap ture could lead to bias due to increased metabolism. water and gradually decreased with increasing sucrose However, mice might thrive at 22C if nesting material concentration. We assume that the decrease in body is provided or if they are able to behaviourally thermo- weight reduction with increasing concentrations of regulate by huddling together with other mice.22–25 sucrose is not attributed to a different water intake/ Diuresis was assessed as the volume of collected diuresis ratio, but rather likely to the effect of increased urine, and it was similar at both tested temperatures. caloric feeding due to the greater sucrose consumption. Since the collected urine was not chilled or collected However, other potentially explanatory parameters under mineral oil during the collection in metabolic such as specific caloric contribution and plasma osmo- cages, an evaporation effect cannot be ruled out; thus, lality were not measured in this study. the results could be biased. From a practical point of The same mice were used in all of the described view, however, it was determined that urine collection experiments. This could have influenced the behaviour in metabolic cages should continue during daylight and of the animals in the metabolic cages. In addition, male at the standard ambient temperature of 22C. mice might have been more profoundly affected In the main part of the study, different concentra- because they were housed more in metabolic cages tions of sucrose solutions were compared to tap water than the female mice. To minimise this bias, recovery as the only liquid provided for drinking in the periods were introduced between the experiments. This metabolic cages. Positive correlations between sucrose way of testing could have led to a slower acclimatisa- intake, liquid intake and diuresis in both genders tion to the stressful environment of a metabolic cage, were found. Moreover, sucrose intake was negatively and the reduced stress could have, in turn, affected the correlated with body weight. The results of Pearson’s urinary output of the animals. However, the groups correlation analysis indicate that liquid intake and that received tap water were exposed to the same 186 Laboratory Animals 53(2)

Figure 5. Effect of acclimatisation on females: (a) bodyweight, (b) food intake, (c) water intake and (d) diuresis during 5 days in metabolic cages. All animals had free access to food for 4 days. On the fifth day food was removed to collect uncontaminated urine. Results are expressed as mean þ SD. *** denotes p < 0.001; ** denotes p < 0.01 in comparison to day 0 or 1, respectively. conditions, and urinary output did not increase signifi- have not yet been studied. Depending on the research cantly with time in those groups. questions being investigated, this slight effect on albu- After measuring urinary albumin and creatinine min/creatinine ratio should be taken into consideration and calculating their ratio, a positive correlation when considering sucrose administration as a way to between sucrose intake and albumin/creatinine ratio decrease the time the animals are housed in metabolic was found. Linear regression analysis showed a slight cages. increase in albumin/creatinine ratios with increasing Next, the experiment conducted previously by concentrations of sucrose, indicating a dose response. Stechman et al. was reproduced with subtle differences However, it remains unknown whether this is a relevant in methodology.7 Mice were acclimatised to housing in effect. A 1% sucrose (hypotonic) solution has been metabolic cages for 5 days. An increase in liquid intake shown to inhibit antidiuretic hormones and reduce in males and food intake in both genders was found on urinary ion concentrations and osmolarity in rats.26 the fourth day, similar to the previously reported Inhibition of antidiuretic hormones might result from results. On the last and fifth day, diuresis increased in a slight reduction in plasma osmolarity or increased both genders with the removal of food. However, blood volume. On the other hand, an isotonic solution throughout the 5 days in the metabolic cages, a consid- resulted in urinary salt and osmotic concentrations erable body weight decrease was observed in both sexes higher than the control. It is not clear how changes in (mean 27% with SD 3.18 for males and mean 21% albuminuria in healthy mice in response to sucrose with SD 2.89 for females). The welfare of the mice administration relate to the pathophysiology of albu- was clearly affected. The harmful effects of stress in minuria in disease models. The mechanisms of the effect metabolic cages are well described.8,28 Some studies of sucrose intake on albuminuria are largely unknown, have shown that mice do not habituate to this kind of with little data available from the literature. Kubota confinement and even claim that the environment does et al. found an increased albumin/creatinine ratio in not meet the common standards of animal welfare. rats on a high-sucrose diet.27 Overall, changes in the The results of this study indicate that the physiological albumin/creatinine ratio induced by sucrose solutions variables affected by metabolic cages were partially Kovalcˇ´ıkova´ et al. 187 stabilised after 3 days of housing (their weight loss sta- electrolytes and urinary electrolyte excretions, the bilised, food and water intake increased, diuresis results of this study do not provide any conclusion increased), even if stress persisted, as highlighted by about the effect of sucrose on renal functions. the fact that mice did not reach their original weight. Moreover, the possibility of using spot urine normal- We propose that it is preferable to house mice in meta- ised by creatinine is worth considering, since it has been bolic cages for short periods because of the stress and previously used with success.29 Further studies should significant weight loss present during the first 3 days, also be aimed at shortening the time spent in metabolic which might affect other physiological parameters. cages with sucrose solutions in animal models of renal Although we did not follow any other stress-related injury and at discovering alternatives to sucrose in the parameters other than weight changes, significant stimulation urine production in experiments where weight reduction was present throughout the 5-day sucrose is not suitable. period in metabolic cages. A standard period for the collection of urine for the assessment of renal param- Acknowledgements eters is 24 h. However, this study has shown a signifi- The authors are grateful to Teri Hreha, PhD (Washington cant weight loss in mice during the initial 24 h in University in St. Louis School of Medicine, St. Louis, MO, metabolic cages (more than 10% weight loss, which is USA) for language editing. All procedures performed in stu- considered severe). This indicates that all concurrent dies involving animals were in accordance with the ethical and subsequent metabolic analyses might be affected standards of the institution or practice at which the studies by this obvious discomfort. Thus, we consider the ini- were conducted. This article does not contain any studies with tial 24 h housing a factor that should be addressed in human participants performed by any of the authors. order to minimise the adverse outcomes in metabolic experiments while preserving the animal’s wellbeing. Declaration of Conflicting Interests In summary, this study found no differences in urin- The author(s) declared no potential conflicts of interest with ary output between mice housed in metabolic cages respect to the research, authorship, and/or publication of this during the day or during the night and between mice article. housed at standard or thermoneutral temperatures. Sucrose in drinking water resulted in increased liquid Funding intake and diuresis in both genders. Acclimatisation only increased liquid intake and diuresis in male mice. The author(s) disclosed receipt of the following financial sup- However, housing in metabolic cages resulted in a port for the research, authorship, and/or publication of this dramatic body weight loss in both genders even after article: This work is the result of the project implementation (ITMS 26240120027) supported by the OPRaD funded by the 1 day. Weight loss continued in animals until the third ERDF. day, when it partially stabilised, but did not reach basal values (Figures 4(a) and 5(a)). Increased sucrose concentrations partially prevented weight loss, at least for ORCID iD 12 h (Figures 2a and 3a), but the slightly higher urinary Peter Celec http://orcid.org/0000-0001-5883-3580 albumin/creatinine ratio in mice drinking sucrose solutions should be taken into account. However, this does References not prevent the use of sucrose solutions, especially if all 1. Kurien BT, Everds NE and Scofield RH. Experimental groups in an experiment are housed the same way. animal urine collection: a review. Lab Anim 2004; 38: Furthermore, if one takes into consideration the stress 333–361. and harm imposed on the animals by housing them in 2. Bartolomucci A, Palanza P, Sacerdote P, et al. Individual metabolic cages, the administration of a sucrose solu- housing induces altered immuno-endocrine responses to psy- tion seems even more reasonable. chological stress in male mice. Psychoneuroendocrinology It is expected that sucrose will induce osmotic diur- 2003; 28: 540–558. esis and thus can change renal haemodynamics and 3. Ishida H, Mitsui K, Nukaya H, et al. Study of active sub- tubular electrolyte handling. In order to confirm the stances involved in skin dysfunction induced by crowding usefulness of sucrose administration as a tool to stress. I. Effect of crowding and isolation on some physiological variables, skin function and skin blood perfusion in decrease the time spent in metabolic changes, it is neces- hairless mice. Biol Pharm Bull 2003; 26: 170–181. sary to perform additional measurements, including 4. Manser CE, Morris TH and Broom DM. An investigation systemic and renal haemodynamic parameters, such into the effects of solid or grid cage flooring on the welfare as glomerular filtration rate and urinary electrolyte of laboratory rats. Lab Anim 1995; 29: 353–363. excretion, and the histological evaluation of the 5. Manser CE, Elliot H, Morris TH, et al. The use of a novel kidney. Therefore, without showing estimates of glom- operant test to determine the strength of preference for erular filtration rate, renal haemodynamics, plasma flooring in laboratory rats. Lab Anim 1996; 30: 1–6. 188 Laboratory Animals 53(2)

6. Zhu S-W, Yee BK, Nyffeler M, et al. Influence of differ- serum and urine creatinine measured from microsamples ential housing on emotional behaviour and neurotrophin by liquid chromatography mass spectrometery (LC/MS) levels in mice. Behav Brain Res 2006; 169: 10–20. vs. Jaffe. J Clin Lab Anal 2014; 28: 405–408. 7. Stechman MJ, Ahmad BN, Loh NY, et al. Establishing 18. Celec P, Hodosy J, Gardlı´k R, et al. The effects of anti- normal plasma and 24-hour urinary biochemistry ranges inflammatory and anti-angiogenic DNA vaccination on in C3H, BALB/c and C57BL/6J mice following acclima- diabetic nephropathy in rats. Hum Gene Ther 2012; 23: tisation in metabolic cages. Lab Anim 2010; 44: 218–225. 158–166. 8. Kalliokoski O, Jacobsen KR, Darusman HS, et al. Mice 19. Noh JY, Han DH, Yoon JA, et al. Circadian rhythms in do not habituate to metabolism cage housing-a three urinary functions: Possible roles of circadian clocks? Int week study of male BALB/c mice. PLoS One 2013; 8: Neurourol J 2011; 15: 64–73. e58460. 20. Lodhi IJ and Semenkovich CF. Why we should put 9. Longhurst PA, Wein AJ and Levin RM. In vivo urinary clothes on mice. Cell Metab 2009; 9: 111–112. bladder function in rats following prolonged diabetic and 21. Abreu-Vieira G, Xiao C, Gavrilova O, et al. Integration non-diabetic diuresis. Neurourol Urodyn 1990; 9: of body temperature into the analysis of energy expend- 171–178. iture in the mouse. Mol Metab 2015; 4: 461–470. 10. Eika B, Levin RM and Longhurst PA. Comparison of 22. Gordon CJ. Thermal physiology of laboratory mice: urinary bladder function in rats with hereditary diabetes Defining thermoneutrality. J Therm Biol 2012; 37: insipidus, streptozotocin-induced diabetes mellitus, and 654–685. nondiabetic osmotic diuresis. J Urol 1994; 151: 496–502. 23. Karp CL. Unstressing intemperate models: how cold 11. Ohnishi N, Horan P, Levin SS, et al. Sucrose diuresis stress undermines mouse modeling. J Exp Med 2012; protects rat bladder from outlet partial obstruction- 209: 1069–1074. induced contractile dysfunction. Urology 1999; 54: 24. Overton JM. Phenotyping small animals as models for 183–187. the human metabolic syndrome: thermoneutrality mat- 12. Arikawe AP, Udenze IC, Akinwolere MF, et al. Effects ters. Int J Obes 2010; 34(Suppl 2): S53–S58. of streptozotocin, fructose and sucrose-induced insulin 25. Kokolus KM, Spangler HM, Povinelli BJ, et al. Stressful resistance on plasma and urinary electrolytes in male presentations: mild cold stress in laboratory mice influ- Sprague-Dawley rats. Nig Q J Hosp Med 2012; 22: ences phenotype of dendritic cells in naı¨ve and tumor- 224–230. bearing mice. Front Immunol 2014; 5: 23. 13. Kobayashi T, Inokuchi T, Yamamoto A, et al. Effects of 26. Walker RL and Olson ME. Renal function in the labora- sucrose on plasma concentrations and urinary excretion tory rat: a student exercise. Am J Physiol 1995; 268: of purine bases. Metabolism 2007; 56: 439–443. S49–S55. 14. Ma LJ and Fogo AB. Model of robust induction of glo- 27. Kubota M, Watanabe R, Kabasawa H, et al. Rice protein merulosclerosis in mice: Importance of genetic back- ameliorates the progression of diabetic nephropathy in ground. Kidney Int 2003; 64: 350–355. Goto–Kakizaki rats with high-sucrose feeding. Br J 15. Lu X, Li N, Shushakova N, et al. C57BL/6 and 129/Sv Nutr 2013; 110: 1211–1219. mice: genetic difference to renal ischemia-reperfusion. 28. Hoppe CC, Moritz KM, Fitzgerald SM, et al. Transient J Nephrol 2012; 25: 738–743. hypertension and sustained tachycardia in mice housed 16. Eisener-Dorman AF, Lawrence DA and Bolivar VJ. individually in metabolism cages. Physiol Res 2009; 58: Cautionary insights on knockout mouse studies: the 69–75. gene or not the gene? Brain Behav Immun 2009; 23: 29. Lorenz JN, Baird NR, Judd LM, et al. Impaired renal 318–324. NaCl absorption in mice lacking the ROMK potassium 17. Askenazi DJ, Moore JF, Fineberg N, et al. Comparison channel, a model for type II Bartter’s syndrome. J Biol of methods, storage conditions, and time to analysis of Chem 2002; 277: 37871–37880.

Re´sume´ Le pre´le`vement repre´sentatif d’urine respectant les normes de bien-eˆtre animal est toujours un proble`me en ne´phrologie expe´rimentale. Les cages me´taboliques commune´ment utiliseés induisent du stress chez les rongeurs. Chez les souris, le volume d’urine pre´leve´ est souvent insuffisant pour permettre une analyse approfondie. Le but de cette expe´rience e´tait d’analyser les effets du moment de la journeé, de la tempe´rature et de solutions de saccharose de 2, 5 ou 10% sur la diure`se, les modifications ponde´rales et la prise de liquide des souris adultes placeés dans des cages me´taboliques pour le pre´le`vement d’urine. Les souris ont e´te´ placeés dans des cages me´taboliques pendant 12 heures au cours de la journeé ou de la nuit a` des tempe´ratures ambiantes standard (22C) et thermoneutres (28C). Pour de´terminer l’effet de l’acclimatation, les souris ont e´te´ placeés dans des cages me´taboliques pendant 5 jours consećutifs. La diure`se augmentait avec les concentrations de saccharose. La re´duction du poids corporel a e´te´ plus rapide dans le groupe recevant l’eau du robinet et diminuait au fur et a` mesure que les concentrations de saccharose augmentaient. Une perte de poids corporel conse´quente a e´te´ observeé chez des souris placeés dans des cages me´taboliques Kovalcˇ´ıkova´ et al. 189 pendant 4 jours consećutifs ayant acce`sa` de l’eau du robinet et de la nourriture, ce qui indique que le temps passe´ dans des cages me´taboliques devrait eˆtre aussi minime que possible, puisqu’un confinement prolonge´ dans les cages me´taboliques semble avoir un effet de´le´te`re sur les souris. L’administration de concentre´ de solutions de saccharose peut potentiellement contribuer au pre´le`vement d’urine chez les souris en re´duisant le temps passe´ dans des cages me´taboliques. La supple´mentation en saccharose a augmente´ le ratio albu- mine/creátinine. Toutefois, sans montrer d’estimations des taux de filtration glome´rulaire, de l’he´modyna- mique reńale, des excre´tions d’e´lectrolytes plasmatiques et d’e´lectrolytes urinaires, les re´sultats de cette e´tude ne fournissent aucune conclusion concernant l’effet du saccharose sur la fonction reńale.

Abstract Eine repra¨sentative Urinsammlung, die den Normen des Tierschutzes entspricht, ist in der experimentellen Nephrologie nach wie vor ein Thema. Die haüfig verwendeten Stoffwechselka¨fige verursachen Stress bei Nagern. Bei Maüsen ist die Menge des gesammelten Urins zuweilen nicht ausreichend fu¨r eine weitere Analyse. Ziel dieses Experiments war es, die Auswirkungen von Tageszeit, Temperatur und 2%, 5% bzw. 10%-ige Saccharoselo¨sungen auf Diurese, Gewichtsvera¨nderung und Flu¨ssigkeitsaufnahme von in Stoffwechselka¨figen zur Urinsammlung untergebrachten erwachsenen Maüsen zu analysieren. Maüse wurden 12 Stunden lang tagsu¨ber oder nachts bei normalen Umgebungstemperaturen (22C) und thermo- neutralen Temperaturen (28C) in Stoffwechselka¨figen gehalten. Um den Effekt der Akklimatisierung zu bestimmen, wurden Maüse 5 Tage hintereinander in Stoffwechselka¨figen untergebracht. Die Diurese stieg mit der Konzentration von Saccharose. Die Ko¨rpergewichtsabnahme war in der mit Leitungswasser versorg- ten Gruppe am schnellsten und nahm mit zunehmender Saccharosekonzentration ab. Ein drastischer Ru¨ckgang des Ko¨rpergewichts wurde bei Maüsen beobachtet, die vier Tage lang in Stoffwechselka¨figen mit Zugang zu Leitungswasser und Nahrung untergebracht waren, was nahelegt, dass die in Stoffwechselka¨figen verbrachte Zeit auf ein Minimum reduziert werden sollte, da ein la¨ngerer Aufenthalt in diesen fu¨rMaüse scha¨dlich sein kann. Die Gabe von konzentrierten Saccharoselo¨sungen kann bei der Urinsammlung von Maüsen helfen, indem die Zeit in den Stoffwechselka¨figen verku¨rzt wird. Die Saccharose-Supplementation erho¨hte den Albumin-Kreatinin-Quotienten. Da jedoch bei dieser Studie Scha¨tzungen der glomerula¨ren Filtrationsrate, der renalen Ha¨modynamik, der Plasma-Elektrolyte und der Harn-Elektrolyt- Ausscheidungen nicht beru¨cksichtigt wurden, lassen die Ergebnisse keinen Ru¨ckschluss auf die Wirkung von Saccharose auf die Nierenfunktion zu.

Resumen La recogida de orina representativa que respeta los estańdares de bienestar animal todavıá supone un problema en la nefrologıá experimental. El uso habitual de jaulas metabo´licas provoca estre´s en los roedores. En ratones, el volumen de orina recogida es a veces insuficiente para poder realizar ana´lisis. El objetivo de este experimento era analizar los efectos del momento del dıá, la temperatura y un 2%, 5% o 10% de soluciones de sacarosa en diuresis, cambio de peso y toma de lı´quidos de ratones adultos en jaulas metabo´licas para la recogida de orina. Los ratones se colocaron en jaulas durante 12 horas durante el dıáo noche a unas temperaturas estańdar ambiente (22C) y termoneutral (28C). Para determinar el efecto de aclimatacioń, los ratones se colocaron en jaulas metabo´licas durante 5 dıás consecutivos. La diuresis aumento´ con concentraciones de sacarosa. La reduccioń del peso corporal fue ma´sra´pido en el grupo al que se dio agua del grifo y disminuyo´ con mayores concentraciones de sacarosa. Se observo´ una caı´da repentina del peso corporal en los ratones colocados en jaulas metabo´licas durante 4 dıás consecutivos con acceso a agua del grifo y comida, lo que indica que el tiempo pasado en jaulas metabo´licas deberıá mantenerse al mıńimo, ya que una reclusioń prolongada en jaulas metabo´licas puede ser dan˜ino para los ratones. La administracioń de soluciones de sacarosa concentrada puede ayudar potencialmente en la recogida de orina de ratones al reducir el tiempo pasado en jaulas metabo´licas. El suplemento de sacarosa aumento´ la relacioń albu´mina/creatinina. Sin embargo, sin mostrar estimaciones del ratio de filtracioń glomerular, hemodina´mica renal, electrolitos de plasma y excreciones urinarias de electrolitos, los resultados de este estudio no ofrecen ninguna conclusioń sobre el efecto de la sacarosa en la funcioń renal. Original Article Laboratory Animals 2019, Vol. 53(2) 190–201 ! The Author(s) 2018 Impact assessment of tail-vein injection Article reuse guidelines: sagepub.com/journals- in mice using a modified anaesthesia permissions DOI: 10.1177/0023677218786982 induction chamber versus a common journals.sagepub.com/home/lan restrainer without anaesthesia

Marlene Resch1, Tania Neels1, Alexander Tichy2, Rupert Palme3 and Thomas Ru¨licke1

Abstract Intravenous (IV) administration in mice is predominantly performed via the lateral tail veins. The technique requires adequate training before it can be used safely and routinely. A novel anaesthesia induction chamber has been developed to simplify the treatment and to facilitate IV injection in mice, particularly for untrained personnel. We have assessed the benefits of the chamber in refining IV injection in isoflurane-anaesthetized mice in direct comparison with the common restrainer method on conscious animals. The body weight, nesting behaviour and concentrations of faecal corticosterone metabolites were taken as indicative of distress induced by the various procedures. The results suggest that both methods of tail-vein injection induce similar levels of momentary stress in the animals, revealed by a short-term increase in the levels of stress hormone metabolites in faeces. A temporary reduction of body weight was observed after IV injection under isoflurane anaesthesia but not for conscious mice injected in the common restrainer. We conclude that the severity of tail-vein injection in mice is ‘mild’ for both methods. There was no evidence that refining the procedure by using isoflurane anaesthesia in the induction chamber was associated with any benefit.

Keywords intravenous injection, mice, isoflurane anaesthesia, induction chamber, restrainer

Date received: 29 January 2018; accepted: 14 June 2018

Introduction of the vessels, tail-vein injection in mice is frequently carried out in combination with a heat treatment. Lateral tail-vein injection is the preferred method Vasodilatation facilitates the procedure by rendering for single or repeated vascular administration of inject- the tail vein more visible, which is especially helpful able substances in laboratory mice. The successful when dark-pigmented mice are to be injected. Local application of IV injection requires adequate practical warming of the tail by infrared exposure or by immer- skills, only attainable with sufficient practice, and spe- sing the tail in warm water, rather than heating the cific equipment for proper immobilization. Currently, tail-vein injection in mice is commonly performed by 1 using a restrainer. Several types of restrainer are on Institute of Laboratory Animal Science, University of Veterinary Medicine Vienna, Austria the market and have proven suitable to restrict the 2Bioinformatics and Biostatistics Platform, University of Veterinary movement of the animal during the procedure, while Medicine Vienna, Austria leaving the tail freely accessible for injection. A cor- 3Unit of Physiology, Pathophysiology and Experimental rectly sized restrainer prevents the animals from turning Endocrinology, University of Veterinary Medicine Vienna, Austria around but allows normal respiration. Furthermore, Corresponding author: rigid restraint devices should enable constant observa- Thomas Ru¨licke, Veterina¨rmedizinische Universita¨t Wien, tion of the restrained animal to minimize the risk of Veterina¨rplatz 1, 1210 Wien, Austria. 1 injury and unnecessary stress. To achieve dilatation Email: [email protected] Resch et al. 191 entire body, are the preferred techniques to minimize in body weight, concentrations of faecal corticosterone stress due to (over-) heating of the animal.2 metabolites and nest-building behaviour were selected The procedure of tail-vein injection with a standard as reliable indicators to evaluate potential impact of the restrainer is routinely performed on conscious mice: no treatments.10–12 The study was conducted separately sedation or anaesthesia is given. Appropriate restraint with experienced and inexperienced personnel. We of the animal is important to avoid injury to the animal expected that use of the novel induction chamber or the handler. A sudden jerking motion by the animal would reduce at least some of the adverse effects of is frequently observed and is a response to the acute tail-vein injection on the animals’ well-being, as well pain caused by the vein puncture. It may complicate the as it would simplify the injection procedure. procedure as abrupt movement of the animal often dis- lodges the inserted needle. Unsuccessful attempts are not unusual for an untrained person and might result Animals, materials and methods in the repetition and prolongation of the procedure. Animals The severity of the intravenous administration of substances via superficial blood vessels is classified as The study was approved by the institutional ethics com- ‘mild’, provided that the injected substance has no or mittee of the University of Veterinary Medicine Vienna only mild impact on the animal and the injection is and an experimental license was granted under performed in accordance with good veterinary practice BMWFW-68.205/0042-WF/II/3b/2014 (Austrian Federal (Directive 2010/63/EU, Annex VIII, Section III, 1.f). Ministry of Science, Research and Economy). The burden However, routine procedures such as restraint and induced by the procedures was prospectively categorized injection are known to cause an acute stress response as ‘mild’. in laboratory mice.3,4 It is unknown whether restraining All mice were purchased from Janvier Labs (France) and IV injection together have a cumulative impact in and the perpetuation of their specific pathogen free terms of increased severity. The use of anaesthesia (SPF) quality according to FELASA recommendations might have the advantage of avoiding discomfort for was confirmed by a sentinel programme.13 Animals tightly restrained animals while simultaneously pre- were housed in a rodent facility (photoperiod 12L/ venting the defence reaction of the mouse, which has 12D, light on at 08:00 am, temperature: 21.3 1.9C, no sensation of pain from the introduction of the relative humidity: 60.3 9.8%). Commercial mouse needle. A novel anaesthesia induction chamber for diet (V1126, Ssniff, Germany) and bottled tap water tail-vein injection of mice (Rothacher Medical GmbH; were available ad libitum. Cages were lined with bed- Heitenried, Switzerland) has recently been recom- ding material (LignocelÕ, heat treated, Rettenmaier mended which claims to refine IV injections in anaes- KG, Austria) and enriched with nesting material thetized mice. The manufacturer promises a simplified (Pur-Zellin 4 5 cm; Paul Hartmann GmbH, Austria). procedure without the need for pre-heating treatment During the adaptation period of 2 wk the mice were to induce vasodilatation and to reduce pain for the housed in groups of five in transparent polycarbonate animals. cages (MakrolonÕ Type III, Tecniplast, Italy). For short-term procedures, inhalation anaesthesia In Part 1 of the study, we evaluated the discomfort represents the method of choice and provides a rapid induced by tail-vein injection in 10 females and 10 onset of induction and recovery. Isoflurane is the most males of each of the two strains C57BL/6NRj and frequently used inhalation anaesthetic in mice because RjORL:SWISS (aged 8–10 wk). C57BL/6N was of its simple use and rapid induction of anaesthesia and chosen because this inbred strain is frequently used as of its comparatively slight cardiodepressive and hemo- a genetic background of mutant mouse models. It is dynamic effects, ensuring a low risk of complications.5,6 presumed that IV injection in dark pigmented mice is Mice generally recover immediately from inhalation more difficult. In contrast, SWISS mice are larger albin- anaesthesia, making additional post care procedures otic outbred mice and their tail veins are more prom- dispensable after minor interventions. However, anaes- inent. The results of both strains were pooled. For the thesia has to be medically indicated due to the adverse investigation period of five days the animals were indi- physiological effects of anaesthetics, which can impair vidually housed in MakrolonÕ type II cages. Because the animal’s well-being or alter experimental data.7–9 separation of mice will induce a transient increase of We have investigated the severity and duration of the faecal corticosterone metabolites (fCM) concentra- possible pain and distress after tail-vein injection car- tion, a five-day period for adaptation to the individual ried out by the standard restrainer method on con- housing condition was included before starting the scious mice, comparing the results with the pain and experiments.14 distress associated with the novel induction chamber The induction chamber is reported to facilitate and in combination with isoflurane anaesthesia. Changes simplify IV injection. This aspect was separately tested 192 Laboratory Animals 53(2) in Part 2 of the study with 16 female C57BL/6NRj mice (1 cm; 60 120 50 mm) with inlet and outlet adapters (aged 8–10 wk), which were housed in groups of eight for fresh and waste anaesthetic gas. The cap has a small in MakrolonÕ type III cages and only temporary sepa- semilunar hole for fixation of the tail. The chamber was rated for the collection of faeces. We choose C57BL/6 connected with the XGI-8 gas anaesthesia system for this practicability test because this strain is predom- (Xenogen, California, USA) and the Fluovac System inantly used by researchers in our facility as genetic (Harvard Apparatus, UK) to absorb anaesthetic background for genetically engineered mutations. vapours from the work area. To induce anaesthesia, the mouse was placed in the chamber and the isoflurane Restrainer and induction chamber vaporizer was initially set at 3%. When the mouse was anaesthetized, which was first indicated by the cessation A stainless steel restrainer (homemade, diameter 3 cm, of body movement for a few seconds, the chamber cap length 12 cm) with a flexible stopper to secure the was briefly opened to test the surgical tolerance of the mouse was used to fix conscious mice (Figure 1). animal by the hind limb withdrawal reflex and to place Before injection the tail was immersed in a 39–40C the tail in the correct position through the semilunar water bath for about a minute to cause sufficient dila- hole, a special feature of the novel device (Figure 2(a)). tation of the lateral tail veins. This hole in the chamber cap is padded with foamed The novel induction chamber for tail-vein injection plastic and after closing the chamber it will compress of mice is made of a special polymethyl methacrylate the tail to induce stasis of the tail veins. In comparison to the common restrainer, dilatation of the lateral tail veins should be induced by a contra pressure to the foamed plastic, making it unnecessary to warm the tail before injection (Figure 2(b)). After closing the chamber, the setting of the vaporizer was reduced to 2% until the end of the procedure. Oxygen (FiO2 ¼ 1.0) was used as carrier gas at 0.5 l min1. The chamber was placed in a filtration fume hood using activated carbon (A18B/AS Captair Flex M391 2C, Fisher Scientific, Austria) to avoid exposing personnel to anaesthetic vapours that leak from the chamber during opening.

Experimental design

Figure 1. A conscious mouse is prepared for IV injection in Part 1. To compare the degree of discomfort induced the stainless steel restrainer. by tail-vein injection with the novel induction chamber

Figure 2. (a) Induction chamber for anaesthesia. The tail of the anaesthetized mouse is placed in the correct position in the semilunar hole to induce stasis of the tail veins. (b) The anaesthetized mouse is prepared for IV injection. Resch et al. 193 against that caused when the standard restraining method Table 1. Time schedule for experimental procedures is used, animals were randomly assigned to four treat- before and after IV injection. ment groups and a control group (eight animals per Day Time Action Test group, two females and two males per mouse strain): 2 05:00 p.m. Pre-op measurement Body weight (0) Control group, without manipulation: mice were Faeces only handled for weighing and faeces collection. collection (1A) Restrainer group, without IV injection: mice 1 05:00 p.m. Pre-op measurement Nest building were fixed in the restrainer and the tail was 0 09:00 a.m. IV injection immersed in a 39–40 C water bath for 1 min. 05:00 p.m. Post-op measurement Body weight (1B) Restrainer group, with IV injection: mice were Faeces fixed in the restrainer, the tail was immersed in collection a 39–40 C water bath for 1 min and saline solu- Nest building tion was administered. 1 05:00 p.m. Post-op measurement Nest building (2A) Chamber group with anaesthesia, without IV 2 05:00 p.m. Post-op measurement Body weight injection: mice were placed in the chamber and anaesthetized by isoflurane until surgical Faeces collection tolerance. (2B) Chamber group with anaesthesia, with IV injection: mice were placed in the chamber, anaesthe- the restrainer group (1B) and chamber group (2B). To tized by isoflurane until surgical tolerance and assess the effects on the animals, body weight and fCM saline solution was administered. were measured. Additionally, we counted the number of attempts per mouse and student until successful Similarly, to the treatment groups, mice of the control injection was performed and the number of termin- group were observed for nest-building activity, weighed ations after the third unsuccessful attempt. and transferred to a new cage for collection of faeces, The IV injection was conducted in both parts of the but were not placed in any device. Assessment of nest study with 0.9% saline solution (Fresenius Kabi building, measurement of body weight and faeces collec- Austria GmbH) at a dosage of 5 ml kg1 body weight tion was performed for all mice in the animal room. The with 30G x 1/2 in injection needles (StericanÕ,B. IV injections were performed in an adjacent procedure Braun, Melsungen, Germany). room. All mouse handling and IV injections were carried Body weight was measured in grams using an elec- out by the same skilled (female) experimenter. Treatment tronic balance (#440-47N, KERN & SOHN GmbH, of the experimental groups and the control animals was Germany). To assess nest building we removed all old evenly distributed over five consecutive weeks. nesting material from the previous night in the morning To assess the effects on the animals we measured (08:00–09:00 a.m.) and offered three fresh pads (Pur- three key characteristics: (a) changes in body weight, ZellinÕ). Nest-building was evaluated at 05:00 p.m. for (b) nest-building behaviour and (c) levels of stress hor- each single-housed mouse. Nest-building activity starts mones (fCM). The time schedule for treatment, assess- soon after nesting material is offered, which corresponds ment and sample collection is given in Table 1. to the high values of nest scores normally observed at the Part 2. As the novel chamber is reported to simplify end of the photoperiod.11 A nest complexity scoring and facilitate IV injection in mice, its applicability to according to Jirkof et al. was used with scores from 0 the procedure was additionally tested by a group of to 5 for the assessment of such measures as the manipu- eight untrained persons (students of veterinary medi- lation of nesting material and nest-building perform- cine, tenth semester). The (female) students were ance.11 Briefly, 0 ¼ pads not manipulated, 1 ¼ most given a short introduction in mouse handling but had pads slightly manipulated, 2 ¼ pads noticeably manipu- no previous experience and had not previously per- lated, shreds spread at a defined place of the cage, formed IV injection in mice. Each participant injected 3 ¼ noticeable arrangement of manipulated nesting a conscious mouse in the standard restrainer and an material, 4 ¼ flat nest with incomplete encasing walls, anaesthetized mouse in the novel chamber. The order 5 ¼ complex nest with heightened walls. of injections was randomly assigned either to start with Secretion of stress hormones was measured via the restrainer or with the chamber method. In the event excreted corticosterone metabolites in faeces. The level of an unsuccessful injection, a second and a (maximum) of circulating corticosterone and of excreted faecal third attempt per mouse was allowed. The study was metabolites follows a robust nycthemeral cycle.15,16 To conducted for all 16 mice in parallel on the same day control for the confounding effects of the 24-hour oscil- according to the time schedule described in Table 1 for lations of stress hormones, faeces samples were collected 194 Laboratory Animals 53(2) at a fixed time of day. In the event of a stressful incident, to day 2 and day þ2 relative to day 0. A significant the corresponding peak of fCM excretion in mice occurs change (increase) of body weight was measured for the 8–10 h after the initial impulse.12 Therefore, 8 h after IV control group (0) and the restrainer group without injection each mouse was placed for 2 h in a separate injection (1A) (group 0 p ¼ 0.003; group 1A p ¼ 0.005; cage (lined with paper sheets instead of wood bedding), Figure 3). Differences between treatment groups were from which fresh faeces were collected and frozen in not significant. It is worth noting that a (non-signifi- atubeat20C. After collection of faeces, the mice cant) decrease in body weight was observed for the were returned to their home cages. For fCM analysis, chamber group with injection (2B) on day 0 relative clean faecal samples from each mouse were dried and to day 2, whereas all other groups showed an increase homogenized and 0.05 g of the dry mass was mixed with of body weight during this period. 1 ml of 80% methanol. The concentrations of corticosterone metabolites were measured by an in-house Stress hormone levels. Because there are sex-based 5a-pregnane-3b,11b,21-triol-20-one enzyme immuno- differences in fCM concentrations in mice, females assay, which has been validated for mice. For further and males were evaluated separately (Figure 4(a) and details of the assay, see two studies by Touma et al. in (b)).16 Our results confirmed the previously published 2003 and 2004.12,16 data that females generally show about twice as high fCM values than males.12 Differences in fCM levels Statistical analysis were observed between females of both strains. However, the higher levels of SWISS females differed All statistical analysis and chart preparation was per- not significantly from those of C57BL/6 females (data formed using IBM SPSS v24 (SPSS Inc. Chicago, IL, not shown). Irrespective of the method applied, the USA). Animals in both parts of the study were ran- impact of treatment was generally more distinct and domly allocated to the treatment groups. We analysed consistent for males, reflecting a short but stressful the effects of tail-vein injection on body weight, fCM event. A significant increase in fCM level was detected and nest-building activity when the mice were immobi- for males within the chamber group 2B with injection lized by different procedures. (day 2:day 0, p ¼ 0.04). The fCM concentration Body weight was analysed as a percentage of change, immediately post injection (day 0) of males of the each relating to the measured individual value 48 h ear- group 2B was also significantly higher in comparison lier (¼ 100%). Differences within each group were to the control group at the same time (p ¼ 0.014; Figure tested with a paired sample t-test, while between- 4(b)). Interestingly, the increase of fCM values on day 0 group differences were tested with a t-test for unpaired (8–10 h post injection) was completely counterbalanced samples. Concentrations of fCM were analysed separ- on day 2 (significant in group 2B, day 0:day 2, p ¼ 0.01), ately for females and males by a General Linear Model suggesting that both immobilization by anaesthesia and (GLM) for repeated measurements, followed by post IV injection are only stressful in the short term. hoc tests using Bonferroni’s alpha-correction procedure. The comparison of fCM concentrations in the con- Nest building. Nesting behaviour was scored most fre- trol group and the treatment groups was performed quently and alike (no significant differences detected) in using simple a priori contrasts (least significant differ- both strains at level ‘3’ and ‘4’ prior to IV injection (day ence). Nest-building activity was assessed by scores 1). Nest-building activity was slightly reduced in all from 0 to 5 and differences between repeated measure- groups, including the controls, on day 0 (after treatments within groups were tested by the Wilcoxon ment). This general reduction might be the result of signed-rank test. All data are graphically depicted as interference from the activities during the morning of box-and-whisker plots with medians (lines in boxes), day 0 in the animal room (transport of cages between IQR (25–75% interquartile range, boxes) and max- animal room and procedure room for IV injection). An imum range (whiskers), excluding extreme values. increase in nest-building activity was detected on day 1, Two-tailed p values of p 0.05 were considered to be which was significant in the chamber group 2A without statistically significant. injection (p ¼ 0.04; Figure 5).

Results Part 2 Part 1 Body weight. A significant decrease in the body weight was observed on day 0 relative to day 2 for animals of Change in body weight. Because of sex- and strain- the chamber group in comparison to the restrainer specific differences, changes in body weight are pre- group (p ¼ 0.049; Figure 6). This weight loss for anaes- sented as percentage of the weight on day 0 relative thetized animals of the chamber group was completely Resch et al. 195

Figure 3. Boxplot for the percentage of change in body weight for female and male C57BL/6 and SWISS mice, which have received an IV injection from a skilled person in the restrainer, or under anaesthesia in the chamber. Body weight was measured 2 days before treatment, at the day of injection and 2 days post manipulation; changes were calculated for the body weight on day 0 relative to day 2 and on day þ2 relative to day 0 (Part 1). Paired t-Test **p 0.01.

Figure 4. Boxplots for faecal corticosterone metabolite concentrations (fCM) in (a) females and (b) males in the different treatments. Faeces were collected 2 days before the experimental manipulation (day 2), at the day of manipulation (day 0) and 2 days post manipulation (day þ2) (Part 1). Comparison using Bonferroni’s alpha correction *p 0.05, **p 0.01. 196 Laboratory Animals 53(2)

Figure 5. Boxplots for nest-building scores (0–5) in the different treatments. Nest building was recorded 1 day before the experimental manipulation (day 1), at the day of manipulation (day 0) and 1 day post manipulation (day þ1) (Part 1). Extreme values more than 1.5 IQR above Q3 or below Q1 (). Wilcoxon Signed Ranks Test *p 0.05. compensated after the treatment, which was shown by Discussion the significant increase of body weight in the 48 h period after IV injection (p ¼ 0.024). The procedure for lateral tail-vein injection consists of different steps depending on the method used. The type of restraint and the techniques to immobilize an animal Stress hormone levels. Similarly, to Part 1, the ana- significantly influence the well-being of the animal, both lysis of fCM revealed an increase of concentrations on during as well as following treatment. We have com- day 0 in both treatment groups. In the chamber group, pared the widely-used restrainer method for IV injection the measured fCM concentration was significantly in conscious mice with a novel induction chamber for higher on day 0 than on day 2 and day 2 tail-vein injection of mice using isoflurane anaesthesia. (p ¼ 0.0004 and p ¼ 0.011). An increased fCM level The outcome of this study generally supports the classi- was also found for animals of the restrainer group on fication of lateral tail-vein injection in mice as severity the day of treatment. The level on day 0 was signifi- degree ‘mild’ for both methods. All tested characteristics cantly higher in comparison to the level measured on revealed either no significant impairment of or only a day 2 post injection (day 0:day 2, p ¼ 0.047, Figure 7). short-term mild impact on the well-being of the animals. In direct comparison to the common restrainer method Practicability. Five of the eight students performed a we found no benefit in using the novel induction cham- successful tail-vein injection with conscious mice in the ber, neither as refinement to improve animal welfare nor restrainer. On average 2.2 attempts were necessary to as a simplification of the procedure. succeed. In contrast, only three of the eight students For direct comparison of both devices we choose performed a successful tail-vein injection with anaesthe- C57BL/6 and SWISS mice. C57BL/6 is the most often tized mice in the induction chamber and all three used genetic background for genetically engineered needed a third (last possibility according to our proto- mice. However, mice of this inbred strain are small, col) attempt to succeed. dark pigmented and their tail veins are less distinctly Resch et al. 197

Figure 6. Boxplots for the percentage of change in body weight in female C57BL/6 mice, which have received an IV injection from untrained persons in the restrainer, or under anaesthesia in the chamber. Body weight was measured 2 days before treatment, at the day of injection and 2 days post manipulation; changes were calculated for the body weight on day 0 relative to day 2 and on day þ2 relative to day 0 (Part 2). Paired t-Test **p 0.05.

visible. In order to test both methods with a second stressful event, simple handling of mice will not have strain, we selected SWISS albinotic mice. These an impact on the changing levels of fCM.21 outbred mice are bigger than C57BL/6 and the larger According to the measured fCM levels in males of diameter of their tails could additionally improve the Part 1 and females of Part 2 of the study, IV injection in stasis of the tail veins after compression in the cap hole the chamber with isoflurane anaesthesia appears to be of the closed chamber. more stressful than it is for conscious mice in a To assess the degree of severity of tail-vein injection common restrainer. This finding is in accordance with we measured three key characteristics: changes in body a previous study that showed that isoflurane anaesthe- weight, the level of faecal metabolites of corticosterone, sia alone induces a short rise in serum cortico- the major glucocorticoid in mice,17 and nest-building sterone.22,23 Although the elevated fCM level on day behaviour. 0 in Part 1 of males in the chamber group without The rate of change in body weight is a recommended injection (group 2A) indicate a previously stressful indicator for previous and prolonged pain or distress, event, fCM increase reached significance only for which is usually manifested in weight loss.10,18 It is fre- anesthetized mice with IV injection (group 2B). This quently used in scoring systems for the assessment of result suggests that anaesthetizing and IV injection animal welfare.19 Mice have a very high motivation together could have a cumulative impact in terms of for nest building and deficits in this activity are often increased severity. Interestingly, the significantly associated with a reduction in well-being.20 The ana- increased fCM levels for the chamber group in Part 2 lysis of corticosterone metabolites in faecal samples are paralleled by a reduction in body weight of these is a non-invasive method to monitor adrenocortical animals which was not observed in the restrainer group. activity in mice and is a well-established technique to This is important because an effect on body weight sug- evaluate pain-induced distress.16 In contrast to a gests a stronger or more prolonged impact of 198 Laboratory Animals 53(2)

Figure 7. Boxplots for faecal corticosterone metabolite concentrations (fCM) in females, which have received an IV injection from untrained persons in the restrainer, or under anaesthesia in the chamber. Faeces were collected 2 days before the experimental manipulation (day 2), at the day of manipulation (day 0) and 2 days post manipulation (dayþ2) (Part 2). Comparison using Bonferroni’s alpha correction *p 0.05, ***p 0.001. anaesthesia on an animal’s well-being, possibly as a day 0 (in the afternoon of the treatment day) in com- consequence of a timely restricted reduction of food parison to day 1 post treatment. intake.24,25 The restrainer was used as it is frequently applied in On average we measured lower and less variable our routines and the novel induction chamber strictly fCM concentrations in Part 2 of the study, in which according to the product description of the manufac- exclusively C57BL/6 females were used. The difference turer. In Part 1 of the study, mice of both strains were can be mainly explained by higher but not significantly successfully injected independent of the device used by different fCM levels for female SWISS outbred mice in the skilled experimenter. In Part 2 we were predomin- the merged groups of Part 1. However, an additional antly interested in the practicability of both methods impact of the housing system (single housing in Part 1 when applied by inexperienced personnel because the and group housing in Part 2) cannot be excluded. novel device is explicitly advertised for a simplified Nevertheless, the average level of measured fCM injection procedure and a considerably lower number values in Part 2 is similar to previously reported data of unsuccessful attempts. Although the rate of success- of C57BL/6 females.12,14 ful injection attempts of untrained persons did not sig- It has been shown that spontaneous nest-building nificantly differ between the two methods, the results behaviour of mice is affected by painful treatments tend to be more favourable for injecting conscious mice and anaesthesia.11 In this study nest-building behaviour in the common restrainer than when anaesthetized mice was not evidently affected by the treatments, suggesting were injected in the induction chamber. The reason may only a mild impact, even if short-term anaesthesia was be the improved vasodilatation of lateral tail veins after applied. One exception was seen in mice of the chamber warming of the tail in the restrainer method. Isoflurane group with anaesthesia (group 2A), in which a signifi- anaesthesia is known to cause low blood pressure,6 cantly lowered nest-building activity was observed on which may reduce the effect of stasis in the tail when Resch et al. 199 pinched in the hole of the chamber cap, reducing the References visibility and accessibility of the vein in anaesthetized 1. Turner PV, Pekow C, Vasbinder MA, et al. mice. Whether additional tail warming would have Administration of substances to laboratory animals: improved the success rate of injections in anaesthetized equipment considerations, vehicle selection, and solute mice or not, we did not test, as this would have neces- preparation. J Am Assoc Lab Anim Sci 2011; 50: 614–627. sitated two additional experimental groups: (a) tail 2. Morton DB, Jennings M, Buckwell A, et al. Refining warming in the chamber groups and (b) anaesthesia procedures for the administration of substances. Report in the restrainer groups. However, from a practical of the BVAAWF/FRAME/RSPCA/UFAW Joint point of view, tail warming of anaesthetized mice in Working Group on Refinement. British Veterinary the chamber would have abolished the accentuated Association Animal Welfare Foundation/Fund for the and beneficial simplification of the procedure with the Replacement of Animals in Medical Experiments/Royal Society for the Prevention of Cruelty to Animals/ novel device. Inhalation anaesthesia for mice in a Universities Federation for Animal Welfare. Lab Anim standard restrainer is not applicable. In order to isolate 2001; 35: 1–41. the effect of anaesthesia on its own in the chamber 3. Meijer MK, Spruijt BM, van Zutphen LF, et al. Effect of group, IV injection on conscious mice would have restraint and injection methods on heart rate and body been necessary. However, this would have harboured temperature in mice. Lab Anim 2006; 40: 382–391. a high risk of injury to the not adequately immobilized 4. Nohara M, Tohei A, Sato T, et al. Evaluation of response mouse and therefore, in our opinion, it was not a prac- to restraint stress by salivary corticosterone levels in adult tically feasible option. male mice. J Vet Med Sci 2016; 78: 775–780. Anaesthesia with inhalants requires appropriate 5. Constantinides C, Mean R and Janssen BJ. Effects of equipment (which may turn out to be costly), including isoflurane anesthesia on the cardiovascular function of an effective scavenging system to protect personnel, as the C57BL/6 mouse. ILAR J 2011; 52: e21–e31. well needing specific clinical experience. At worst, pro- 6. Janssen BJ, De Celle T, Debets JJ, et al. Effects of anes- longed exposure to significantly higher doses of anaesthetics on systemic hemodynamics in mice. Am J Physiol thetic may result in mortality of the animals via Heart Circ Physiol 2004; 287: H1618–H1624. respiratory and cardiovascular depression. Even in pro- 7. Ding F, Zheng L, Liu M, et al. Ciproxifan, an H3 recep- fessional clinical practice, experimental results can be tor antagonist, improves short-term recognition memory significantly influenced by side-effects of anaesthesia. impaired by isoflurane anesthesia. J Anesth 2016; 30: 684–690. We found no refinement of the injection procedure 8. Lin D and Zuo Z. Isoflurane induces hippocampal cell with the induction chamber and isoflurane anaesthesia. injury and cognitive impairments in adult rats. Although both methods are categorized at the lowest Neuropharmacol 2011; 61: 1354–1359. degree of severity, we assess IV injection with the 9. Yonezaki K, Uchimoto K, Miyazaki T, et al. common restrainer as the more suitable method, Postanesthetic effects of isoflurane on behavioral pheno- although the novel chamber might have advantages types of adult male C57BL/6J mice. PLoS One 2015; 10: for the induction of anaesthesia and for short-time e0122118. transport of anaesthetized mice for use in other experi- 10. Brain PF, Brugere H, Clausing P, et al. Pain and distress mental procedures. in laboratory rodents and lagomorphs. Report of the Federation of European Laboratory Animal Science Acknowledgements Associations (FELASA) Working Group on Pain and We acknowledge Sebastian Gehrig for providing the induc- Distress. Lab Anim 1994; 28: 97–112. tion chamber, Edith Klobetz-Rassam for fCM analysis and 11. Jirkof P, Fleischmann T, Cesarovic N, et al. Assessment Nina Hornacek, Dalma Batkay and Graham Tebb for editing of postsurgical distress and pain in laboratory mice by the figures and text of the manuscript. nest complexity scoring. Lab Anim 2013; 47: 153–161. 12. Touma C, Palme R and Sachser N. Analyzing cortico- Declaration of Conflicting Interests sterone metabolites in fecal samples of mice: a noninva- The author(s) declared no potential conflicts of interest with sive technique to monitor stress hormones. Horm Behav respect to the research, authorship and/or publication of this 2004; 45: 10–22. article. 13. Mahler CM, Berard M, Feinstein R, et al. FELASA recommendations for the health monitoring of mouse, rat, Funding hamster, guinea pig and rabbit colonies in breeding and experimental units. Lab Anim 2014; 48: 178–192. The author(s) received no financial support for the research, authorship and/or publication of this article. 14. Kolbe T, Palme R, Tichy A, et al. Lifetime Dependent Variation of Stress Hormone Metabolites in Feces of ORCID iD Two Laboratory Mouse Strains. PLoS One 2015; 10: e0136112. Thomas Ru¨licke http://orcid.org/0000-0002-2121-9496 200 Laboratory Animals 53(2)

15. Sollars PJ, Weiser MJ, Kudwa AE, et al. Altered entrain- evokes an increase in adrenocortical and neural activity ment to the day/night cycle attenuates the daily rise in in mice. Horm Behav 2016; 81: 97–105. circulating corticosterone in the mouse. PLoS One 2014; 22. Hohlbaum K, Bert B, Dietze S, et al. Severity classifica- 9: e111944. tion of repeated isoflurane anesthesia in C57BL/6JRj 16. Touma C, Sachser N, Mostl E, et al. Effects of sex and mice-Assessing the degree of distress. PLoS One 2017; time of day on metabolism and excretion of cortico- 12: e0179588. sterone in urine and feces of mice. Gen Comp 23. Jacobsen KR, Kalliokoski O, Teilmann AC, et al. Endocrinol 2003; 130: 267–278. The effect of isoflurane anaesthesia and vasectomy on 17. Spackman DH and Riley V. Corticosterone concentra- circulating corticosterone and ACTH in BALB/c mice. tions in the mouse. Science 1978; 200: 87. Gen Comp Endocrinol 2012; 179: 406–413. 18. Ullman-Cullere MH and Foltz CJ. Body condition scor- 24. Jacobsen KR, Kalliokoski O, Teilmann AC, et al. ing: a rapid and accurate method for assessing health Postsurgical food and water consumption, fecal cortico- status in mice. Lab Anim Sci 1999; 49: 319–323. sterone metabolites, and behavior assessment as noninva- 19. Bugnon P, Heimann M and Thallmair M. What the lit- sive measures of pain in vasectomized BALB/c mice. erature tells us about score sheet design. Lab Anim 2016; J Am Assoc Lab Anim Sci 2012; 51: 69–75. 50: 414–417. 25. Ratsep MT, Barrette VF, Winterborn A, et al. 20. Van de Weerd HA, Van Loo PL, van Zutphen LF, et al. Hemodynamic and behavioral differences after adminis- Preferences for nesting material as environmental enrich- tration of meloxicam, buprenorphine, or tramadol as ment for laboratory mice. Lab Anim 1997; 31: 133–143. analgesics for telemeter implantation in mice. JAm 21. Mallien AS, Palme R, Richetto J, et al. Daily exposure to Assoc Lab Anim Sci 2013; 52: 560–566. a touchscreen-paradigm and associated food restriction

Re´sume´ L’administration intraveineuse (IV) chez les souris est principalement effectueé par les veines caudales late´rales. La technique nećessite une formation ade´quate avant de pouvoir eˆtre utiliseé en toute sećurite´ et de manie`re routinie`re. Une nouvelle chambre d’induction d’anesthe´sie a e´te´ de´veloppeé pour simplifier le traitement et faciliter les injections en IV chez les souris, particulie`rement par les employe´s non forme´s. Nous avons e´value´ les avantages de la chambre pour raffiner les injections en IV chez les souris anesthe´sieés par de l’isoflurane en la comparant directement avec la me´thode habituelle de retenue d’animaux conscients. Le poids corporel, le comportement de nidification et les concentrations de me´tabolites de corticoste´rone dans les matie`res fećales ont e´te´ pris en compte comme indicatifs de la de´tresse provoqueé par les diffe´r- entes proce´dures. Les re´sultats sugge`rent que les deux me´thodes d’injection dans la veine caudale induisent des niveaux semblables de stress momentane´ chez les animaux, qui sont re´ve´le´s par une augmentation a` court terme des niveaux de me´tabolites des hormones du stress dans les matie`res fećales. Une re´duction temporaire de poids corporel a e´te´ observeé apre`s les injections en IV chez les souris anesthe´sieés par de l’isoflurane, mais pas chez les souris conscientes injecteés dans les dispositifs de retenue habituels. Nous en concluons que la gravite´ de l’injection dans la veine caudale est « le´ge`re » dans le cas des deux me´thodes. Aucune preuve ne sugge`re que raffiner la proce´dure en utilisant une anesthe´sie par isoflurane dans la chambre d’induction n’ait e´te´ associeéa` un quelconque beńe´fice.

Abstract Intraveno¨se (IV) Injektion erfolgt bei Maüsen hauptsa¨chlich u¨ber die lateralen Schwanzvenen. Das Verfahren erfordert ausreichend U¨bung, bevor es sicher und routinema¨ßig angewendet werden kann. Daher wurde eine neuartige Ana¨sthesie-Injektionskammer entwickelt, um die IV-Injektion bei Maüsen, insbesondere fu¨r ungeu¨btes Personal, zu erleichtern. Wir haben die Vorteile der Kammer als zur IV- Injektion mit Isofluran- Ana¨sthesie in direktem Vergleich mit der herko¨mmlichen Restrainer-Methode bei wachen Maüsen bewertet. Ko¨rpergewicht, Nestbauverhalten und Konzentrationen von fa¨kalen Kortikosteron-Metaboliten wurden zur Belastungsbeurteilung beider Verfahren ausgewertet. Die Ergebnisse legen nahe, dass beide Methoden der Schwanzveneninjektion bei den Tieren einen a¨hnlichen momentanen Stress hervorrufen, was anhand eines kurzfristigen Anstiegs der Stresshormon-Metaboliten im Kot sichtbar wird. Eine voru¨bergehende Reduktion des Ko¨rpergewichts wurde nach der IV Injektion unter Isofluran-Ana¨sthesie beobachtet, jedoch nicht bei Maüsen, die im wachen Zustand im Restrainer injiziert wurden. Schlussfolgernd stellen wir fest, dass der Belastungsgrad einer Schwanzveneninjektion bei Maüsen fu¨r beide Methoden ‘‘leicht’’ ist. Es gab keine Resch et al. 201

Hinweise darauf, dass das ‘Refinement’ des Verfahrens durch den Einsatz von Isofluran-Ana¨sthesie in der Narkosekammer mit einem Vorteil verbunden wa¨re.

Resumen La administracioń intravenosa (IV) en ratones se realiza predominantemente a trave´s de las venas laterales de la cola. La tećnica requiere una capacitacioń adecuada antes de poder usarse de manera segura y rutinaria. Se ha desarrollado una novedosa ca´mara de induccioń de anestesia para simplificar el tratamiento y facilitar la inyeccioń intravenosa en ratones, en concreto para el personal no capacitado. Hemos evaluado los beneficios de la ca´mara para perfeccionar la inyeccioń IV en ratones anestesiados con isoflurano en comparacioń directa con el me´todo de jaula comuń en animales despiertos. El peso corporal, el comportamiento de anidamiento y las concentraciones de metabolitos de corticosterona fecal se tomaron como indicativos de la angustia inducida por los diversos procedimientos. Los resultados sugieren que ambos me´todos de inyeccioń en la vena de la cola inducen niveles similares de estre´s momentańeo en los animales, demostrado por un aumento a corto plazo en los niveles de los metabolitos de la hormona del estre´s en las heces. Se observo´ una reduccioń temporal del peso corporal despue´s de la inyeccioń IV bajo anestesia con isoflurano, pero no para ratones despiertos inyectados en la jaula comuń. Concluimos que la severidad de la inyeccioń en la vena de la cola en ratones es ‘‘leve’’ en ambos me´todos. No hubo evidencia de que el perfeccionamiento del procedimiento mediante el uso de anestesia con isoflurano en la ca´mara de induccioń estuviera asociado a alguń beneficio. felasa TECH DAY clasa June 12, 2019

LATE REGISTRATION FROM APRIL 1, 2019

Animal Technician and Carer Day on June 12, 2019 Programme

At the Congress you can look forward to the Tech Day You can also look forward to programme on Wednesday offering various education the very interesting Congress sessions. Programme offering special Main sessions include: education sessions and workshops. • Harmonisation of Education, Training and CPD for laboratory animal caretakers and technicians The scientific programme consists • Novel elements to the culture of animal (rodent) care of six streams: • Minimising pain, suffering, distress and long-lasting Education and Training, Reproducibility and Translation, harm to animals in practice Model Management, Defining Good Care, Compliance and • Focus on the welfare and quality of livestock species Communication, Severity Classification to be used in research To view the “Preliminary Programme”, visit our congress Do not miss the Tech Day on Wednesday! website www.felasa2019.eu

Prague

Prague, in Czech “Praha”, is the fourteenth largest city in the European Union. The city is home to about 1.3 million people. Prague, often called “Golden” or “Hundred-spired”, belongs to the architecturally unique European towns, attracting tourists from around the world. This lovely, romantic and favourite place is a city of very high standards, offering both historical and modern conference venues, hotels, restaurants and places of interest.

www.felasa2019.eu Expert in the world of research diets

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. Special Diets Services

Special Diets Services PO Box 705, Witham, Essex, England CM8 3AD Telephone: +44 (0) 1376 511260 SDS www.sdsdiets.com Fax: +44 (0) 1376 511247 the essential resource for quality research diets Email: [email protected]

SDS8222 SDS Corporate Ad LAJ 280x210mm.indd 1 19/01/2017 10:40

News Laboratory Animals 2019, Vol. 53(2) 205 ! The Author(s) 2019 Don’t forget to register for the FELASA Article reuse guidelines: sagepub.com/journals- 14th Congress permissions DOI: 10.1177/0023677219830935 journals.sagepub.com/home/lan Jean-Philippe Mocho

The FELASA 14th congress is being held in the Wednesday 12th will see four sessions dedicated to beautiful city of Prague, from 10 to 13 June 2019. animal technicians and caretakers, and another set of As usual there will be a comprehensive programme four sessions dedicated to animal research reporting of seminars and workshops to cover most areas of organised in collaboration with Laboratory Animal interest and expertise. These streams include education journal. Some focus on experimental design will take and training, reproducibility and translation, model place on Thursday 13th. management, defining good care, compliance and com- Not only is a FELASA congress a good way to munication, and severity classification, as well as spe- undertake continuing professional development cial events. (CPD), it is also an ideal opportunity to discuss new More specifically, some sets of sessions are dedicated ideas and views with others working within your indus- to ‘‘hot topics’’. For example, on Monday 10th, there try, as well as an excellent chance for networking – or will be special highlights on large animals. On Tuesday just meet up with old friends and acquaintances. 11th, you can be briefed on the latest advances on fish All information and booking details may be found husbandry. Or you may consider reflecting on severe on the special website www.felasa2019.eu suffering in collaboration with the RSPCA. We look forward to seeing you there!

Chair of the FELASA 2019 Scientific Committee

Corresponding author: Jean-Philippe Mocho Email: [email protected]

Contributions to the News section are not subject to peer review and reflect the opinion of your subscribed society. Explore the Laboratory Animals Handbooks

NEW EDITION!

The Design of Animal Experiments Health and Safety in Laboratory Parasites of Laboratory Animals nd (2 Edition) Animal Facilities Dawn G Owen Reducing the Use of Animals Edited by in Research through Better Margery Wood and Experimental Design Maurice W. Smith Michael Festing

Find out more and buy online at uk.sagepub.com/lahandbooks

Laboratory Animals

The international journal of laboratory animal science, medicine, technology and welfare, Laboratory Animals publishes peer-reviewed original papers and reviews on all aspects of the use of animals in biomedical research. The journal promotes improvements in the welfare or well-being of the animals used, particularly focusing on research that reduces the number of animals used or which replaces animal models with in vitro alternatives.

Read the latest content at journals.sagepub.com/home/lan

journals.sagepub.com/home/lan

N8J2049 LAN Flyer Update.indd 1 09/01/19 5:14 PM AALAS Heads to Denver for the 70th National Meeting

Join us for the 70th AALAS National Meeting in Denver, Colorado. Each fall since 1950, the American Association for Laboratory Animal Science has held its annual National Meeting. During the five days of the meeting, members and nonmembers come together to enjoy the workshops, lectures, poster sessions, and exhibits. The program is designed to have topics relevant to the entire membership. Exhibitors have an opportunity to interact with AALAS members from the academic community, research institutions, government organizations, and commercial companies. The AALAS National Meeting is the largest gathering in the world of professionals concerned with the production, care, and use of laboratory animals.

This will be an event to remember.

American Association for Laboratory Animal Science Phone: (901) 754-8620 Fax: (901) 753-0046 [email protected] • www.aalas.org

2019 Denver ad_210x280_LAJ.indd 1 3/5/2019 8:32:14 AM News Laboratory Animals 2019, Vol. 53(2) 208–209 ! The Author(s) 2018 Article reuse guidelines: Expert information from the Working sagepub.com/journals- permissions DOI: 10.1177/0023677219830548 Group on Hygiene: Harmonisation of journals.sagepub.com/home/lan Health Monitoring Reports

Werner Nicklas and Karin Seidel

The Working Group on Hygiene, one of the nine per- Both recommendations have similar objectives; how- manent committees of the GV-SOLAS, recently pub- ever, they differ slightly in some aspects. The FELASA lished an expert information on the ‘Harmonisation recommendations give concrete information on the of Health Monitoring Reports’.1 The publication is implementation of the testing programme, whereas available in both English and German in order to the requirements for the layout of the health monitor- reach German-speaking staff involved in animal experi- ing report are less precise. In contrast to this, the ments as well as readers in non-German-speaking FELASA-AALAS recommendation puts forward the countries.1,2 creation and formatting of a health monitoring The harmonisation of health monitoring reports report. An easy to recognize layout, as well as a precise becomes increasingly important, as health monitoring list of microorganisms to be specified in a health moni- reporting is not yet standardised and a variety of dif- toring report are compiled. These lists are not complete, ferent formats is used to date. however, as, for example, in immunodeficient animals, It is well known that the increasing international or under certain circumstances, additional microorgan- transfer of genetically modified mice is associated isms can be considered (e.g. opportunistic microorgan- with higher risks pertaining to the dissemination of isms, dermatophytes etc.). microorganisms. Although by now – also due to Many formats for the description of health monitor- animal welfare reasons – preferentially cryo-conserved ing programmes as well as for the respective health sperm or embryos are shipped, shipping of live animals monitoring reports exist nationally and internationally. is often the standard procedure, as many academic In Europe, health monitoring reports are increasingly institutions do not have the infrastructure to conduct generated using the design of the FELASA recommen- embryo transfer. Therefore, an exact review of the dations for the health monitoring of rodents (2014), health status of mice as well as the documentation of which have a long tradition since 1994. Many institu- this reviewing process is extremely important to prevent tions, particularly in Non-European countries, often the introduction of unwanted microorganisms into an supply test reports from testing laboratories without animal facility. For this, the receiving institutions need any explanation or summary. These can be difficult to reliable information on the current health status of the read and understand and do usually not provide suffi- animals. An interpretation based solely on laboratory cient information to assess the infectious status of spe- reports holds many risks. cific animals. This inconsistent reporting holds many The publication describes the information on the risks, including creating uncertainty when interpreting health monitoring programme that is necessary for a results and wasting time in daily routine because of correct interpretation of health monitoring reports. necessary further enquiries with senders, and can there- European animal facilities increasingly apply the rec- fore cause delays when importing important mouse ommendations of the Federation of European Laboratory Animal Science Associations (FELASA) Die Gesellschaft fu¨r Versuchstierkunde/Society of Laboratory for the health monitoring of rodents. A joint working Animal Science (GV-SOLAS) group of FELASA and the American Association for Laboratory Animal Science (AALAS) has published a Corresponding author: position statement to discuss the potential for an inter- Karin Seidel Email: [email protected] national harmonised health monitoring reporting format. This would be highly appreciated in an inter- Contributions to the News section are not subject to peer review national context. and reflect the opinion of your subscribed society. Nicklas and Seidel 209 strains. As the interpretation of health reports is very – A contact person should be specified. complex, respective experts are needed for a proper – Changes in taxonomy and nomenclature should interpretation. be adjusted, i.e. Pasteurellaceae The format shall enable the compilation of a standardised health monitoring report and thus simplify its The GV-SOLAS Working Group on Hygiene expli- interpretation. The decision how, and for which micro- citly recommends the use of this format in order to organisms, tests will be performed is dependent on the improve and facilitate the interpretation of health respective institution and the use of the animals in monitoring reports when importing rodents in a research. In addition to that, a standardised health European and International context. This is a further monitoring report shall facilitate to determine which step towards the harmonisation of experimental animal microorganisms were actually tested for. research at a European level. The proposed format for a health monitoring report by the AALAS–FELASA Working Group can be Declaration of Conflicting Interests downloaded for mice and rats in an Excel format The author(s) declared no potential conflicts of interest with (Microsoft). The format can be adapted respectively. respect to the research, authorship, and/or publication of this The following important information shall be con- article. sidered when using the format: Funding – The list of microorganisms and their respective The author(s) received no financial support for the research, order should be maintained. authorship, and/or publication of this article. – Additionally detected and relevant microorganisms as well as newly described and important References microorganisms should be listed. They can be 1. GV-SOLAS, Working Group on Hygiene. Harmonisation inserted e.g. under ‘additional agents’. of Health Monitoring Reports, http://www.gv-solas.de/ – If no tests were performed for some microorgan- fileadmin/user_upload/pdf_stellungnahme/2018stell_hyg- isms, they should be listed and NOT TESTED Harmonisation_e.pdf (accessed March 2018). should be noted. 2. GV-SOLAS, Ausschuss fu¨r Hygiene. Harmonisierung – If microorganisms are detected, confirmation tests von Gesundheitszeugnissen, http://www.gv-solas.de/ shall be noted and specified. fileadmin/user_upload/pdf_stellungnahme/2018stell_hyg- – Ideally, the health monitoring programme should Harmonisation.pdf (accessed March 2018). be described precisely. together we thrive Ø connuing educaon Ø latest news & informaon Ø representaon at EU Ø improved animal welfare Ø advancing scienfic knowledge

ESLAV is the only European Scienfic Associaon represenng all veterinary professionals in the field of animals used for Webinar Series scienfic purposes hp://eslav- eclam.org/educaon/ www.eslav-eclam.org eslav-webinar-series/

ESLAV acvely represents its more than 400 members as a stakeholder at diﬀerent levels of internaonal decision-making bodies like the European Commission and AAALAC Internaonal

Join today Get more informaon on how to join by vising hp://eslav-eclam.org/eslav/membership/

Summer School hp://eslav-eclam.org/ educaon/eslav-eclam- summer-school/ From the heart of southern Spain it is a pleasure to invite you to the forthcoming XV National Congress of the Spanish Society for Laboratory Animal Sciences which will take place in Seville on November 6th-8th 2019. Technical workshops will run on November 5th. We are confident that the attractive conference program along with the beauty of our well- known city will attract large numbers of participants.

The Congress venue has6th, well-appointed 7th conference y 8th and posterde rooms November surrounded by a splendidly 2019 luminous area where our industry partners will exhibit their latest products, equipment or services and where delegates can enjoy networking with other professionals in the field.

The Congress covers a wide range of Laboratory Animal Science, in particular: - New technologies relating to laboratory animals. - Staff motivation and conflict resolution. - Efficient management of laboratory animal facilities. - Advances in the 3Rs. - Panel: legislation, education and training. - Challenges in non-conventional research environments. - Philosophical basis of animal protection and research. - Reproducibility and translation of experiments. - Transparency of animal research. - Veterinary care in severe experiments. - Brief presentations of their newest products from commercial vendors

More information: www.secalsevilla2019.com

Publi_ingles_SECAL_XV.indd 1 31/1/19 12:41 212...... Calendar of events

Meetings of interest to laboratory animals scientists and technicians: references to Laboratory Animals are for further details. 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.

2019

7–10 April TT2019 (International Society for Transgenic Technologies) – 15th Annual meeting, Kobe, Japan. For further information visit https://www.transtechsociety.org/index.php?src=news&srctype=detail&category= Announcements&refno=13 8–10 April Assessment, prevention and alleviation of pain and distress in laboratory animals workshop, Newcastle upon Tyne, England. For further information visit http://www.lasa.co.uk/meetings/ 8–11 April Microbiology Society Annual Conference 2019, Belfast, UK. For further information visit https://microbiolo- gysociety.org/event/annual-conference/annual-conference.html 9–12 April Institute of Animal Technology (IAT) Congress, North West England, UK. For further information visit http:// www.iat.org.uk/congress 14–19 April Applied In Vitro Toxicology Course (ESTIV/ASCCT), Bucharest, Romania. For further information visit https://estiv2019.com/ 15–17 April British Toxicology Society (BTS) Annual Congress 2019, Cambridge, UK. For further information visit http:// www.thebts.org/bts-annual-congress-2019/ 8–10 May Animal Genetics and Diseases conference 2019, Cambridge, UK. For further information visit https://course- sandconferences.wellcomegenomecampus.org/our-events/animal-geneticsdiseases-2019/ 10 May The emotional cost of caring: Impact on animal welfare, London, UK, For further information visit https://www. eventbrite.com/e/the-emotional-cost-of-caring-impact-on-animal-welfare-tickets-55399738108 13–17 May FELASA Laboratory Animal Science Course on Primates, Gottingen, Germany. For further information visit https://www.nc3rs.org.uk/sites/default/files/documents/Events/AnnouncementProgram_FELASACourse_ May2019 21–24 May JRC Summer School 2019: Non-Animal Approaches in Science, Ispra, Italy. For further information visit https://ec.europa.eu/jrc/en/event/conference/jrc-summer-school-non-animalapproaches-science 22–24 May 13th Minipig Research Forum (MRF), Vienna, Austria. For further information visit https://minipigresearch- forum.org/meetings/ 4–5 June Advances in Cell and Tissue Culture 2019, Cardiff, UK. For further information visit https://theactc.com/ 10–13 June 14th FELASA Congress, Prague, Czech Republic. For further information visit http://www.felasa2019.eu/ 24–28 June ESLAV/ECLAM Summer School (Module 4), Stockholm, Sweden. For further information visit:http://eslav- eclam.org/education/eslav-eclam-summer-school/module-4-experimental-and-surgical-techniques-design- and-conduct-of-research/ 15–18 July The 15th International Congress of Toxicology (ICTXV), Honolulu, Hawaii. For further information visit https://www.toxicology.org/events/ict/index.asp 8–11 September EUROTOX Congress, Helsinki, Finland. For further information visit http://www.eurotoxcongress.com/2019/ 30 September. Deadline for the second 2019 announcement for LAL grants 30 September–11 October 5th LAS EU Functions Course, Function ABCD, mice and rats, Athens, Greece. For further information visit http://hsblas.gr/lasfunctionscourse/ 2–4 October 45th AFSTAL Conference, La Rochelle, France. For further information visit https://www.colloque-afstal.com/ 2019/ 10–13 October 22nd European Congress on Alternatives to Animal Testing / 19th Annual Congress of EUSAAT, Linz, Austria. For further information visit http://eusaat-congress.eu/ 13–17 October AALAS National meeting, Denver, Colorado, USA. For further information visit https://www.aalas.org/ national-meeting/general-information/future-meetings 6–8 November. SECAL National Congress. Sevilla. Spain. For further information visithttps://www.secalsevilla2019.com 26–28 November LASA Annual meeting, Birmingham, UK. For further information visit http://www.lasa.co.uk/meetings/

Index to Advertisers APRIL 2019

AALAS 207 ESLAV 210 PFI Systems Ltd 122 Altromin International OBC AnLab Ltd 204 FELASA 202 SECAL 211 AVID plc 120 Fine Science Tools GmbH 112 Special Diets Services 203 ssniff Spezialdiðten GmbH 119 Bell Isolation Systems Ltd 116 Laboratory Animals Ltd (LAL) 206 Laboratory of Pharmacology and Toxicology Tecniplast SpA IBC CEDS 204 GmbH & Co KG 115 Transnetyx Inc 111 Charles River Laboratories IFC LBS 118 Vet-Tech Solutions Ltd 124 Datesand Ltd 123 Marshall BioResources 117 ZOONLAB 121 *compared to previous generation Tunnel GP Your global operating experts in lab animal diets.

$VDQΖ62FHUWLȴHGFRPSDQ\ ZLWKRYHU\HDUVRIH[SHULHQFH ZHR΍HU\RX VWDQGDUGL]HGDQLPDOQXWULWLRQDOVROXWLRQV DQGFXVWRPL]HGVSHFLDOGLHWV

altromin Spezialfutter GmbH & Co. KG Im Seelenkamp 20 D- 32791 Lage Made in Germany www.altromin.com International