review open access anaesthesia and physiological

Tremoleda et al. EJNMMI Research 2012, 2:44http://www.ejnmmires.com/content/2/1/44

REVIEW Open Access

Anaesthesia and physiological monitoring duringin vivo imaging of laboratory rodents:considerations on experimental outcomesand animal welfareJordi L Tremoleda1*, Angela Kerton2 and Willy Gsell1

Abstract

The implementation of imaging technologies has dramatically increased the efficiency of preclinical studies,enabling a powerful, non-invasive and clinically translatable way for monitoring disease progression in real time andtesting new therapies. The ability to image live animals is one of the most important advantages of thesetechnologies. However, this also represents an important challenge as, in contrast to human studies, imaging ofanimals generally requires anaesthesia to restrain the animals and their gross motion. Anaesthetic agents have aprofound effect on the physiology of the animal and may thereby confound the image data acquired. It istherefore necessary to select the appropriate anaesthetic regime and to implement suitable systems for monitoringanaesthetised animals during image acquisition. In addition, repeated anaesthesia required for longitudinal studies,the exposure of ionising radiations and the use of contrast agents and/or imaging biomarkers may also haveconsequences on the physiology of the animal and its response to anaesthesia, which need to be considered whilemonitoring the animals during imaging studies. We will review the anaesthesia protocols and monitoring systemscommonly used during imaging of laboratory rodents. A variety of imaging modalities are used for imagingrodents, including magnetic resonance imaging, computed tomography, positron emission tomography, singlephoton emission computed tomography, high frequency ultrasound and optical imaging techniques such asbioluminescence and fluorescence imaging. While all these modalities are implemented for non-invasive in vivoimaging, there are certain differences in terms of animal handling and preparation, how the monitoring systems areimplemented and, importantly, how the imaging procedures themselves can affect mammalian physiology. Themost important and critical adverse effects of anaesthetic agents are depression of respiration, cardiovascularsystem disruption and thermoregulation. When anaesthetising rodents, one must carefully consider if these adverseeffects occur at the therapeutic dose required for anaesthesia, if they are likely to affect the image acquisitions and,importantly, if they compromise the well-being of the animals. We will review how these challenges can besuccessfully addressed through an appropriate understanding of anaesthetic protocols and the implementation ofadequate physiological monitoring systems.

Keywords: Preclinical imaging, Anaesthesia, Physiological monitoring

* Correspondence: [email protected] Imaging Centre (BIC), Medical Research Council (MRC) ClinicalScience Centre, Imperial College London, Hammersmith Campus, CyclotronBuilding, Du Cane Road, London, W12 0NN, UKFull list of author information is available at the end of the article

© 2012 Tremoleda et al.; licensee Springer. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproductionin any medium, provided the original work is properly cited.

mailto:[email protected]

http://creativecommons.org/licenses/by/2.0

Tremoleda et al. EJNMMI Research 2012, 2:44 of 23http://www.ejnmmires.com/content/2/1/44

ReviewIntroductionThe use of imaging technologies is increasing in biomed-ical research due to their great scope for non-invasivelystudying biochemical and biological processes in the liv-ing animal. Their application represents a major impacton the refinement of in vivo studies in animal models, inparticular for allowing longitudinal monitoring of theonset and the progression of disease within the sameanimal, and studying the biological effects of drug candi-dates and their therapeutic effectiveness. They provide avery useful set of tools for a more rapid, efficacious andcost-effective use and characterisation of animal diseasemodels, with great potential for translational research.Indeed, small animal imaging is extensively used as apreclinical experimental tool in many animal models ofhuman diseases including cardiovascular [1], neurode-generative [2] and musculoskeletal disorders [3] andcancer studies [4]. More recently, it has been applied forthe development of stem cell-based therapies [5,6].There have also been advances wherein researchers areincreasingly using imaging for phenotyping and charac-terisation of transgenic disease models [7].The main challenge for in vivo imaging remains the

‘biological motion’ not only regarding the physical re-strain, but also the respiratory and cardiac activitiesaffecting the quality of the images. In contrast to humanstudies, imaging of small animals generally requires an-aesthesia which helps to restrain the animals and theirgross motion, but there is still the need to control car-diac and respiratory motion. During anaesthesia, there isan inevitable autonomic nervous system depressionwhich induces cardiovascular and respiratory depressionand induces hypothermia. In addition, other conditionssuch as repeated anaesthesia required for longitudinalstudies, the exposure of ionising radiations and the useof contrast agents and/or imaging biomarkers will alsohave consequences on the physiology of the animal andtherefore will need considerations. All these factors willhave a profound effect on the animal’s homeostasis andmay thereby confound the image quality and interpreta-tions. Therefore, it is important to implement adequatenon-invasive monitoring systems appropriately suitedfor the different imaging modalities.Anaesthesia plays a key role in animal imaging, and

thus, investigators who are planning imaging experi-ments are required to be familiarised with the multitudeof anaesthesia protocols commonly used for imagingand the physiological monitoring systems available. Vari-ous imaging modalities are used for small animal im-aging, including magnetic resonance imaging (MRI),computed tomography (CT), positron emission tomog-raphy (PET), single photon emission computed tomog-raphy (SPECT), high frequency ultrasound and optical

imaging techniques such as bioluminescence and fluor-escence imaging. While all these modalities are imple-mented for in vivo imaging, there are certain differencesin terms of animal handling and preparation for imaging,how the monitoring systems are implemented and, im-portantly, how the imaging procedures themselves canaffect mammalian physiology.We will review the anaesthesia protocols commonly

used in laboratory animals during in vivo imaging, pro-viding a brief description of the most commonly usedimaging modalities, discussing different anaesthesiaprotocols and general procedures associated with thepreparation of the animal (e.g. transport of animals toimaging facility, acclimatisation period, fasting and diet,administration of any contrast agent) as well as review-ing the challenges associated with the implementationof these physiological monitoring systems during pre-clinical imaging and how this can be addressed to min-imise the impact on the well-being of the animals andits effects on image acquisition and outcome of thestudies.

Imaging modalities: considerations for animal handlingA number of imaging modalities are available for pre-clinical research, all of them providing a common ad-vantage of allowing longitudinal non-invasive serialimaging studies within the same animals and ability toinvestigate any anatomical-structural and/or functionalalterations within the tissue/organs of the animal. How-ever, it is important to understand how the implemen-tation of these techniques may influence the animal’sphysiology and its handling during imaging. We willinitially revise the particularities of each modality onthe impact of animal handling and its physiologicalmonitoring (for detailed comparison on preclinical im-aging technologies, see reviews from Koba et al. [8] andKagadis et al. [9]).CT is an X-ray-based imaging modality in which an X-

ray source and a detector mounted on a gantry rotatearound the specimen to be imaged. The animal isexposed to multiple X-rays through different angles toproduce three-dimensional (3D) images of the anatom-ical structures [10]. Currently available preclinical CTscanners can achieve high resolution with an isotropicvoxel size of as low as a few micrometres (down to 5 to10 μm), providing good contrast images for mineralisedtissues (bone), but it is poor for imaging other non-mineralised tissues. However, the use of contrast agentsthat are based on heavy elements such as iodine or bar-ium allows good imaging enhancement of the differentsoft-tissue anatomical compartments [11].CT technology remains extensively used in ortho-

paedic research and also for characterising anatomicalphenotypes in transgenic animals. It is also used for co-


registering anatomical imaging with functional data fromPET/SPECT modalities and for attenuation correctionfor these radionuclide-based techniques.One of the main challenges of this technology is to

achieve a reasonable resolution and sensitivity in smallfield of view within a short acquisition time, accountingfor time that animal is required to be under anaesthesia,and also to minimise radiation exposure. Most of theacquisitions should not require high radiation; however,levels could become critical during high-resolution im-aging with higher energy X-rays, longer acquisition timesand higher frequency imaging. Small animal CT causesradiation doses ranging from 70 to 400 mGy [12]. Dosesbetween 6.5 to 7 Gy are lethal to mice, but even lowerdoses have shown biological effects (e.g. stimulationDNA repair, free-radical detoxification). Therefore,researchers should be aware of the risk of interferencedue to radiation during imaging.The speed of acquisition is also important when con-

trast agents are used since most of these agents are rap-idly cleared away from the bloodstream by the kidneysas rodents typically have a high heart rate (heart rate ofmice ranges between 400 and 600 bpm; in rats, between250 to 400 bpm compared with 60 to 80 bpm for adulthumans) and a very short circulation time (approxi-mately 10 s in mice compared with approximately 30 sin humans). Animal imaging technology is alreadyimplemented with gating-based acquisition systems, inwhich image acquisition is acquired simultaneously andtriggered at some predefined physiological signals orprocessed post-acquisition. This allows controlling forthe interference effects due to the physiological move-ment created through the cardiac and respiratory cycleand has remarkably improved the quality of the imagesacquired. Ongoing challenges are focusing on the devel-opment of ultrafast acquisitions and further improve-ments on detector sensitivity. The benefits of applyinggating during image acquisitions have been well reportedin micro-CT imaging of rodents [13].MRI is a non-ionising 3D imaging technique that has

advantages over other methods that use ionising radi-ation such as CT, SPECT and PET, especially for serialimaging. This technology uses strong magnetic field toalign the spins of hydrogen nuclei within the tissues andthen the application of pulses of radio waves to system-atically alter this alignment, causing the hydrogen nucleito produce a rotating magnetic field that is detectable bythe scanner. This signal is then manipulated by add-itional magnetic fields to be digitalized and to build upenough information to construct an image of the tar-geted area of the body [14]. Nowadays, preclinical sys-tems can routinely achieve a resolution of 100 μm in alldimensions in living animals, providing high-quality ana-tomical imaging with good structural detail. Moreover,

this technology can also provide information on chem-ical composition (spectroscopy) and other physiologicalfunctional parameters such as blood flow velocity [15].These techniques are extensively used for anatomical,functional and physiological characterisations of tissues/organs in preclinical models.The ongoing challenges for preclinical MRI are also

related to the relative small anatomical size of rodentscompare to humans; for a mouse image to retain thesame anatomical definition as that achieve in humanimages, the acquisition must be with a voxel volume(3D imaging pixel unit volume) approximately 3,000times smaller than that of a human and somehow com-pensating for the signal deficit. The two most import-ant ways to overcome these problems are by optimisingthe radio-frequency coil that surrounds the sample andelicits and receives the nuclear magnetic signal and byemploying longer acquisition times. Another method toincrease the signal is to work with higher magneticfields, and generally, the preclinical systems are work-ing on the range of 7 to 9.4 T, commonly going up to11.7 T [14].The handling of animals and their anaesthesia regimes

becomes challenging when carrying out functional MRI,in which the response to a stimulus is assessed throughalterations in the animal physiology. In such studies, it iscrucial to maintain the animal within stable and narrowwell-defined physiologic parameters to allow for the crit-ical detection of a physiologic response associated to thestimulus. Anaesthesia can markedly affect such func-tional imaging procedures by affecting the blood flow,blood oxygenation levels and cardiac and respiratoryfunctions [16]. The need to use longer imaging times forsome MRI acquisitions represents a major challenge forthese studies, and researchers must ensure that adequatephysiological monitoring is supported with gating sys-tems to minimise biological motion effects and that theanimal’s homeostasis through the scanning period ismaintained. Because of the high magnetic field, monitor-ing equipment must be specifically built-in without con-taining any ferromagnetic material. There are fewsuppliers that provide MRI-compatible equipment in-cluding electrocardiogram (ECG), temperature, bloodpressure and respiration monitoring equipment [17,18].All the anaesthesia and monitoring equipment usedshould not emit radio frequencies that interfere with thescan. MRI-compatible systems using fibre-optic orcarbon fibre cabling avoid this problem. Main powersupplies can also carry interference through the radio-frequency screen; therefore, monitoring equipmentshould use an adequately filtered and isolated powersource or be battery powered [19].PET and SPECT rely on the detection of photons

emitted from radiolabelled tracers in the body. SPECT


systems record gamma rays directly after radionuclideemission through collimated detectors rotating aroundthe animal. Most preclinical SPECT scanners areequipped with multi-pinhole collimators which allowacquisitions with higher spatial resolution and sensitivity.A tomographic data reconstruction is applied, yielding a3D dataset that can then be manipulated to show anyparticular axis of the body [20]. PET systems also detectgamma rays. The radionuclides emit positrons whichrapidly annihilate with an electron in the tissue whileemitting two photons in opposite directions. The twosimultaneously released gamma rays are registered ascoincidences by external detectors placed around thesubject providing good sensitivity for the detectionof the biotracer with a resolution within the range of 1to 2 mm.SPECT systems use radiopharmaceuticals that have a

longer half-life and are widely used in clinics, being morereadily available and less costly than PET tracers, whichrequire a cyclotron for their production [21]. Neverthe-less, the high sensitivity of PET tracers and their integra-tion as biomarkers with multi-capability applicationsmake them very well suited for small animal imagingand tracers. Examples of currently used radiomarkers are18 F-FDG, an analogue of glucose, and 18 F-FLT, a thymi-dine analogue, which are extensively used as a biomarkerof tissue metabolic activity and inflammation and cellproliferation, respectively. These techniques providegreat potential to investigate physiological functions,track metabolic processes and quantify receptor density.The main limitations of the SPECT preclinical systems

include lower sensitivity and spatial resolution. The PETsystems provide higher sensitivity, with a greater abilityto quantify the tissue concentration of the tracer; how-ever, the tracers have very short half-life and, as with theSPECT systems, do not provide anatomical informationof the tissue/organ images. The advantage of both tech-niques is the high sensitivity in biological systems downto the picomolar range.Imaging acquisitions may take up to 2 h, requiring ad-

equate physiological monitoring to maintain goodhomeostasis which is crucial for the assessment of func-tional/metabolic parameters measured through theradiotracers. Temperature monitoring remains criticalduring PET/SPECT acquisitions as the body temperaturestrongly confounds the body’s metabolic functions. It isimportant to note that vascular access (e.g. cannulationtail vein) is required to inject the tracer through the gen-eral circulation to ensure its distribution through the tis-sues/organs during imaging. Furthermore, to investigatethe kinetics of the tracer uptake, serial blood sampling isrequired for extracting an accurate arterial input func-tion of the tracer used and also for assessing theintegrity of the parent radiolabelled tracer. Such

measurement poses a significant challenge in mice be-cause blood volume is smaller and the heart rate is muchhigher in mice than in larger animals or humans. Severaltechniques have been proposed such as the use of beta-probe measurements, microblood sampling, arterial-venous shunts, factor analysis of dynamic structures andimage-based measurement from a left ventricular regionof interest [22]. The arterial blood sampling in smallamounts at timed intervals is often considered the refer-ence standard for true blood activity determination.Animals may get exposed to radiation when imaged

with PET or SPECT. The radiation exposure will de-pend on the dosage of the tracer injected. The tissueactivity concentration in laboratory rodents is rela-tively higher than in humans, resulting in higher radi-ation dose. However, the large surface area comparedto the small body volume of the animals allows for alarger escape of radiation. PET tracer doses typicallyrange from 18.5 to 74 MBq (0.5 to 2.0 mCi) for ratsand 1.85 to 7.4 MBq (50 to 200 μCi) for mice. There isa large variation in organ exposure related to theradionuclide, biodistribution and clearance of thecompound.Preclinical optical imaging includes bioluminescence

and fluorescence imaging [23]. Fluorescence imaginguses external dyes or fluorescent markers which emitphotons after excitation. Fluorescent probes are exten-sively used and can be found expressed in reporter pro-teins (GFP, RFP), in microspheres or as dyes that arewidely used for labelling cells or for monitoring geneexpressions. Bioluminescence relies on the productionand emission of light that resulted from an enzymatic re-action between the luciferase enzyme and its substrate,luciferin, to produce light. The principal advantage ofin vivo bioluminescence is that the equipment is highlysensitive and allows the direct detection of low emis-sions of light, with very little background, yielding a highsignal-to-noise ratio. However, the technique requiresthe animal to be injected with the substrate. Fluores-cence imaging involves no additional animal treatmentbut instead requires an external excitation source andmultiple filters to obtain a spectrum of the light emittedwhich can compromise the sensitivity as there is a rela-tively high level of background-emitted light. Reductionof the autofluorescence is important to enable detectionof weak fluorescent signals, increasing signal sensitivity.This is generally achieved by minimising skin fluores-cence using nude animals, avoiding pigmented rodentsand removing the fur around the area of interest. Re-cently, this has been minimised by the use of infraredspectrum fluorophores that guarantees acquisition withless background autofluorescence and also greater tissuepenetration and sensitivity. These techniques are exten-sively used for monitoring the expression of transgenes,


tumour growth and metastasis, infections and gene ther-apy assessments.The advantage of these technologies is that they use

low-energy, non-ionising radiation with high sensitivitydown to microns and subpicomolar concentrations.Moreover, the acquisition times are short, thus minimis-ing the time that animals are maintained under anaes-thesia. The main challenges rely on the properties of thelight sources, their wavelength, diffusion and absorption,which influence the penetration depth of the light intotissues, as well as the resolution. Also, this technologyhas no direct equivalent in the clinics, which compro-mises the translation of preclinical data.High frequency ultrasound is based on the propaga-

tion of sound waves through the soft tissues that arebeing imaged. The preclinical systems are dedicatedultrasound systems that use high frequency modes (20to 50 MHz) to provide good spatial resolution and ad-equate penetration for anatomical, functional andhemodynamic real-time information on preclinical mod-els. These systems are relatively inexpensive and providereal-time imaging but are limited by the tissue depth(maximum penetration is around 15 mm) and the struc-tures (artefacts caused by bone or air) that can bereached and visualised. This technique is extensivelyused in cardiovascular research, prenatal developmentand for monitoring tumour growth and metastasis pro-gression. Acquisitions are generally quick, but it is im-portant to continue monitoring body temperature,especially in rodents, to minimise any confoundingeffects on the cardiovascular function.

Animal preparation for imaging proceduresDifferent procedures will be required depending on theimaging modality that will be used, the tissue/organs tobe imaged and the in vivo experimental protocolrequired (e.g. fasting, artery/vein cannulation, ECGprobes, tracheal intubation). All these procedures will in-fluence the physiological parameters in the animals andwill have to be accounted for when carrying out in vivoimaging.

Transport of animals and acclimatisationAnimals will have to be transported to the imaging facil-ity. Such facilities should have an animal housing area toallow for the acclimatisation of the animals beforeimaging and also for monitoring their recovery after theexperimental procedure. Not all the facilities are neces-sarily fully integrated within a centralised animal unit,and therefore, it is important that researchers are awareof the requirements for transporting the animals withinthe imaging facilities. These include (1) health screen-ings to minimise potential disease transfer and (2) accli-matisation periods to reduce transport stress-related

responses for the animals. Furthermore, it is also import-ant that animals are housed in an area near but sepa-rated from a procedure room in which all the animalpreparation for imaging (e.g. vein cannulation) is carriedout, as recommended by many codes of practice for thehousing and care of animals used in research.The standard recovery period after transport for

rodents varies between 2 and 7 days, depending on theprocedure and transport period. In general, it has beenreported that there was an increase of glucocorticoidconcentrations for up to 2 days, decrease in body weightand immune response suppression of up to 48 h aftertransport in rodents [24]. Similarly, a recovery periodshould also be considered after imaging, depending onthe type of anaesthetic, the length of anaesthesia and ex-perimental procedures carried out (e.g. significant bloodvolume withdrawal, surgically induced procedure, bonemarrow reconstitution).

Health checksThe health and general conditions of the animal willaffect its physiological response to the anaesthesia andthe outcome of the imaging procedures. Researchersmust be aware that rodents are prey species, so theymay hide or compensate for distress and/or disease con-ditions to such extent as they appear to be healthily nor-mal. It is important that, prior to imaging and inductionof anaesthesia, animals are carefully evaluated for goodhealth. Wherever possible, within the constraints of thestudy, animals in good general condition should be used.This is also important for good experimental study de-sign as it minimises the variation between individuals,leading to more accurate, reproducible results and thusreduction in animal numbers used.The pre-anaesthetic exam should contain but is not

limited to confirmation of animal’s identification, sex,age, body weight, body condition, skin condition, esti-mate hydration, colour of mucous membranes, heartrate and rhythm, respiratory rate, signs of diarrhoea,body temperature, evidence of normal food and waterconsumption and normal production of urine and faecesin the cage. Overall, the body condition of the animal isvery important to anaesthesia induction, maintenanceand recovery (e.g. very obese animals may respondslowly to drugs).

Food withdrawalRodents do not require compulsory fasting prior to an-aesthesia as they do not possess a vomiting reflex. How-ever, fasting may be required for some imagingprocedures. Fasting can be an effective way to ensureuniformity of PET imaging with 18 F-FDG (Figure 1), bydecreasing the blood glucose levels associated with foodintake and thus allowing for a more controlled

Figure 1 Example of PET/CT acquisitions showing fasting effects on biodistribution of 18 F-FDG-PET tracer in C57BL/6 mice. (A, B)Maximum intensity projection, and (A′, B′) two-dimensional coronal view. The fasted animals displayed a more targeted and selective uptake ofthe glucose analogue tracer throughout the body, avoiding the general uptake throughout the whole digestive system due to food ingestion.


assessment of metabolic rate of glucose throughout thebody. In rodents, fasting times of up to 6 h are effectiveto clear the stomach food [25], but longer fasting periodswill have important detrimental side effects in the ani-mal conditions including loss of body mass and decreaseof blood glucose and fatty acid levels [26]. Generally,researchers may fast the animals overnight, removingthe food on the evening prior to the imaging day. How-ever, Levine and Saltzmann [27] observed loss of bodyweight, depletion of liver glycogen, decrease in bloodglucose and loss of amino acids in rats that were starvedovernight and reported that feeding sugar overnightmaintained metabolic homeostasis in rats and is prefer-able to overnight starvation. If anaesthesia is requiredfor a pregnant animal, the need to withhold food shouldbe carefully evaluated.Rodent food consumption is associated with the ani-

mals’ circadian cycle, and thus, the removal of food over-night (dark hours) will have a stronger impact on theircaloric intake as compared with that for light hours. It isimportant to control the length and timing of fasting be-fore imaging studies and to ensure that experimental

protocols are well standardised to ensure the reproduci-bility between imaging studies. Similarly, it is also veryimportant to consider the effects of fasting if any specificdietary requirements are being used and how this caninfluence the outcome of the imaging studies (e.g. somediets can induce strong background autofluorescence inmice [28]).

PremedicationPremedication refers to the administration of any drugsin the period prior to induction of anaesthesia. A widevariety of drugs can be used during the preparation ofanimals for imaging studies include the following: anti-cholinergic drugs which are used to decrease oral andrespiratory secretions, maintain the heart rate and de-crease gut motility; tranquillisers which are used to re-lieve anxiety, produce calmness, aid restraint, assist witha calm postoperative recovery and reduce dose of anaes-thetic needed; narcotics which are used to sedate an ani-mal. They reduce the amount of anaesthesia needed,induce smooth recovery and provide postoperative painrelief. Other analgesics that may be given are non-


steroidal anti-inflammatory drugs, such as carprofen, toprovide pain relief for up to 24 h after the procedure ifminor surgery has been performed.

AnaesthesiaAnaesthesia is generally required during imaging experi-ments in order to ensure humane restrain of the ani-mals. Anaesthesia is described as a state of lack ofawareness of a part or the whole body of the animalinduced by the administration of specific drugs that de-press nervous tissue activity. It involves a triad effect in-cluding analgesia (pain relief ), loss of consciousness andimmobilisation (muscle relaxation) [29]. While most ofthe imaging procedures are non-invasive, it is very im-portant to ensure that anaesthetised animals remain in astable physiological state with consistent cardiovascularoutput, body temperature and blood oxygenation. In-deed, many studies (e.g. functional MRI) are highly sen-sitive to these parameters. Therefore, it is very importantto address the physiological effects that different anaes-thetic regimes are likely to have during imaging.In order to achieve the effects sought for anaesthesia

in rodents, there is an inevitable autonomic nervous sys-tem depression which induces cardiovascular depression(e.g. reduction in cardiac output, blood pressure, alteredblood flow) and respiratory depression (e.g. hypoxia,hypercapnia, related acidosis) and induces hypothermia,affecting body metabolism. Moreover, rodents are preyspecies with small body size, high body surface/weightratio and high metabolic rate which compromises thepharmacological efficacy of injectable agents and theirbody temperature regulation. Consequently, high dosesof agents are required to induce unconsciousness, whichhas also a detrimental effect on autonomic depressions.In this vein, the use of inhalation anaesthesia in rodentsallows these adverse effects to be more controlled due totheir rapid onset and recovery time and low metabolism.Hence, it is highly recommended to use inhalation an-aesthesia for imaging purposes.It is also important to appreciate the variation in re-

sponse to anaesthetics between different animal strains.Therefore, it is important to reassure and adjust the an-aesthesia protocol to the particular needs of the strainand experimental set-up.

Anaesthetic regimes and equipmentInjectable and inhaled anaesthetics are commonly usedin rodents. The latter method is most suitable for im-aging; due to its rapid onset and recovery times, it allowsa much greater control of the anaesthetic doses andmaintenance times of anaesthesia than when usinginjectables. Moreover, inhaled agents are eliminatedquicker via the lungs, whereas injectable agents need tobe metabolised by the liver and excreted by the kidneys

[30]. This allows the animals to recover quicker, which isimportant in regaining normal physiology to controlpost-operative hypothermia and fluid or electrolyte im-balance. Some of the newer injectable agents have spe-cific reversal agents which speed recovery and help toovercome many of these potential problems.If injectable agents are used for long-term anaesthesia,

as can be the case in MRI acquisitions, there is an in-creasing risk of developing hypoxia, especially if oxygensupplementation is not provided, which could lead to re-spiratory depression and hypercapnia and acidosis ifprolonged.Injectable anaesthetics. Anaesthetic dose rates for in-

jectable agents will depend on species used, administra-tion route, age, sex, strain, body condition, environment,experimental set-up, previous drug treatments and thelevel of anaesthesia required. During the initial period ofuse, it is important to monitor animals closely and makeany adjustments necessary for future use.Increasing the length of anaesthesia with a given drug

may require a second or further injection at intervalsduring the procedure. However, increasing the initialdose will result in an increase in depth at the time ofpeak action with the risk of reaching a point of an over-dose. Also, giving intermittent top-up doses of the drugwill cause the depth of anaesthesia to vary considerably.This may be overcome by administering it as a continu-ous infusion so that steady plasma concentrations of theanaesthetic are maintained. However, this can be chal-lenging depending on the pharmacodynamics of the an-aesthetic agent and the experimental set-up. In general,repeated doses will have progressively greater effect and,in addition to extending the duration of surgical anaes-thesia, can prolong the sleeping/recovery time followinganaesthesia. This is particularly critical when usingopioids. If the animal does eventually wake up, the re-sidual effects of the drug may persist for a long period oftime [31].The two anaesthesia combinations that are widely used

in laboratory rodents are fentanyl-based and ketamine-based. Agents available are as follows:

1. Fentanyl/fluanisone (Hypnorm™; mostly incombination with benzodiazepine). Hypnorm™

(tradename) consists of two components: fentanyland fluanisone. Fentanyl is a potent, short-term,opioid agonist analgesic. Fluanisone is abutyrophenone tranquilliser and is used to suppresssome of the undesirable effects of the narcotic(fentanyl) such as vomiting and excitement. Thisneuroleptanalgesic combination induces moderaterespiratory depression and a poor degree of musclerelaxation. Hypotension and bradycardia may also benoted.


Fentanyl/fluanisone (Hypnorm™, Vetapharma, Leeds,UK) is licensed for sedation in rodents, rabbits andguinea pigs. As there is poor induction of musclerelaxation, it cannot be used on its own for surgicalprocedures. Fentanyl/fluanisone in combination withbenzodiazepine (midazolam or diazepam) is licensedfor surgical anaesthesia in rodents, rabbits andguinea pigs. This combination reduces the dose ofHypnorm™ by 50% to 70%, and benzodiazepineproduces good muscle relaxation. The mostcommonly used benzodiazepine tranquilliser ismidazolam (Hypnovel™) as it is water soluble.Depending on the animal’s depth and length ofanaesthesia, a ‘top-up’ administration may benecessary. While the effects of midazolam ordiazepam can last for several hours, anaestheticduration is lengthened in the first instance by givingadditional Hypnorm™. However, due toenterohepatic recirculation of the metabolisedanaesthetic products, a relapse of the anaesthesiaeffects may occur, which means that Hypnorm™/Hypnovel™ combinations can have a prolongedrecovery time or ‘sleep’ time (average >4 h forrodents). Monitoring must continue during thisperiod; the body temperature, maintained. Thiscombination also appears to sensitise rodents toauditory stimuli.

2. Ketamine. Ketamine is one of the most commonlyused anaesthetics in the veterinary field, and thecommercially available injectable versions ofketamine hydrochloride are known as Vetalar™ orKetaset™. Ketamine produces a state of ‘dissociativeanaesthesia’ in which there is profound analgesia,light sedation and muscle rigidity (stage ofcatalepsy). It does not depress the central nervoussystem (CNS), and thus, reflexes remain intact.Some side effects are as follows: eyes remain open,and ophthalmic ointment is recommended; presenceof spontaneous movements and muscles are tensewhich causes an initial increase in blood pressurealthough there is also a peripheral vasoconstriction.In contrast to other anaesthetics, ketamine does notdepress respiration or cardiac output [32].Used on its own, ketamine does not achieve surgicalanaesthesia but is extremely useful whenadministered in combination with xylazine,medetomidine or diazepam for the production ofsurgical anaesthesia. Ketamine is contraindicated foruse in animals with renal or hepatic disease [33].During ketamine recovery, animals are hyper-responsive and ataxic. Recovery may also beassociated with behaviour alterations [34].Combinations of ketamine for general anaesthesiaare as follows:

a) Ketamine + alpha-2 adrenergic agonist sedatives(alpha 2s): medetomidine (DomitorW, PfizerAnimal Health, Tadworth, UK) or xylazine(RompunW, Animal Bayer Health, Newbury, UK).Ketamine/alpha-2 combinations are suitable forsurgical anaesthesia but are significantlyhypotensive and can induce profound bradycardiaand respiratory depression. These combinationsprovide, on average, 30 min of surgicalanaesthesia. The sedative can be reversed usingthe alpha-2 antagonist atipamezole (AntisedanW,Pfizer Animal Health), but this should not begiven until the ketamine has worn off, at least 30min after injection. Atipamezole is a highlyspecific antagonist for medetomidine but hassome effect in reversing xylazine as well. Notethat the combination of ketamine/medetomidine + buprenorphine as preoperativeanalgesia has proved toxic in rats [35], andcaution is advised when considering this regime,but this is strain dependent. This adverse effecthas not been observed when buprenorphine isadministered post operatively. Ketamine-xylazinecombinations may affect the brainhaemodynamics, causing a reduction of thecerebral blood flow and affecting brainoxygenation which can have confounding effectson nuclear magnetic resonance perfusion imaging[36]. Also, ketamine combinations with xylazinehave substantial cardiovascular effects, manifestedby low pulse rates and hypotension [37].

b) Ketamine and benzodiazepine combinations.Ketamine can also be combined withbenzodiazepines, such as midazolam, to provide20 to 30 min of light anaesthesia in rodents. Thedepth of anaesthesia may only be sufficient topermit minor procedures to be performed butcould be used for immobilisation. The degree ofrespiratory and cardiovascular depression is lessthan when ketamine is combined with an alpha-2agonist.

3. Alfaxalone (Alfaxan). Alfaxalone (AlfaxanW

Vetoquinol UK Ltd, Buckingham, UK) is ananaesthetic steroid. The fast metabolism of this drugenables it to be used for long periods of anaesthesiaby continuous infusion in rodents and maintain afairly rapid recovery following the lastadministration, but it has to be administeredintravenously (i.v.) in rodents [38]. Renal and hepaticperfusion and respiratory function is wellmaintained.

4. Propofol (RapinovetW, DiprivanW). Propofol is anisopropylphenyl compound that is available for


intravenous use as a 1% solution in soybean oil,glycerol and egg phosphatide. Propofol exerts itsCNS effects via modulation of the gamma-aminobutyric acid (GABA) channels throughdifferent sites than the barbiturates, steroids orbenzodiazepines. It rapidly induces unconsciousness;recovery is more rapid and complete with minimalresidual CNS effects compared with thiopentone,holding good potential as an anaesthetic regime forfunctional MRI (fMRI) studies [39]. Its rapidelimination from the body and lack of cumulativeeffects make it particularly suitable for continuousinfusion, either alone for restraint or in combination(e.g. ketamine, an opioid or isoflurane) for surgicalanaesthesia. However, it does not induce analgesia,and thus, analgesics should be considered whenpainful manipulations or procedures are performed.Cerebral blood flow, perfusion pressure andintracranial pressure decrease following propofoladministration [40]. It is a potent respiratorydepressant, and apnoea is common on inductionunless the drug is given slowly.

5. Barbiturates. Barbiturates modulate GABAtransmission, the most common inhibitorytransmitter in the mammalian nervous system,inducing depression activity in the reticularformation (necessary for maintenance ofwakefulness). In addition, barbiturates selectivelydepress transmission at the sympathetic ganglia,which may contribute to decreased blood pressurefollowing their administration. These agents arehighly metabolised by the liver, and metabolitesaccumulate in the body over time. Many of theseagents will produce sedation associated with severerespiratory depression and general anaesthesia aslarger doses are administered. At high doses, theseagents are utilised for euthanasia. Somebarbiturates are caustic substances (high alkalinepH) when injected into the living tissue and somust be given intravenously or diluted and givenintraperitoneally (i.p.) (only for pentobarbitone).Subcutaneous or intra-muscular routes should beavoided:

a) Pentobarbitone. This is a barbiturate with amedium length of action that can be given i.p. ori.v. in a range of species. Its main effect is one ofhypnosis, with poor analgesia and musclerelaxation except at near lethal dose rates. Thereis profound cardiovascular and respiratorydepression particularly in rats. As pentobarbitoneis highly metabolised by the liver and hascumulative effects, it is not ideal for long-termanaesthesia.

b) Thiopentone and methohexitone. Short-actingbarbiturates are commonly given intravenously;they have poor analgesic effects yet reasonablemuscle relaxation. Respiratory function isdepressed and may be temporarily suspendedduring induction. Recovery from single doses ofthiopentone and methohexital is due toredistribution of the drug from brain tonon-nervous tissues (fat, primarily viscera andskeletal muscle). In the case of methohexital,rapid hepatic metabolism contributes remarkablyto recovery. It induces a transient small decreasein arterial blood pressure that is compensatedfor by an increase in heart rate. Myocardialdepression is minimal and far less than wouldoccur with volatile inhalants. Thiopentalanaesthesia has minimal effect on theneurochemical profile in the rat brain butsubstantially increases brain glucose content inthe cortex as detected by in vivo 1H MRS [41].Thiopentone acts as an irritant outside the veinand can induce perivascular necrosis.Methohexitone has a shorter half-life and arapid metabolism, so it is suitable for long-terminfusion. It is not an irritant outside the vein,but tremors can occur if given withoutpremedication.

6. Miscellaneous. The following are the miscellaneousagents:

a) Chloral derivatives (e.g. chloral hydrate,α-chloralose)

� Chloral hydrate is a reliable sedative hypnotic,but it has poor analgesic properties.Consequently, it has frequently been used incombination with some other drugs such asmagnesium sulphate when general anaesthesiawas the desired endpoint [42]. Chloral hydrateis medium acting (1 to 2 h). Intraperitonealadministration of chloral hydrate to rats isassociated with paralytic ileus. For this reason,it is recommended that this agent be only usedfor non-recovery procedures [43].

� α-Chloralose is a hypnotic frequently used inneuroscience experiments [44]. This drug hasno analgesic activity but can be combined withlocal or regional anaesthesia (novocaine,marcaine, bupivicaine), to produce sufficienttranquilisation and anaesthesia to allow minorsurgical procedures. The duration of action ofα-chloralose is prolonged (8 to 10 h), makingit a possible choice for long-term anaesthesia.


Since this drug is probably a hypnotic ratherthan a true anaesthetic, with unprovenanalgesic potency, it appears that the mostethically and scientifically acceptable use ofthis agent in rodents is to provide long-lastinganaesthesia for procedures involving no painfulsurgical intervention. The primary advantageof this drug is the minimal cardiopulmonarydepression seen at the normal doses (highdoses can cause severe respiratory depression).However, it is very irritating to the GI tract,causing a dynamic ileus if given i.p. and ulcersif given orally. Therefore i.v. use is the onlyroute recommended. This drug should not beused if any other alternative is available. α-chloralose has been used for fMRI studiesproviding consistent and reproducible bloodoxygen level dependent (BOLD) signals, but itis important to monitor the animal closelyduring recovery of anaesthesia [45].

b) Urethane. Urethane produces long periods (8 to10 h) of anaesthesia, has a wide safety marginand has little effect on normal blood pressureand respiration. It produces sufficient analgesiato allow surgical manipulations [46]. However,the drug should be handled with ‘extreme care’as it is considered to be cytotoxic, carcinogenicand immunosuppressive. Due to the potentialhazards to staff, other anaesthetics should beassessed and an alternative regimen usedwherever possible. Because of its carcinogeniceffects urethane should only be used inanimals that will not recover followinganaesthesia.

c) Tribromoethanol (AvertinW). Tribromoethanol iswidely used as a short anaesthetic agent for embryo-transfer surgery for the generation of transgenicmice [47]. Potential side effects such as localirritation, fibrous adhesions in the abdominal cavityand mortalities have been reported.Tribromoethanol has many features of a goodanaesthetic: a wide margin of safety, rapid inductionand recovery, simple route of administration (i.p.),provides good muscle relaxation and loss of reflexactivity. However, there have been conflictingreports of toxic side effects [48]. See Table 1 forsummary of properties of injectable anaesthetics.

Inhalation anaesthesia. Inhalation anaesthesia is themethod of choice for general imaging protocols for la-boratory rodents. Highly volatile anaesthetic agents aretransported through a carrier gas to the animals via abreathing circuit with an integrated vaporizer which

allows regulation of the anaesthetic concentration andflow into the animal.Oxygen is most commonly used as the carrier gas for

the inhalation anaesthetic agent. It is supplied in pres-surised cylinders and delivered via a pressure regulatorto a flow metre that provides adequate flow rates (0.5 to1.5 l/min) through an integrated vaporiser. The lungcapacity of the animal and to a lesser extent the effi-ciency of the anaesthesia circuit used will determine thegas flow rate needed. The circuit is also necessary to re-move exhaled gases and to provide a method for assist-ing or controlling ventilation. Non-rebreathing circuitsare usually used in laboratory animals to ensure mini-mum dead space and resistance. Also, they provide aknown inspired concentration as the fresh gas inlet goesdirectly to the animal, allowing for a good regulation ofthe anaesthesia concentration in the inspired gases. Themost commonly used anaesthetic circuits for rodents arethe Bain’s co-axial T-piece facemask in which the freshgas inflow pipe runs inside the reservoir limb and theopen facemask system. One of the drawbacks of the lat-ter is that removal of waste anaesthetic gases is difficultbecause gas escapes all around the mask, and this couldresult in serious health and safety issues for the operator.

� Inhalation anaesthetic agents. Inhaled anaestheticsare volatile compounds that have specific effects onthe CNS. Different inhalation anaesthetics requiredifferent concentrations to induce and maintainanaesthesia, each agent differing in potency andefficacy. The desirable inhalant must producecomplete anaesthesia, provide a rapid inductionand recovery and be safe, non-irritant and non-explosive.

1. Isoflurane and halothane. These are the mostcommonly used inhalation anaesthetics in laboratoryanimals. These agents are broadly similar in speed ofonset/offset and potency (Table 2). Isoflurane isextensively used due to the fact that it is minimallymetabolised (<0.17%) by the liver and therefore isless toxic to the animal’s metabolism as comparedwith injectable anaesthetics. However, both agentscan be rapidly fatal if overdosed, and therefore,animals should be individually induced andtransferred to a lower maintenance percentageimmediately once they are unconscious, i.e. have losttheir righting reflex.
Isoflurane itself has become a preferred generalanaesthetic agent for cardiovascular studies becauseit causes less depression of cardiac function thaninjectable general anaesthetics [49]. However, itdecreases blood pressure by reducing peripheralresistance [50]; see Table 2 for further properties.

Table 1 Properties of main anaesthetic agents used in preclinical research: halothane and isoflurane

Agents Advantages Disadvantages Dosing

Halothane Potent anaesthetic Highly metabolised (hepatotoxic) Induction 3% to 4% Maintenance 1%to 2% (rats and mice)

High therapeutic index Cardiovascular depressant

Rapid induction andrecovery (1 to 3 min)

Moderate hypotension: reduction incardiac output and peripheralvasodilatation)

Adequate muscle relaxation Respiratory depressant

Non-irritant, non-flammablenor explosive

Halothane sensitises the heartto catecholamines (sympatheticstimulation)

Easy to vaporise

Isoflurane Similar physicalproperties to halothane

Decreases arterial blood pressure(vasodilatation)

Induction 3% to 4% (rats and mice)Maintenance: 1.5% to 2% (mice)1.5%to 2.5% (rats)

Rapid induction and recovery More expensive than halothane

Low toxicity and metabolicactivity: highly safe

Strong smell: aversive

Suitable for high frequencyand long-term anaesthesia

More potent respiratorydepressant than halothane

Minimal cardiovasculardepression

Moderate respiratorydepression

Good muscle relaxation


2. Enflurane, desflurane (I653) and sevoflurane are allused in human anaesthesia; some offer nosignificant advantages over isoflurane for themajority of procedures while being considerablymore expensive.Sevoflurane was first synthesised in the late 1960sbut was not developed for approval until 1990 inJapan because of concerns about degradation withsoda lime (to produce compound A, which canproduce renal toxicity) and release of fluoride ionduring metabolism (which can also producenephrotoxicity). Neither of these concerns has beenproven to be clinically important in the human orveterinary field. It is less soluble than isoflurane,which means that inductions and recoveries areeven faster. Sevoflurane also appears to be bettertolerated by face mask and chamber inductionbecause it has low pungency and low airwayirritability. Since sevoflurane is less potent thanisoflurane (minimum alveolar concentration forsevoflurane is 2.4% for dogs), the doses used areusually higher than those used for isoflurane.However, there have been recent reports thatisoflurane and sevoflurane provided an equallyreliable anaesthesia in laboratory mice [51].

3. Nitrous oxide. Nitrous oxide (N2O) is not potentenough for induction of anaesthesia and is,therefore, unsuitable for use alone. It is an analgesic,relaxant and, to a certain extent, tranquilliser.

Nitrous oxide has minimal cardiovascular andrespiratory effects. It is generally used incombination with other volatile agents to reduce therequired concentration of that agent, allowing forreduction in dose of other agents, reducing costs,speeding up induction and providing additionalmuscle relaxation. However, it is important to beaware that the oxygen percentage of the inspired gasis lowered, and there is a risk for ‘second gas effect’,in which after prolonged anaesthesia, oxygen can bedisplaced from the lungs by the nitrous oxide as it isbreathed off, causing diffusion hypoxia and can leadto suffocation. It is important to supplement with100% oxygen after N2O is discontinued to protectagainst diffusion hypoxia.

4. Diethyl ether. Diethyl ether is occasionally used inlaboratory animals. It is a good analgesia andpromotes muscle relaxation with minimalrespiratory and cardiovascular depression. However,it is flammable, explosive and aversive, causingnauseas and vomiting. Most units have prohibitedthe use of this agent due to serious health and safetyconcerns and the fact that superior anaestheticagents are now available.

5. Methoxyflurane. Methoxyflurane is the least volatileand the most metabolised of the inhalants. Itproduces very slow induction and recovery and isused occasionally in laboratory animals (e.g. usefulfor neonatal rodents).

Table 2 Summary of the properties of injectable anaesthetics

Anaesthetic agent Advantages Disadvantages Dosing

Fentanyl/fluanisone(Hypnorm™)-basedcombination

Good analgesic effect Cardiovascular and respiratorydepression

10 ml/kg (mouse), 2.7 ml/kg (rat) i.p. mixtureHypnorm™/Hypnovel™ (midazolam)/water mixture(1:1:2 volume) (120 to 140 min sleep time)Hypnorm™ top-up 0.3 ml/kg (mouse), 0.1ml/kg(rat) i.p. (30 to 40 min sleep time)

Sedative Poor muscle relaxation alone

Possible to top-up forlong-term anaesthesia

Risk of enterohepatic recirculation:relapse

Reversal of sedativeeffect withbuprenorphineto speeds up recovery time

Prolonged recovery time

Hypersensitivity to noise

Ketamine-basedcombination

Analgesic effects Muscle rigidity +++ unlesscombined with other agents.

Ketamine+medetomidine 75 mg/kg+ 0.5 to1 mg/kg i.p.(Ketamine + xylazine 75 to 100 mg/kg)/(10mg/kg i.p.)(60 to 120 min sleep time ) (mouseand rats)Atipamezole 1mg/kg i.p.Light sedation Increases intracranial pressure.

Wide safety margin Recovery often involved withataxia and hyper responsiveness.

Can increase blood pressure

Alfaxalone (Alfaxan) Minimal respiratory orCVS depression

Administration route i.v. (rodents,cats) or i.m. (primates)

15 to 20mg/kg (mouse), 10 to 12mg/kg (rat) i.v.(10 to 15 min sleep time after bolus)0.25 to0.75 mg/kg/min i.v. infusion (long term)

Rapidly metabolised,repeated doses donot accumulate.

Suitable for long-termanaesthesia in rodents

Propofol(RapinovetW,DiprivanW)

Rapidly metabolised,continuous infusion possiblefor long-term anaesthesia.

i.v. use only 26 mg/kg (mouse), 10 to 12mg/kg (rat) i.v.(10 to15 min sleep time after bolus)2 to 2.5 mg/kg/mini.v. infusion (long-term, mouse) 0.5 to 1 mg/kg/mini.v. infusion (long-term, rat)

No analgesic properties

Rapid recovery Severe respiratory depression

Can be used in animalswith hepatic orrenal impairment.

Apnoea can occur after i.v. bolus

Barbituratesproducts

Sedative effect No analgesic properties Pentobarbitone, 40 to 50mg/kg i.p. (mouse) (120to 180 min sleep)

Good hypnoticeffect

Severe respiratory depression andhypotensive

Reasonable musclerelaxation

Easy to overdose Thiopentone, 30 mg/kg i.v. 15 min sleep (rat)

Metabolites accumulate with time

Caustic substances (thiopentoneonly i.v. route)

Chloral hydrate Sedative effect No analgesic properties 300 to 400 mg/kg i.p. (1 to 2 h sleep time) (mouseand rats)

Good hypnoticeffect

Paralytic ileus noted in rats

Minimal CVS andrespiratory depression

Terminal/non-recovery work only

α-Chloralose Sedative effect No analgesic properties 50 to 60 mg/kg i.p. (rats), 120 mg/kg i.p. (mouse)(8 to 12 h for non-recovery only)50 mg/kg i.v. bolusfollowed by 25 to 40 mg/kg/h (rats)Good hypnotic

effecti.v. use only

Suitable forlong-term anaesthesia

Slow induction and recoveryassociated with involuntaryexcitement

Minimal CVS and respiratorydepression


Urethane Suitable for long-termanaesthesia.

Carcinogenic: only allowed to beused with special justification

0.8 to 1.3 g/kg i.p. (mouse and rats)Duration ofaction 8 to 10 h (non-recovery only)


Table 2 Summary of the properties of injectable anaesthetics (Continued)

Minimal CVS and respiratorydepression


AvertinW

(tribromoethanol)Wide safety margin Local irritation/peritonitis 0.015 ml/g body wt of 2.5% i.p.

Good muscle relaxation Handling and storage safety issues 30 min; supplemental doses of anaesthesia:minimum of 1/2 of the initial dose up to 1 mlmaximum volume per animalRapid induction and recovery Toxic effects


Monitoring anaesthesia during imaging proceduresWhile the ideal anaesthetic agent would produce revers-ible unconsciousness and analgesia with no otherhomeostatic effects, the reality is that all anaestheticdrugs affect other body systems to some extent. Themost important effect of anaesthetic drugs is depressionof respiration, the cardiovascular system and thermo-regulation. Therefore, one must seriously assess thetherapeutic dose required for anaesthesia, what the like-lihood is to affect image acquisitions and the outcome ofscientific data acquisition interference during the pro-cedure. It is also important to clearly understand thatextreme deviations from maintaining normal physiologyduring anaesthesia can be life threatening for the animal.While it is not clearly defined how frequently one

should monitor the animals during anaesthesia, it is ob-vious that the more invasive the procedure and/or thelonger time under anaesthesia, the more likely it is tointerfere with normal homeostasis and thus greater theneed for monitoring. A general approach is to continu-ously monitor cardiovascular and respiratory function asthe minimum standard applicable to veterinary anaes-thesia. Indeed, for laboratory animals, this may requirespecialised monitoring equipment.The position of the animals is also of importance as

physiological parameters can be quite affected by thebed systems used in different imaging modalities. It iscrucial that the neck and head are well positioned withthe airway maintained well and open at all times, with-out restricting breathing or/and cardiovascular function,e.g. to minimise pressing equipment on the thoracic cav-ity. Body and extremities should be positioned withoutrestricting the circulation and avoiding any bruising,strains or avulsion in the body structures.Generally, the monitoring procedure should cover the

four body systems, namely central and peripheral ner-vous, respiratory, cardiovascular and musculoskeletalsystems. Basic parameters should include the following:

a. Respiratory rate, depth and characterb. Heart rate, rhythm and pulse intensityc. Body temperatured. Colour of mucous membrane, extremities and

capillary refill time (peripheral perfusion)e. Central nervous system (muscle tone and reflexes)

Direct access and visualisation of the animal whileplaced in the scanner during imaging can be challenging,especially for assessing parameters such as capillary refilland muscle tone and reflexes.

Respiratory systemRespiratory monitoring can be based on clinical observa-tions, including observation of the chest motion, orassessed through detection of breathing motion by com-pressions of a respiratory sensor placed in contact withthe chest. Respiratory motion detection systems are ex-tensively used during imaging as they are highly compat-ible with and safe to use for most of the imagingmodalities (Figure 2).Motion artefacts due to breathing which can seriously

compromise imaging acquisitions can be eliminatedfrom the acquired images by employing gating techni-ques, which allow the synchronisation of data collectionwith the respiratory cycle. Typically, the largest move-ment of the diaphragm and abdomen occur between in-spiration and expiration; thus, selective acquisitionduring the still part of the respiratory cycle can be effect-ive in reducing breathing artefacts.Other advanced respiratory monitoring systems in-

clude digital systems that can monitor and detect thechanges in gas temperature and/or compositions be-tween inspiration and expiration at the endotrachealtube connector or face masks:

1. Capnograph measures the CO2 level through ahighly sensitive infrared spectroscopy CO2 sensor inthe inhaled and exhaled gas, based on the CO2

values in the venous return to the heart and theeffectiveness of breathing. This equipment is a verygood indicator of the respiratory function, providingcontinuous measurements of the CO2 levels. TheCO2 levels rise rapidly during the first part ofexhalation and then flatten off. If the pulmonaryfunction is compromised or there is poor lungperfusion, the flatten-out phase will disappear. Thelevel of CO2 at the end of expiration (etCO2) isnormally within a few millimetres of mercury of thearterial CO2, and it is very useful for assessingadequacy of ventilation during both spontaneousand artificial ventilation [52].

Figure 2 Image displaying some of examples of respiratory monitoring equipment. (a) Respiratory sensor (VioHealthcare, Uckfield, UK); (b)Capnographs used for rodents: (i) type 340 Capnograph system (Harvard Apparatus, Holliston, MA, USA) with specifications for working withmouse or rats; (ii) MRI adaptable system: V9004 Capnography/Pulse oximeter (Harvard Apparatus); and (iii) PhysioSuite (Kent Scientific Corporation,Torrington, CT, USA) CapnoScan to measure the end tidal CO2.


One problem with this equipment is that the gassampling rate can be quite fast and removesexcessive amount of gas from the airway. There aresome adapted capnographs for rodents that utilisesmall volumes and are equipped with a pressuresensor for measuring tracheal pressure and apneumotach with differential pressure transducer formeasuring respiratory airflow. These sensors aremounted directly on the tracheal or intubationcannula and connected to the control unit tominimise the dead space. The tracheal pressure andthe respiratory flow are important signals to monitormechanical ventilation when using a respirator,aiding in avoidance of barotraumatic ventilation.There are several capnographs with specificationsfor working with mouse or rats, but unfortunately,the equipment tend to be not MRI compatible.There are few MRI adaptable versatile capnographson the market which can include pulse oximetry tomeasure PO2, PCO2 and fractional inspired oxygenmeasurements. (Figure 2).

2. Blood gas analysis. Arterial blood gas analysis is thegold standard for assessing respiratory function. ThepH, PO2 and PCO2 levels of the blood sample aremeasured with specific electrodes and thebicarbonate concentration is also calculated.Through these measurements, we can obtain adetailed picture of the state of the respiratorysystem:

� PO2. It is standard for measuring bloodoxygenation; decrease in PO2 is due tohypoventilation, inspiration of hypoxic gasmixtures or impaired gas exchange.

� PCO2. Levels of PCO2 are established by thebalance between CO2 production andelimination. The PCO2 is fairly constant, and thearterial level is determined by elimination.

� HCO3− concentration. Base excess and pHmeasurements provide information of acid andbase balance. These measurements may be veryrelevant when imaging animals with metabolicdisorders which may have acid and basedisturbances.

A blood gas analyser is not a routine piece of monitor-ing equipment, and interpretation of the results requiresknowledge of pulmonary and renal physiology. Theequipment needs to be near the imaging equipment, andin particular, for the rodents, one of the main limitationsis the volume of blood that can be removed and the fre-quency of sampling [53]. However, the system providesinvaluable monitoring data for animals used for cardio-vascular and respiratory research. It is also standard touse it for functional MRI studies which are based onmeasuring changes in the blood oxygenation levels inmicrocirculation in the tissues [54,55] (Figure 3).

Cardiovascular systemBasic cardiovascular monitoring includes assessment ofthe following:

� Heart rate. This is generally assessed throughpalpation of the chest wall or auscultation using astethoscope, but during imaging, the animal isgenerally placed within a restricted area with limitedaccess (e.g. imaging platform). Hence, most of themodalities will have an electrocardiograph systemimplanted with specific electrodes adapted for eachimaging modality.

� Pulse. The pulse is generally measured using apressure cuff or an oximeter (see below).

� Capillary refill time. It is measured by applyingpressure to a pink extremity or a mucous membrane(e.g. gums) and by checking the time for pinkness tore-invade the pressurised whitened areas (1 to 2 s).

Figure 3 Changes in arterial O2 saturation and PO2 and examples of whole pattern activation. (a) Changes in arterial O2 saturation andPO2 under different O2 gas concentrations during anaesthesia. These parameters can affect the BOLD response during MRI; therefore, it isimportant to monitor blood gas levels during such functional MRI studies [54]. (b, c) Example of whole pattern activation in a rat brain erraticallyinduced by too deep isoflurane anaesthesia. Such a ‘bad’ activation pattern in the whole brain (c) would mask any specific activation due to apharmacological and/or sensorial stimulus, and in this case, the time course shows (b) a general decline in signal, and no effect of the challengeis visible.


� Colour of mucous membranes (gums, perianal area,conjunctiva) or hairless extremities.

The following are advanced cardiovascular monitoringequipment:

1. Electrocardiograph. The ECG monitors the electricalactivity of the heart. With each beat the atria andventricles depolarize and repolarise. Thedepolarisation and repolarisation are synchronised ineach chamber, and thus, the action potentials fromeach fibre summate, producing a signal that is largeenough to be measured at the surface of the body.The electrical signal is picked up by electrodes,amplified and displayed on a screen. The ECG isalways measured as the difference in voltagebetween two electrodes. Depending upon theplacement of the electrodes, the ECG has differentshapes. If the electrodes are placed on each arm, alead I waveform is obtained. Lead II is measuredfrom the right arm to the left leg, and lead III ismeasured from the left arm to the left leg. Foranaesthetic monitoring, the lead configuration islargely irrelevant because detailed measurements ofthe complex heights are not generally needed,requiring a waveform that contains all the maincomponents. A lead II type trace, with positive P, Rand T waves, is usually chosen. In theory, only two

electrodes are needed to record an ECG since it isthe voltage difference between the two electrodes. Inpractice, a third ground electrode is needed to rejectinterference. Different ECG electrode systems usedin imaging modalities are shown in Figure 4.The electrocardiogram only monitors the electricalactivity of the heart, and the heart rate is derivedfrom this. They do not provide information aboutthe mechanical function of the heart or the state ofthe circulation. However, they are important for thediagnosis and treatment of arrhythmias. They areuseful in critical situations where the blood pressurehas fallen so low that peripheral pulses are notpalpable. During imaging, it is crucial to get a goodsignal for the gating systems to work efficiently.Motion artefacts during MR imaging can beeliminated or greatly reduced by employing gatingsystems during acquisitions of the MR data. Bysynchronising MR data collection with theelectrocardiogram, images can be obtained atspecific times during the animal’s cardiac cycle.

2. Pulse oximeter. The pulse oximeter measures arterialoxygenation. Pulse oximeters are very useful as theydetect arterial hypoxemia that may be related toproblems that occur in anaesthesia such ashypoventilation, airway obstruction and equipment-related problems. Hypoxia is a common and seriousproblem, and a pulse oximeter will detect

Figure 4 ECG electrodes systems. (a) System BioVet™ (©m2m Imaging Corp, Newark, USA): the carbon fibre electrodes are applied directly incontact with the cleaned and shaved chest skin and applied with gel electrode so that a minimal impedance electrical connection is made withthe electrode. (b) Model 1025 small animal monitoring and gating system (Small Animal Instruments, Inc., Stony Brook, NY, USA): the ECG systemused for the MRI scanners uses sub-dermal needle electrodes, pads or surface electrodes. The placement of the electrodes is typically in or on theright forepaw and the left hind pore, or electrodes are placed in the forepaw as long as it is across the heart plane. All the wire bundles withinthe scanner should be taped to eliminate unwanted movement from the gradient vibration and/or air flow. (c) Schematic representation of thefMRI setting: all the equipment needs to be non-ferromagnetic, and it is connected to a module system which allows gated acquisition ofimages, avoiding interferences from motion due to breathing and /or heart beating. Body temperature is also regulated through a heatingmodule (small rodent heater system; Small Animal Instruments, Inc.) to monitor and control the animal temperature during imaging. The systemsoftware continuously processes the temperature measurements and sends an optical control signal to the heater control module. The rate ofchange of temperature is monitored, and heater control is adjusted to regulate temperature changes. Mouse temperature variations of less than±0.1°C can typically be obtained during magnetic resonance (MR) examination.



hypoxemia long before the animal becomescyanosed. The pulse oximeter consists of a probe, alight-emitting diode and a photo-detector, which canbe applied above a superficial vessel such that thelight is transmitted across the tissues. The emittedlight alternately transmits red light at two differentwavelengths (in the visible and infrared regions)which are absorbed to different degrees byoxyhaemoglobin and deoxyhaemoglobin. Theintensity of transmitted light reaching the photo-detector is converted to an electrical signal. Thisinformation is processed, and the absorption due tothe tissues and venous blood, which is static, issubtracted from the best-to-beat variation to displaythe peripheral oxygen saturation both as a waveform and a digital reading. The waveform can alsobe interpreted to give a reading of heart rate. Strictlyspeaking, the pulse oximeter relies on and thereforegives information about both circulatory andrespiratory systems.For laboratory animals, the light sensor clips havebeen adapted to be placed onto the animal’s thigh,foot, neck (mostly for mice) or tail for rats andlarger rodents (Figure 5), and assesses the colour ofarterial blood close to the surface. The systemprovides real-time continuous measurements ofarterial O2, saturation, pulse strength, breath rate,blood flow and effort to breathe.The absolute accuracy of pulse oximeters seems tobe a little lower for animals than for humans [56],but the trends are still important, and overall, thisequipment is a very valuable tool for monitoringanimals. In general, a saturation of >95% is good. Ifit falls to <90%, then the anaesthetist should takenote, but corrective action may not be necessary,especially if the cause is known and self-limiting.When the saturation falls below <80%, action shouldbe taken to improve oxygenation. Some pulse

Figure 5 Images displaying the clip sensors used by the pulse oximetfoot in rat. The MouseOxW murine pulse oximeter system from Starr Life Scsaturation, pulse rate, respiration and pulse and breathe distension. (c) Profat 100% and 21% O2 during inhalation anaesthesia with isoflurane.

oximeter systems are also MRI compatible withadapted non-ferrous-magnetic clips and cables formonitoring animals while imaged in small borepreclinical MRI scanners [57] (Figure 5).

3. Blood pressure. The blood pressure measurement isone of the best methods to monitor thecardiovascular system during anaesthesia [58]. Themean arterial pressure (MAP) is the best indicator ofhow well tissues are perfused. As the MAP falls, vitalorgan auto-regulation and perfusion is lost. Thesusceptibility of tissues to poor circulation variestremendously depending upon their metabolic rate.The MAP in anaesthetised mice drops is around80 ± 10 mmHg [59], with average systolic pressuresof 112 ± 10 and mean diastolic pressures of 42 ± 12mmHg. Combinations of injectable anaestheticssuch as ketamine and xylazine induce markeddecrease of the MAP and the pulse rate [37].Similarly, isoflurane also has coronary vasodilatationproperties and thus can alter blood flow parametersin rodents [60]. Therefore, blood pressuremonitoring remains a very important processespecially when undertaking functional MRI studies[61] or when assessing the perfusion of radiotracersthrough specific tissue/organs during PET/SPECTimaging.

There are two basic methods for measuring arterialblood pressure, direct and indirect. The direct bloodpressure measurement involves placing a catheter in anartery and connecting it to a transducer. In rodents, thisprocedure is invasive as the artery, femoral or carotidgenerally has to be exposed surgically. Once the catheterhas been placed in the vessel and secured, it is eitherconnected to a pressure transducer via a fluid-filled lineor can be directly connected through the use of indwel-ling catheters and pressure sensing transducers [62]. Forthe fluid filled systems, specific small lines are used for

er systems. (a) In the base of the mouse or (b) in the centre of theiencesW Corp. (Oakmont, PA, USA) provides measurements of O2

ile of arterial O2 saturation measurement in rat during MRI acquisitions

Figure 6 Several systems available for monitoring blood pressure. They include Millar probes (Millar Mikro-TipW, Millar Sensors Systems,ADInstruments Ltd., Oxford, UK), fibre optic transducer, (which are most suitable for MRI imaging (e.g. Samba Preclin catheterW; Samba SensorsAB, Västra Frölunda, Sweden) and pressure transducers (TSD104A blood pressure transducer; Biopac Systems, Inc., Goleta, CA, USA). Theperivascular flow probes (Transonic Systems, Inc., Ithaca, NY, USA) provide good precision blood pressure measurements for small animal vessels(0.25 mm) without the need to expose the lumen of the vessels, although the procedure still requires surgical exposure of the vessels to localiseand fix the transducer around the vessel wall [64]. Indirect blood pressure method (non-invasive) involves inflating a cuff around the tail whichworks as a pressure transducer measuring the blood flow in the tail artery.


rodents to minimise mobilising large volumes of blood/fluid to minimise any site effects on the animal’s homeo-stasis [26] An alternative system can be used for bloodpressure in conscious mice. It consists of a pulse trans-ducer and inflatable tail pressure cuff configured for usewith rodents. The system records both the cuff pressureand pulse transducer signal from the tail artery [63]. Sev-eral systems available are shown in Figure 6.

Body temperatureMost anaesthetic agents profoundly depress thermoregu-lation, and laboratory rodents with a high metabolic rateand large surface area-to-body weight ratio are likely tosuffer hypothermia. This is particularly critical whenanaesthetising mice and can be a source of considerablemortality; therefore, core temperature monitoring ishighly recommended. Temperature can be monitoredusing a digital rectal thermometer or thermocouple;there are several systems available in the market for allthe imaging modalities. Due to muscle relaxation, rectaltemperature may be up to 1°C less than the core bodytemperature. Hypothermia can be treated using an

external heat source, with several systems available in-cluding circulating hot water blankets, blowing air sys-tems, electric pads, bubble wrap or infrared lamps.Hyperthermia is as dangerous as hypothermia; thus, theexternal heat source should not only be ‘thermostaticallycontrolled’ but should be linked to the core temperature.It is important to minimise heat losses during the animalpreparation before imaging (e.g. hair removal, alcohol-based surgical preparation) using regulated heatingpad and/or by covering the animal, in particular, the tail,and insulating it from a cold surface. If fluid replacementis required, the isotonic solutions should be warmedto 37°C.

Colour mucous membrane, capillary refill time andassessment of the central nervous system awarenessThese parameters are generally challenging to assessduring imaging while the animals are placed within thescanners, which are not accessible to researchers. Duringinduction of anaesthesia, most animals will becomeataxic, lose their righting reflex and eventually becomeimmobile. At this depth of anaesthesia, they can easily


be roused by painful stimuli, so anaesthesia must beallowed to deepen until such responses to pain are ab-sent. To assess the depth of anaesthesia, the researchercan check the muscle tone and the response to reflexesof the anaesthetized animal. Some other regularlychecked reflexes include the foot pinch stimulus (loss ofthe pedal withdrawal response) and assessing the blinkresponse (palpebral reflex) of the animals.

Long-term anaesthesia and monitoring recoverySome imaging protocols, in particular for MR imaging,may require long acquisition times, and thus, the animalwill have to be under anaesthesia for a long time. Whenanimals are anaesthetised for a long period of time, ad-verse effects caused by poor technique and/or sideeffects of many anaesthetic drugs become more apparentand increasingly critical for the animal’s well-being andthe experimental outcome. Therefore, it is very import-ant that all the monitoring procedures abovementionedare well implemented and thoroughly applied duringlong-term imaging.Anaesthesia can be prolonged by maintaining the ani-

mal on inhalation anaesthesia or by repeating standardinjectable regimes. Metabolites will accumulate in an es-calating manner, so it is important to assess how thiscan interfere with the purpose of the experiment.To minimise systemic accumulation of anaesthetic

metabolites and detrimental effects of long-term use ofanaesthetic drugs, three regimes are generally used forlong-term anaesthesia:

1) Using minimally metabolised volatile anaestheticgases (e.g. isoflurane). There is very little practicaldifference between administering an inhalation agentfor 30 min or for 8 h. Once the animal has beenanaesthetised, it will remain anaesthetised if suppliedwith an appropriate maintenance concentration ofanaesthetic. After 1 or 2 h, the maintenanceconcentration of anaesthetic can usually be reduced;further reductions may be possible, particularly ifthe animal is stable, and no further painful stimulusis given. The major problems encountered whenusing volatile anaesthetics for prolonged periods isusually due to anaesthetic techniques or equipmentfailure. It is essential to select an anaesthetic circuitwith minimal dead space and minimal circuitresistance: this is particularly critical for the MRsystems in which the gas circuit is generally quitelong to access the animal in the magnet bore.

2) Continuous intravenous infusion with rapidlymetabolised and eliminated agents such as propofol.There are many problems associated withprolonging anaesthesia using repeated doses of aninjectable agent at frequent time intervals. Giving

intermittent doses of the drug will cause the depthof anaesthesia to vary considerably, although thismay be overcome by administering it as acontinuous infusion so that steady plasmaconcentrations of the anaesthetic are maintained.However, this will require good monitoring of theanimals, familiarisation with the anaesthetic regimeand the pharmacodynamics of the anaesthetic agent.Animals will have to be catheterised, and a longinfusion line may be required to reach the animalwhile placed inside the scanner during imaging (e.g.for MRI). In general, repeated doses of anaestheticdrug will increase the depth and duration of theanaesthesia state, which could potentially prolongthe recovery time. It is preferable to administerincremental doses of the drug by intravenous routeas the effects are seen more rapidly, and also, thedosing/effect can be more accurately adjusted.Administration by other routes is possible, but thedepth of anaesthesia will vary less predictably.

3) Long-acting injectable anaesthetic agents. Analternative to administering repeated doses or acontinuous infusion of short-acting agents is toselect an anaesthetic with a prolonged duration ofaction (e.g. urethane, α-chloralose).

It is also critical to maintain normal physiology duringlong-term anaesthesia. In addition to the general moni-toring guidelines already mentioned, there are some im-portant procedures that should be followed to ensurethe homeostasis of the long-term anaesthetised animal:

1) Regular repositioning. To avoid any damage that canbe caused to muscles and nerves by the long-termimmobilised position and/or the adoption of anabnormal position during long-term anaesthesia, it isprudent to reposition the animal regularly (every fewhours) to minimise local pressure effects and aidgeneral circulation. Unfortunately, this may notalways be feasible in some imaging protocols inwhich the animal is required to maintain the sameposition for the acquisitions. In this case, it isimportant to ensure that animals are placed in aproper imaging bed. The animal can be immobilised,but excessive pressure should not be placed on it,and normal physiological movements (e.g. breathing)should not be compromised.

2) Eye protection. The eyes should be protected bothfrom exposure keratitis (by regular application ofophthalmic ointment e.g. Lacrilube) and from retinaldamaged caused by exposure to excessive lightintensity in albino strains.

3) Temperature. Core temperature monitoring andmaintenance is essential during long-term


anaesthesia; hypo- or hyperthermia will interferewith many electrophysiological outcomemeasurements. The extremities and tail should becovered whenever possible.

4) Blood sampling. Blood sampling for blood gases,electrolytes, pH, haemoconcentration etc. is ideal forlong-term monitoring, and correction for anyabnormality by means of the appropriate electrolyteadministration, but this is not always feasible insmall rodents. However, end tidal capnographyshould be used, and the ventilator adjusted to keepend expired CO2 as near normal as possible. If theanimal is kept at normal hydration and temperature,its kidneys should maintain normal homeostasis. Inthe extreme situation of extensive long-termanaesthesia (after 24 h), the risk for oxygen toxicity(muscle fasciculation, convulsion) is increased, andin this case, the inspired oxygen should be keptbelow 40%.

5) Fluid status and urine output. Physiological fluidssuch as Hartmann’s (e.g. lactated Ringer’s alternatingwith glucose-saline) should be infused in order tomeet the daily fluid requirement and to maintainurine output at about 1to 2 ml/kg/h and to allow forrespiratory losses. Unfortunately, this may be quitechallenging due to the position of the animal in theimaging bore and experimentally required blood/fluid collection.

6) Ventilation. Some studies will require the animal tobe mechanically ventilated. Small animal ventilatorsare designed to achieve controlled ventilation of theanimal’s lungs by the application of intermittentpositive pressure to the airway, providing the tidalvolume and physiologically relevant respiratoryrates. During mechanical ventilation, the ventilatorapplies a positive pressure to the anaesthetic gases toovercome airway resistance and elastic recoil of thechest, and flow occurs into the lungs (as comparedto the negative intrathoracic pressure that draws airin during spontaneous ventilation). The technique isusually referred to as intermittent positive pressureventilation. In both, spontaneous and mechanicalventilation expiration occurs by passive recoil of thelungs and chest wall.

When using the ventilator, it is important to set theright parameters for the animal species and size,accounting for the following: the breathing rate, themaximum inspiratory pressure and/or volume and theinspiratory-to-expiratory ratio (usually set at a 1:1 or1:2). The latter can be increased to 1:3 or 1:4 to induceless cardiac depression, while maintaining inflationpressure below 15 cm H2O. This is because cardiacoutput is decreased when the thoracic pressure is

positive (i.e. during inspiration when mechanicallyventilated).It is important to understand that mechanical ventila-

tion reverses natural breathing, with important effectson thoracic haemodynamics and also overrides auto-nomic reflex control of breathing, which normally main-tains blood gas homeostasis. When monitoring aventilated animal, it is important to realise that carbondioxide is the main stimulus for respiration and that, ifthe arterial carbon dioxide level is reduced below 30 to35 mmHg, the animal will not attempt to breathe. Acommon mistake is to under-ventilate so that the animalfights the ventilator; this may be misinterpreted as theanimal being not deep enough under anaesthesia andmore anaesthetic is given. The end result is a hypercap-nic animal that is too deeply anaesthetised. Ideally, oneshould have oxygen/carbon dioxide monitoring equip-ment (e.g. pulse oximeter, end tidal capnograph). Theventilation settings can then be adjusted to try to main-tain physiological levels of O2/CO2. Blood gas and elec-trolyte analysis are useful but confer intermittentmonitoring, and a local analyser is needed. If the ani-mal’s kidneys are functioning normally, (i.e. not shutdown due to dehydration, blood or heat loss), the animalcan maintain its blood electrolyte, and acid and base bal-ance, provided that the ventilation rate is adjusted tomaintain the CO2 level. For long-term protocols, thismay involve fluid infusion. For long-term mechanicalventilation, it may be worth considering humidifying theinspired gas mix to minimise airway drying and irrita-tion. Regular tracheal aspiration may be needed to re-move respiratory tract exudates from the endotrachealtube in larger laboratory animal species, or alternatively,an anti-muscarinic agent such as atropine could beincorporated as a premedication.

RecoveryMonitoring should be carried out until the animal hasfully recovered and there is confidence that all essentialphysiological functions are well recovered, in particular,normal respiratory and cardiovascular function, bodytemperature and general behaviour. It is very importantto carry out such assessments before the animal is trans-ferred back to its housing cage.

� Assessment of peripheral perfusion. Assessment ofperipheral perfusion through the colour mucousmembrane and capillary refill time since perfusiondecreases with anaesthesia. If the colour of themucous membranes is pale or cyanotic, it isimportant to check the cardiovascular andrespiratory systems as abovementioned. Excessiveloss of fluids through bleeding or evaporation if abody cavity is surgically opened will rapidly lead to


complications such as circulatory shock. This inturn will affect the function of many organs in amanner that could have an impact on research data.If substantial fluid loss is anticipated, calculation offluid replacement requirements can be estimated byweighing the animal before and after the procedure.Lost fluids can be replaced using solutions such as0.9% saline or lactated Ringer’s solution, preferablywarmed to 37°C and administered subcutaneouslyor intraperitoneally.

� Assessment of central nervous system awareness. Thisis important for assessing the depth of theanaesthesia and, in particular, the level ofconsciousness and immobilisation/responsiveness ofthe anaesthetised animal during recovery. Afteradministration of some anaesthetic agents, especiallyinjectable agents, animals may become ataxic, showpoor coordination and poor reflex response,especially on the back limbs during the recovery,and therefore, it is important that they aremonitored closely until they gain normal behaviour.

During recovery, it is important to maintain the coretemperature of the animal, placing the animals on theirright side or in sternal recumbency, in a warm recoverycage with no sawdust or suchlike bedding that may beinhaled. If the animals have been under anaesthesia for arelatively long time, it may be worth considering to sup-ply oxygen and fluids and nutrition supplement (oral orparenteral support e.g. high-energy moist foods such asnutrient agar, jelly or crushed rodent pellets mixed withwater) presented in a petri dish or other manner thatdoes not require the animal to reach up high. Additionalanalgesia should be provided if the animal has under-gone a painful procedure or if there appears to be anysigns of pain. It is very important to assess for any signsof pain, e.g. guarding (e.g. limping) or licking/rubbingthe affected area, quietness, isolation from the group,failure to explore, hunched posture, piloerection andweight loss [64].

ConclusionsRodent imaging tools permit new investigations to bemade into fundamental biology, and answers to thesequeries often reveal a need for more imaging tools withwhich to pursue additional new knowledge, that is, newtools and new findings reinforce each other, leading toan ever-advancing frontier of new knowledge abouthuman health and disease obtained through imaging re-search with rodents. Laboratory rodent imaging is a rela-tively new field which is expanding very rapidly andbeing integrated into a wide variety of biomedical disci-plines. Therefore, it is important to consolidate soundprotocols for handling, anaesthesia and monitoring the

animals to ensure suitability for all the specific needs ofa wide range of imaging experiments. Obtaining accurateand consistent images remains one of the major chal-lenges in preclinical imaging, in particular, for longitu-dinal studies. Success relies extensively on choosing theappropriate anaesthetic regime and using suitable moni-toring equipment to assist in maintaining good homeo-stasis. While anaesthetic monitoring equipment forrodent imaging is in its infancy, it is anticipated that thisrapidly advancing field holds great opportunities forfurther research and technology development. Indeed,further refinement of animal handling protocols, imple-mentation of new anaesthetic regimes and developmentin monitoring equipment are needed to maximisethe potential of current preclinical imaging systems,minimising any confounding effects on the imagingoutcomes and the studied biological processes. It alsorepresents an important step towards improvinganimal welfare and the three Rs’ principles of hu-mane experimental technique: replacement, refine-ment and reduction.

Competing interestsThe authors declare that they have no competing interests.

Authors’ contributionsJLT performed some of the reported animal experiments, reviewed currentliterature, provided the concept for the review and wrote the manuscript. AKprovided very valuable intellectual advice on rodent anaesthesia andcritically edited and reviewed the manuscript. WG reviewed the manuscriptand supported the conception of this review. All the authors read andapproved the final manuscript.

AcknowledgementsThe authors would like to express their gratitude to Dr. Florine Morrison forher advice and assistance during the concept development of this reviewand Miss Leigh Brody for her help with the editing of the manuscript.

Author details1Biological Imaging Centre (BIC), Medical Research Council (MRC) ClinicalScience Centre, Imperial College London, Hammersmith Campus, CyclotronBuilding, Du Cane Road, London, W12 0NN, UK. 2Central Biological Services,Imperial College London, South Kensington, London, SW7 2AZ, UK.

Received: 23 April 2012 Accepted: 16 July 2012Published: 9 August 2012

References1. de Kemp RA, Epstein FH, Catana C, Tsui BM, Ritman EL: Small-animal

molecular imaging methods. J Nucl Med 2010, 51(Suppl 1):18S–32S.2. Sharma S, Ebadi M: SPECT neuroimaging in translational research of CNS

disorders. Neurochem Int 2008, 52:352–362.3. Schambach SJ, Bag S, Schilling L, Groden C, Brockmann MA: Application of

micro-CT in small animal imaging. Methods 2010, 50:2–13.4. Reddy S, Robinson MK: Immuno-positron emission tomography in cancer

models. Semin Nucl Med 2010, 40:182–189.5. Lee Z, Dennis JE, Gerson SL: Imaging stem cell implant for cellular-based

therapies. Exp Biol Med (Maywood) 2008, 233:930–940.6. Long CM, Bulte JW: In vivo tracking of cellular therapeutics using

magnetic resonance imaging. Expert Opin Biol Ther 2009, 9:293–306.7. Henkelman RM: Systems biology through mouse imaging centers:

experience and new directions. Annu Rev Biomed Eng 2010, 12:143–166.8. Koba W, Kim K, Lipton ML, Jelicks L, Das B, Herbst L, Fine E: Imaging

devices for use in small animals. Semin Nucl Med 2011, 41:151–165.


9. Kagadis GC, Loudos G, Katsanos K, Langer SG, Nikiforidis GC: In vivo smallanimal imaging: current status and future prospects. Med Phys 2010,37:6421–6442.

10. Bouxsein ML, Boyd SK, Christiansen BA, Guldberg RE, Jepsen KJ, Müller R:Guidelines for assessment of bone microstructure in rodents usingmicro-computed tomography. J Bone Miner Res 2010, 25:1468–1486.

11. Hallouard F, Anton N, Choquet P, Constantinesco A, Vandamme T:Iodinated blood pool contrast media for preclinical X-ray imagingapplications–a review. Biomaterials 2010, 31:6249–6268.

12. Taschereau R, Chow PL, Chatziioannou AF: Monte Carlo simulations ofdose from microCT imaging procedures in a realistic mouse phantom.Med Phys 2006, 33:216–224.

13. Bartling SH, Kuntz J, Semmler W: Gating in small-animal cardio-thoracicCT. Methods 2010, 50:42–49.

14. Driehuys B, Nouls J, Badea A, Bucholz E, Ghaghada K, Petiet A, Hedlund LW:Small animal imaging with magnetic resonance microscope. ILAR J 2008,49:35–53.

15. Muir ER, Shen Q, Duong TQ: Cerebral blood flow MRI in mice using thecardiac-spin-labeling technique. Magn. Reson. Med 2008, 60:744–748.

16. Ferrari L, Turrini G, Crestan V, Bertani S, Cristofori P, Bifone A, Gozzi A: Arobust experimental protocol for pharmacological fMRI in rats and mice.J Neurosci Methods 2012, 204:9–18.

17. Data Acquisition System - BIOPAC Systems, Inc. http://www.biopac.com/magnetic-resonance-imaging-mri-compatible.

18. SAII Small Animal Instruments, Inc. http://www.i4sa.com/web_app/main/defaultProduct.aspx?ID=76&PT=3.

19. Peden CJ, Menon DK, Hall AS, Sargentoni J, Whitwam JG: Magneticresonance for the anaesthetist. Part II: anaesthesia and monitoring in MRunits. Anaesthesia 1992, 47:508–517.

20. Riemann B, Schäfers KP, Schober O, Schäfers M: Small animal PET inpreclinical studies: opportunities and challenges. Q J Nucl Med MilImaging 2008, 52:215–221.

21. Khalil MM, Tremoleda JL, Bayomy TB, Gsell W: Molecular SPECT imaging:an overview. Int J Mol Imaging 2011, 2011:796025.

22. Laforest R, Sharp TL, Engelbach JA, Fettig NM, Herrero P, Kim J, Lewis JS,Rowland DJ, Tai YC, Welch MJ: Measurement of input functions inrodents: challenges and solutions. Nucl Med Biol 2005, 32:679–685.

23. Hickson J: In vivo optical imaging: preclinical applications andconsiderations. Urol Oncol 2009, 27:295–297.

24. Obernier JA, Baldwin RL: Establishing an appropriate period ofacclimatization following transportation of laboratory animals. ILAR J2006, 47:364–369.

25. Vermeulen JK, De Vries A, Schlingmann F, Remie R: Food deprivation:common sense or nonsense? Animal Tech 1997, 2:45–54.

26. Hedrich H: (Ed): The Laboratory Mouse: The Handbook of ExperimentalAnimals. San Diego: Elsevier; 2004.

27. Levine S, Saltzman A: Feeding sugar overnight maintains metabolichomeostasis in rats and is preferable to overnight starvation. Lab Animals2000, 34:301–306.

28. Mc Nally JB, Kirkpatrick ND, Hariri LP, Tumlinson AR, Besselsen DG, GernerEW, Utzinger U, Barton JK: Task-based imaging of colon cancer in the Apo(min/+) mouse model. Appli Opt 2006, 45:3049–3062.

29. Flecknell PA: Laboratory Animal Anaesthesia. 3rd edition. London: AcademicPress; 2009.

30. Wolfensohn S, Lloyd M: Laboratory Handbook of Laboratory AnimalManagement and Welfare. In Anaesthesia of laboratory animals. 3rd edition.Edited by Wolfensohn S, Lloyd M. Oxford: Blackwell Publishing; 2003.

31. Lovell DP: Variation in pentobarbitone sleeping time in mice. 2. Variablesaffecting test results. Lab Anim 1986, 20:91–96.

32. Hau J, Van Hoosier GL(E): Handbook of Laboratory Animal Science. BocaRaton: CRC Press; 2003.

33. Marietta MP, White PF, Pudwill CR, Way WL, Trevor AJ: Biodisposition ofketamine in the rat: self-induction of metabolism. J Pharm. Exp.Therapeutics 1975, 196:536–544.

34. Wright M: Pharmacological effects of ketamine and its use in veterinarymedicine. J Am Vet Med Association 1982, 180:1462–1472.

35. Hedenqvist P, Roughan JV, Flecknell PA: Effects of repeated anaesthesiawith ketamine/medetomidine and of pre-anaesthetic administration ofbuprenorphine in rats. Lab Anim. 2000, 34:207–211.

36. Lei H, Grinberg O, Nwaigwe CI, Hou HG, Williams H, Swartz HM, Dunn JF:The effects of ketamine-xylazine anesthesia on cerebral blood flow and

oxygenation observed using nuclear magnetic resonance perfusionimaging and electron paramagnetic resonance oximetry. Brain Res 2001,913:174–179.

37. Buitrago S, Martin TE, Tetens-Woodring J, Belicha-Villanueva A, Wilding GE:Safety and efficacy of various combinations of injectable anesthetics inBALB/c mice. J Am Assoc Lab Anim Sci. 2008, 47:11–17.

38. Smith W: Responses of laboratory animals to some injectableanaesthetics. Lab Anim 1993, 27:30–39.

39. Griffin KM, Blau CW, Kelly ME, O’Herlihy C, O’Connell PR, Jones JF, KerskensCM: Propofol allows precise quantitative arterial spin labelling functionalmagnetic resonance imaging in the rat. Neuroimage 2010, 51:1395–1404.

40. Bruns A, Kunnecke B, Risterucci C, Moreau JL, von Kienlin M: Validation ofcerebral blood perfusion imaging as a modality for quantitativepharmacological MRI in rats. Magn Reson Med 2009, 61:1451–1458.

41. Lei H, Duarte JM, Mlynarik V, Python A, Gruetter R: Deep thiopentalanesthesia alters steady-state glucose homeostasis but not theneurochemical profile of rat cortex. J Neurosci Res 2010, 88:413–419.

42. Field FJ, White WJ, Lang CM: Anaesthetic effects of chloral hydrate,pentobarbitone and urethane in adult male rats. Laboratory Animals 1993,27(3):258–269.

43. Guidelines for Chloral Hydrate. http://vetmed.duhs.duke.edu/guidelines_for_chloral_hydrate.htm.

44. Williams KA, Magnuson M, Majeed W, LaConte SM, Peltier SJ, Hu X, KeilholzSD: Comparison of α-chloralose, medetomidine and isofluraneanesthesia for functional connectivity mapping in the rat. MagneticResonance Imaging 2010, 28:995–100.

45. BdeC A, Makarova T, Hess A: On the use of α-chloralose for repeatedBOLD fMRI measurements in rats. J Neurosci Methods 2011, 195:236–240.

46. Olson ME: Problems associated with urethane anaesthesia. CALAS (CanAssoc Lab Anim Sci.) News 1985, 17:115–119.

47. Hogan B, Beddington R, Costanti F, Lacy E: Manipulating the Mouse Embryo.2nd edition. Plainview: Cold Spring Harbor Laboratory Press; 1994.

48. Silverman J, Schocker AE, Anestidou AE, Villarreal Acosta Y: Anaesthetics inGEM: does TBE make the grade? Lab Animals Mag NY 2003, 32:19–21.

49. Kober F, Iltis I, Cozzone PJ, Bernard M: Cine-MRI assessment of cardiacfunction in mice anesthetized with ketamine/xylazine and isoflurane.MAGMA 2004, 17:157–161.

50. Kersten JR, Schmeling TJ, Hettrick DA, Pagel S, Gross GJ, Warltier D:Mechanism of myocardial protection by isoflurane: role of adenosinetriphosphate-regulated potassium (KATP) channels. Anesthesiology 1996,85:794–807.

51. Cesarovic N, Nicholls F, Rettich A, Kronen P, Hassig M, Jirkof P, Arras M:Isoflurane and sevoflurane provide equally effective anaesthesia inlaboratory mice. Lab Anim 2010, 44:329–336.

52. Larach DR, Schuler G, Skeehan TM, Derr JA: Mass spectrometry formonitoring respiratory and anaesthetic gas waveforms in rats. J ApplPhysiol 1988, 65:955–963.

53. NC3Rs - Dosing and sampling. http://www.nc3rs.org.uk/category.asp?catID=10.

54. Borsook D, Becerra L, Hargreaves R: A role for fMRI in optimizing CNS drugdevelopment. Nat Rev Drug Discov 2006, 5:411–424.

55. Tremoleda JL, Wylezinska MM, Habib J, Stuckey D, Price AN, Gsell W:Myocardial T2* is increased by transient oxygen challenge and reflects oxygenextraction via hemoglobin saturation at the capillary level. http://www.wmicmeeting.org/2011/2011abstracts/data/papers/P430.html.

56. Erhardt W, Lendl C, Hipp R, von Hegel G, Wiesner G: The use of pulseoximetry in clinical veterinary anaesthesia. Ass Vet Anaesth 1990, 17:30–31.

57. Pulse Ox. http://starlifesciences.com/mri.html.58. Wagner AE, Brodbelt DC: Arterial blood pressure monitoring in

anesthetized animals. JAVMA 1997, 210:1279–1285.59. Rao S, Verkman AS: Analysis of organ physiology in transgenic mice. Am J

Physiol Cell Physiol 2000, 279:C1.60. Hartley CJ, Reddy AK, Madala S, Michael LH, Entman ML, Taffet GE: Effects

of isoflurane on coronary blood flow velocity in young, old and ApoE(−/−) mice measured by Doppler ultrasound. Ultrasound Med Biol 2007,33:512–521.

61. Klomp A, Tremoleda JL, Wylezinska M, Nederveen AJ, Feenstra M, Gsell W,Reneman L: Lasting effects of chronic fluoxetine treatment on the latedeveloping rat brain: age-dependent changes in the serotonergicneurotransmitter system assessed by pharmacological MRI. Neuroimage2012, 59:218–226.

http://www.biopac.com/magnetic-resonance-imaging-mri-compatible

http://www.biopac.com/magnetic-resonance-imaging-mri-compatible

http://www.i4sa.com/web_app/main/defaultProduct.aspx?ID=76&PT=3

http://www.i4sa.com/web_app/main/defaultProduct.aspx?ID=76&PT=3

http://vetmed.duhs.duke.edu/guidelines_for_chloral_hydrate.htm

http://vetmed.duhs.duke.edu/guidelines_for_chloral_hydrate.htm

http://www.nc3rs.org.uk/category.asp?catID=10

http://www.nc3rs.org.uk/category.asp?catID=10

http://www.wmicmeeting.org/2011/2011abstracts/data/papers/P430.html

http://www.wmicmeeting.org/2011/2011abstracts/data/papers/P430.html


62. Mikro-Tip pressure catheters for use with mice and rats. http://www.adinstruments.com/products/research/transducers-%26-accessories/mikro-tip-electrophysiology-catheters.

63. Krege JH, Hodgin JB, Hagaman JR, Smithies O: A noninvasivecomputerized tail-cuff system for measuring blood pressure in mice.Hypertension 1995, 25:1111–1115.

64. Roughan JV, Flecknell PA: Evaluation of a short duration behaviour-basedpost-operative pain scoring system in rats. Eur J Pain 2003, 7:397–406.

doi:10.1186/2191-219X-2-44Cite this article as: Tremoleda et al.: Anaesthesia and physiologicalmonitoring during in vivo imaging of laboratory rodents: considerationson experimental outcomes and animal welfare. EJNMMI Research 20122:44.

Submit your manuscript to a journal and benefi t from:

7 Convenient online submission

7 Rigorous peer review

7 Immediate publication on acceptance

7 Open access: articles freely available online

7 High visibility within the fi eld

7 Retaining the copyright to your article

Submit your next manuscript at 7 springeropen.com

http://www.adinstruments.com/products/research/transducers-%26-accessories/mikro-tip-electrophysiology-catheters



review open access anaesthesia and physiological

Documents