ICRP Publication 153: Radiological Protection in Veterinary Practice

(33) In radiotherapy, high radiation doses delivered to the target tissue induce tissue reactions that ultimately prevent cancerous cells from further out-of-control multiplication. That said, effects such as skin burns and ocular effects are potential undesired effects on healthy tissues in some therapeutic procedures and cannot always be avoided (Gillette et al., 1995; Collen and Mayer, 2006; Pinard et al., 2012). Overexposure in radiotherapy, however, can result in severe tissue reactions that are very painful and can also lead to a variety of long-term complications. The specific complication will depend on the technology being used, dose fraction, total dose, and the organ(s) in the target volume. These complications can be benign to severe, and tend to be irreversible (e.g. fibrosis, necrosis, chronic inflammation) and difficult to treat, with detrimental impact on the patient’s quality of life in both animals and humans (Gillette et al., 1995; Collen and Mayer, 2006; Balter and Miller, 2014; Hall and Giaccia, 2019; Mayer et al., 2019a). An interesting (trivial) consequence of radiation exposure associated with radiotherapy can be the change in colour of an animal’s fur (leukotrichia; Fig. 3.1) (Inomata et al., 2009; Gerard et al., 2010; Mayer et al., 2019a; Lee et al., 2020).

(34) In interventional procedures, lesions such as radiation-induced skin burns in the area where the primary radiation beam enters the body may appear within weeks (Balter et al., 2010), particularly when complex procedures requiring prolonged fluoroscopy times are performed on larger animals. Most of such injuries can be managed and are self-limiting, but it is important to remember that unnecessary tissue reactions result in needless suffering.

(35) Although occupational doses received in veterinary practice are generally too low to observe tissue reactions, some nuclear medicine, interventional, or brachytherapy procedures, combined with poor practice, incidents, or accidents, have the potential to result in skin burns or effects on the lens of the eye based on experience in human medicine (Miller et al., 2010; Dauer, 2014; ICRP, 2018b).

(37) For humans, although there are indications of an increase in the risk of cancer for exposed children, including after in-utero exposures (Wakeford and Bithell, 2021) at lower doses, an increase of the risk of cancer in exposed members of the public at doses below approximately 100 mSv cannot be firmly demonstrated by epidemiological surveys alone. However, when combined with a deliberately prudent interpretation of radiation physics and radiation biology data, the Commission considers a linear-non-threshold model to be the best practical approach to managing risk from radiation exposure (ICRP, 2007a). This model assumes a linear relationship between dose and stochastic risk, which means that any increase in dose may result in an increase in the stochastic risk, bearing in mind that risks are increasingly uncertain at lower doses (ICRP, 2021a). It is challenging to develop definitive risk predictions for radiogenic stochastic effects at low doses because there are a variety of factors that contribute to overall risk, as well as additional modifying factors that can influence the promotion or progression of the disease (NRC, 2006; McLean et al., 2017). The risk indicator used by the Commission for humans is the ‘radiation detriment’, which is sex- and age-averaged over a composite reference population. It is determined from the lifetime risk of cancer, and considers severity in terms of lethality, quality of life, and years of life lost. It also considers heritable effects based on information from animal studies (ICRP, 2007a; Ban et al., 2022; ICRP, 2022).

(38) With respect to exposed animals, a common misconception is that an animal with a shorter life span than a human will not experience radiogenic cancer. However, it has been widely observed since the 1970s that, across species, neither increased body size nor longer life span is associated with an increase in the overall risk of cancer, as expected from the associated increase in the number of cells or cellular divisions, respectively (Abegglen et al., 2015; Vincze et al., 2022). This provided foundational insight for the modern recognition that the physiological factors influencing the responses of organisms to carcinogens are varied and complex.

(39) Cancer patterns in mammals are similar and, in general, are relative to life span (Albert et al., 1994; Schiffman and Breen, 2015); in other words, ‘risk of cancer in old age is not vastly different in species with very different life-spans’ (Peto, 2016). Latency periods are less than that in humans for many animals with shorter, physiologically compressed life spans (NRC, 1991; Backer et al., 2001; Cagan et al., 2022). Of interest to veterinary practice is the observation that dogs demonstrate a comparatively greater risk of developing cancers overall (Dobson, 2013; Abegglen et al., 2015; Schiffman and Breen, 2015). For cancer prevention in dogs, it has been stated explicitly in the literature that ‘dogs should be exposed to radiation only when the expected benefits will outweigh the risks’ (Kelsey et al., 1998), consistent with the principle of justification in radiological protection. A large number of studies investigating radiation carcinogenesis (and other morbidities) in experimental animals have been conducted, and dose–response relationships vary (Broerse et al., 1985; Duport et al., 2012; Tang et al., 2017; Spatola et al., 2021).

(40) Inheritance of radiation-induced abnormalities was reported by Hermann Muller in 1927 based on studies with x-ray irradiation of Drosophila (fruit flies) (Pontecorvo, 1968). Since then, ample evidence of radiogenic hereditary effects in other animals and plants has been reported (e.g. UNSCEAR, 2001; Russell, 2013). Of note, radiation exposure can only increase the incidence of the same mutations that occur spontaneously in a population (Hall and Giaccia, 2019). This makes potential heritable radiogenic effects in humans difficult to study because of the high natural incidence of the same mutations. Thus, hereditary effects in humans have not been definitively or reliably shown to be induced by ionising radiation exposure (Boice, 2020; NCRP, 2021). Humans are likely also susceptible to these effects, but with risk much lower than that for carcinogenesis (UNSCEAR, 2001, 2014).

(41) In humans, the likelihood of developing cancer in response to exposure to a carcinogenic agent depends on a variety of factors including, but not limited to, age; sex; environmental, socio-economic, and lifestyle factors; and genotype (Colditz et al., 1996). Individual variability in radiosensitivity to carcinogenesis is acknowledged but not fully understood (Rajaraman et al., 2018). However, there are some clear, population-level attributes, such as age and sex, that influence susceptibility to radiation-induced carcinogenesis (NRC, 2006; Preston et al., 2007). This risk is higher overall for the fetus, children, and adolescents, due to their longer life ahead and the comparative sensitivity of developing organs and tissues (ICRP, 2013b); and for females, primarily due to the radiosensitivity of the breast (Boice et al., 1991; NRC, 2006). This age and sex dependence of risk should be considered in the process of justification and optimisation, particularly with respect to children. For example, in veterinary practice, children and young adolescents are excluded from assisting in radiological examinations as the exposure is not justified. Similarly, the potential presence of individuals who are or may be pregnant needs careful consideration with respect to justification when radiological procedures are being performed; this has to do with both the radiosensitivity of the unborn child (Section 3.2.3) and the possible sensitivity of the breast tissue in some stages of preparing for lactation (Ronckers et al., 2004). The justification process for any such exposure should bear in mind that the dose limit for the unborn child (1 mSv during pregnancy; see Table 3.1) is not to be exceeded. If the presence of a pregnant or possibly pregnant individual is deemed justified, and informed consent is given, radiological protection measures need to be optimised. This could be achieved by providing instructions on where to stand, how to behave, what protective equipment to use, etc. Strategies for optimisation are discussed further in Section 6.

(42) It has been shown in laboratory animals that age at exposure and sex influence the risk of carcinogenesis, although to a varying extent (Benjamin et al., 1991; Shuryak et al., 2010; Haley et al., 2011; Tang et al., 2017). These risk dependencies are thus also relevant considerations for animal patients, as some groups receive exposures from a young age (e.g. dysplasia screening in puppies) (Dziuk, 2007) or presale examinations of performance horses (Judy, 2013). Being mindful of these risks is especially important when determining if such exposures are justified (discussed further in Section 6.2).

(47) Three exposure categories are identified: occupational exposure (received at work as a result of situations that can reasonably be regarded as being the responsibility of the operating management), medical exposure (received as a patient/research volunteer or from a patient as a non-occupational comforter/carer), and public exposure (received apart from occupational and medical) (ICRP, 2007a, 2014b). As the Recommendations are currently written (ICRP, 2007a,b), the medical exposure category appears to apply solely to human medicine. Veterinary applications of ionising radiation are, to a very large extent, comparable to human medical exposures; in fact, the only distinction is that the exposures are aimed at animals in one case and at humans in the other. In both cases, occupational and public exposures may occur. As veterinary practice involves subjects other than humans, local governments and regulatory agencies manage exposures received in a veterinary setting in different ways. Where veterinary practice is considered – from a regulatory perspective – to be comparable with an industrial application of ionising radiation rather than a medical application, this may lead to an approach whereby the animal is considered a mere object, without consideration of its characteristics as a sentient living creature.

(48) Environmental exposure (i.e. exposure to the living environment) is a fourth type of exposure, although it is not defined explicitly as an exposure category in Publication 103 (ICRP, 2007a). Thus far, ICRP has focused on the natural environment, with the goal of maintaining biological diversity, conserving species, and maintaining the health status of associated habitats, communities, and ecosystems (ICRP, 2003b, 2008, 2009c, 2014a, 2017b, 2021b), although there is ongoing discussion on potentially broadening this perspective (Clement et al., 2021).

(50) Dose constraints are prospective, source-related restrictions on individual dose to workers and/or members of the public intended to serve as the upper bound of the optimisation goal for that source (Fig. 3.2). Note that dose constraints are not intended to be hard limits. Rather, consistent with the core value of justice, dose constraints are intended to serve as a mechanism for limiting potential inequity that could result from differences in value judgements when implementing the optimisation process. In fact, interpreting constraints as rigorous limits can distort the outcome of the optimisation process (ICRP, 2007a). Dose constraints are used initially within the optimisation process at the planning stage to establish an appropriate level of protection for a given situation, and develop corresponding protective actions. The numerical value taken for a dose constraint will depend on the situation at hand.

(51) After the planning stage, dose constraints can uncover discrepancies between planning and implementation, or reveal potential changes that warrant additional consideration. If set properly, a dose constraint can play a role in revealing a departure from normal situations. For example, if a process or procedure is known to consistently and appropriately result in an effective dose of 0.5 mSv over 3 months, and recently that procedure resulted in 2 mSv over 3 months, a review would be conducted to discern the root cause of the increase. The increase may have been warranted, in which case no further action is necessary, or it may demonstrate a lapse in proper technique, problem with equipment, or other issues that need to be addressed.

(52) The principle of optimisation of protection for human patients is unique in the system of radiological protection. In diagnostic procedures, it is the same person who gets the benefit and suffers the risk. The imposition of individual restrictions on patient dose could also be counterproductive to the medical purpose of the procedure. Source-related dose constraints for the individual are therefore not relevant, and thus DRLs for a particular procedure, which apply to groups of similar patients rather than individuals, are used. Radiation therapy is also very different from other situations in that the dose is intentional; cell killing is the purpose of the treatment. In this case, optimisation becomes an exercise in minimising doses (and/or their deleterious effects) to surrounding tissues without compromising the pre-determined and intentionally lethal dose and effect to the target volume.

(53) Intuitively, these ideas could also apply to animal patients (Pentreath, 2016), although if and how these patients fit within the principle of optimisation has not been defined explicitly. Thus, management strategies are inconsistent between different countries (HERCA, 2012). In many countries, veterinary medicine is considered to be an industrial rather than a medical practice, the latter of which is considered to include human medicine alone. Unfortunately, this philosophy often neglects considerations associated with unique but necessary aspects of veterinary practice, such as safety of animal patients under sedation or anaesthesia, or situation-dependent risk management consistent with a graded approach (IAEA, 2018) (i.e. the implementation of the system of protection in a way that is proportionate to the associated risk, the complexity of the exposure situation, and the prevailing circumstances).

(54) Dose limits do not apply to the patient in medical exposures so as not to interfere with necessary, medically indicated diagnostic or therapeutic procedures; generic dose limits might jeopardise the effectiveness of the diagnosis or treatment, thereby doing more harm than good. Emphasis is therefore placed on the justification of the procedures in the first place, on the optimisation of protection, and, for diagnostic procedures, on the use of DRLs, which are not seen as limits, but instead indicate if a dose received from an imaging procedure is unusually high or low to guide the optimisation process and thus help manage patient exposures (ICRP, 2007b, 2017a). The Commission recommends that a similar, proportionate approach to that applied for human medical exposures should be developed and applied for veterinary practice to include a quality dose management programme that allows for periodic audits, continuous peer learning, and use of incident reporting systems that capture incidents and near-misses [e.g. Safety in Radiation Oncology (SAFRON), Safety in Radiological Procedures (SAFRAD), Radiation Oncology Safety Education and Information System (ROSEIS)] ⁴ .

(55) In environmental radiological protection, derived consideration reference levels (DCRLs), rather than limits, are used to inform the appropriate level of management or control of an exposure. DCRLs are absorbed dose rates above which, for a given taxonomic class, there is the potential for deleterious effects on individuals of a species that may lead to population-level consequences. DCRLs can be used as points of reference to optimise the level of effort expended on environmental protection, dependent upon the overall management objectives and the relevant exposure situation (ICRP, 2014a). As such, although relevant to animals in general, the concepts developed for radiological protection of the environment do not suffice for the adequate protection of individual animals exposed in veterinary settings.

(56) Emergency and existing exposure situations utilise reference levels rather than limits, because what defines a reasonable or tolerable exposure is strongly dependent on the prevailing circumstances of the exposure in these situations. The current work on radiological protection in veterinary practice focuses on planned exposure situations, although there may potentially be veterinary concerns in the other exposure situations as well (e.g. emergency exposures following a large-scale nuclear accident).

(60) Significantly limiting the duration of an exposure is not always feasible because a certain amount of time is usually required to perform a given task. However, detailed work plans with practice runs beforehand (without the source) can help to reduce overall exposure time. If practical, splitting tasks(s) between personnel or rotating through personnel can also reduce an individual’s exposure time. Another example of optimising time is the use of pulsed fluoroscopy in both fluoroscopy and interventional procedures, in combination with last image hold which can effectively reduce the time of exposure while keeping required image guidance ⁵ .

(61) Where reasonably possible, maximising distance from a radiation source is a simple and practical principle for dose reduction. The use of handling tools (e.g. tweezers, tongs) and hand carts should be considered, along with working at arm’s length and taking one step back where feasible (Fig. 3.3). However, note that these three basic rules should be used in conjunction with each other, as it could be that using a device such as tongs could increase the time spent handling the source (at a greater distance), whereas a short, quick manoeuvre closer to the source may result in less dose. Also, consideration should be given to individuals working for long periods of time in awkward or uncomfortable positions (e.g. working behind shields, etc.), which may create an ergonomic/orthopaedic hazard with potential for fatigue-induced mistakes or, again, an increase in the time to complete the task. Where safely applicable, the use of sedation or anaesthesia may considerably reduce the time that people need to spend in close proximity to an animal; their radiation exposure would then be reduced by the combination of a shorter exposure time and a greater distance from an animal seen as a radiation source. The more fractious an animal, the more personnel will typically have to ‘lean in’ to keep it in position during imaging, and sedation/anaesthesia can make it easier to work at arm’s length or take a step back. Furthermore, sedation/anaesthesia will ease patient positioning, reducing the need for retakes, which will reduce the total exposure time for the personnel involved in restraining animals.

(62) The most appropriate type of shielding to employ is dependent on the circumstance as well as the type and energy of the radiation involved. For example, the electrons (i.e. beta particles) produced in beta decay will interact with their surroundings and produce bremsstrahlung (‘braking radiation’). Bremsstrahlung refers to the photons produced when the path of a free electron is decelerated by an atomic nucleus; the more protons in a nearby nucleus, the more bremsstrahlung there will be. It is therefore better to shield beta emitters with low atomic number (Z) material (e.g. plastic or acrylic glass) as this will block the electrons while producing less bremsstrahlung than high Z material. High Z material is good at shielding photons, so lead shielding can be added on the outside of the primary container to shield the resultant photons while storing or transporting.

(63) Lead is commonly used to shield gamma- and x-ray radiation; however, in practice, any dense material (e.g. tungsten, steel, concrete, etc.) can attenuate these photons sufficiently if it is thick enough. For example, some high activity sources are housed in basement facilities to make use of the natural, earthen shielding. Photon attenuation is exponential, although for broad beam or poor geometry conditions, scattered radiation can result in ‘build-up’ and an exposure higher than that predicted purely by exponential attenuation. PPE frequently employs lead (e.g. aprons, gloves) or leaded glass (e.g. eyeglasses) to protect radiosensitive organs (Mayer et al., 2018). Care should be given that use of PPE optimises protection and safety (e.g. considering level of transmission, ergonomic issues, influence on the time required to perform a task, etc.). For example, lead aprons are not appropriate for use in PET studies, as the transmission of annihilation photons (511 keV) through a typical apron is >90% (Martinez et al., 2012); the increase in work time associated with wearing an apron negates this fairly trivial reduction in exposure.

(64) Shielding should start with assessment of the collective work environment (concrete walls, leaded doors and windows, standing or ceiling-suspended shields, etc.), and should be complemented with appropriate PPE worn by staff and assisting persons.

(66) Internal exposure to radionuclides is possible through inhalation, ingestion, or absorption through open wounds or even intact skin in the case of a high-mobility radionuclide. Internal radiological protection and contamination prevention measures focus on preventing or minimising the intake of radionuclides into the body and the deposition of radioactive substances on the body. Such protective measures (e.g. confine, contain, enclose) are consistent with general industrial hygiene measures, and generally include strategies for maintaining control of the source and the environment in which the source is handled and used, as well as using PPE when appropriate (see Section 3.4). Additionally, consistent with the justification principle, the amount of radiopharmaceutical administered to a patient should be selected such that no more is used than that needed to achieve the optimal diagnostic or therapeutic result. This optimises protection and safety of the patient, as well as that of workers, the public, and the environment.

(94) Decision support tools, such as referral guidelines or appropriateness criteria (EC, 2014; Subramaniam et al., 2019), could also be particularly useful in veterinary medicine to ensure Level 2 justification in the absence of direct veterinary radiologist input. These guidelines should be easily accessible, free of charge, and easy to use (e.g. ideally through integration into the electronic medical record system) to ensure their widespread adoption. Such guidelines, however, need to be developed collaboratively by national or international professional veterinary radiology societies, in conjunction with veterinary professional bodies, and animal health and regulatory authorities, and would require a substantial commitment of time and resources to their creation and periodic update.

(95) Of note, referral guidelines or appropriateness criteria not only contain information on different radiological imaging procedures (e.g. plain x ray, CT scanning), but also on imaging modalities that do not make use of ionising radiation – ultrasound and magnetic resonance imaging (MRI) in particular. From the point of view of justification, but only after careful consideration of all other factors (e.g. availability, cost, etc.) that come into play and are judged equal, the imaging method that can provide the required information for the lowest exposure – or no exposure to ionising radiation at all – should be preferred, if available.

(96) When new types of radiological equipment are considered and introduced in veterinary practice, an assessment of their potential implications for radiological protection should also be made. Recently, there has been an increase in radiological equipment dedicated to veterinary medicine on the market, and this equipment may not always comply with the imaging quality and/or radiological protection standards required for medical devices. Vigilance, both from potential buyers and regulating authorities, is therefore required to ensure that the adoption of such new equipment can be justified.

(97) Level 3 justification requires that the radiological procedure is required for the management of the individual patient. A diagnostic procedure should be able to answer a given clinical question, and have an impact on the patient’s diagnosis, prognosis, or treatment. Consideration should also be given to alternative modalities with less exposure to ionising radiation – or no exposure at all – such as replacing CT with MRI or ultrasound. One part of justification is determination of the most appropriate examination, within the constraints of available modalities. The radiology request should contain sufficient clinical information that a radiologist or an internal or external auditor can assess whether the particular examination is justified.

(99) For non-medically-indicated radiological procedures, Level 2 justification is therefore important. Particular attention should be paid to the clinical evidence for the usefulness of the procedure, ensuring that the chosen imaging procedure is suitable both for the detection of the condition in question, and for screening a potentially large number of animals (Bladon and Main, 2003). Furthermore, there should be a demonstrable relationship between the imaging findings and the goal of the screening. For example, for a breeding suitability examination, the trait in question should have a sufficient degree of heritability, as well as a prevalence in the population that makes it a relevant discriminator between potential breeding animals (e.g. Bruun et al., 2020). For presale examinations, the results of the imaging should be predictive of the animal’s future performance (e.g. Bladon and Main, 2003; RIRDC, 2009). Again, appropriateness criteria could be developed by professional veterinary speciality organisations, in conjunction with professional veterinary societies, regulatory authorities, breed societies, insurance companies, and industry representatives or other stakeholders, as appropriate.

(101) Benefits for exposed animals include a direct benefit from improved diagnosis and treatment in the case of animal patients, while the results from a presale or breeding suitability examination may help to ensure that the animal is suited for its intended purpose, and will not suffer negative health consequences from its future use. Screening examinations of asymptomatic animals may help to detect subclinical disease, and such early diagnosis may potentially lead to improved treatment results. In addition to welfare benefits for the individual animal, the welfare of animal populations may also be improved through breeding suitability examinations, if undesirable traits or medical conditions can be reduced in the population based on the imaging results.

(102) Benefits for veterinary staff from the appropriate use of ionising radiation include the ability to provide the best possible diagnosis and treatment to their patients, customer satisfaction, and financial revenue from the radiological procedures (and any follow-up treatment). Owners and handlers may benefit both emotionally and economically from improved diagnosis and appropriate treatment of their animals.

(103) In addition to individual animals, owners, and veterinary staff benefiting directly from the appropriate use of ionising radiation – or its alternatives – in veterinary practice, society at large will also benefit from such use. Animal and human health are interlinked, and radiological procedures that contribute to animal health may also improve public health, particularly when they contribute to the control of zoonotic diseases (Viljoen and Luckins, 2012; see also Annex B). Furthermore, a healthy population of working animals and livestock will also benefit society, both in terms of public health and the economy. Other industries, such as the racing and showing industries, would likely also benefit economically from improved animal health. In the case of rare or endangered species, conservation efforts may also sometimes benefit from the use of radiological procedures to diagnose and/or treat disease in zoo animals or wild animals. Moreover, with increasing societal concern over the ethics of the use of animals for production and entertainment, ensuring good health in these animals could also be seen as a prerequisite for the social acceptance of such use.

(104) Radiation risks to exposed animals include both stochastic effects and tissue reactions. While the risk of stochastic effects (compared with tissue reactions) predominates in plain radiography, high-dose diagnostic procedures, such as CT-guided and other interventional procedures, are performed increasingly in veterinary medicine, and could potentially result in tissue reactions, particularly in the skin (Balter et al., 2010; Cléroux et al., 2018). Furthermore, in veterinary radiation therapy, adverse effects associated with tissue reactions are encountered frequently in normal tissues, and the probability of their occurrence must be balanced carefully against the clinical benefits of tumour control or palliation (e.g. Collen and Mayer, 2006; Gieger and Nolan, 2017). On the other hand, lack of access to appropriate diagnostic imaging or therapy, or inappropriate choice of diagnostic or therapeutic modality could lead to adverse health effects for the animal due to misdiagnosis or inappropriate treatment.

(105) Veterinary staff receive most of the radiation doses associated with veterinary radiological procedures, either when operating radiological equipment; holding image detectors; restraining animals during diagnostic procedures; performing or assisting in nuclear medicine, interventional, or therapeutic procedures; or caring for animals after nuclear medicine diagnostics or therapy with sealed or unsealed radioactive sources. Most doses to staff will be low (see Section 6.2.2), but over time could potentially contribute to the development of stochastic effects. Additionally, epidemiologic studies of radiation workers in human medicine note a higher incidence of cataracts in both interventional proceduralists and nuclear medicine technologists who receive chronic low-dose exposures (ICRP, 2012b). Higher dose procedures, such as long interventional procedures, potential spills in nuclear medicine, or accidents relating to therapeutic procedures could potentially lead to deterministic effects. Owners or handlers exposed to radiation when assisting in radiological procedures or caring for animals after nuclear medicine procedures may also be at risk, albeit low, mainly for stochastic effects. The assistance of members of the public in radiological procedures is currently a subject of debate, and will be discussed further in Section 6.2.1.

(106) Environmental contamination may also occur after diagnostic or therapeutic nuclear medicine procedures, either through releases from the veterinary facility where the procedure is carried out, or through radioactivity eliminated from the animal after its discharge from the veterinary facility. While releases at the veterinary facility are often well controlled and regulated, uncertainty exists after the animal is discharged. These uncertainties will depend on the level of radioactivity in the animal at the time of discharge, the mechanisms of elimination of the radionuclide used, the veterinary practices’ recommendations for isolation of the animal and management of its waste, and the degree of owner or handler compliance with these recommendations. Environmental contamination may lead to radioecological effects, as well as to human exposure through external radiation or internal contamination. The nature and extent of the consequences of environmental contamination will depend on the type, amount, and duration of the contamination event (ICRP, 2014a).

(110) Radiological examinations are now common practice from a young age (e.g. screening tests) and for the life of certain animals. Pet animals tend to live much longer than they used to due, in part, to robust veterinary care, including earlier diagnosis and specialised medicine (Cozzi et al., 2017). With the increase in prevalence and frequency of radiological examinations, and the increased life span of companion animals, there is a need for more attention to optimisation of protection and safety in veterinary procedures that explicitly include radiological protection considerations with regard to the exposed animals.

(111) Optimisation can generally be achieved by: (1) appropriate design and construction of installations, and careful selection of equipment; and (2) day-to-day strategies such as adequate and regularly updated education and training of staff, clarity with regard to their exact roles and responsibilities, routine performance tests of equipment, and systematic application of procedural rules, all embedded in a safety culture at organisational level. The optimisation process may also include dose constraints based on reasonable, good practice, particularly as veterinary staff often perform various services (IAEA, 2021). This approach is consistent with what is advocated for the practice of human medicine (ICRP, 2007b). At present, no guidance is available related to, for instance, animal-specific DRLs or similar imaging guidance parameters. However, considering the societal value of animals, optimisation strategies relevant for human patients should also be valid for animal patients.

(112) Optimisation should not be confounded with dose minimisation. Too much focus on dose reduction alone may impede the diagnostic or therapeutic quality of the procedure, and result in suboptimal care or necessitate a repeat procedure. This clear distinction between dose optimisation and dose minimisation is critical in radiotherapy, where underdosage may lead to insufficient tumour control, and even optimal procedures may result in the inevitable appearance of early or late side effects. Moreover, risk induced by radiation exposure is only one of the elements to be taken into consideration, and optimisation of protection and safety therefore needs a holistic view comprising not only broad animal welfare considerations but also general safety aspects for staff members and members of the public.

(113) Veterinarians and associated staff face many occupational challenges and hazards, of which exposure to ionising radiation is just one. For example, other hazards such as bites, scratches, or kicks may be more important, and certainly more acute, issues. Thus, the optimisation process for veterinary workers should broadly encompass consideration of risk, benefit, and practicality. In other words, the level of protection should be optimised in a way that most reasonably accounts for the given circumstances, as consistent with other exposure situations. Gloves might be worn when handling a patient prone to biting, but if the patient is afraid of gloves to the extent that an examination cannot be conducted, it may be prudent to leave them off and consider an alternate strategy.

(114) Similarly, sedation and anaesthesia are frequently advocated from a radioprotection point of view, but in some cases, the associated detrimental impact on the animal’s health may lead to the conclusion that this may not be the best option for patient restraint. Where permitted, optimisation could then consist of having the owner restrain the animal, even though this might result in some radiation exposure to this person, which, in turn, should be mitigated by providing clear instructions and – where applicable – adapted protective equipment.

(115) Optimisation clearly also applies to members of the public, defined in the system of protection as individuals who receive an exposure to ionising radiation that is neither occupational nor medical (ICRP, 2007a). With respect to veterinary practice, the public may include pet owners/handlers, clients in a waiting area, farm hands assisting with an equine examination, etc. In some countries, members of the public will not be allowed to assist in veterinary radiological procedures. In such instances that they can be allowed to assist in some procedures (see Para. 120, for example), the following conditions should be fulfilled: (1) the procedure is justified; (2) the person’s presence is overall beneficial from a ‘holistic’ perspective, as discussed above (see Para. 111); (3) the person, after having received relevant information regarding potential risks, agrees to undergo some limited exposure; and (4) after having been instructed on how to behave (where to stand, where to put their hands, possibly what protective equipment to use, etc.) in order to minimise their exposure. Children and individuals who are pregnant should not be allowed to assist in such radiological procedures.

(116) Given the great number and diversity of elements to consider in any specific case, optimisation needs to be tailored to best fit, within the boundaries of what is prudent and reasonable, and the needs of each individual case. This individual approach should first consider the clinical needs, but also the whole environment in which the procedure takes place (e.g. owners’ wishes, location and transport facilities, available equipment, etc.).

(117) Prudence is highly relevant to the process of optimisation, consistent with other areas of veterinary practice that use potentially harmful substances or principles; for example, if 50 mg of a drug would suffice to obtain the desired clinical effect, it would not make sense to use 100 mg. Considering the wide variety of risk factors present in a given circumstance and making value judgements as to the most reasonable choice necessarily involves prudence. In situations that are unfamiliar, rare, or without precedent [as may be the case with exotics or zoo animals (e.g. Adkesson and Ivančić, 2019; Schilliger et al., 2020)], it may be prudent to consult a qualified expert (board-certified veterinary radiologist, radiation physicist, safety officer, or other individual with recognised competence in radiation safety) in advance of the procedure for guidance.

(119) Radiologic procedures should be performed in an adequately safe environment. The room should be spacious enough to allow people to keep sufficient distance from the radiation sources, and it should be equipped with shielding commensurate with the procedures performed. Hazards may arise when a room initially conceived for occasional standard small animal radiographic procedures has become a room in which CT or interventional procedures are performed, or simply when the number of procedures performed rises well beyond that which was taken into consideration when the room was first conceived and constructed.

(120) A designated area for radiological examinations or therapeutic procedures should be established and physically demarcated with warning signs (e.g. Fig. 6.1) to limit unnecessary public exposure. As many equine radiography examinations are performed in stables with mobile generators, additional measures should be taken to delineate the exposure area to avoid unforeseen exposure of members of the public who are not involved in the examination. Performing such procedures in stables with solid concrete or brick walls should be preferred where this is possible, because of the shielding offered. Placing signage at the entrance can then suffice. If procedures need to take place in the open field, delineating the designated area with appropriate signage is much more demanding as it needs to consider all risks involved, not just radiological hazards.

(121) In general, members of the public should be kept outside areas where radiation activities are performed, and in the small animal veterinary setting, pet owners should typically not be asked to help during radiological procedures. However, there may be some circumstances in which an owner’s presence comforts the animal in a significant way, resulting in a more efficient and, in some cases, physically safer examination. This, in turn, could reduce the overall exposure to the technologist, for instance by reducing the need for repeat exposures. In other instances, it might be inappropriate to include members of the public or owners/handlers due to the nature and frequency of the exposure, and/or the characteristics of the person considered. For example, a young person working at a stable may wish to assist in every horse’s radiograph series, yet this would likely do more overall harm than good. The decision on whether to allow lay-person assistance in an examination results from a balancing exercise of pros and cons, and is similar to that in human paediatrics (i.e. parents and carers, but owners and handlers in the veterinary setting) (ICRP, 2013b), and needs to be made prudently, focused on beneficence and non-maleficence, and considering the prevailing circumstances. Children and individuals who are pregnant need specific consideration and, in some countries, may be legally excluded from participating in such activities. If the presence of members of the public is judged to be required or useful, then rotation of these persons may be considered in order to limit the exposure of any single individual. Similarly, when deciding on where to perform a procedure, it may be that leaving an animal in a familiar environment (e.g. a horse in its stable) may bear more risk of radiation exposure, yet will be beneficial overall, and thus the most reasonable and optimised choice.

(122) Any individual involved in a radiological examination should avoid – as much as possible – being exposed where the intensity of the radiation field is highest, such as in the primary radiation beam. Where reasonable, positioning and immobilisation aids and/or patient sedation/anaesthesia should be considered to reduce staff and owner/handler exposures. Similarly, when possible, personnel should stand behind fixed or mobile protective shields; for example, exposure of the head, neck, and upper body of the veterinarian performing an interventional procedure can be reduced considerably by the adequate use of a ceiling-suspended shield. Optimally, neither portable x-ray generators nor the associated cassettes should be handheld. A recent study simulating exposure from scattered radiation in equine radiography to the hands of workers holding the cassette or the x-ray generator found that doses ranged from 0.26 to 2.64 µGy/study⁻¹ and from 0.84 to 12.09 µGy/study⁻¹, respectively, without the use of hand shielding. Hand shielding reduced doses by at least 98%, depending on the type of glove used (Belotta et al., 2022).

(123) In some cases, such as with interventional radiology, it is necessary for staff to perform a variety of tasks within the radiation field for varying times and at different distances from the source. Where external radiation exposure is a concern, in addition to the use of protective shields, the use of shielding PPE should be considered, including protective wraparound aprons, hand/forearm protectors, thyroid collars, and eye protection (e.g. lead safety glasses), depending on the specific circumstance. Note that open-palm hand shields or mitts have limited use; in general, fully enclosed leaded gloves should be used rather than open-palm shields during manual restraint of an animal patient, and it should be borne in mind that even fully enclosing gloves only provide limited protection when the hands are positioned in the primary x-ray beam, which should always be avoided (Mayer et al., 2019b). Shielding properties of the PPE selected for a procedure should be balanced against other workplace hazards. For example, the weight of a lead apron can result in orthopaedic issues such as strain on the lower back if worn for long periods of time (Martin and Sutton, 2015; Alexandre et al., 2017). This, along with the restriction of movement, can increase working time as well as result in physical injury; as such, a vest/skirt configuration or the use of lighter aprons, made of so-called ‘lead-equivalent materials’, may be preferred. Similarly, wearing radiation protective gloves while working close to the animal’s irradiated body volume will reduce dose to the extremities and is frequently warranted (Stoeckelhuber et al., 2005). However, use of these gloves will have a negative effect on dexterity and range of motion, which may lead to safety concerns associated with increased muscle fatigue and working time (Martin and Sutton, 2015).

(124) With respect to equipment, optimisation of radiological protection involves ensuring that radiological equipment is suitable for the task at hand, and that exposure parameters are adequately tailored to animal patients and veterinary working routines. The use of radiological equipment in a veterinary setting may be off-label (i.e. not used as originally intended or designed) for new or refurbished medical equipment, or dedicated to veterinary practice by design. For all types, it is recommended that the manufacturers should continue to maintain the equipment and that no modification occurs that would decrease image quality and/or radioprotection properties (e.g. inner shielding, collimator). For equipment designed specifically for veterinary use, the manufacturer is often able to alter the components of the equipment for which medical standards are no longer legally ‘needed’ (no established standards of installation and performance). Such changes often have a positive impact on the selling price of the equipment, yet possibly a negative impact on image quality; output stability; and radiological protection of the animal patient, the veterinary professional, and members of the public. For example, reduced inner shielding of portable radiography units results in significantly increased amounts of leakage and scattered radiation. In a number of countries, industrial standards are applied when dealing with veterinary equipment, and this may be insufficient both from a veterinary care perspective and a radiological protection perspective. The Commission therefore recommends that adequate, fit-for-purpose standards should be applied on all equipment marketed and used in veterinary applications of ionising radiation, and suggests that responsible authorities should consider applying appropriate standards for the accreditation of the equipment and the credentialling of staff members. Ideally, these standards would be recognised internationally, as manufacturers often sell equipment in multiple countries. Of note, equipment standards should also include requirements on the device connections that allow installation in a dedicated veterinary room (e.g. light signalling at the room entrances, emergency stops, door switches).

(125) Optimisation measures for patient protection in veterinary diagnostic radiology, for the same image quality, should be discussed with the manufacturer and installation engineer, and implemented when possible. This would include considerations such as limitation of views to those necessary for common diagnostic protocols and technique charts for the range of animal sizes relevant to the facility. Similar strategies apply for CT examinations with a special procedure for auditing repeat examinations and requests for systematic whole-body imaging. Standardisation of national referral guidelines for when and what imaging should be done for common situations, and standard protocols that describe how to perform the imaging examination would greatly aid veterinary practice worldwide in caring for animals and increasing radiation safety. Examinations should not be repeated if no clinical benefit would be obtained. In other words, aesthetically pleasing images should not be the preponderant consideration; rather, the image quality needs to be sufficient to confidently make a diagnosis or proceed with an interventional procedure with the lowest possible exposure. The priority for a diagnostic image is that it is interpretable, which relies not just on the physics of the image (e.g. resolution and contrast) but also on factors such as how and where the data are displayed, the ambient environment, and the experience of the person reading the images. Reasonable reduction of the animal dose and improvement of study quality contribute to the optimisation of protection and safety by reducing doses received by both the animal patient and staff.

(126) A highly important step in optimisation of radiographic procedures of any kind is to limit the exposed tissue volume to what is relevant for the clinical case at hand. In standard diagnostic radiology and interventional fluoroscopy, this should be achieved by appropriate beam collimation; in CT, this should be achieved by scan length limitation. These simple measures lower patient dose and – by reducing radiation scatter generated in the exposed tissues and materials – improve image quality and reduce the exposure of professionally exposed persons as well as members of the public.

(127) A prerequisite for optimisation is a thorough knowledge of the doses associated with a given exposure situation, as well as the factors that influence this dose. Reported doses per image to persons participating in radiographic examinations of small and large animals (Ackerman et al., 1988; Seifert et al., 2007; Hupe and Ankerhold, 2008 , 2011; Barber and McNulty, 2012; Eckert et al., 2015), or per examination for personnel present during standing CT examinations of the equine head (Dakin et al., 2014), fall in the range from 0.1 to 34 µSv. Doses towards the higher end of the range are typically encountered when thicker body parts are being radiographed, such as the abdomen in large dogs, or the equine head, spine (especially thoracic and lumbar regions), and proximal extremities. While several of the above studies state that estimated annual doses will be well below regulatory limits for a given caseload, other studies of occupational doses in veterinary medicine have found that personnel doses may approach annual dose limits recommended by ICRP (Table 4.1) (Hernández-Ruiz et al., 2012; Canato et al., 2014).

(128) Recently, dosimetric data have been published for veterinary interventional radiology and intra-operative fluoroscopically guided surgery, where there is close proximity between personnel and animal patients during exposure, often for extended periods (Sung et al., 2018; An et al., 2019; Hersh-Boyle et al., 2019). Reported operator dose levels may approach or even exceed regulatory limits, which emphasises the need for both quantitative radiation monitoring and the use of appropriate protective measures during these procedures.

(129) With regards to dose to the animal patient, few dosimetric studies have been published. Primary beam doses or entrance surface skin doses, typically of the order of 1 mGy, have been reported with the aim of assessing their contribution to personnel dose (Veneziani et al., 2010; Barber and McNulty, 2012). However, dosimetric publications aimed at the radiation protection of the animal patient are emerging. Nemanic et al. (2015) addressed the potential of lead shielding to reduce dose to the animal during elbow radiography in dogs, and Hersh-Boyle et al. (2019) reported radiation exposure of dogs and cats undergoing intra-operative fluoroscopic procedures. In the latter study, doses up to 617.5 mGy were reported. However, systematic reporting of dose descriptors such as the dose area product and CT dose index for clinically relevant protocols, both within and between institutions, are lacking in veterinary medicine; hence DRLs or similar do not exist. Furthermore, while the relationship between these dose descriptors and radiation risk in the form of effective dose has been established in human medicine through the use, for example, of anthropomorphic or patient-based voxel phantoms (e.g., ICRP, 2009a) and Monte Carlo simulations, such links still have to be established in veterinary medicine (although, as mentioned earlier, some phantoms such as these have been developed for animals, including canines). The number of different species involved, as well as the range of patient sizes within a species, may be relevant challenges in veterinary medicine.

(130) More dosimetric data are needed, both for personnel and animal patients, particularly for potentially high-dose procedures, such as interventional radiology and fluoroscopically guided surgical procedures. Furthermore, as CT interventional procedures become more prevalent in veterinary medicine, dosimetric aspects of these procedures should also be addressed. Systematic reporting of dose descriptors for clinically relevant protocols will be necessary to compare such protocols, both within and between institutions, and thus to optimise such protocols with respect to dose. The relationship between dose descriptors, organ doses, and associated radiation risk must also be determined for veterinary medicine.

(132) Safety measures to prevent contamination with radioactive substances can be implemented at the source or the worker, and are consistent with general industrial hygiene practices for protecting workers from other types of contaminants. Example methods for confining or containing a radioactive source include storing radioactive material in a secure, shielded location; limiting the handling of radioactive materials to well-defined areas within a practice (e.g. a secure drawing up area with appropriate mobile shielding); using secondary containment (e.g. trays, buckets) to limit the consequences of possible spills; and using a ventilated hood with sufficient and consistent air flow. Good housekeeping practices (i.e. cleanliness and organisation), regular radiological surveys, and detailed record keeping are also important for the prevention of contamination.

(133) External radiation safety measures follow those described in Section 3.4.1. Specific examples in nuclear medicine include using an appropriately shielded syringe, using lead containers and/or hand carts to transport the radiopharmaceutical to the receiving patient, and taking one step back from the injected patient where possible.

(134) The PPE used is essentially aimed at preventing contamination risks by the radioactive material involved. For example, when injecting, radiopharmaceutical impermeable gloves, a long-sleeved laboratory coat, and a face mask or shield should be worn to limit skin exposure in the situation of back pressure when injecting into a catheter.

(135) With respect to the patient, it is important to be aware of the potential for deterministic effects in patients undergoing certain nuclear medicine procedures. These effects may be unavoidable to some extent (e.g. therapy). In nuclear medicine therapy, there may well be side effects; for example, effects on salivary glands when treated for thyroid cancer with radioiodine. Of course, there are also the potential consequences of extravasation (i.e. when the radiopharmaceutical ends up next to the vein through which it was supposed to enter the body) (van der Pol et al., 2017).

(136) It should be kept in mind that the administered activity of a given radioisotope or radiopharmaceutical will, to a large extent, determine the radiation risks to the animal itself, to any humans involved, and to the environment. Prudence can provide insight into whether additional dose (activity) should be used to speed up a nuclear medicine procedure, or whether longer sedation or anaesthesia would be appropriate. Different situations require different approaches, always considering the ALARA principle. For example, there are two standard protocols in PET imaging, based on the timing of radiopharmaceutical injection and induction of anaesthesia. The protocol in which anaesthesia is induced prior to injection has a longer anaesthesia time (up to approximately 2 h) but lower radiation doses to personnel compared with the protocol in which anaesthesia is induced after injection. Reported doses per procedure to personnel participating in veterinary PET studies range from 0 to 30 µSv (Martinez et al., 2012; Hetrick et al., 2015), and estimates of the total annual effective doses to personnel associated with the latter protocol are well within the annual occupational dose limit (maximum approximately 5 mSv assuming 100 animal patients per year) (Martinez et al., 2014). Other considerations beyond anaesthesia time and radiation dose include keeping the animal as still as possible during the radiopharmaceutical uptake period in order to avoid unwanted uptake in active muscles.

(137) In order to protect staff, members of the public, and the environment from the consequences of radionuclide administration to an animal, particularly after therapeutic procedures, the animal may need to be hospitalised so that its excrements may be collected and treated as radioactive waste. The risk of contamination from the animal itself usually subsides rather rapidly because of natural elimination, mostly by the kidneys through urine. However, it may take several days or even weeks before the dose rate emitted by the animal has fallen below the threshold values for its release and return home (Davila, 2019).

(138) Hospitalisation, particularly for long durations, needs to be considered as a potential welfare issue for both the animal and its owner or carer (Graf, 1999; Boland et al., 2014; Johansson et al., 2014). Again, radiological protection concerns need to be balanced against and considered together with all other values at stake. Hospitalisation creates a stressful situation, especially in small animal pets (dogs, cats), as it has now been shown that animals have feelings, likely evolved to protect primary needs (Hewson, 2003; Lloyd, 2017). With the progressively more prominent place that animals, particularly companion animals, have gained in human society, it can be as stressful for pet owners to have their animal in the hospital for a long duration as it is for the animals themselves (McConnell et al., 2011 , 2017; Amiot et al., 2016).

(139) In view of the complexity of nuclear medicine procedures on animals, resulting, in part, from the need to manage external exposure and contamination risks simultaneously, veterinary nuclear medicine should only be performed by veterinarians and staff members that have completed specialist training programmes successfully. This is even more compelling for therapeutic applications.

(141) The high doses and dose rates applied also have the potential to cause serious risks to staff members involved in these procedures. Blocked sources in remote after-loading or accidental ‘beam on’ situations in teletherapy could generate these types of risks, whereby other deterministic effects than just skin burns cannot be excluded. Strict procedures must be in place to allow the most optimal and safe use. Such complex and high-risk procedures should only be performed by veterinarians who have completed extensive education and training in radiological protection. From a veterinary care perspective, it may be preferable that the radiological practitioners responsible for these procedures are diplomates of speciality education and training programmes, bearing in mind the current curricula may be insufficient when it comes to addressing the radiation hazards specifically. The Commission therefore recommends that the providers of such education and training programmes better embrace radiological protection as an indispensable and integrated element of quality care.