What features are best interpreted as indicators of pain or distress in rodents?
can influence the stress response include age, sex, physical fitness, experience, disease, and medication.
Neural and endocrine substances are released in response to a stressor and play an important role in the initiation and coordination of the behavioral, cardiovascular, and immunologic responses to stress. The site of origin of those substances influences the physiologic stress response, which has two categories of duration: a rapid increase in circulating neurally derived substances with a short duration of action and a slower increase in endocrine-derived substances with a longer duration of action.
Acute stress—surgery, postoperative pain, burns, anesthesia, cardiac arrest, exercise, and sometimes headache—results in the hypothalamic secretion of corticotropin-releasing factor (CRF). CRF then stimulates the pituitary to cosecrete adrenocorticotropic hormone (ACTH), which promptly stimulates the release of corticosteroids from the adrenal cortex and the opioid peptides β-lipotropin and β-endorphin. In humans, plasma concentrations of these hormones can increase by a factor of 2–5 during stress. The roles of β-lipotropin and β-endorphin in the stress response are poorly understood. β-Endorphin might play a role in pain modulation (Hargreaves et al., 1983, 1987; Pickar et al., 1983; Szyfelbein et al., 1985). A potential target of circulating β-endorphin could be inflamed tissue (Joris et al., 1987).
The major example of neurally derived responses to acute stress is the activation of the sympathoadrenal system. The pituitary-adrenal axis responds with a prompt increase in catecholamines from the adrenal medulla, which is usually followed by an increase in corticosteroids. Plasma concentrations of catecholamines increase considerably after surgery or postoperative pain, because of increased circulating concentrations of epinephrine, primarily from the adrenal medulla, and norepinephrine, primarily from sympathetic nerve terminal activation (Goldstein, 1987). Those different components of the sympathoadrenal system can be regulated separately. The sympathoneural release of norepinephrine evokes a regionally selective effect, whereas the adrenal medulla produces an increase in systemic plasma concentrations of epinephrine. The physiologic consequences of activation of those systems include increased cardiac output, increased skeletal muscle blood flow, cutaneous vasoconstriction, reduced gut motility, and increased glucose availability.
Hormonally derived responses to stress result from changes in concentrations of growth hormone, prolactin, glucagon, insulin, vasopressin, neuropeptides, and other hormones. Acute stress causes increased secretion of growth hormone and prolactin, both from the anterior pituitary. Prolactin participates in metabolic processes and might modify nociception, but concentrations remain in the normal range in the presence of chronic stress. The significance of the short-lasting growth hormone increase is not known. Glucagon increases in response to acute stress, and insulin decreases. The decrease in insulin—in combination with increased prolactin, growth hormone, glucagon, and epinephrine—contributes to the development of hyperglycemia.
Ensuring that laboratory rodent pain is well managed underpins the ethical acceptability of working with these animals in research. Appropriate treatment of pain in laboratory rodents requires accurate assessments of the presence or absence of pain to the extent possible. This can be challenging some situations because laboratory rodents are prey species that may show subtle signs of pain. Although a number of standard algesiometry assays have been used to assess evoked pain responses in rodents for many decades, these methods likely represent an oversimplification of pain assessment and many require animal handling during testing, which can result in stress-induced analgesia. More recent pain assessment methods, such as the use of ethograms, facial grimace scoring, burrowing, and nest-building, focus on evaluating changes in spontaneous behaviors or activities of rodents in their home environments. Many of these assessment methods are time-consuming to conduct. While many of these newer tests show promise for providing a more accurate assessment of pain, most require more study to determine their reliability and sensitivity across a broad range of experimental conditions, as well as between species and strains of animals. Regular observation of laboratory rodents before and after painful procedures with consistent use of 2 or more assessment methods is likely to improve pain detection and lead to improved treatment and care—a primary goal for improving overall animal welfare.
Abbreviations: CFA, complete Freund adjuvant; CPP, conditioned place preference; CPA, conditioned place avoidance; HPA, hypothalamic-pituitary-adrenal gland; MGS, mouse grimace scale; RGS, rat grimace scale; TINT, time to integrate to nest; USV, ultrasonic vocalization
The scientific study of pain in humans and animals and its underlying mechanisms have been the subject of extensive research for over 200 y.109 Despite this, theories underpinning our current understanding about pain signal transduction and related mechanisms were developed only 50 y ago, with Melzack and Wall's proposal of a gate control theory to describe the mechanism of spinal cord transmission and regulation of pain signaling (reviewed by82,89). With improved understanding about the underlying mechanisms of pain came improved recognition of pain as a disorder in its own right,113 and a desire for human clinicians to improve assessment and treatment of pain in their patients.82 Development of assessment methods for pain in domestic animals mirrored those in human species, although it was not until the late 20th century that pain management of veterinary species and research animals were prioritized by veterinarians and regulatory agencies. Today, it is recognized that accurate identification and assessment of pain are essential for refining care of research animals undergoing painful procedures and improving the validity of translational pain research.90,108 Animal ethics committees and regulatory authorities require researchers and laboratory animal veterinarians to assess and manage pain in animal subjects. However, even today, this is seldom a straightforward task—particularly for research rodents.138
Defining Pain and Nociception
Pain and nociception are often confusingly described and poorly defined, leading to their incorrect usage or the assumption that they are synonyms. For clarity throughout this paper, the definitions of the International Association for the Study of Pain54 are used:
Pain: “An unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage.”
Nociception: “The neural process of encoding noxious stimuli.” This process may have autonomic (for example elevations of heart rate or blood pressure) or behavioral (for example withdrawal reflex or more complex nocifensive behavior, such as licking or rubbing) consequences. Importantly, nociception can occur without the sensation of pain.”
Challenges to Observing and Monitoring Pain in Laboratory Rodents
Because animals are nonverbal, assessment of pain involves observing surrogate measures of pain and signs of animal wellbeing and then making a judgment about the animal's condition based on the interaction between these 2 sets of data. Assessing the presence or severity of pain from only 2 subjective measures provides only an approximation of a painful state. Many commonly used measures of pain are indirect, including activity, heart rate, blood pressure, and body weight, all of which can be altered by factors other than pain, thus confounding their interpretation. Furthermore, many measures require a sound knowledge of species-typical behavior for accurate interpretation. Even in humans, the primary factors driving the experience of pain; that is, motor, sensory, and psychologic, were rarely studied or assessed together until recently.13 This means that approaches used for the assessment of pain in laboratory rodents may be suboptimal, and thus multiple measures of pain assessment should be used together to improve accuracy.153
To further complicate monitoring pain in laboratory rodents, a significant confounding variable may be the sex of the individual who is assessing the animal. Direct exposure of mice and rats to men or male scent increased defecation rates in control animals of both sexes, as well as decreasing perceived pain intensity, as detected by various standard assays, such as tail and paw withdrawal tests and von Frey fiber thresholds (see Figure 1).131 This sex-based stress-induced analgesia may be responsible for conflicting opinions about animal wellbeing in studies with painful outcomes.
Open in a separate window
Common methods to assess analgesic efficacy in rodents.
Finally, because rodents are nocturnal, more accurate pain assessments would be likely if animals were monitored for pain during the most active times in their circadian cycle. Nociception is most acute in the dark phase of their day/night cycle,20,86 concurrent with the time of their lowest responsiveness to opiate analgesics,105 and peak activity of inflammatory signaling pathways, as controlled by their internal molecular clocks.16 Animal monitoring and assessment during the dark phase would require widespread changes in practice, including housing management, and there are practical difficulties in achieving this.
Pain Assessment and Monitoring Methods in Laboratory Rodents
Many methods have been proposed for assessing and monitoring pain in mice and rats (see Figure 1 for an overview of methods), including examining the patient directly and making a subjective assessment of animal comfort based upon hair coat, posture, general activity level, and degree of alertness.1,65 Unfortunately, very simple assessment methods may fail to detect subtle changes in animal behavior and comfort, and a nonstandardized approach to pain assessment is likely to result in wide variations of opinion between individuals looking at the same group of animals. This has been demonstrated to occur for larger animals, such as cattle and sheep, and is not expected to be any different for much smaller animals, such as mice and rats, which are harder to observe.64,140 Handling and restraint increase the potential for stress-induced analgesia, so pain assessments that can be conducted with unrestrained animals are preferred. However, the question remains: how should the clinician and researcher evaluate pain monitoring methods in the face of such a wide array of pain assessment techniques?
Validation of Assessment Methods
Authors of papers that include a pain assessment scale often use the term ‘validated’ to claim that the scale can be relied upon to measure the outcome of interest. Unfortunately, this is often not the case. The scale may have never been formally validated, the context (for example, the specific pain model) in which it was originally developed may differ from the subsequent clinical use, or the scale may have been modified in some way by other users, such that the effect(s) of this change on scale validity are unknown. In general, pain scale development and validation should be viewed as a continuum rather than a discrete process. For example, it is common to assume that a scale developed in one strain of mice or rats under specific conditions and applied by a particular group of users will perform equally well under different circumstances. This assumption that scale performance is a fixed property is an oversimplification and fails to account for the multitude of factors affecting scale function, and therefore, makes it difficult to compare studies.
An overview of the key steps in scale validation are briefly presented as a background to the discussion of different pain assessment methods. The key concepts of scale development are validity and reliability. For more in-depth coverage of this topic, interested readers are referred to the review by Streiner and colleagues.136
A simple definition of validity is: does a scale measure what it claims to measure?103,136 It is important to know if the items making up a scale are all necessary and important (face and content validity), how a scale compares to a gold standard should one exist (criterion validity), and whether the scale can identify meaningful changes (construct validity).
Content validity is concerned with the scale items and whether they fully capture the measure of interest. Content validity is frequently established through face validity, which uses the opinions of experts to determine the scale items. Initially, it is often useful, to begin with, a large number of potential items, reducing the list based on discussion and consensus. More formal methods to assess utility of potential items include evaluating the relationships between items and the effect of item omission on scale performance.
Criterion validity is the comparison of a scale with an accepted standard. In adult humans capable of verbal communication, the self-report is usually considered the ‘gold standard’. In contrast, where self-reporting is impossible, such as in animals and neonates, it is often argued that criterion validity cannot be performed. Therefore, it is common to employ construct validity as a substitute for criterion validity in pain scale development in animals.Construct validity is experimental testing of a hypothesis based on what is known (or assumed) about a construct like pain, anxiety or depression. An example of this is the assumption that giving an analgesic will reduce the pain scale score or that pain scale scores will increase immediately after surgery and return to baseline as inflammation subsides.90 Because construct validity is only limited by the number of hypotheses that can be generated, it should be viewed as an ongoing rather than static process. Additional useful components of validity are sensitivity and responsiveness. These terms describe the ability of a scale to detect a change, and although they are often used interchangeably, sensitivity can be considered detection of any change whereas responsiveness is detection of an important or relevant change.75
Reliability is the amount of error associated with a measurement scale; that is, the reproducibility of the results.143 For a scale to be useful, this statistical error should be appropriate for the encountered range of observations. For example, a weight scale for use with cats with accuracy reported to ± 0.1 kg would be acceptable but would be of no use for weighing rats. Similarly, the measurement error of a pain assessment scale should be smaller than the range of scores reflecting an important change in pain level. This error can be assessed for various situations, including differences between different users (interrater reliability), and differences within the same user (intrarater reliability). Commonly employed measures of reliability are the intraclass correlation coefficient (ICC) and κ coefficient. For both measures, the higher the level of correlation observed the better the reliability. As reliability decreases, the harder it becomes to detect small differences (the user must decide if these differences are worth trying to detect). Therefore, reliability sets the limit for validity, and if measurement error is large, it will not be possible to detect important differences. Importantly, reliability does not refer to the scale, as in the reliability of the scale, but rather, to the results obtained with the scale when applied in a given situation.
As this discussion indicates, describing a scale as validated is both misleading and an oversimplification. Validity and reliability are all limited by the population and context in which the testing was performed. Therefore, the best that can be said is that a scale is valid for the population studied within a defined context.136
Use of Algesiometry Assays for Assessment of Pain
Algesiometry assays are standardized tests of evoked response reflexes that can be used as models of nociception in rodents and other species and to determine the potency of a given analgesic agent or regimen.8,71 Examples include the tail-flick, hot plate, acetic acid writhing, and von Frey tests (see Figure 1). During research, these tests are widely used because the assays are relatively simple to perform, inexpensive to run, and many can be automated.94,107 While they have been used for decades to assess aspects of nociception,36 these assays likely represent an oversimplification of the condition of pain as it is understood today. For example, classic algesiometry assays all require the observer to remove the mouse or rat from its home enclosure and manually restrain them prior to initiating the test which can result in stress-induced analgesia. The results from these tests are sex- and strain-dependent, and are only effective in reflecting changes in a limited number of chronic pain models, for example, paw swelling but not chronic back pain or migraines. Likewise, interpretation of the results can be confounded by many parameters, such as impaired animal locomotor ability, rapid learning of the expected test response by the animal, observer sex and its influence on animal behavior,131 and social housing conditions and interanimal cues.68
More recently developed tests for assessing spontaneous pain in mice and rats are likely to be of increased utility, and many of these are discussed in subsequent sections below. These include behavioral assessments of nonspecific or specific activities (ethograms), ultrasonic vocalizations, facial expressions of pain (that is, pain grimacing), burrowing, open field tests (in which more anxiety and less exploration are seen in a number of animal models of pain), free-choice thermal preference assays, voluntary wheel running tests, conditioned place preference tests, and conditioned place avoidance tests (for a review, see reference 138). While more difficult to design, and interpret, these tests are potentially more sensitive and specific measures for detecting pain in mice and rats. As novel methods for assessing pain are developed, it will be necessary to characterize their relationship with traditional (evoked) measures to determine their utility.
Behavioral Assessments of Pain
When implementing any metric for assessing pain in rodents, it is imperative to first understand the baseline for the particular patient population, inclusive of species, strain, sex, age, and health status of the animals. Also, factors that could potentially confound the assessments chosen should be considered, such as the presence of a human observer. Additional confounders may stem from the experiment itself, such as residual effects of anesthetics or debilitation to motor function or cognition as a component of the model. Because of the long list of variables that can impact pain assessments, a triangulation approach, using multiple methods that assess various components of the pain experience, such as, evoked, nonevoked, physiologic, clinical assessments may be needed.
The presence of an observer is an often underappreciated confounder to animal welfare assessments. Prey species have a natural drive to suppress pain behaviors in the presence of another animal, especially if it is perceived that it could pose a threat. Mice demonstrate the ability to suppress grimacing and guinea pigs have been shown to suppress pain-specific behaviors in the presence of a human observer.85,102 Therefore, methods that can be conducted indirectly, without the presence of a human, such as via remote video, can be particularly helpful for pain assessments in rodents.
Understanding the impact of anesthetics and analgesics on pain assessments is also critical for accurate identification of unalleviated pain. Depending on the species of the patient, anesthetic and analgesic drugs may suppress or enhance measurements of postprocedural pain. For example, buprenorphine can cause sedation resulting in a decrease in ambulation in guinea pigs, and hyperactivity in mice; in both instances behaviors observed in pain-free mice, such as rearing and grooming, are reduced.23,49,79,84,102,119,151 Interestingly, buprenorphine does not influence mouse grimace scores, whereas isoflurane can increase grimace scores in both mice and rats, due to unknown mechanisms.84,85 Given the varying direct effects of anesthesia and analgesia on pain assessments, it is helpful to understand the half-life of administered drugs and what effects they will have on the assessments used. Comparing known effects of anesthetics and analgesics on the postprocedural condition can highlight pain-specific changes in behaviors. If evaluating change from baseline or change from an anesthesia/analgesia condition, a smaller change indicates fewer differences between the nonpainful and potentially painful conditions and therefore better pain control.
Ethograms capture 2 major types of nonevoked responses useful for pain assessment: loss of normal behaviors, such as rearing and ambulation, and presence of new pain-specific behaviors, such as back-arching (Figure 2 A), writhing (Figure 2 B), weight shifting and staggering.120-123,150,151 General ethograms are available for use, or custom ethograms can be created for a particular patient population or model system by crafting clear, detailed descriptions for all behaviors for inclusion in the assessment.35,41 Normal behaviors generally occur with greater frequency and may be easier to score using automated software.151 However, new pain-specific changes in behavior, such as increased paw licking in certain cancer pain models in mice,5 may be subtle, performed quickly, and occur more fleetingly, lending themselves more to manual scoring.35,123
Open in a separate window
Examples of pain-associated behaviors demonstrated after a laparotomy in an adult male Sprague–Dawley rat. (A) Back arch—this behavior is described as a vertical cat-like stretch upward. (B) Writhe - this behavior is described as the contraction of the abdominal muscles (arrow).
In some cases, a surrogate indicator can be used to assess animal wellbeing or pain. Nesting behavior and burrowing are examples of surrogate behaviors that can be used when assessing pain in mice. Grooming is another surrogate behavior that can be used to indirectly assess painful states. This can be achieved through the Grooming Transfer Test (Figure 3).104 The Grooming Transfer Test takes advantage of the fastidious nature of mice and their highly patterned grooming behavior. A nontoxic, inert powder that fluoresces under black light is suspended in mineral oil and applied to the top of a mouse's head. As the mouse grooms, they transfer the fluorescent signal to additional body locations, the cage environment, and nesting material. Eventually, the mouse's normal grooming behavior will completely remove the oil/powder suspension. This behavior is conserved across inbred and outbred mice, is performed by both males and females, and allows for individual assessment of mice, even when socially housed. The Grooming Transfer Test exhibits construct validity as grooming behavior is delayed after abdominal surgery and is restored with appropriate analgesia.104
Open in a separate window
The Grooming Transfer Test allows indirect assessment of a mouse's grooming behavior. If grooming well, the mouse will transfer the fluorescent signal from the top of their head to additional body locations, and in time, completely groom away from the signal. The latency to progressive grooming scores is increased in mice with unalleviated postlaparotomy pain. Reproduced with permission from AALAS.51
While general activity evaluations and formal behavioral scoring have significant potential to uncover both known and novel expressions of pain in mice and rats, these techniques are laborious to conduct and require training of observers to ensure intra- and interrater consistency. Observations must be conducted and acted upon in real-time to be useful for effective analgesia administration. For these reasons, open field tests (OFT) have also been explored for assessing analgesia efficacy in mice18 and rats.106 OFT can examine aspects of mobility, anxiety and exploration behaviors in a relatively short period of time, and observation collection and processing can be automated. This test has only been studied for a limited number of painful conditions in rodents, and thus is currently of limited utility for widespread clinical assessment of pain.
Conditioned Place Preference Test
The conditioned place preference test (CPP) is unlikely to be used for spontaneous pain assessment in rodents on a day-to-day basis within vivaria; however, it has been used to demonstrate analgesic efficacy in models of rodent pain. In this assay, 2 outer chambers that are distinguishable by visual or olfactory cues are linked by a neutral middle chamber. Animals are preconditioned with free access to all 3 chambers, and then provision of an analgesic is paired with a particular chamber. In pain models using both mice and rats animals, strong preference can be seen for the chamber in which the analgesia was administered.50,66,137 This model is based on positive reinforcement, a system that requires that the animal have intact contextual memory of the reward.
Conditioned Place Aversion Test
The conditioned place aversion test (CPA) is similar to the CPP test, but uses a neghative stimulus rather than a reward. This test has been used for models of pedal inflammation in rodents.156 The test paradigm consists of a similar 3-chamber set-up with visual cues except for the floor of the chambers consists of mesh. At regular intervals, von Frey fibers of a diameter to invoke a noxious stimulus in an animal with pedal inflammation, are poked through the mesh into the animal's paw in one of the outer chambers. Over time, the animal learns to avoid the chamber in which they receive the stimulation to the inflamed paw.9 This mechanical hypersensitivity and chamber aversion are partially or completely abolished after administration of an efficacious analgesic agent. As with the CPP Test, this assay requires an intact memory, and the test is only useful for very specific models of pain in rodents.
The vast majority of pain assessment assays in laboratory animals continue to rely on evoked responses; therefore, exploring the relationship between these assays and burrowing behavior, a nonevoked response, is valuable in terms of characterizing what different endpoints may be measuring and providing an indication ofthe validity of an endpoint compared with the human experience, and their temporal relationship. Unfortunately, few studies have directly compared these measures.52,69,95,126,128
An important appeal of burrowing as an outcome measure of pain is that it requires simple equipment and is relatively easy to implement. Observations are easily performed in the home cage, reducing potential interference from stress, and methods employing latency to begin burrowing, burrowing duration and total volume displaced have been described.3,29,52,58-60 Of these, the measurement of the total volume displaced is the most common. Burrowing represents a goal-directed behavior that laboratory mice and rats are motivated to perform, and the technique can be applied as a research tool, as well as for clinical assessment.29,30,32,34,129,131 While it can be used to specifically assess pain,3,12,46,52,59,61,69,95,111,124-126,128,130,145,147 reductions in burrowing activity can also identify the influence of a range of factors, not all of which may be associated with pain, such as cognitive dysfunction, anxiety, systemic bacterial and viral infections, and inflammation.24,27,29,31-34,47,55,57,98,144 As such, burrowing may be viewed as a surrogate measure of pain in rodents, and also as a reflection of instrumental activities of daily living (IADL), an outcome used in humans to reflect the impact of disabilities such as pain on day-to-day activities (for example, general mobility, care of others, maintaining the living space).88 Factors other than pain that result in a reduction in burrowing behavior have been better characterized in mice and these include neurodegenerative disease, anxiety, and systemic infection or inflammation, thus interpretation of changes in burrowing behavior needs to be case-specific.24,27,29,32,33,39,47,60,78 In addition, the assessment of burrowing can be confounded by variations in housing, including type of flooring,7 familiar surroundings,60 presence of conspecifics,58 diet,70 and estrous cycle.19 Further, strain differences exist in mice, with reduced burrowing behavior observed in CBA, 129-substrains and Egyptian spiny mice (Acomys cahirinus).29 In rats, scoring of burrowing behavior has been employed successfully in Hooded Lister, Wistar and Sprague–Dawley strains.3,29,52,69
Burrowing is reduced in a wide range of pain models in rats and mice, including laparotomy,4,58-60 colitis and mucositis,57,74,144 neuropathic pain,3,52,69,95,124 inflammation,3,46,95,130,139,144,147 and arthritis.12,125,126 In addition, many of these studies have confirmed responsiveness, showing either an improvement or return to baseline burrowing behavior following the administration of antiinflammatories and analgesics.
Detailed information regarding burrowing tube dimensions, construction, substrate, and test paradigm are readily available.3,28,29,126,147 In rats, gravel is most commonly used, though sand has also been reported (Figure 4).125,147 In mice, earth, sand, bedding, and food pellets have been used.28,29 In both mice and rats, individual variability in burrowing behavior exists, and this is an important fact to consider in its evaluation.3,28,29 Under experimental conditions, burrowing behavior can be encouraged with social facilitation, pairing good and poor burrowers to encourage burrowing behavior.3,28,29,147 In tracking individual changes, animals could serve as their own controls.
Open in a separate window
Burrowing behavior in an adult female Sprague–Dawley placed in a burrowing tube with 2-5 mm gravel.
While burrowing behavior has been useful for experimental studies of pain and analgesia efficacy in mice and rats, there are important considerations in applying it to a more clinical setting. These include individual variability in the propensity to burrow, the current reliance on individual testing, and occasional failure of the assay. In a recent international, multicenter randomized, blinded study (8 different laboratories across 4 countries) of burrowing in rats (adult Sprague–Dawley and Wistar strains) that used a well-characterized model of inflammation (Complete Freund adjuvant injection into a paw), the ability to reproduce the expected inflammation-induced suppression of burrowing varied considerably between centers.147 The underlying reasons are unknown, but this suggests that more investigation of the technique is needed.
Ultrasonic vocalization (USV) has been suggested as a means of going beyond simple evoked reflex measures to reflect an integrative behavioral pain response. Rodents are capable of producing USV, defined as vocalizations at frequencies greater than 20 kHz. These vocalizations are used for communication, but are also triggered by drugs producing either a negative or a positive affective state.10,11,149 These USV are typically categorized into 2 distinct frequency bands centered approximately at 22 kHz (generally associated with negative situations) and 50 kHz, which are associated with positive situations.10,99,142,149 Specifically, noxious or painful stimuli have been associated with the 22 kHz band.14,62,67
Although initial promising results documented USV in response to electrical shocks to the tail in rats62 and an arthritis model in which Freund adjuvant was injected into into tail,14 recent studies have revealed several important concerns and potential limitations with USV as an outcome measure. These include variable correlation with traditional evoked reflex responses in rats and mice,142 a lack of specificity and sensitivity for emission of USVs in mice,146 a failure of USV emissions to identify likely chronic pain states63,142 and a confound of the presence or absence of a conspecific on the expression of USVs.14,63
Employing 3 well-characterized models of inflammation (formalin injection in the hind paw), neuropathy (partial sciatic nerve ligation) and referred pain (bladder inflammation following instillation with turpentine and olive oil), Wallace and colleagues did not find any correlation between USV and withdrawal responses evoked by thermal (heat and cold) or touch (von Frey filament) stimuli.142 These experiments were performed in male Wistar rats (partial sciatic nerve ligation), female Wistar rats (bladder inflammation) and male Wistar rats and C57BL/6 mice (formalin test). The expression of USV was limited to the initial presentation of the testing chamber during habituation, before model induction. After this, no USV were recorded either at baseline or following model induction despite predicted withdrawal responses occurring. These findings indicate failure to achieve adequate construct validity. In a study of weanling mice (male and female, 21 to 28 d old, B6;129S6-Stat5b) undergoing tail snipping or ear notching for DNA testing, Williams and colleagues found the incidence of USV to be highly variable, with 65% of animals not vocalizing in response to either procedure.146 Furthermore, of the mice that did produce USV, audible vocalizations (less than 20 kHz) occurred concurrently in all but one animal. This suggests that during these potentially painful procedures, USV could not be reliably elicited and was not superior to monitoring audible vocalizations alone. Testing an acute (carrageenan injection into the hind paw) and 2 chronic (Freund adjuvant injection into the base of the tail and diabetic neuropathy induced with IP streptozotocin) pain models in male Sprague–Dawley rats, Jourdan and colleagues found a nonsignificant tendency to reduced USV in painful animals when in the presence of another rat, but no USV when animals were tested alone.63 These findings highlight the critical role of experimental design in affecting results and study interpretation.
A possible explanation for some of the observed differences between studies is difficulty in separating responses to pain from distress or anxiety.67,96 In studies that have reported USV, the numbers of animals that emitted vocalizations were low, suggesting, as that it is not a sensitive indicator of nociception or pain.62,146 However, an important consideration when interpreting USV studies is the central role of vocalization in communication and the impact of single compared with group housing.10 The tendency to vocalize may be linked to the presence of conspecifics, a situation that is seldom present during experimental testing but may occur during assessment of pain in colony conditions with social housing.14,63
Physiologic Assessments of Pain in Laboratory Mice and Rats
Body Weight Changes.
Reductions in body weight and growth rate are commonly used as indicators of pain and distress and as humane endpoints in research rodent studies.15,27,53,76,92,93,100,101,136 While loss of body weight may reflect behavioral changes associated with pain, it is a nonspecific indicator that can also reflect compromised wellbeing, malaise and adverse environmental or social conditions.27,53,100,101,118,135 Weight loss due to reduced body mass (rather than dehydration) may also occur with chronic disease states that may or not be painful, such as cancer or infection.101,115,118
In experimental models when changes are slow and progressive, the number of consecutive days of ongoing weight loss or a weight loss maintained for several days may be a more sensitive measure of deterioration than absolute weight change alone.115,148 An upper limit of 15% to 20% is often cited as an endpoint for weight loss;91,148 however, a recent evaluation of 90 rat toxicity studies from 13 pharmaceutical companies and contract research organizations found that the maximum tolerated dose (defined as when dosing had to be stopped or animals were lost through death or euthanasia) was exceed in 12/13 studies in which a 20% weight loss was allowed. As a result, the authors suggested using a weight loss threshold of greater than 10% to trigger a decision regarding study continuation.17
An elegant study that combined several approaches to assess pain and wellbeing (conditioned place preference, automated behavior identification, evoked response [heat], weight loss) in a mouse model (female, C3H/HeN) of bladder cancer showed that weight loss of approximately 5% was closely and significantly associated with increasing preference for the morphine-associated CPP chamber.118 Importantly, the link between weight loss and morphine seeking occurred well before animals approached the typical 15% to 20% weight loss range applied in many studies.
Thus, in consideration of this work, particularly where more specific and sensitive measures of pain are available, weight loss should not be used as the sole measure of pain, but may be used to complement other assessments as a global reflection of deterioration of animal welfare.92 Furthermore, where weight loss is used, consideration should be given to the appropriate change for the model used.
Alterations in the Hypothalamic-Pituitary-Adrenal Gland Axis.
With acute stress, including acute pain, the hypothalamic-pituitary-adrenal gland axis (HPA) is activated, releasing stress hormones that then induce increases in heart rate, respiratory rate blood pressure, and body temperature.4,21,56,110,114 However, the HPA axis and associated physiologic changes are impacted by many other factors, which can make it challenging to use as a single measure for pain assessment. Stress from handling or restraint (when necessary for some forms of algesiometry testing) also significantly stimulates the HPA access and can either conceal or exacerbate changes attributable to pain.118,121 Similarly, other variables that can cause significant and persistent perturbations in heart rate and blood pressure include routine cage changes, social housing, various experimental manipulations, and the provision of in-cage resources, such as running wheels.1,43,65,127,133 Therefore, remote assessment of variables such as heart rate, respiratory rate, body temperature, and blood pressure is preferable, whenever possible. This can be achieved through telemetry, and some physiologic parameters, such as respiratory and heart rates, can even be measured by automated ‘smart caging’.44,77 Assessment of these physiologic variables should never be used alone when attempting to assess pain in mice and rats.
Facial Grimace Scales for Assessment of Pain in Rodents.
The use of grimace scales in mice, subsequently described in numerous species, was introduced in 2010.68 This work and the subsequent description of the Rat Grimace Scale (RGS) demonstrates many of the features expected in validation studies (Figure 5).132
Open in a separate window
The Rat Grimace Scale (A) Rat depicted with ‘pain’ (left) and with ‘no pain’ (right). The ‘pain’ rat has 1) folded ears that are angled away from the front of the face, 2) partial eye closure, 3) a flattened and elongated nose and 4) whiskers that are bunched together and directed away from the face. The ‘no pain’ rat has 1) rounded ears that face forward, 2) no eye closure, 3) a rounded nose and cheeks and 4) whiskers that are fanned and droopy at ends. (B) Image depicts the face of a normal male Wistar rat with no pain. Its eyes are round and open. Its ears are rounded, facing forward and roughly perpendicular to the top of its head. Its nose and cheeks are rounded with an evident bulge and crease between the nose and cheeks. Lastly, the whiskers are spread apart and droop downward at the ends. (Action unit scores—Eyes: 0, Ears: 0, Nose/cheek: 0, Whiskers: 0). (C) Image depicts an adult male Wistar rat grimacing with orbital tightening, nose/cheek flattening with only a slight crease between the nose and cheeks and straightened whiskers that are pulled toward the cheeks. Its ears are curled and slightly rotated outwards. (Action unit scores—eyes: 1, ears: 1, nose/cheek: 1, whiskers: 1, overall score of 1 [from average of 4 action units]). (D) Image depicts an adult male Wistar rat grimacing with an overall score of 2. It has a tightly closed eyelid. Its nose and cheeks are flattened with the nose appearing elongated. The nose and cheek flatten with no crease evident between them. The whiskers are straightened, bunched together and horizontal to the cheeks. Its ears are rotated outwards and curled inwards. (Action unit scores—eyes: 2, ears: 2, nose/cheek: 2, whiskers: 2). (E) Image depicts an adult male Wistar rat grimacing with an overall score of 1.75. Its eyelids are tightly closed. The ears are curled and rotated away from the front of the rat's face. Its nose and cheeks are flattened with no crease evident between them. The whiskers are straight and pulled toward the cheeks. (Action unit scores—eyes: 2, ears: 2, nose/cheek: 2, whiskers: 1). Illustration by Dr Vivian SY Leung.
Both the mouse grimace scale (MGS) and RGS show construct validity and reliability. Construct validity was most comprehensively demonstrated in the MGS using responses to analgesia, different levels of pain (by generating dose-response curves during induction of pain models) and following temporal changes as models progressed. For the RGS, construct validity was shown with temporal changes in RGS scores over time (predicted increase followed by decrease in scores as induced inflammation resolved) and an analgesic dose-response curve. Content (face) validity is grounded in the proposal by Darwin (first published in 1872) that facial expressions revealed emotions in humans and animals, with overlap across the species.25 Furthermore, as grimace scales have been developed in other species, similar facial features have proved to be robust as signalers of pain.
Grimace scale interobserver reliability has been rated as ‘good’ to ‘very good’ by several research groups, indicating that error associated with the scale is small for differences between treatments typically examined in these studies.68,103,118,132 Importantly, reliability is not uniform across features included in facial expression evaluation. For example, reliability associated with scoring whisker shape and position is frequently low.103 It is unclear if this represents inherent difficulty in scoring whiskers or difficulty in collecting images of sufficient quality to be scored.72 Linked to interobserver reliability, though seldom directly addressed, is the question of observer training. The overwhelming majority of MGS and RGS papers are from research groups with the time and personnel to invest in achieving proficiency in use of the MGS and RGS, though substantial variability between observers in this setting has been reported.87 Therefore, while some studies have supported the use of naïve observers, it is largely unknown how these scales might perform in the hands of casual users, such as animal care staff or clinical veterinarians.117 One study investigating the role of training has shown that the combination of practice scoring images alongside structured discussions is more effective than practice scoring alone.155 This may explain why adoption of the MGS and RGS by the laboratory animal veterinary community has been limited.
A comparison between grimace scales and a standard/traditional assessment method has been evaluated with both the MGS and RGS.26,68,98,134 In these cases, mechanical hypersensitivity testing with von Frey filaments has been performed with interesting results. With the MGS, in an inflammation model (zymosan injection into the hind paw or ankle joint), an analgesic effect of acetaminophen could be detected; however, the same dose of acetaminophen (300 mg/kg, SC) did not significantly reduce mechanical hypersensitivity.68 In the case of the RGS, the duration of mechanical hypersensitivity resulting from inflammation (induced with intraplantar injections of CFA or carrageenan) was considerably longer-lasting than that of facial expression changes26 Interestingly, the findings of the RGS study mirror a report of the experiences of a pain researcher who inadvertently self-injected CFA into a finger.45 In this case, the pain experience was relatively short-lived (subsiding at 48 h) compared with the presence of mechanical hypersensitivity (42 d). Similarly, in a model of induced orofacial pain, mechanical hypersensitivity persisted beyond the changes in RGS.2,134 Taken together, these findings raise important questions about the relevance of traditional hypersensitivity testing, particularly in light of the relative unimportance and infrequent occurrence of hypersensitivity in human chronic pain states.90,108 There have been limited comparisons of facial grimace scales to behavior-based systems of pain assessment, with a study in mice undergoing vasectomy surgery showing strong correlation between the MGS and a suite of behaviors altered in the presence of pain.72 In contrast, in a rat model of mucositis induced with IP 5-fluorouracil, an ethogram consisting of writhing, twitching, and back-arching showed increased frequencies of these behaviors in the presence of mucositis, without concurrent changes in the RGS.145 This finding suggests that this that more research is needed for reliable conclusions to be drawn.
One of the major challenges in providing analgesia to laboratory rodents is the identification of efficacious doses of analgesic drugs.37 Grimace scales have been applied to reevaluate commonly used analgesics.81,132 These studies have shown that historic dosing strategies may not provide adequate analgesia in the models and strains studied. More research is needed in this area to gather perform confirmatory studies that show the effectiveness of grimace scales alongside other methods of pain assessment.
The application of MGS and RGS to chronic and neuropathis pain and the role of confounding factors during their application have not been resolved. After the development of the MGS, it was assumed that grimace scales could not accurately detect chronic pain states,68 and this assumption has continued as scales were developed for other species. The basis for this assumption was the absence of changes in MGS scores in 2 well-characterized chronic pain models, spared nerve injury and chronic constriction injury. Subsequently, others have reported that chronic pain may be identified using grimace scales in other models, notably colitis (dextran sodium sulphate-induced), cervical radiculopathy (surgical compression model), neuralgia (chronic constriction injury of the infraorbital nerve), orofacial pain (movement or load-induced), spinal cord injury (cord impact model) and migraine (nitroglycerin-induced) in rats and mice.2,6,48,74,80,112,134,152 These data suggest that limiting application of the MGS and RGS to acute pain may be premature.
Collection of images for grimace scoring and later assessment has limited use where a quick evaluation to guide clinical decision-making is desired. Two potential solutions to shorten the process while maintaining scoring integrity are real-time and automated scoring.73,132,141 The feasibility of real-time scoring, where the observer assigns a score based on direct observation of the animal, has been shown with the RGS.73 Scores based on continuous or discrete observations over periods as short as 2 min were able to identify predicted differences as a result of analgesic treatment. Importantly, the presence of the observer did not affect the RGS. Most recently, Tuttle and colleagues have shown machine learning to be a promising means of automated scoring.141 Following training a neural net with close to 6000 images, automated scoring was able to identify pain and no-pain states with an accuracy (compared with human-generated scores) of 83 to 93%. Critically, confidence associated with scores (as determined by the neural net) was greatest with images representing the greatest changes in MGS, at either extremity of the scale. Further work is required to improve machine-based scoring across the encountered spectrum of images.
A key step in adoption of the MGS and RGS as tools for rapid evaluation of animals and facilitating decisions about care is development of an intervention threshold. One has been derived for the RGS (greater than 0.67/2, sensitivity; 85%, specificity; 89%).103 An intervention threshold identifies the score above which a rat is more likely to be painful and the score should be viewed as a guide for rescue analgesic treatment rather than an absolute rule. Increasing the threshold will increase specificity at the expense of sensitivity and vice versa.103
Limitations in the use of the MGS and RGS are still being defined. In addition to their roles in identifying chronic pain (as described above), exposure to general inhalant anesthetics temporarily inflates RGS scores and should be considered when assessing animals in the early postoperative period.83 Olfactory cues, such as exposure to clothing worn by men, can result in stress-induced analgesia and a consequent reduction in the MGS.131
Use of Nest-Building Behaviors to assess Pain in Laboratory Rodents.
Nesting behavior is a major evolutionary driver for most rodents, particularly mice. Evaluations of mouse health and wellbeing through nest-building have been in continual evolution for at least a decade. The advantage of this approach is that it is easily incorporated into standard husbandry practices and takes advantage of an animal engaging in intrinsically motivated behavior. It may, therefore, be more sensitive to subtle aspects of the pain experience and may better reflect pain as it affects quality of life.104 In its first iterations, evaluations of cage structure were used, assessing if the mice engaged in organizing their cage space to include a dedicated sleeping area separate from their toilet area.4 Next the focus shifted to the complexity of the nest with scores ranging from zero, for an unmanipulated compressed cotton square, to a score of 5, if a bowl with well-defined walls was built.60 The type and amount of nesting material which would optimally allow for the construction of a complete nest followed, facilitating a progression of nest scoring (Figure 6).51 A complete nest has a full dome that encloses the center of the nest and allows the mouse to create a warm, dark, microenvironment that can be up to 10 °C degrees warmer than the rest of the cage environment.42 Mice that underwent surgery had significantly lower nest scores than those that underwent sham procedures or had surgery with adequate pain management.60
Open in a separate window
When given the appropriate amount and material, mice can build a full domed nest and a nest complexity score can be assigned. Reproduced with permission from AALAS.51
Whereas all of these nesting and cage organization assessments were valuable steps in the evolution of cageside pain assessments in mice and added a quantitative component to otherwise qualitative measures, they were also relatively subjective. Therefore interobserver reliability could be problematic. In addition, while the mice may have had effective analgesia in the immediate postoperative period, subsequent changes in their pain state may go unrecognized because the nest complexity or cage structure score does not change once the nest is formed. Therefore, the next iterations of nest assessments corrected for these limitations by creating on-demand assessments of nest consolidation. In the time-to-integrate-to-nest or TINT test, mice are given a small piece of nesting material in the opposite end of the cage from their existing nest which they must retrieve, and then return the piece to their nest and integrate it. This was found to occur in 9 of 10 mouse strains within 10 min after the provision of the new nesting material. The likelihood of a negative TINT (taking greater than 10 min to retrieve) was significantly increased after having undergone a painful procedure.116 The benefit of this approach is that the mice can be recovered in their home cage with an existing nest, supporting positive welfare in the postprocedural time period. In addition, TINT can be assessed on demand and in repetition, so the arc of recovery or absence of effective analgesia can be assessed over time. However, this method is assessed in a binary fashion, making it difficult to create a gradient with which to assess relative analgesic efficacy. So, while it is a good initial tool for identifying which cages require additional veterinary attention, it may not reveal smaller changes in pain severity or alleviation. The zone clearance test is a similar assessment of how quickly a mouse will retrieve pieces of nesting material around the cage.97 Using this test, a mouse can retrieve 6 pieces of cotton nesting within 100 min at baseline, but likely fails this test after a painful procedure. However, normal retrieval and consolidation behavior is observed once analgesia is provided. The benefit of this assessment is that a greater range of scores are possible, allowing for a greater ability to tease apart differences in efficacy between analgesic regimens, as well as differences in the response to treatment. The challenge is that this method requires mice to recover from a procedure in a cage without an existing nest, and importantly, requires that the mouse chooses to build their nest in a corner such that they are clearing zones as they gather the nesting material. However, dependent on a number of caging and animal factors, mice may choose to build a nest in a location other than the corner, significantly reducing the number of zones cleared, even if the mouse is clearly exhibiting nest consolidation behavior. As published, this test would be used for a one-time assessment as it requires starting without an existing nest; this limits its use at multiple time points without significantly altering the home-cage environment.
The most recent iteration of nest consolidation tests combines the strengths of the prior 2 assessments, TINT and zone clearance, by allowing mice to recover in a home cage with an existing nest and facilitates on-demand assessment that allows for minimal disturbance and longitudinal assessment while also providing more sensitivity in scoring to allow for gradations of scores.104 In addition, this new assessment allows the nest to be built anywhere in the home cage, rather than only in corners. The Nest Consolidation Test requires mice be provided 4 pieces of cotton nesting material to be placed either in the 4 corners of the cage if no nest is already present in the home cage, or on the opposite end of the cage from an existing nest (Figure 7). The mice are then timed for how long it takes to retrieve the pieces of nesting material and consolidate by moving the pieces at least half of the cage length and/or width to be joined with another piece of nesting material or within one inch of the existing nest. This test was found to identify postprocedural pain in males, females, inbred and outbred mice, and was minimally impacted by anesthetics and analgesics. It not only identified unalleviated postoperative pain but could differentiate between different analgesic regimens allowing for better drug discrimination than traditional measures of mechanical threshold and clinical measures such as body weight loss. One important consideration for all nesting assessments for pain in mice is that to assess individual animals, they must be singly housed.104 Single housing of animals that could otherwise be socially housed is not desirable as it can contribute to unnecessary stress in a recovering animal.
Open in a separate window
The Nest Consolidation Test allows mice to retrieve one of 4 pieces of nesting material, either with or without an existing nest. The pieces must be consolidated to within a specific distance of one another or within the existing nest. The nest can be built anywhere in the home cage. Reproduced with permission from AALAS.104
Recommendations for Ongoing Monitoring of Pain in Laboratory Rodents.
If it is unclear whether pain or a confounder is underlying the pain assessment score, a tried and true approach in both human and veterinary medicine is to assess the response to analgesia (assuming that an effective dose of analgesic is administered).40 Assessing animals before and after analgesic administration or comparing animals that received different agents or routes can often reveal whether the analgesic regimen is effective (Figure 8).22 With this approach, it is important to understand the direction in which the chosen assessment parameter will change if pain is alleviated. For example, for paw withdrawal in response to mechanical pressure, one would expect latency to increase if the animal becomes more comfortable. In contrast, with nest consolidation scoring, alleviation of pain will be expected to shorten the latency to nest consolidation as the animal becomes more comfortable and regains normal behavior.
Open in a separate window
The postsurgical rat on the left (A) received a single analgesic, carprofen, whereas the rat on the right (B) that underwent the same procedure received multimodal analgesia: carprofen and tramadol. When comparing the 2 animals it is clear that the animal that received multimodal analgesia is more alert, paying attention to the observer, and in addition, its fur is lying flat, and its eyes are wide open, all indicating this animal is more comfortable than the rat that only received only carprofen. Reproduced with permission.22
Assessments should be made starting from before pain is anticipated to begin, allowing the observer to identify when the pain starts. This facilitates early treatment and minimizes the period of untreated pain. Once pain is present, the observer should estimate the length of time that the pain is anticipated to continue – whether it is acute with a rapid recovery or the start of a long-term disease process, like arthritis or tumor induction and progression.
Once pain is present, the frequency of assessments should be tailored to match the expected duration of analgesic therapy. In human medicine, pain is expected to lessen within 30 min of analgesic drug administration.38 In veterinary patients, this may not be likely in all situations because of analgesic pharmacokinetics,40 but is a useful rule of thumb to consider when treating animals. Thus, assessments should be made before and after the provision of an analgesic to ensure that the agent is achieving the desired effect, and to provide additional analgesic if needed. Assessments should be repeated, based on the known pharmacokinetics of the drug to determine if pain has returned, and whether additional doses or different therapies are required. In human medicine, to be considered clinically useful, a minimal 33% change in an outcome measure is sought after treating patients with additional rescue medication for acutely painful conditions.38 This clinical cut-off point was developed recognizing that there are no objective measures of painful experiences in human patients and individuals show wide variability in response to interventions. While a worthy goal, the utility of a clinical cut-off point is untested in veterinary medicine.
Analgesic therapy for laboratory rodents must be performed with a clear goal in mind. Dynamic pain occurs only when the animal is engaged in a particular behavior or when it adopts a particular body posture. Dynamic pain is often less severe and may affect more of the nonevoked measures of pain, such as nest building and grooming. Less potent analgesics,40 lower doses or shorter regimens of analgesic may be sufficient to allow an animal to be engage in these higher-level, spontaneous behaviors. Alternatively, static pain occurs when the animal is at rest. Static pain is likely to prevent even basic maintenance behaviors, such as eating and drinking. While it may not be possible to alleviate all pain, goals for analgesic therapy should be to prevent static pain at a minimum, while helping the animal to return normal spontaneous behavior. This allows them to create and maintain a microenvironment that further supports recovery, such as normal nesting, burrowing and social behaviors with cagemates. Static pain conditions may require more potent analgesics, higher doses, and a longer course of therapy to maintain the animal in a comfortable state.
The search for novel measures to assess pain in laboratory rodents that do not rely on traditional evoked-response reflex testing is important for the evolution of translational pain research and for enhancing laboratory animal welfare. Many advances have occurred in rodent pain assessment techniques, but additional work is needed to understand the range of circumstances for which each test is useful. Future work should focus on the development of additional nonevoked cageside measures of pain that do not require handling or even the presence of an observer to aid in accurate identification of rodents in need of veterinary care. For mice, facial grimacing, nest building, and grooming have served this need under some experimental conditions. However, these evaluations should be more broadly implemented in formal clinical pain assessments, through institutional training programs for animal ethics committees, research groups, and technical personnel. Unfortunately, these approaches have not been widely tested for other laboratory rodents, although alternate assays, such as analysis of burrowing behavior, may be appropriate. Identifying behaviors that these species will readily engage in, that are significantly altered by a painful stimulus, and that can be restored by analgesia must be developed. Advancements in technology, such as home cage ethogram analysis, automated facial grimace analysis, and smart cage read-outs of animal physiology and activity may also assist with discovery of new or more efficacious analgesic treatment regimens for different rodent conditions. However, even if better assessment tools can be developed, a major challenge remains: how to provide individualized animal assessments and pain mitigation when large numbers of rodents are on study at any given time. This ethical issue merits further consideration by the community at large.
We thank Lon Kendall for assistance with Figure 1.
1. Adamson TW, Kendall LV, Goss S, Grayson K, Touma C, Palme R, Chen JQ, Borowsky AD. 2010. Assessment of carprofen and buprenorphine on recovery of mice after surgical removal of the mammary fat pad. J Am Assoc Lab Anim Sci 49:610–616. [PMC free article] [PubMed] [Google Scholar]
2. Akintola T, Raver C, Studlack P, Uddin O, Masri R, Keller A. 2017. The grimace scale reliably assesses chronic pain in a rodent model of trigeminal neuropathic pain. Neurobiol Pain. 2:13–17. 10.1016/j.ynpai.2017.10.001. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
3. Andrews N, Legg E, Lisak D, Issop Y, Richardson D, Harper S, Pheby T, Huang W, Burgess G, Machin I, Rice AS. 2012. Spontaneous burrowing behaviour in the rat is reduced by peripheral nerve injury or inflammation associated pain. Eur J Pain 16:485–495. 10.1016/j.ejpain.2011.07.012. [PubMed] [CrossRef] [Google Scholar]
4. Arras M, Rettich A, Cinelli P, Kasermann HP, Burki K. 2007. Assessment of post-laparotomy pain in laboratory mice by telemetric recording of heart rate and heart rate variability. BMC Vet Res 3:1–10. 10.1186/1746-6148-3-16. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
5. Asai H, Ozaki N, Shinoda M, Nagamine K, Tohnai I, Ueda M, Sugiura Y. 2005. Heat and mechanical hyperalgesia in mice model of cancer pain. Pain 117:19–29. 10.1016/j.pain.2005.05.010. [PubMed] [CrossRef] [Google Scholar]
6. Asgar J, Zhang Y, Saloman JL, Wang S, Chung MK, Ro JY. 2015. The role of TRPA1 in muscle pain and mechanical hypersensitivity under inflammatory conditions in rats. Neuroscience 310:206–215. 10.1016/j.neuroscience.2015.09.042. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
7. Bangsgaard Bendtsen KM, Krych L, Sørensen DB, Pang W, Nielsen DS, Josefsen K, Hansen LH, Sørensen SJ, Hansen AK. 2012. Gut microbiota composition is correlated to grid floor induced stress and behavior in the BALB/c mouse. PLoS One 7:1–13. 10.1371/journal.pone.0046231. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
8. Barrot M. 2012. Tests and models of nociception and pain in rodents. Neuroscience 211:39–50. 10.1016/j.neuroscience.2011.12.041. [PubMed] [CrossRef] [Google Scholar]
9. Boyce-Rustay JM, Zhong C, Kohnken R, Baker SJ, Simler GH, Wensink EJ, Decker MW, Honore P. 2010. Comparison of mechanical allodynia and the affective component of inflammatory pain in rats. Neuropharmacology 58:537–543. 10.1016/j.neuropharm.2009.08.008. [PubMed] [CrossRef] [Google Scholar]
10. Brudzynski SM. 2005. Principles of rat communication: quantitative parameters of ultrasonic calls in rats. Behav Genet 35:85–92. 10.1007/s10519-004-0858-3. [PubMed] [CrossRef] [Google Scholar]
11. Brudzynski SM. 2015. Pharmacology of ultrasonic vocalizations in adult rats: significance, call classification and neural substrate. Curr Neuropharmacol 13:180–192. 10.2174/1570159X13999150210141444. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
12. Bryden LA, Nicholson JR, Doods H, Pekcec A. 2015. Deficits in spontaneous burrowing behavior in the rat bilateral monosodium iodoacetate model of osteoarthritis: an objective measure of pain-related behavior and analgesic efficacy. Osteoarthritis Cartilage 23:1605–1612. 10.1016/j.joca.2015.05.001. [PubMed] [CrossRef] [Google Scholar]
13. Butera KA, Fox EJ, George SZ. 2016. Toward a transformed understanding: From pain and movement to pain with movement. Phys Ther 96:1503–1507. 10.2522/ptj.20160211. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
14. Calvino B, Besson JM, Boehrer A, Depaulis A. 1996. Ultrasonic vocalization (22-28 Hz) in a model of chronic pain, the arthritic rat: effects of analgesic drugs. Neuroreport 7:581–584. [PubMed] [Google Scholar]
15. Carstens E, Moberg GP. 2000. Recognizing pain and distress in laboratory animals. ILAR J 41:62–71. 10.1093/ilar.41.2.62. [PubMed] [CrossRef] [Google Scholar]
16. Carter SJ, Durrington HJ, Gibbs JE, Blaikley J, Loudon AS, Ray DW, Sabroe I. 2016. A matter of time: study of circadian clocks and their role in inflammation. J Leukoc Biol 99:549–560. 10.1189/jlb.3RU1015-451R. [PubMed] [CrossRef] [Google Scholar]
17. Chapman K, Sewell F, Allais L, Delongeas JL, Donald E, Festag M, Kervyn S, Ockert D, Nogues V, Palmer H, Popovic M, Roosen W, Schoenmakers A, Somers K, Stark C, Stei P, Robinson S. 2013. A global pharmaceutical company initiative: an evidence-based approach to define the upper limit of body weight loss in short term toxicity studies. Regul Toxicol Pharmacol 67:27–38. 10.1016/j.yrtph.2013.04.003. [PubMed] [CrossRef] [Google Scholar]
18. Cho H, Jang Y, Lee B, Chun H, Jung J, Kim SM, Hwang SW, Oh U. 2013. Voluntary movements as a possible nonreflexive pain assay. Mol Pain 9:1–9. 10.1186/1744-8069-9-25. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
19. Christensen SL, Petersen S, Sørensen DB, Olesen J, Jansen-Olesen I. 2016. Infusion of low dose glyceryl trinitrate has no consistent effect on burrowing behavior, running wheel activity and light sensitivity in female rats. J Pharmacol Toxicol Methods 80:43–50. 10.1016/j.vascn.2016.04.004. [PubMed] [CrossRef] [Google Scholar]
20. Christina A, Merlin N, Vijaya C, Jayaprakash S, Murugesh N. 2004. Daily rhythm of nociception in rats. Circadian Rhythms 2:1–3. 10.1186/1740-3391-2-2. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
21. Cinelli P, Rettich A, Seifert B, Bürki K, Arras M. 2007. Comparative analysis and physiological impact of different tissue biopsy methodologies used for the genotyping of laboratory mice. Lab Anim 41:174–184. 10.1258/002367707780378113. [PubMed] [CrossRef] [Google Scholar]
22. Ciuffreda MC, Tolva V, Casana R, Gnecchi M, Vanoli E, Spazzolini C, Roughan J, Calvillo L. 2014. Rat experimental model of myocardial ischemia/reperfusion injury: an ethical approach to set up the analgesic management of acute post-surgical pain. PLoS One 9:1–8. 10.1371/journal.pone.0095913 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
23. Cowan A, Doxey JC, Harry EJ. 1977. The animal pharmacology of buprenorphine, an oripavine analgesic agent. Br J Pharmacol 60:547–554. 10.1111/j.1476-5381.1977.tb07533.x. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
24. Cunningham C, Campion S, Teeling J, Felton L, Perry VH. 2007. The sickness behaviour and CNS inflammatory mediator profile induced by systemic challenge of mice with synthetic double-stranded RNA (poly I:C). Brain Behav Immun 21:490–502. 10.1016/j.bbi.2006.12.007. [PubMed] [CrossRef] [Google Scholar]
25. Darwin C. 2009. The expression of emotions in man and animals. London (United Kingdom): Penguin Classics. 10.1017/CBO9780511694110 [CrossRef] [Google Scholar]
26. De Rantere D, Schuster CJ, Reimer JN, Pang DS. 2015. The relationship between the Rat Grimace Scale and mechanical hypersensitivity testing in 3 experimental pain models. Eur J Pain 20:417–426. 10.1002/ejp.742. [PubMed] [CrossRef] [Google Scholar]
27. de Sousa AA, Reis R, Bento-Torres J, Trévia N, Lins NA, Passos A, Santos Z, Diniz JA, Vasconcelos PF, Cunningham C, Perry VH, Diniz CW. 2011. Influence of enriched environment on viral encephalitis outcomes: behavioral and neuropathological changes in albino Swiss mice. PLoS One 6:1–12. 10.1371/journal.pone.0015597. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
28. Deacon R. 2012. Assessing burrowing, nest construction, and hoarding in mice. J Vis Exp 59:1–12. [PMC free article] [PubMed] [Google Scholar]
29. Deacon RM. 2006. Burrowing in rodents: a sensitive method for detecting behavioral dysfunction. Nat Protoc 1:118–121. 10.1038/nprot.2006.19. [PubMed] [CrossRef] [Google Scholar]
30. Deacon RM. 2009. Burrowing: a sensitive behavioural assay, tested in 5 species of laboratory rodents. Behav Brain Res 200:128–133. 10.1016/j.bbr.2009.01.007. [PubMed] [CrossRef] [Google Scholar]
31. Deacon RM, Cholerton LL, Talbot K, Nair-Roberts RG, Sanderson DJ, Romberg C, Koros E, Bornemann KD, Rawlins JN. 2008. Age-dependent and -independent behavioral deficits in Tg2576 mice. Behav Brain Res 189:126–138. 10.1016/j.bbr.2007.12.024. [PubMed] [CrossRef] [Google Scholar]
32. Deacon RM, Croucher A, Rawlins JN. 2002. Hippocampal cytotoxic lesion effects on species-typical behaviours in mice. Behav Brain Res 132:203–213. 10.1016/S0166-4328(01)00401-6. [PubMed] [CrossRef] [Google Scholar]
33. Deacon RM, Raley JM, Perry VH, Rawlins JN. 2001. Burrowing into prion disease. Neuroreport 12:2053–2057. 10.1097/00001756-200107030-00052. [PubMed] [CrossRef] [Google Scholar]
34. Deacon RM, Rawlins JN. 2005. Hippocampal lesions, species-typical behaviours and anxiety in mice. Behav Brain Res 156:241–249. 10.1016/j.bbr.2004.05.027. [PubMed] [CrossRef] [Google Scholar]
35. Dunbar ML, David EM, Aline MR, Lofgren JL. 2016. Validation of a behavioral ethogram for assessing postoperative pain in guinea pigs (Cavia porcellus). J Am Assoc Lab Anim Sci 55:29–34. [PMC free article] [PubMed] [Google Scholar]
36. Evans WO. 1964. A critical review of some new methods in animal analgesiometry. J New Drugs 4:179–187. 10.1002/j.1552-4604.1964.tb00199.x. [PubMed] [CrossRef] [Google Scholar]
37. Faller KM, McAndrew DJ, Schneider JE, Lygate CA. 2015. Refinement of analgesia following thoracotomy and experimental myocardial infarction using the Mouse Grimace Scale. Exp Physiol 100:164–172. 10.1113/expphysiol.2014.083139. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
38. Farrar JT, Berlin JA, Strom BL. 2003. Clinically important changes in acute pain outcome measures: A validation study. J Pain Symptom Manage 25:406–411. 10.1016/S0885-3924(03)00162-3. [PubMed] [CrossRef] [Google Scholar]
39. Filali M, Lalonde R, Rivest S. 2011. Subchronic memantine administration on spatial learning, exploratory activity, and nest-building in an APP/PS1 mouse model of Alzheimer's disease. Neuropharmacology 60:930–936. 10.1016/j.neuropharm.2011.01.035. [PubMed] [CrossRef] [Google Scholar]
40. Foley PL, Kendall LV, Turner PV. 2019. Clinical management of pain in rodents. J Am Assoc Lab Anim Sci. EPub ahead of print. [Google Scholar]
41. Garner Laboratory, Stanford School of Medicine. 2019. [Internet]. Mouse ethograms. [Cited 25 March 2019]. Available at: http://mousebehavior.org/ethogram/
42. Gaskill BN, Gordon CJ, Pajor EA, Lucas JR, Davis JK, Garner JP. 2013. Impact of nesting material on mouse body temperature and physiology. Physiol Behav 110-111:87–95. 10.1016/j.physbeh.2012.12.018. [PubMed] [CrossRef] [Google Scholar]
43. Gilmore AJ, Billing RL, Einstein R. 2008. The effects on heart rate and temperature of mice and vas deferens responses to noradrenaline when their cage mates are subjected to daily restraint stress. Lab Anim 42:140–148. 10.1258/la.2007.06030e. [PubMed] [CrossRef] [Google Scholar]
44. González-Sánchez C, Fraile JC, Pérez-Turiel J. 2016. Capacitive sensing for noninvasive breathing and heart monitoring in nonrestrained, nonsedated laboratory mice. Sensors (Basel) 16:1–16. 10.3390/s16071052. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
45. Gould HJ., 3rd 2000. Complete Freund's adjuvant-induced hyperalgesia: a human perception. Pain 85:301–303. 10.1016/S0304-3959(99)00289-4. [PubMed] [CrossRef] [Google Scholar]
46. Gould SA, Doods H, Lamla T, Pekcec A. 2016. Pharmacological characterization of intraplantar Complete Freund's Adjuvant-induced burrowing deficits. Behav Brain Res 301:142–151. 10.1016/j.bbr.2015.12.019. [PubMed] [CrossRef] [Google Scholar]
47. Guenther K, Deacon RM, Perry VH, Rawlins JN. 2001. Early behavioural changes in scrapie-affected mice and the influence of dapsone. Eur J Neurosci 14:401–409. 10.1046/j.0953-816x.2001.01645.x. [PubMed] [CrossRef] [Google Scholar]
48. Harris HM, Carpenter JM, Black JR, Smitherman TA, Sufka KJ. 2017. The effects of repeated nitroglycerin administrations in rats; modeling migraine-related endpoints and chronification. J Neurosci Methods 284:63–70. 10.1016/j.jneumeth.2017.04.010. [PubMed] [CrossRef] [Google Scholar]
49. Hayes KE, Raucci JA, Jr, Gades NM, Toth LA. 2000. An evaluation of analgesic regimens for abdominal surgery in mice. Contemp Top Lab Anim Sci 39:18–23. [PubMed] [Google Scholar]
50. He Y, Tian X, Hu X, Porreca F, Wang ZJ. 2012. Negative reinforcement reveals non-evoked ongoing pain in mice with tissue or nerve injury. J Pain 13:598–607. 10.1016/j.jpain.2012.03.011. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
51. Hess SE, Rohr S, Dufour BD, Gaskill BN, Pajor EA, Garner JP. 2008. Home improvement: C57BL/6J mice given more naturalistic nesting materials build better nests. J Am Assoc Lab Anim Sci 47:25–31. [PMC free article] [PubMed] [Google Scholar]
52. Huang W, Calvo M, Karu K, Olausen HR, Bathgate G, Okuse K, Bennett DL, Rice AS. 2013. A clinically relevant rodent model of the HIV antiretroviral drug stavudine induced painful peripheral neuropathy. Pain 154:560–575. 10.1016/j.pain.2012.12.023. [PubMed] [CrossRef] [Google Scholar]
53. Institute for Laboratory Animal Research. 2011. Guide for the care and use of laboratory animals, 8th ed Washington (DC): National Academies Press. [Google Scholar]
54. International Association for the Study of Pain. [Internet]. 1994. [Cited 5 July 2019]. IASP terminology. Available at: https://www.iasp-pain.org/terminology?navItemNumber=576.
55. Jirkof P. 2014. Burrowing and nest building behavior as indicators of well-being in mice. J Neurosci Methods 234:139–146. 10.1016/j.jneumeth.2014.02.001. [PubMed] [CrossRef] [Google Scholar]
56. Jirkof P. 2017. Side effects of pain and analgesia in animal experimentation. Lab Anim (NY) 46:123–128. 10.1038/laban.1216. [PubMed] [CrossRef] [Google Scholar]
57. Jirkof P, Cesarovic N, Rettich A, Arras M. 2013. Housing of female mice in a new environment and its influence on postsurgical behaviour and recovery. Appl Anim Behav Sci 148:209–217. 10.1016/j.applanim.2013.08.006. [CrossRef] [Google Scholar]
58. Jirkof P, Cesarovic N, Rettich A, Fleischmann T, Arras M. 2012. Individual housing of female mice: influence on postsurgical behaviour and recovery. Lab Anim 46:325–334. 10.1258/la.2012.012027. [PubMed] [CrossRef] [Google Scholar]
59. Jirkof P, Cesarovic N, Rettich A, Nicholls F, Seifert B, Arras M. 2010. Burrowing behavior as an indicator of postlaparotomy pain in mice. Front Behav Neurosci 4:1–9. 10.3389/fnbeh.2010.00165. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
60. Jirkof P, Fleischmann T, Cesarovic N, Rettich A, Vogel J, Arras M. 2013. Assessment of postsurgical distress and pain in laboratory mice by nest complexity scoring. Lab Anim 47:153–161. 10.1177/0023677213475603. [PubMed] [CrossRef] [Google Scholar]
61. Jirkof P, Leucht K, Cesarovic N, Caj M, Nicholls F, Rogler G, Arras M, Hausmann M. 2013. Burrowing is a sensitive behavioural assay for monitoring general wellbeing during dextran sulfate sodium colitis in laboratory mice. Lab Anim 47:274–283. 10.1177/0023677213493409. [PubMed] [CrossRef] [Google Scholar]
62. Jourdan D, Ardid D, Chapuy E, Eschalier A, Le Bars D. 1995. Audible and ultrasonic vocalization elicited by single electrical nociceptive stimuli to the tail in the rat. Pain 63:237–249. 10.1016/0304-3959(95)00049-X. [PubMed] [CrossRef] [Google Scholar]
63. Jourdan D, Ardid D, Eschalier A. 2002. Analysis of ultrasonic vocalisation does not allow chronic pain to be evaluated in rats. Pain 95:165–173. 10.1016/S0304-3959(01)00394-3. [PubMed] [CrossRef] [Google Scholar]
64. Kaler J, Green LE. 2008. Recognition of lameness and decisions to catch for inspection among sheep farmers and specialists in GB. BMC Vet Res 4:41 10.1186/1746-6148-4-41. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
65. Kendall LV, Wegenast DJ, Smith BJ, Dorsey KM, Kang S, Lee NY, Hess AM. 2016. Efficacy of sustained-release buprenorphine in an experimental laparotomy model in female mice. J Am Assoc Lab Anim Sci 55:8. [PMC free article] [PubMed] [Google Scholar]
66. King T, Vera-Portocarrero L, Gutierrez T, Vanderah TW, Dussor G, Lai J, Fields HL, Porreca F. 2009. Unmasking the tonic-aversive state in neuropathic pain. Nat Neurosci 12:1364–1366. 10.1038/nn.2407. Erratum: Nat Neurosci 2010. 13: 1033. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
67. Knapp DJ, Pohorecky LA. 1995. An air-puff stimulus method for elicitation of ultrasonic vocalizations in rats. J Neurosci Methods 62:1–5. 10.1016/0165-0270(95)00044-5. [PubMed] [CrossRef] [Google Scholar]
68. Langford DJ, Bailey AL, Chanda ML, Clarke SE, Drummond TE, Echols S, Glick S, Ingrao J, Klassen-Ross T, Lacroix-Fralish ML, Matsumiya L, Sorge RE, Sotocinal SG, Tabaka JM, Wong D, van den Maagdenberg AM, Ferrari MD, Craig KD, Mogil JS. 2010. Coding of facial expressions of pain in the laboratory mouse. Nat Methods 7:447–449. 10.1038/nmeth.1455. [PubMed] [CrossRef] [Google Scholar]
69. Lau W, Dykstra C, Thevarkunnel S, Silenieks LB, de Lannoy IA, Lee DK, Higgins GA. 2013. A back translation of pregabalin and carbamazepine against evoked and non-evoked endpoints in the rat spared nerve injury model of neuropathic pain. Neuropharmacology 73:204–215. 10.1016/j.neuropharm.2013.05.023. [PubMed] [CrossRef] [Google Scholar]
70. Lavin DN, Joesting JJ, Chiu GS, Moon ML, Meng J, Dilger RN, Freund GG. 2011. Fasting induces an antiinflammatory effect on the neuroimmune system which a high-fat diet prevents. Obesity (Silver Spring) 19:1586–1594. 10.1038/oby.2011.73. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
71. Le Bars D, Gozariu M, Cadden SW. 2001. Animal models of nociception. Pharmacol Rev 53:597–652. [PubMed] [Google Scholar]
72. Leach MC, Klaus K, Miller AL, Scotto di Perrotolo M, Sotocinal SG, Flecknell PA. 2012. The assessment of postvasectomy pain in mice using behaviour and the Mouse Grimace Scale. PLoS One 7:1–9. 10.1371/journal.pone.0035656. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
73. Leung V, Zhang E, Pang DS. 2016. Real-time application of the Rat Grimace Scale as a welfare refinement in laboratory rats. Sci Rep 6:1–12. 10.1038/srep31667. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
74. Leung VS, Benoit-Biancamano MO, Pang DS. 2019. Performance of behavioral assays: the Rat Grimace Scale, burrowing activity and a composite behavior score to identify visceral pain in an acute and chornic colitis model. Pain Rep 4:e718 10.1097/PR9.0000000000000712. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
75. Liang MH. 2000. Longitudinal construct validity: establishment of clinical meaning in patient evaluative instruments. Med Care 38 (9 Suppl):II84–II90. 10.1097/00005650-200009002-00013. [PubMed] [CrossRef] [Google Scholar]
76. Lidster K, Jefferys JG, Blümcke I, Crunelli V, Flecknell P, Frenguelli BG, Gray WP, Kaminski R, Pitkänen A, Ragan I, Shah M, Simonato M, Trevelyan A, Volk H, Walker M, Yates N, Prescott MJ. 2016. Opportunities for improving animal welfare in rodent models of epilepsy and seizures. J Neurosci Methods 260:2–25. 10.1016/j.jneumeth.2015.09.007. [PubMed] [CrossRef] [Google Scholar]
77. Lim MA, Defensor EB, Mechanic JA, Shah PP, Jaime EA, Roberts CR, Hutto DL, Schaevitz LR. 2019. Retrospective analysis of the effects of identification procedures and cage changing by using data from automated, continuous monitoring. J Am Assoc Lab Anim Sci 58:126–141. 10.30802/AALAS-JAALAS-18-000056. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
78. Line SJ, Barkus C, Coyle C, Jennings KA, Deacon RM, Lesch KP, Sharp T, Bannerman DM. 2011. Opposing alterations in anxiety and species-typical behaviours in serotonin transporter overexpressor and knockout mice. Eur Neuropsychopharmacol 21:108–116. 10.1016/j.euroneuro.2010.08.005. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
79. Lofgren J, Miller AL, Lee CCS, Bradshaw C, Flecknell P, Roughan J. 2017. Analgesics promote welfare and sustain tumour growth in orthotopic 4T1 and B16 mouse cancer models. Lab Anim 52:351–364. 10.1177/0023677217739934. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
80. Long H, Liao L, Gao M, Ma W, Zhou Y, Jian F, Wang Y, Lai W. 2015. Periodontal CGRP contributes to orofacial pain following experimental tooth movement in rats. Neuropeptides 52:31–37. 10.1016/j.npep.2015.06.006. [PubMed] [CrossRef] [Google Scholar]
81. Matsumiya LC, Sorge RE, Sotocinal SG, Tabaka JM, Wieskopf JS, Zaloum A, King OD, Mogil JS. 2012. Using the Mouse Grimace Scale to reevaluate the efficacy of postoperative analgesics in laboratory mice. J Am Assoc Lab Anim Sci 51:42–49. [PMC free article] [PubMed] [Google Scholar]
82. Meldrum ML. 2003. A capsule history of pain management. JAMA 290:2470–2475. 10.1001/jama.290.18.2470. [PubMed] [CrossRef] [Google Scholar]
83. Miller AL, Golledge HD, Leach MC. 2016. The influence of isoflurane anaesthesia on the Rat Grimace Scale. PLoS One 11:1–8. 10.1371/journal.pone.0166652. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
84. Miller A, Kitson G, Skalkoyannis B, Leach M. 2015. The effect of isoflurane anaesthesia and buprenorphine on the mouse grimace scale and behaviour in CBA and DBA/2 mice. Appl Anim Behav Sci 172:58–62. 10.1016/j.applanim.2015.08.038. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
85. Miller AL, Leach MC. 2015. The Mouse Grimace Scale: a clinically useful tool? PLoS One 10:1–10. 10.1371/journal.pone.0136000. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
86. Minett MS, Eijkelkamp N, Wood JN. 2014. Significant determinants of mouse pain behaviour. PLoS One 9:1–7. 10.1371/journal.pone.0104458. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
87. Mittal A, Gupta M, Lamarre Y, Jahagirdar B, Gupta K. 2016. Quantification of pain in sickle mice using facial expressions and body measurements. Blood Cells Mol Dis 57:58–66. 10.1016/j.bcmd.2015.12.006. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
88. Mlinac ME, Feng MC. 2016. Assessment of activities of daily living, self-care, and independence. Arch Clin Neuropsychol 31:506–516. 10.1093/arclin/acw049. [PubMed] [CrossRef] [Google Scholar]
89. Moayedi M, Davis KD. 2013. Theories of pain: from specificity to gate control. J Neurophysiol 109:5–12. 10.1152/jn.00457.2012. [PubMed] [CrossRef] [Google Scholar]
90. Mogil JS, Crager SE. 2004. What should we be measuring in behavioral studies of chronic pain in animals. Pain 112:12–15. 10.1016/j.pain.2004.09.028. [PubMed] [CrossRef] [Google Scholar]
91. Morton CM, Reid J, Scott EM, Holton LL, Nolan AM. 2005. Application of a scaling model to establish and validate an interval level pain scale for assessment of acute pain in dogs. Am J Vet Res 66:2154–2166. 10.2460/ajvr.2005.66.2154. [PubMed] [CrossRef] [Google Scholar]
92. Morton DB. 2000. A systematic approach for establishing humane endpoints. ILAR J 41:80–86. 10.1093/ilar.41.2.80. [PubMed] [CrossRef] [Google Scholar]
93. Morton DB, Griffiths PH. 1985. Guidelines on the recognition of pain, distress and discomfort in experimental animals and an hypothesis for assessment. Vet Rec 116:431–436. 10.1136/vr.116.16.431. [PubMed] [CrossRef] [Google Scholar]
94. Mulder GB, Pritchett K. 2004. Rodent analgesiometry: The hotplate, tasil flick and von Frey hairs. Contemp Top Lab Anim Sci 43:54–55. [PubMed] [Google Scholar]
95. Muralidharan A, Kuo A, Jacob M, Lourdesamy JS, Carvalho LM, Nicholson JR, Corradini L, Smith MT. 2016. Comparison of burrowing and stimuli-evoked pain behaviors as end-points in rat models of inflammatory pain and peripheral neuropathic pain. Front Behav Neurosci 10:1–9. 10.3389/fnbeh.2016.00088. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
96. Naito H, Nakamura A, Inoue M, Suzuki Y. 2003. Effect of anxiolytic drugs on air-puff-elicited ultrasonic vocalization in adult rats. Exp Anim 52:409–414. 10.1538/expanim.52.409. [PubMed] [CrossRef] [Google Scholar]
97. Negus SS, Neddenriep B, Altarifi AA, Carroll FI, Leitl MD, Miller LL. 2015. Effects of ketoprofen, morphine, and к opioids on pain-related depression of nesting in mice. Pain 156:1153–1160. [PMC free article] [PubMed] [Google Scholar]
98. Nunamaker EA, Goldman JL, Adams CR, Fortman JD. 2018. Evaluation of analgesic efficacy of meloxicam and 2 formulations of buprenorphine after laparotomy in female Sprague–Dawley rats. J Am Assoc Lab Anim Sci 57:498–507. 10.30802/AALAS-JAALAS-17-000129. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
99. Nyby J, Whitney G. 1978. Ultrasonic communications of adult myomorph rodents. Neurosci Biobehav Rev 2:1–14. 10.1016/0149-7634(78)90003-9. [CrossRef] [Google Scholar]
100. Olfert E, Bhasin J, Latt R, Macallum E, McCutcheon K, Rainnie D, Schunk M. 1998. The CCAC guidelines on: choosing an appropriate endpoint in experiments using animals for research, teaching and testing. Ottawa (ON): Canadian Council on Animal Care. [Google Scholar]
101. Olfert ED, Godson DL. 2000. Humane endpoints for infectious disease animal models. ILAR J 41:99–104. 10.1093/ilar.41.2.99. [PubMed] [CrossRef] [Google Scholar]
102. Oliver VL, Athavale S, Simon KE, Kendall LV, Nemzek JA, Lofgren JL. 2017. Evaluation of pain assessment techniques and analgesia efficacy in a female guinea pig (Cavia porcellus) model of surgical pain. J Am Assoc Lab Anim Sci 56:425–435. [PMC free article] [PubMed] [Google Scholar]
103. Oliver V, De Rantere D, Ritchie R, Chisholm J, Hecker KG, Pang DS. 2014. Psychometric assessment of the Rat Grimace Scale and development of an analgesic intervention score. PLoS One 9:1–7. 10.1371/journal.pone.0097882. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
104. Oliver VL, Thurston SE, Lofgren JL. 2018. Using cageside measures to evaluate analgesic efficacy in mice (Mus musculus) after surgery. J Am Assoc Lab Anim Sci 57:186–201. [PMC free article] [PubMed] [Google Scholar]
105. Oliverio A, Castellano C, Puglisi-Allegra S. 1982. Opiate analgesia: evidence for circadian rhythms in mice. Brain Res 249:265–270. 10.1016/0006-8993(82)90060-9. [PubMed] [CrossRef] [Google Scholar]
106. Parent AJ, Beaudet N, Beaudry H, Bergeron J, Bérubé P, Drolet G, Sarret P, Gendron L. 2012. Increased anxiety-like behaviors in rats experiencing chronic inflammatory pain. Behav Brain Res 229:160–167. 10.1016/j.bbr.2012.01.001. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
107. Patel PK, Sahu J, Chandal SS. 2016. A detailed review on nociceptive models for the screening of analgesic activity in experimental animals. International Journal of Neurologic Physical Therapy 2:44–50. [Google Scholar]
108. Percie du Sert N, Rice AS. 2014. Improving the translation of analgesic drugs to the clinic: animal models of neuropathic pain. Br J Pharmacol 171:2951–2963. 10.1111/bph.12645. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
109. Perl ER. 2007. Ideas about pain, a historical review. Nat Rev Neurosci 8:71–80. 10.1038/nrn2042. [PubMed] [CrossRef] [Google Scholar]
110. Peterson NC, Nunamaker EA, Turner PV. 2017. To treat or not to treat: The effects of pain on experimental parameters. Comp Med 67:469–482. [PMC free article] [PubMed] [Google Scholar]
111. Pfeiffenberger U, Yau T, Fink D, Tichy A, Palme R, Egerbacher M, Rülicke T. 2015. Assessment and refinement of intra-bone marrow transplantation in mice. Lab Anim 49:121–131. 10.1177/0023677214559627. [PubMed] [CrossRef] [Google Scholar]
112. Philips BH, Weisshaar CL, Winkelstein BA. 2017. Use of the Rat Grimace Scale to evaluate neuropathic pain in a model of cervical radiculopathy. Comp Med 67:34–42. [PMC free article] [PubMed] [Google Scholar]
113. Raffaeli W, Arnaudo E. 2017. Pain as a disease: an overview. J Pain Res 10:2003–2008. 10.2147/JPR.S138864. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
114. Raşid O, Chirita D, Iancu AD, Stavaru C, Radu DL. 2012. Assessment of routine procedure effect on breathing parameters in mice by using whole-body plethysmography. J Am Assoc Lab Anim Sci 51:469–474. [PMC free article] [PubMed] [Google Scholar]
115. Redgate ES, Deutsch M, Boggs SS. 1991. Time of death of CNS tumor-bearing rats can be reliably predicted by body weight-loss patterns. Lab Anim Sci 41:269–273. [PubMed] [Google Scholar]
116. Rock ML, Karas AZ, Rodriguez KB, Gallo MS, Pritchett-Corning K, Karas RH, Aronovitz M, Gaskill BN. 2014. The time-to-integrate-to-nest test as an indicator of wellbeing in laboratory mice. J Am Assoc Lab Anim Sci 53:24–28. [PMC free article] [PubMed] [Google Scholar]
117. Roughan JV, Bertrand HG, Isles HM. 2016. Meloxicam prevents COX2-mediated post-surgical inflammation but not pain following laparotomy in mice. Eur J Pain 20:231–240. 10.1002/ejp.712. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
118. Roughan JV, Coulter CA, Flecknell PA, Thomas HD, Sufka KJ. 2014. The conditioned place preference test for assessing welfare consequences and potential refinements in a mouse bladder cancer model. PLoS One 9:1–16. 10.1371/journal.pone.0103362. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
119. Roughan JV, Flecknell PA. 2000. Effects of surgery and analgesic administration on spontaneous behaviour in singly housed rats. Res Vet Sci 69:283–288. 10.1053/rvsc.2000.0430. [PubMed] [CrossRef] [Google Scholar]
120. Roughan JV, Flecknell PA. 2003. Evaluation of a short duration behaviour-based post-operative pain scoring system in rats. Eur J Pain 7:397–406. 10.1016/S1090-3801(02)00140-4. [PubMed] [CrossRef] [Google Scholar]
121. Roughan JV, Flecknell PA. 2004. Behaviour-based assessment of the duration of laparotomy-induced abdominal pain and the analgesic effects of carprofen and buprenorphine in rats. Behav Pharmacol 15:461–472. 10.1097/00008877-200411000-00002. [PubMed] [CrossRef] [Google Scholar]
122. Roughan JV, Flecknell PA, Davies BR. 2004. Behavioural assessment of the effects of tumour growth in rats and the influence of the analgesics carprofen and meloxicam. Lab Anim 38:286–296. 10.1258/002367704323133673. [PubMed] [CrossRef] [Google Scholar]
123. Roughan JV, Wright-Williams SL, Flecknell PA. 2009. Automated analysis of postoperative behaviour: assessment of HomeCageScan as a novel method to rapidly identify pain and analgesic effects in mice. Lab Anim 43:17–26. 10.1258/la.2008.007156. [PubMed] [CrossRef] [Google Scholar]
124. Rutten K, Gould SA, Bryden L, Doods H, Christoph T, Pekcec A. 2018. Standard analgesics reverse burrowing deficits in a rat CCI model of neuropathic pain, but not in models of type 1 and type 2 diabetes-induced neuropathic pain. Behav Brain Res 350:129–138. 10.1016/j.bbr.2018.04.049. [PubMed] [CrossRef] [Google Scholar]
125. Rutten K, Robens A, Read SJ, Christoph T. 2014. Pharmacological validation of a refined burrowing paradigm for prediction of analgesic efficacy in a rat model of sub-chronic knee joint inflammation. Eur J Pain 18:213–222. 10.1002/j.1532-2149.2013.00359.x. [PubMed] [CrossRef] [Google Scholar]
126. Rutten K, Schiene K, Robens A, Leipelt A, Pasqualon T, Read SJ, Christoph T. 2014. Burrowing as a nonreflex behavioural readout for analgesic action in a rat model of sub-chronic knee joint inflammation. Eur J Pain 18:204–212. 10.1002/j.1532-2149.2013.00358.x. [PubMed] [CrossRef] [Google Scholar]
127. Sharp J, Azar T, Lawson D. 2014. Effects of a complex housing environment on heart rate and blood pressure of rats at rest and after stressful challenges. J Am Assoc Lab Anim Sci 53:52–60. [PMC free article] [PubMed] [Google Scholar]
128. Shepherd AJ, Cloud ME, Cao YQ, Mohapatra DP. 2018. Deficits in burrowing behaviors are associated with mouse models of neuropathic but not inflammatory pain or migraine. Front Behav Neurosci 12:1–11. 10.3389/fnbeh.2018.00124. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
129. Sherwin CM, Haug E, Terkelsen N, Vadgama M. 2004. Studies on the motivation for burrowing by laboratory mice. Appl Anim Behav Sci 88:343–358. 10.1016/j.applanim.2004.03.009. [CrossRef] [Google Scholar]
130. Smith M, Taylor C, Weerasinghe N, Koutsikou S, Lumb B, Murrell J. 2016. Does inflammation induced by ultraviolet B and heat rekindling alter pain-related behaviour in rats. Vet Anaesth Analg 43:579–585. 10.1111/vaa.12349. [PubMed] [CrossRef] [Google Scholar]
131. Sorge RE, Martin LJ, Isbester KA, Sotocinal SG, Rosen S, Tuttle AH, Wieskopf JS, Acland EL, Dokova A, Kadoura B, Leger P, Mapplebeck JC, McPhail M, Delaney A, Wigerblad G, Schumann AP, Quinn T, Frasnelli J, Svensson CI, Sternberg WF, Mogil JS. 2014. Olfactory exposure to males, including men, causes stress and related analgesia in rodents. Nat Methods 11:629–632. 10.1038/nmeth.2935. [PubMed] [CrossRef] [Google Scholar]
132. Sotocinal SG, Sorge RE, Zaloum A, Tuttle AH, Martin LJ, Wieskopf JS, Mapplebeck JC, Wei P, Zhan S, Zhang S, McDougall JJ, King OD, Mogil JS. 2011. The Rat Grimace Scale: a partially automated method for quantifying pain in the laboratory rat via facial expressions. Mol Pain 7:55. [PMC free article] [PubMed] [Google Scholar]
133. Späni D, Arras M, Konig B, Rulicke T. 2003. Higher heart rate of laboratory mice housed individually vs in pairs. Lab Anim 37:54–62. 10.1258/002367703762226692. [PubMed] [CrossRef] [Google Scholar]
134. Sperry MM, Yu YH, Welch RL, Granquist EJ, Winkelstein BA. 2018. Grading facial expression is a sensitive means to detect grimace differences in orofacial pain in a rat model. Sci Rep 8:1–10. 10.1038/s41598-018-32297-2. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
135. Stokes WS. 2002. Humane endpoints for laboratory animals used in regulatory testing. ILAR J 43 Suppl:S31–S38. [PubMed] [Google Scholar]
136. Streiner DL, Norman GR. 2008. Health measurement scales: a practical guide to their development and use. Oxford (United Kingdom): Oxford University Press; 10.1093/acprof:oso/9780199231881.001.0001 [CrossRef] [Google Scholar]
137. Sufka KJ. 1994. Conditioned place preference paradigm: a novel approach for analgesic drug assessment against chronic pain. Pain 58:355–366. 10.1016/0304-3959(94)90130-9. [PubMed] [CrossRef] [Google Scholar]
138. Tappe-Theodor A, Kuner R. 2014. Studying spontaneous and ongoing pain in rodents—challenges and opportunities. Eur J Neurosci 39:1881–1890. 10.1111/ejn.12643. [PubMed] [CrossRef] [Google Scholar]
139. Teeling JL, Felton LM, Deacon RM, Cunningham C, Rawlins JN, Perry VH. 2007. Sub-pyrogenic systemic inflammation impacts on brain and behavior, independent of cytokines. Brain Behav Immun 21:836–850. 10.1016/j.bbi.2007.01.012. [PubMed] [CrossRef] [Google Scholar]
140. Tunstall J, Mueller K, Grove White D, Oultram JWH, Higgins HM. 2019. Lameness in beef cattle: UK farmers’ perceptions, knowledge, barriers, and approaches to treatment and control. Front Vet Sci 6:1–14. 10.3389/fvets.2019.00094. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
141. Tuttle AH, Molinaro MJ, Jethwa JF, Sotocinal SG, Prieto JC, Styner MA, Mogil JS, Zylka MJ. 2018. A deep neural network to assess spontaneous pain from mouse facial expressions. Mol Pain 14:1–9. 10.1177/1744806918763658. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
142. Wallace VCJ, Norbury TA, Rice ASC. 2004. Ultrasound vocalisation by rodents does not correlate with behavioural measures of persistent pain. Eur J Pain 9:445–452. 10.1016/j.ejpain.2004.10.006. [PubMed] [CrossRef] [Google Scholar]
143. Watson PF, Petrie A. 2010. Method agreement analysis: a review of correct methodology. Theriogenology 73:1167–1179. 10.1016/j.theriogenology.2010.01.003. [PubMed] [CrossRef] [Google Scholar]
144. Whittaker AL, Lymn KA, Nicholson A, Howarth GS. 2015. The assessment of general wellbeing using spontaneous burrowing behaviour in a short-term model of chemotherapy-induced mucositis in the rat. Lab Anim 49:30–39. 10.1177/0023677214546913. [PubMed] [CrossRef] [Google Scholar]
145. Whittaker AL, Lymn KA, Wallace GL, Howarth GS. 2016. Differential effectiveness of clinically-relevant analgesics in a rat model of chemotherapy-induced mucositis. PLoS One 11:1–19. 10.1371/journal.pone.0158851. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
146. Williams WO, Riskin DK, Mott AK. 2008. Ultrasonic sound as an indicator of acute pain in laboratory mice. J Am Assoc Lab Anim Sci 47:8–10. [PMC free article] [PubMed] [Google Scholar]
147. Wodarski R, Delaney A, Ultenius C, Morland R, Andrews N, Baastrup C, Bryden LA, Caspani O, Christoph T, Gardiner NJ, Huang W, Kennedy JD, Koyama S, Li D, Ligocki M, Lindsten A, Machin I, Pekcec A, Robens A, Rotariu SM, Vo BS, Segerdahl M, Stenfors C, Svensson CI, Treede RD, Uto K, Yamamoto K, Rutten K, Rice AS. 2016. Cross-centre replication of suppressed burrowing behaviour as an ethologically relevant pain outcome measure in the rat: a prospective multicentre study. Pain 157:2350–2365. 10.1097/j.pain.0000000000000657. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
148. Workman P, Aboagye EO, Balkwill F, Balmain A, Bruder G, Chaplin DJ, Double JA, Everitt J, Farningham DA, Glennie MJ, Kelland LR, Robinson V, Stratford IJ, Tozer GM, Watson S, Wedge SR, Eccles SA, Committee of the National Cancer Research Institute 2010. Guidelines for the welfare and use of animals in cancer research. Br J Cancer 102:1555–1577. 10.1038/sj.bjc.6605642. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
149. Wright JM, Gourdon JC, Clarke PB. 2010. Identification of multiple call categories within the rich repertoire of adult rat 50-kHz ultrasonic vocalizations: effects of amphetamine and social context. Psychopharmacology (Berl) 211:1–13. 10.1007/s00213-010-1859-y. [PubMed] [CrossRef] [Google Scholar]
150. Wright-Williams SL, Courade JP, Richardson CA, Roughan JV, Flecknell PA. 2007. Effects of vasectomy surgery and meloxicam treatment on faecal corticosterone levels and behaviour in 2 strains of laboratory mouse. Pain 130:108–118. 10.1016/j.pain.2006.11.003. [PubMed] [CrossRef] [Google Scholar]
151. Wright-Williams S, Flecknell PA, Roughan JV. 2013. Comparative effects of vasectomy surgery and buprenorphine treatment on faecal corticosterone concentrations and behaviour assessed by manual and automated analysis methods in C57 and C3H mice. PLoS One 8:1–13. 10.1371/journal.pone.0075948. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
152. Wu J, Zhao Z, Zhu X, Renn CL, Dorsey SG, Faden AI. 2016. Cell cycle inhibition limits development and maintenance of neuropathic pain following spinal cord injury. Pain 157:488–503. 10.1097/j.pain.0000000000000393. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
153. Younger J, McCue R, Mackey S. 2009. Pain outcomes: A brief review of instruments and techniques. Curr Pain Headache Rep 13:39–43. 10.1007/s11916-009-0009-x. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
154. Zegre Cannon C, Kissling GE, Goulding DR, King-Herbert AP, Blankenship-Paris T. 2011. Analgesic effects of tramadol, carprofen or multimodal analgesia in rats undergoing ventral laparotomy