3

New and Emerging Methods, Approaches, and Technologies for Detecting Pain and Its Causes

This chapter addresses the committee’s task to identify potential new and emerging methods, approaches, and technologies for detecting hoof and pastern pain and its causes (see Chapter 1, Box 1-3 for the full statement of task). This chapter begins with a discussion of factors that affect pain perception and the expression of pain. This is followed by a review of pain detection methods and technologies based on horse behavior and physiological parameters and a discussion of how these methods could be used to improve the detection of soreness in horses during inspections for compliance with the Horse Protection Act (HPA).

Detection of pain in horses is complex and requires adequate training and experience. A thorough clinical exam is the foundation of veterinary diagnosis, and its value for grading pain and lameness is supported by an abundance of scientific evidence. Palpation of the painful area remains the gold standard for detecting soreness, though behavioral changes and facial expressions can also help identify a painful individual. Human health care practitioners commonly use grimace scales as an adjunctive method to grade pain, and their use in horses is promising. Thus far most of the research has looked at facial expressions in horses with or without clinical pain under controlled conditions. The standardized protocol used during show inspections of the Tennessee walking horse (TWH) offers a unique opportunity to study whether the facial grimace could be adopted as a noninvasive low-cost method to improve detection of soreness.

INTRODUCTION

Pain is a vital sensory modality that detects certain types of threats to homeostasis—the tendency of the body’s various systems to remain at equilibrium and maintain optimal functioning. Behavioral reactions to pain act to defend the animal against potential injury and include efforts to escape from, cope with, avoid, or remove the source of pain. The sensation of a harmful chemical, mechanical, or thermal stimulus activates peripheral pain receptors, called nociceptors. The neural signal is transmitted to the dorsal horn of the spinal cord, where the primary afferent nerve axon signals neurons in the spinal cord, initiating a withdrawal reflex; the signal is also passed along to the brain, which leads to the actual perception of pain.

Pain perception in horses can be influenced by extraneous factors in the environment as well as by horses’ individual differences in pain sensitivity, coping style, and history. For example, compared with sensitive horses (i.e., active coping style), stoic horses (i.e., passive coping style) tend to demonstrate less behavioral change with pain (Ijichi et al., 2014). This may be an important factor to consider when assessing pain in TWHs, which have been bred for the qualities of docility and stoicism. Furthermore, individual differences among horses in sensitivity to pain, personality, and training history might cause some sore horses to display pain behaviors while others under the same conditions might not. Furthermore, extraneous factors such as stress and distractions—as would be present at a horse show—can help explain why the same horse may respond differently to pain from one moment to the next.

Context and Environment

Situational factors can facilitate or inhibit pain expression and thus contribute to scoring and decision errors during an evaluation for pain. The modulation of pain behavior by environmental stressors, distractions, other sources of pain, and habituation is discussed below.

Stressors

Pain and stress are closely related but operationally distinct constructs. Pain is one type of stress that threatens homeostasis, but not every stressor is painful. Behavioral responses to pain may be similar to and confounded with responses to other causes of distress (Rietmann et al., 2004). To accurately assess pain and avoid confounding pain and stress responses, pain assessment procedures are typically conducted in an environment with as few extraneous stressors as possible. For example, in scoring a horse’s facial expression of pain using the Horse Grimace Scale (HGS), Dalla Costa et al. (2014), who developed the scale, recommend that the horse should be observed in a quiet location without outside interference from observers. To increase the accuracy of the score, the authors also suggest videotaping and repeating observations, particularly if the goal is to monitor changes in pain state, such as following surgery (see discussion of the HGS in the section on Behavioral Assessment of Pain).

The effect of stress on pain sensitivity is complex and depends on the type of stressor, on its duration and intensity, and on individual differences in the stress response. Research in horses suggests that pain thresholds increase when stressors are present in the environment, thereby inhibiting pain expression. This phenomenon is called stress-induced analgesia or hypoalgesia and is considered typical in prey animals. Even a mild stressor has been shown to suppress pain behavior in horses. In one study, horses moved and paced more when in a stressful situation (social isolation) and were less active in response to mild somatic pain (a neck skin pinch). Horses were restless in the combined stressor–pain condition, indicating that stressors can moderate pain behavior (Reid et al., 2017). In another study, the mere presence of a person was enough to inhibit pain. Horses in a hospital with orthopedic pain showed significantly fewer discomfort behaviors when a caretaker was present than when the horse was alone (Torcivia and McDonnell, 2020). Discretely observing the horse in a quiet environment—for example, by using video—is the ideal standard, but it is not possible in the context of an inspection during a horse show. It is important to consider, however, that even mild signs suggesting pain observed in an environment with distractions should be taken seriously, since the threshold for pain perception and expression may be markedly increased.

Responses to stress and pain can be inextricably confounded in some cases. Studies of pain in the ridden horse recognize that distress behaviors can be caused by either pain—for example, from a tight noseband or an ill-fitting saddle—or by anxiety from environmental stressors (Dyson et al., 2018; Gleerup et al., 2018). To accurately identify pain, the assessment protocol should minimize environmental stressors and discriminate between responses caused by pain and those caused by stress.

Discriminant validation conducted on several human behavioral and facial expression pain scales has confirmed that stress and pain are distinct. In one study, the Wong-Baker Scale featuring simple cartoon pain faces accurately discriminated between self-reports of pain and fear in young children (Garra et al., 2013). In another study, behavioral responses for five commonly used neonatal behavioral pain scales were found to discriminate between a painful experience (heel lance) and a stressful experience (nappy change), whereas physiological measures such as heart rate, blood pressure, and respiration rate measures did not accurately discriminate between the pain and stress (Kappesser et al., 2019). Distinguishing between behavioral expression of pain and stress is a relatively unexplored area of research in horses. In one pilot study (Dalla Costa et al., 2017) researchers scored facial expressions of horses in four potentially distressing situations using the HGS, which was designed to grade pain. Only horses that were

startled by an umbrella opening (the fear condition) trended to score higher on two of six facial indicators, “ears held stiffly backward” and “prominent chewing muscles.” On the basis of these findings the researchers concluded that the assessment tool was a specific indicator of pain. Further discriminant validation research of this sort is needed to distinguish pain from other sources of stress in horses.

Distractions

Horses are inspected for violations of the Horse Protection Regulations at show grounds which have a wide range of environmental distractions, including other horses, exhibitors and spectators, and noises. To reduce distractions, 9 C.F. R. § 11.5(a)(2) states that:

Section 11.6 describes the inspection space and facility requirements and states:

Section 11.21(a)(4) also discourages handlers from distracting the horse, and states that:

Distractions created by horse custodians can contribute to unexplained variance in pain assessment during an inspection and across inspectors. The committee’s observation of 61 inspection videos revealed that many exhibitors adhered to Horse Protection Regulations when holding a horse for inspection, but others did not. Horse custodians inadvertently or intentionally held reins closer than 18 inches from the bit shank, touched the horse or the bit, held the reins taut (in some cases above the level of the horse’s mouth), jiggled or jerked on the reins, and stood in front of the horse in a dominant stance. The custodian may have been trying to control or correct an unruly horse, but these distractions can draw the horse’s attention away from the digital palpation; a shift in attention has been shown to suppress pain expression (Hoegh et al., 2019; Torcivia and McDonnell, 2020).

Conditioned Pain Modulation

Pain inhibits pain. Conditioned pain modulation (CPM) occurs when two painful stimuli are presented together, either simultaneously or sequentially. The pain of interest is in one location, but the response to that pain is inhibited by pain induced in a different location (Kennedy et al., 2016). Both distraction and CPM suppress pain, but they appear to work by two independent mechanisms (Hoegh et al., 2019). In humans, CPM is known to inhibit the withdrawal reflex at the level of spinal activity via “differential recruitment of the muscles involved in the protective behavior” (Jure et al., 2019, p. 259). To the committee’s knowledge CPM has not been studied in horses, although in humans it is a hypothesized mechanism for exercise-induced analgesia, a phenomenon whereby pain is inhibited by vigorous exercise (Lima et al., 2017). One study with endurance horses confirmed that lower limb pain was less immediately after competition than it was before competition (Schambourg and Taylor, 2020).

Through CPM, a horse’s withdrawal response to the digital palpation of a painful pastern could be inhibited by pain in a different location. When a horse is held for inspection, pain in the oral cavity will evoke an evasive response. The biomechanics of forces created by movement of the bit, tension in the reins, or the reins raised sharply can cause pain. A shank bit is used on most competition horses undergoing inspection. As a result of lever action, any force applied by movement of the reins will be amplified. When the reins are lifted upward, the direction of bit rotation is opposite to their direction during riding, putting pressure on sensorily naïve tissue. To relieve pain in the oral cavity, a horse is likely to raise the head and neck and brace backward (O. Doherty, International Association of Equitation Science Council, personal communication, April 20, 2020). In its review of 61 inspection videos, the committee observed some horses reacting this way during digital palpation, creating uncertainty about the source: Was it a reaction to the palpation of a painful pastern, to pain in the oral cavity, or to some other stressor?

Habituation and Peripheral Sensitization

Reflex strength can be reduced by repeated stimulation through the process of habituation and is a potential source of variability in responses to digital palpation during inspections. Any initial responses to a stimulus, such as pressure, applied to a nonpainful area are expected to habituate and therefore to decrease with repeated stimulation. Responses to pressure applied to a painful area, however, are not expected to habituate. Information about response habituation could be incorporated into the inspection training in order to reduce the misattribution of potentially stressful, but nonpainful, handling procedures as a pain response.

Peripheral sensitization can result in the expansion of pain response to mechanical stimuli, such as digital palpation, to an area of uninjured tissue adjacent to the source of injury (Woolf, 1989). Nociceptors

generally have high thresholds that are only activated by intense stimuli, but tissue injury and peripheral sensitization result in a decreased pain threshold. Thus, digital palpation of a painful area of the pastern could elicit a withdrawal response over a broad area.

Individual Differences

Horses differ in their sensitivity and response to pain due to differences in genetics, personality, past experiences, and training history. Individual differences result in variations between horses and can help explain why some sored horses, as determined by physical evidence such as a violation of the scar rule or inflammation that is apparent with thermography, may not display pain behaviors. Individual differences in sensitivity, coping style, and training history and their potential effect on pain behavior are described below.

Pain Sensitivity

Some individuals are inherently more responsive to pain induced by a stimulus than others because of genetically based differences in nociceptor sensitivity and activity. Previous painful insults can also have long-lasting effects on nociceptor sensitivity. Research on the development of chronic pain has provided information about the neuroplasticity of pain. For example, repeated injury can exacerbate a painful stimulus and experience through an increase in the number and activity of pain receptors (Woolf, 1989). This can lead to hyperalgesic priming, which is an increased sensitivity to subthreshold stimuli, and in extreme cases to allodynia, where pain is caused by a stimulus that does not normally elicit pain, such as the light touch of clothes on sunburned skin (Latremoliere and Woolf, 2009).

Coping Style

Coping style refers to an individual’s manner of responding to perceived danger, a stressful situation, or an environmental challenge, and, like other dimensions of personality, coping style is stable across situations and time (Coppens et al., 2010; Ijichi et al., 2014). Coping style is modeled as a continuum, with proactive and reactive types as anchors (Koolhaas et al., 1999; Koolhaas and Van Reenen, 2016). Proactive individuals have an active coping style, exerting control to remove themselves from the situation (flight) or to remove the source of danger (fight). Reactive individuals tend to be passive, responding to stressors by freezing and emotional blunting (lack of emotional expression) (Koolhaas et al., 1999). Individual differences in coping style can muddle the link between the intensity of a painful stimulus and the observed pain response (Squibb et al., 2018).

Coping style has been linked to personality (Koolhaas and Van Reenen, 2016). Bold personality types tend to have a proactive coping style and shy personality types tend to have a reactive coping style. Breed differences in personality, notably anxiousness and excitability, have been reported in horses (Lloyd et al., 2008). Although personality has not been systematically studied in the TWH, the breed is characterized as having a “gentle disposition” and a “calm, docile temperament” (TWHBEA, 2020), traits consistent with a shy personality and reactive coping style. Despite their quiet, compliant demeanor, individuals with a reactive coping style have a more pronounced physiological response to stressors (Coppens et al., 2010), raising welfare concerns.

Training History

Compliance during a stressful or painful handling experience is not always a reliable indicator of a horse’s underlying affect, physiological state, or level of physical discomfort but may instead reflect its

training history. In one study, compliance, as measured by latency to cross a tarp or walk through streamers, did not correlate with physiological indicators of stress, including heart rate variability, infrared eye temperature, and core body temperature (Squibb et al., 2018). The researchers hypothesized that compliance (such as standing still and following) in trained horses may depend more on previously learned cues than on the horse’s level of distress and that these previously learned cues could “overshadow inherent emotional responses” (Squibb et al., 2018, p. 37).

For practical and safety reasons, horses are generally trained to defer to a handler rather than to react to events in the environment. When training involves the application of pressure through a lead rope or rein, the horse seeks to escape from the discomfort, and behavior such as halting is reinforced by the release of pressure (McGreevy and McLean, 2009).

Horses quickly learn to anticipate and respond to cues that predict pain or pressure. Pressure applied to the bit can cause oral pain that may overshadow the limb withdrawal response during palpation of the pastern. Through associative learning, cues that predict bit pressure or pain, such as a movement of the hand, reins, or halter, can also come to overshadow pain responses to palpation, possibly through an extension of CPM (Kennedy et al., 2016).

The intensity and urgency of coping with a stressor can be mitigated in the presence of a calm, competent handler (Ijichi et al., 2018). Having some degree of control over pain and stressors can also mitigate many of their negative effects. Sustained tension on the reins during training or inspections, however, causes acute uncontrollable and inescapable pain, and learned helplessness may result (Hall et al., 2008). Research has not been done with horses, but seminal work with dogs (Seligman and Maier, 1967; Maier and Seligman, 1976) and rats (Seligman and Beagley, 1975) provides a model for learned helplessness, indicating that it is an outcome of uncontrollable stress and pain; in this research dogs exposed to inescapable shock became apathetic and, in subsequent trials, made no effort to escape from pain, and the effect persisted over time (Seligman et al., 1975). Learned helplessness resulting from aversive training methods has been suggested in horses, a species that displays a surprising level of compliance under stressful and painful conditions (Waran et al., 2002).

VARIABILITY OF PAIN EXPRESSION

To accurately determine the amount of pain an individual is experiencing requires having both reliable assessment methods and an agreement among raters about how to implement those methods. For most pain assessment scales, reliability, validity, and inter-rater agreement are known and published, having been calculated as part of the scale development and validation process. For example, as mentioned above (see section on Nociceptive Withdrawal Reflex) scores on the Composite Pain Scale (CPS) item “Response to palpation of the painful area” had good to excellent agreement between raters as calculated by Cohen’s kappa statistic (κ) (Bussieres et al., 2008), which is a widely accepted measure for determining inter-rater reliability. An assessment method with low inter-rater reliability is generally not used in practice.

The validity of a behavioral assessment procedure is called into question when there are scoring discrepancies among raters. Low agreement can occur when one or some combination of the following occurs: (1) the assessment method is unreliable; (2) extraneous factors create inconsistencies in the behavior being scored; and (3) raters apply the method differently or inconsistently, often due to inadequate training or conflict of interest.

The reliability of a behavioral scale can be compromised if it is used in a new context, if it is used by an untrained individual, or if it is applied inconsistently. When a scale is developed in one context but applied in a different context, its validity and reliability may differ from the published values, and additional research must typically be carried out in the new context. For example, the CPS (Bussieres et al., 2008) was developed in horses with induced orthopedic pain and then validated later for use in horses

with laminitis (van Loon and Van Dierendonck, 2019). In addition, clinical scales are expected to be used by a large number of raters; inter-rater agreement, and thus the validity of the assessment method, is ensured through standardized training and consistent application across raters. Uncontrolled extraneous factors can also introduce error into the assessment. Some behaviors are robust against, and others more easily modulated by, extraneous variables. Pain behavior can be inhibited or facilitated by extraneous variables. Factors that influence variability in the expression of pain are discussed below.

As Sator-Katzenschlager (2014) wrote, “The amount of pain perceived … is assumed to be directly proportional to the extent of injury” (p. 699). Pain perception, however, is subjective and does not necessarily correlate with the degree of injury. In addition, responses may depend on the severity of pain. While veterinarians generally agree in their assessments of severe pain, their assessments tend to differ for moderate and chronic pain (Price et al., 2002; Rietmann et al., 2004).

Finding 3-1: Individual horses differ in perception and expression of pain. These differences are influenced by such factors as distractions and stressors in the immediate environment and the horse’s genetics, training history, temperament, and coping style.

Finding 3-2: Research has shown that horses’ responses to environmental stressors tend to overshadow their responses to pain. Hence, pain assessment scales used in veterinary research and practice recommend observing the horse in a quiet environment to ensure that the findings are valid and reliable.

Finding 3-3: Observation of 61 inspection videos revealed that some inspections were conducted in relatively quiet locations during a show whereas others were conducted in locations with loud noises and with large numbers of people and other horses moving around nearby.

Finding 3-4: The “pain inhibits pain” effect (i.e., conditioned pain modulation) occurs when the pain of interest is inhibited by a pain induced in a different part of the horse’s body. During inspection, it is possible that pain in the lower limb and hoof that is being evaluated could be inhibited if the horse also experiences pain because of how it is being restrained by the custodian.

Finding 3-5: Observation of 61 inspection videos revealed numerous incidents of stewarding during the standing inspection that were not dealt with by the inspector. Stewarding may have simply been out of habit or to prevent or control the horse’s restless behavior. Examples of stewarding included holding the reins closer than 18 inches from the bit, often just below or on the shank. In some cases, the horse was restrained with constant tension, often with the reins held in an upward direction, or the reins were pulled sharply. These restraint tactics create a distraction during the palpation procedure and can induce pain in the oral cavity, and they violate Horse Protection Regulations.

Conclusion 3-1: Environmental distractions present during horse inspections can result in the inspector reaching inaccurate conclusions regarding soreness. Distractions and stressors can inhibit a horse’s sensitivity to and expression of pain, such that detection of soreness would be missed, or a horse's reaction to distractions could be incorrectly attributed to pain. Moreover, when more than one inspector examines the horse, its behavior may differ between the two inspections if the number and type of distractions and stressors at that location and time also differ.

Conclusion 3-2: Pain or discomfort can be caused by restraint during an inspection. Some restraint methods create acute oral cavity pain that can inhibit limb and hoof pain. How a horse is restrained during an inspection may differ between inspectors and potentially result in different observations and conclusions about the same horse.

BEHAVIORAL ASSESSMENT OF PAIN

The goal of inspections is, as described in Chapter 1 of this report, to examine a horse to determine compliance with or violation of the HPA. Designated qualified persons (DQPs) and, less often, veterinary medical officers (VMOs) examine horses entered in show classes for lower limb pain, scars and lesions, and prohibited substances that contribute to or mask soreness. The Horse Protection Regulations allow for the use of visual methods to determine whether a horse is in violation or is compliant (9 C.F.R. § 11.21) and further states that the inspector should “observe for responses to pain in the horse” (9 C.F.R. § 11.21(a)(2); see Chapter 2). However, the Horse Protection Regulations do not specifically mention examining behaviors that can be indicative of pain, nor is behavior included as a category on official inspection forms used by DQPs and VMOs.

A horse’s behavior can inform an assessment of physical pain and distress. A valid pain assessment method should produce a consistent response that corresponds to the level of perceived pain. Pain perception, however, involves a subjective element that does not always correlate perfectly with the degree of physical insult (Reid et al., 2018).

Most objective clinical pain scales include a behavioral component. Being able to recognize a patient’s pain experience aids in making decisions about diagnosis and appropriate palliative care. In nonverbal humans and animals, for whom self-report measures are not reliable or possible, grading pain relies heavily on observing behavior. To accurately judge an animal’s pain state requires being familiar with the species and the individual, using a reliable assessment method, and controlling factors that contribute to variation in the perception and expression of pain. This section presents clinical scales for evaluating orthopedic pain and laminitis in horses that include behavioral indicators, and it reviews the factors that can facilitate or inhibit pain expression.

Pain Sensation, Perception, and Expression

As noted above, pain is a vital sensory modality that detects certain types of threats to homeostasis. Initial behavioral responses to acute pain are mediated by descending motor pathways. The nociceptive withdrawal reflex (NWR) is a relatively simple flexor reflex produced entirely by neural pathways that lie within the spinal cord. A familiar example of the NWR is the automatic withdrawal of a hand after touching a hot stove burner.

Pain signals are further transmitted to the brain via ascending afferent neurons in the spinal cord. Pain perception occurs in the brain, bringing pain into conscious awareness, localizing the pain, and adding an emotional component. Pain perception is complex and is modulated by an individual's past experiences and coping style. An individual’s perception of pain is the amount of pain an individual subjectively experiences at a given moment, which does not always correspond with the absolute magnitude of the stimulus causing the pain. Behaviors that reflect a horse’s perceived level of pain include facial expressions and voluntary motor behaviors, such as posture, pawing, and head movements.

Pain Behavior Scales

The choice of an assessment tool for diagnosing and grading physical pain in horses depends on the source of pain (e.g., visceral, orthopedic, traumatic)¹ and the intended use of the assessment. Pain scales intended for research or inpatient hospital use tend to be time-consuming to complete and complicated to score and often require extensive training. In the case of some tools, repeated observations (e.g., baseline measures) are also needed before a determination about the animal’s pain state can be made, limiting

___________________

¹ For a comprehensive review of pain assessment tools organized by type of pain (nonspecific pain, abdominal pain, and limb and foot pain) see Ashley et al. (2005).

the tools’ clinical application (de Grauw and van Loon, 2016). Pain assessment conducted in the field tends to employ scales that are simpler, take less time to complete and score, and yield rapid results. These features facilitate diagnosis and treatment in real-life conditions and when time is limited.

The development of clinical scales for assessing pain in animals lags behind, but parallels, the development of these scales in humans. Pain assessment in infants (Riddell et al., 2013) is particularly relevant. As is the case with infant pain scales, equine pain scales generally include physiological (e.g., heart rate, blood pressure, respiration rate) and behavioral (e.g., facial expression, posture, discrete behaviors) indicators.

The equine pain scales presented in this chapter meet the following criteria: (1) The scale was developed or validated, or both, in horses with orthopedic pain or laminitis—two types of clinical pain that are most similar to the pain a sore horse might experience; (2) the scale includes at least some behavioral indicators of pain that could be freely expressed by a horse during a show inspection; other behaviors, such as lying down, might also be included on a scale but are not relevant to the show context; and finally (3) the scale has been validated to some extent for specificity, sensitivity, and/or inter-rater reliability. When considering scale validation, the committee was mindful that most equine scales have been developed for clinical application. In this context, false negative results, meaning that the assessment method does not detect some individuals that do have pain, create a serious treatment and welfare concern. In the context of HPA horse inspections, however, false positive results raise an equally serious concern because they can potentially result in a reported soring violation when there was none, with unwarranted penalties for the exhibitor.

BEHAVIORAL INDICATORS OF PAIN

Nociceptive Withdrawal Reflex

NWR is a behavioral response to palpation of a painful area (see Chapter 2 for a discussion of palpation). Limb withdrawal responses to palpation are graded as positive if the horse displays the NWR or if the horse avoids the pressure by lifting its foot, attempting to paw the ground or stamp, flexing the limb, or attempting to walk off (Luna et al., 2015).

Research provides evidence that the limb withdrawal response to mechanical stimulation is an accurate and valid method for assessing pain. In a study by Luna et al. (2015), nociceptive thresholds to mechanical, thermal, and electrical stimuli were measured in eight horses. The stimuli were applied to the thorax and lower limb, and thresholds were scored by multiple raters at two time points separated by months. The researchers found that a mechanical stimulus applied to the hoof had the highest inter-observer agreement (100 percent), sensitivity (100 percent), and specificity (94–97 percent), and they concluded that the stimuli “were easy to apply, aversive responses were consistent and easy to interpret, and all tests were reliable, sensitive and specific” (Luna et al., 2015, p. 613).

The limb withdrawal response is included as an item on the CPS (see Table 3-1), which was developed in horses with induced orthopedic pain (Bussieres et al., 2008). Horses showing little or no response to palpation were rated as having little or no pain, and those resisting palpation or showing a violent reaction were rated as having more severe pain. This item has been found to have good to excellent reproducibility across raters. Scores for the “response to palpation” item also had high sensitivity and specificity, meaning that the item accurately discriminated between horses with and without pain.

The NWR is reliably elicited by palpation of a painful forelimb pastern. As a relatively simple and invariant behavior, this flexor reflex is readily identified by different inspectors. Although horses can be trained not to respond to pain and being in an unfamiliar environment can dampen the pain response, the NWR is less affected than other behavioral indicators of pain by training, extraneous environmental factors, and individual differences. For these reasons, palpation of the potentially painful pastern is an indispensable element of the HPA inspection protocol for detecting pain in sored horses.

TABLE 3-1 Score Sheet for the EQUUS-COMPASS Composite Pain Scale

SOURCE: Adapted by van Loon and Van Dierendonck (2019) from Bussieres et al. (2008).

Weight Off-Loading and Lameness

Force applied to a painful limb will cause a horse to shift weight away from the pain, causing it to adopt an abnormal limb position and head and neck movement, which results in lameness, defined as an abnormal stride during locomotion. Observation for lameness is included in the inspection procedures detailed in Chapter 2 (section on Observation of Horse Movement and Appearance). In addition, reluctance to lead, gait abnormalities or problems with locomotion, shifting weight to the rear legs, and stepping forward with the rear limbs while the front limbs remain lightly planted are aspects of lameness and weight off-loading that are included on a list of indicators of pain in the Animal Care, Horse Protection Program DQP training material from the U.S. Department of Agriculture’s Animal and Plant Health Inspection Service (APHIS).²

Behavioral scales developed for horses with laminitis and orthopedic pain typically include items describing abnormal posture, weight bearing, and movement. Behavioral indicators of discomfort are described and illustrated in a recently developed ethogram using horses with orthopedic pain in a hospital (Torcivia and McDonnell, 2020). Another clinical scale, the Obel Method (Meier et al., 2019), originally developed in 1948 (Obel, 1948), is widely used for grading discomfort and lameness associated with laminitis (Table 3-2). In an evaluation procedure that is similar to that used with the Obel scale, DQP and VMO

___________________

² This document was provided by APHIS to the committee. A copy can be requested from the Public Access Records Office of the National Academies of Sciences, Engineering, and Medicine.

TABLE 3-2 Obel Laminitis Grades for Rating a Horse’s Withdrawal from Pressure/Palpation of Localized Area

SOURCE: Adapted from Meier et al. (2019).

inspectors evaluate a horse’s gait while it is walking on a straight line and turning in accordance with 9 C.F.R. § 11.21 9(a)(1). Recognizing gait abnormalities depends on having a standard for comparison, but there are no available gait analyses performed in padded or flat-shod competition horses that have never been subjected to the practice of soring; it would be valuable if such analyses could be carried out in future research (see Chapter 2 section on Observation of Horse Movement and Appearance).

Facial Grimace

In humans, pain scales based on facial expressions offer objective, quick, and simple tools for use in clinical practice. As they are used in human medicine, facial expressions reliably convey information about a patient’s perceived pain and its severity, and both facial expressions and limb withdrawal are commonly used to grade pain in children and infants for whom verbal self-report is unreliable or impossible (Garra et al., 2013).

The way that pain is expressed in the face has features that are similar in a number of mammals and is referred to as a “facial grimace” or “pain face.” Grimace scales have been developed and validated to assess pain in animals that are used in laboratory research, including mice (Langford et al., 2010), rats (Sotocinal et al., 2011), and rabbits (Keating et al., 2012). The APHIS Animal Care, Horse Protection Program training material for DQPs includes “abnormal reactions of the eye, ears, and head in response to palpation” in a list of pain indicators, but no further information is provided. In the past decade several scales have been developed that describe facial features indicative of pain in horses (Dalla Costa et al., 2014; Gleerup et al., 2015; van Loon and Van Dierendonck, 2015). These scales have not been psychometrically compared with one another in a systematic way, but all describe a similar facial expression indicative of pain.

A horse in pain shows distinctive and likely involuntary facial expressions (Dalla Costa et al., 2014; Gleerup et al., 2015; Wathan et al., 2015). In the upper half of the face the horse’s ears rotate backward to focus caudally with increased distance between them. Tension is apparent in the muscles above the eye with a pronounced zygomatic process, and the horse has a withdrawn gaze and a reduced blink rate. In the lower half of the face the horse’s nostrils are dilated, the muzzle is tense with pursed lips, and the chewing muscles along the cheeks are tense. The overall appearance is a flattened facial profile (Table 3-3; Figures 3-1 and 3-2).

The HGS (Figure 3-1) was developed for use in research and clinical practice using a sample of horses that were undergoing routine castration (Dalla Costa et al., 2014), and the scale was later validated on horses diagnosed with acute laminitis (Dalla Costa et al., 2016) and with dental pain (Coneglian et al., 2020). The equine pain face (Gleerup et al., 2018; shown in Figure 3-2) and the Equine Utrecht University

TABLE 3-3 Facial Features of Horses in Pain

Scale for Facial Assessment of Pain (EQUUS-FAP; van Loon and Van Dierendonck, 2015; Table 3-4) also describe grimace-like facial expressions in horses with pain. The equine pain face was developed by comparing facial action units (FAUs) of horses in a control condition and two pain-induction conditions: a chemical burn caused by the topical application of capsaicin on the antebrachium, and ischemic pain caused by a blood pressure cuff. The EQUUS-FAP was developed (van Loon and Van Dierendonck, 2015) and validated (Van Dierendonck and van Loon, 2016) in horses diagnosed with acute colic, and follow-up studies further validated its application to horses with facial pain (van Loon and Van Dierendonck, 2017) and orthopedic pain (van Loon and Van Dierendonck, 2019). In its review of 61 HPA-compliant and noncompliant inspection videos, the committee found that many horses displayed a facial grimace during digital palpation (Figure 3-3), indicating that the palpation was painful. Facial expressions consistent with pain were often observed concurrently with changes in the horses’ posture and focus, including reduced movement of head and neck; ventral positioning of the head, with head positioned forward or turned slightly away from the inspector; and an inward focus of attention.

Scoring the HGS, equine pain face, and EQUUS-FAP scales to grade pain requires some training but is sufficiently simple and quick for the scales to be used in clinical practice. Items for both the HGS (Figure 3-1) and EQUUS-FAP (Table 3-4) scoresheets are assigned a value 0, 1, or 2. Higher values indicate greater pain characterized by increasing tension and internal focus/withdrawal. The EQUUS-FAP scale includes additional facial behavior categories, such as yawning and teeth grinding. Training and use of these objective scales can potentially improve accuracy of pain diagnosis and grading. In one study, dental pain was rated on a scale ranging from 0 (no pain) to 3 (severe pain) from photographs (Coneglian et al., 2020). The research found high agreement among veterinarians who were trained to use the HGS and poor agreement among equine veterinarians who evaluated pain subjectively based on experience.

Mobile apps are currently available for scoring facial expression and other behavioral indicators of pain in horses. The Equine Pain and Welfare App (EPWA) was developed by researchers and veterinarians at Utrecht University in the Netherlands for Android and Apple operating systems. The measurement of pain using facial expressions is based on the EQUUS-FAP scale (van Loon and van Dierendonk, 2015). The Horse Grimace Scale HGS app, developed by AWIN WP4 for Android operating systems, includes an informational video, a training session on how to use the HGS (Della Costa et al., 2014), and a session for scoring horses. The apps offer a convenient, simple, and accurate way to clinically assess behavioral indicators of pain.

Biomedical research has applied computer technology to identify and integrate FAUs that correspond to a level of perceived pain. EquiFACs (Equine Facial Action Coding System) is an emerging method modeled after a human facial action coding system (Ekman and Friesen, 1978). Using technology, the movement and position of facial muscles are categorized and functionally linked to equine affective states (Wathan et al., 2015). A study using this system found the FAU scores of the equine grimace, as coded by the HGS, correlated with the pain state of the animal (Dalla Costa et al., 2018).

Other Behavioral Indicators of Pain

Behavioral responses to pain involve characteristic postures and movements that act to alleviate pain. These behaviors are included in the CPS (Table 3-1). The CPS assessment tool was initially developed on a sample of horses with induced synovitis pain in the tarsocrural joint of the hock (Bussieres et al., 2008) and later clinically validated in horses presenting with acute orthopedic pain (van Loon and Van Dierendonck, 2019).

The CPS is a multifactorial scale that includes physiological measures, spontaneous behaviors, and evoked responses to stimuli (Table 3-1). Each of the 13 items is assigned a score from 0 to 3, giving a total pain score ranging from 0 (no signs of pain) to 39 (maximal pain score). During scale development in horses with orthopedic pain, researchers compared the CPS scores of each horse with and without anesthesia. Posture was found to have the greatest diagnostic sensitivity and specificity for pain, and the authors recommended it be included in a composite pain scale. The descriptions of postures indicative of orthopedic pain were “non-weight bearing positions and abnormal weight distribution” and “analgesic posture, prostration, muscle tremors.” Pawing the ground was also strongly associated with pain state, but its utility in HPA inspections may be limited because horses are often unable to paw or prevented from pawing the ground. The CPS item “interactive behavior” had high specificity but low sensitivity for pain, and “head movement” had low specificity. Consequently, the authors suggested that these items should not be included in a composite pain scale.

While spontaneous pain behaviors are prominently represented in the CPS, the scale also includes two “response to treatment” items—“interactive behavior” and “palpation of the painful area.” These items are of particular interest because of their direct relevance to the inspection procedure. “Response to palpation of the painful area” was previously discussed (see Nociceptive Withdrawal Reflex section above). “Interactive behavior” refers to a horse’s attention and behavior toward the environment. The CPS grades a horse that pays attention to people or shows an “exaggerated response to an auditory stimulus” as having little or no pain. Interestingly, the scale grades both a horse that overreacts to or shows an aggressive response to an auditory stimulus and a horse that does not respond to an auditory stimulus and appears to be in a “stupor” as experiencing a high degree of pain. This may be explained by the fact that an individual horse’s personality is linked to its expression of pain, which adds to the complexity of pain assessment (Ijichi et al., 2014).

The pros and cons of behavioral assessment scales discussed in this section are summarized in Table 3-5.

Finding 3-6: DQPs are directed to observe the horse for responses to pain during the inspection process in 9 C.F.R. § 11.21. Some information about behavioral indicators of pain appear in the APHIS training material for DQPs. However, the training material lists “abnormal reactions of the eye, ears, and head in response to palpation.” The term “abnormal” is unnecessarily vague, given that specific facial expressions indicative of pain have been described in clinical research literature.

Finding 3-7: Pain can be detected accurately and consistently when it is assessed using physical, physiological, and behavioral parameters that are based on validated clinical scales.

Finding 3-8: Clinical research in horses under veterinary care for laminitis and orthopedic injuries has confirmed that pain assessment using the withdrawal response to palpation is an accurate and reliable method for identifying pain, with very high agreement between raters.

TABLE 3-4 Score Sheet for the Equine Utrecht University Scale for Facial Assessment of Pain (EQUUS-FAP) Scale

SOURCE: van Loon and Van Dierendonck (2017, supplementary table S1).

TABLE 3-5 Behavioral Assessment Scales Basis, Pros, and Cons

Finding 3-9: Horse Protection Regulations do not include current information about equine pain behavior and its application to clinical practice. Facial grimace scales have long been used in human medicine to assess pain in infants and young children and are currently used in laboratory animal research and veterinary care to assess pain and welfare state.

Finding 3-10: Some horses displayed a facial grimace during standing inspection in the 61 videos provided to the committee. However, the videos also showed that various factors, such as dim lighting, a horse’s dark color, and an inspector’s body position and direction of gaze while palpating the limb, may prevent a single inspector from simultaneously palpating the forelimb and observing the horse’s facial expression.

Conclusion 3-3: A common set of objective criteria grounded in behavioral science, including facial expressions indicative of pain, is lacking from inspector training. Thus, an inspector’s interpretation of a horse’s behavior is subjective, but it can influence a determination of soreness.

Conclusion 3-4: Research is needed to determine the utility of assessing facial expression of pain in TWHs as part of the inspection procedure before use of facial expressions can be proposed as an additional method for detecting soreness. It is important to know if facial grimace can be reliably identified by different inspectors. It is also important to determine the extent to which the facial expressions of pain correspond to current evidence of soreness during inspections, such as withdrawal responses to digital palpation and findings of noncompliance with the scar rule criteria.

Conclusion 3-5: One practical limitation to including facial expressions to assess pain during digital palpation is the challenge an inspector might have of simultaneously observing the horse’s face and forelimb.

Conclusion 3-6: In clinical research, agreement between raters on horses’ responses to digital palpation is consistently high. While agreement may be lower when palpation is carried out in a horse show environment, differences between inspectors’ findings are more likely to result from inadequate training and

inconsistent application of technique than from the validity of the pain assessment procedure itself. Another factor might be conflict of interest, which the USDA OIG 2010 audit found was an influence on how DQPs conducted inspections.

PHYSIOLOGICAL ASSESSMENT OF PAIN

Physiological measurements have been used extensively in assessing pain in horses and humans, both in clinical practice and in applied research. The factors that are measured include, but are not limited to, heart rate and heart rate variability, respiratory rate, body temperature, ocular temperature, blood pressure, and various endogenous substances such as beta-endorphins, cortisol, serotonin, dopamine, substance P, and oxytocin. This section includes a discussion of physiological parameters that are used to assess or indicate pain as well as of the biomarkers and noninvasive techniques that have been explored for their utility in pain and stress assessment. To the committee’s knowledge, these parameters are not currently included in the TWH inspection process and may warrant further investigation for such a purpose.

Physiological Parameters as Indicators of Pain and Stress

The advantages of physiological values over other methods to assess pain are that they are objective, are noninvasive, and can be measured relatively easily and repeatably. Heart rate, respiratory rate, and temperature are routinely measured during a physical exam; endogenous substances can be measured from blood samples. Heart rate variability, ocular temperature, and blood pressure measurements require specialized equipment and are therefore not routinely measured during a physical exam, but they are frequently included as part of research on the physiology of pain and stress. The major disadvantages of these measurements are: (1) they have been shown to have low specificity for pain (Rietmann et al., 2004), (2) baseline measures may vary across individuals, and (3) they fluctuate greatly from measurement to measurement. The results of a laboratory analysis of blood, for example, can depend on the precise timing of the draw; this is the case for cortisol, for instance, which has a diurnal pattern. Furthermore, because blood samples are analyzed in independent labs, the results are not available immediately, and performing the test is an added expense. Finally, endocrine levels do not reliably or only weakly correlate with other measures of pain (Rietmann et al., 2004).

Most physiological measures do not discriminate between pain and other sources of autonomic arousal—in particular, stress. Stress responses and pain responses are both characterized by elevated heart rate, blood pressure, respiration, and body temperature (Rietmann et al., 2004) and by elevated ocular temperature. Moreover, physiological measures fail to distinguish or discriminate between arousal elicited by stimuli with negative valence and those with positive valence. For example, heart rate will increase with pain but also with exercise, excitement, stress, dehydration, hyperthermia, and certain medications. Thus, the horse show environment includes many triggers leading to physiological changes that mirror those seen in pain.

Biomarkers

Substance P (SP) is a neuropeptide active in pain perception that is actively being investigated as a potential biomarker for pain in animals, and recent research suggests that SP may increase in proportion to the amount of perceived distress. For example, in one study calves undergoing routine castration without the use of local anesthesia had 30 percent higher serum SP levels than calves undergoing sham castration, while there was no difference in serum cortisol levels between the two groups. Serum SP and

cortisol levels are used as a biomarker for the stress response (Coetzee et al., 2008). Interestingly, vocalization by calves during the procedure was significantly correlated with levels of SP but not with cortisol levels. In another study, serum SP was found to be higher in dogs with fractures or medial patella luxation than in healthy controls that underwent the same clinical procedures (Yoon et al., 2019). Furthermore, SP levels were significantly higher in those dogs with a fracture than in dogs with patella luxation, suggesting that SP may be sensitive to levels of perceived pain.

Noninvasive Techniques for Pain Assessment

Objective physiological assessment measures are commonly recorded in standardized pain assessment scales, such as the Composite Pain Scale for horses (Bussieres et al., 2008) (Table 3-1). However, physiological parameters alone have generally been found not to be valid for diagnosing orthopedic pain (Raekallio et al., 1997). For example, increases in noninvasive blood pressure (NIBP) are thought to be significantly correlated with behavioral pain scores, but NIBP recorded in standing horses tends to underestimate blood pressure, and the precision and accuracy of the NIBP measures are low, putting into question the utility of NIBP as a physiological indicator of pain in horses (Heliczer et al., 2016).

Another attractive noninvasive technique for measuring stress and pain is ocular infrared thermography, which measures temperature changes on the surface of the eye. Findings from a research study in calves (Coetzee et al., 2008) suggest that ocular thermography has the potential to discriminate between pain and distress. Calves undergoing castration showed increased eye temperature with stress and decreased eye temperature with pain. In horses, ocular thermography has been used to quantify stress during athletic performance and with the use of tight nosebands (Fenner et al., 2016; Cravana et al., 2017). However, the committee is not aware of any studies specifically differentiating pain from stress in horses, and this may warrant further research in TWHs.

A recent study explored the effect of stacked wedge pads and chains applied to the forefeet of TWHs on behavioral and biochemical indicators of pain. This study was conducted on 20 sedentary TWHs (10 horses shod with stacks and chains, 10 control horses that were flat shod) at the flat walk on a walker for a 5-day period, with the testing done after only a 5-day acclimation period (Everett et al., 2018). Considering the facts that none of these horses were actually sored and that the conditions of the study did not replicate the conditions under which the horses are shown (ridden running walk, with shoes and chains applied for an extended period of time [months to years]), it is not too surprising that no significant changes were found in any of the biochemical parameters evaluated (fibrinogen, SP, plasma cortisol).

Physiological predictors are often included in composite pain scales to bolster their validity and reliability; however, as previously mentioned, physiological parameters should not be used in isolation to detect pain. Instead, they should be integrated in a multimodal approach that includes observational and objective measures, visual inspection for signs of trauma and an antalgic stance, changes in facial expressions captured in composite pain scales (see section on Behavioral Assessment of Pain), palpation of limbs and other potential sensitive areas, and gait evaluation (see Chapter 2).

Finding 3-11: Physiological parameters (e.g., heart rate, respiratory rate, body temperature, and blood pressure) have been used extensively to assess pain in horses and humans. They are objective and can be measured easily and repeatably; however, they have low specificity for pain, vary across individuals, and fluctuate between measurements.

Finding 3-12: Most physiological measures do not discriminate between pain and other sources of autonomic arousal. Changes in physiological parameters, while indicative of pain, may also be due to physical exertion, excitement, stress, dehydration, hyperthermia, or certain medications.

Finding 3-13: Ocular thermography has been shown to discriminate between pain and distress in calves undergoing castration. It has also been used to quantify stress in horses during athletic performance and in horses that wear tight nosebands.

Conclusion 3-7: The show environment and other conditions during inspections may cause physiological changes in horses that mirror those seen in pain, thus limiting utility of physiological parameters to help detect if a horse is experiencing soreness.

Conclusion 3-8: Although often included as predictors in composite pain scales to bolster their validity and reliability, physiological parameters are not meant to be used in isolation to detect pain, but instead should be integrated with other measures in a multimodal approach.

Conclusion 3-9: The potential of ocular thermography to help differentiate between pain and stress in TWHs and its utility in detecting soreness warrant further investigation.

CLINICAL ASSESSMENT OF PAIN

Pain recognition in horses is complex and typically involves a multimodal approach including observational and objective measures, visual inspection for signs of trauma and an antalgic stance, changes in facial expressions captured in composite pain scales (see section on Behavioral Assessment of Pain in this chapter), physiological parameters (see section on Physiological Assessment of Pain in this chapter), and palpation of limbs and other potential sensitive areas and gait evaluation (see Chapter 2). Identifying pain in horses is not intuitive, particularly for those unfamiliar with normal breed-specific or individual behaviors (Taylor et al., 2002).

Horses notoriously hide pain well so as to mask weakness, as is the case with other prey animals as well. From an evolutionary standpoint, prey cannot afford to show potential predators that they are injured, as they are likely to draw attention to themselves and hence be attacked (Seksel, 2007; Allweiler, 2020). This tendency can make it difficult to reliably detect pain in horses. Complicating the issue even further is the existence of individual differences in pain tolerance, which have been demonstrated in people and animals and which play an important role in the identification and management of pain. For example, the TWH, praised for its stoic and docile nature, may have a higher pain tolerance than other horses (although that does not make it any less necessary that the horses get treated for whatever underlying conditions led to the pain). The result is that the identification and diagnosis of pain in horses—and in TWHs in particular—is challenging and, as pointed out in Chapter 2, requires extensive training, ideally by experienced equine veterinarians.

In determining the musculoskeletal health of horses—which is a major component of athletic soundness at a competition—it is crucial that one observe the horses’ pain behavior at rest and during exercise and also palpate for pain (Tabor et al., 2020). These actions are the basis for horse inspections at all official international equestrian competitions and are strictly regulated by the international equestrian governing body, the International Federation for Equestrian Sports (FEI). The FEI enforces the Code of Conduct for the Welfare of the Horse which is to “acknowledge and accept that at all times the welfare of the horse must be paramount. The welfare of the horse must never be subordinated to competitive or commercial influences” (FEI, 2020). The FEI Limb Sensitivity Testing Procedure is discussed in Box 2-2 in Chapter 2.

Visual Inspection for Signs of Pain

It is important to remember that general pain behavior in the horse is influenced by temperament, age, sex, breed, and environment (de Grauw and van Loon, 2016). The fact that environment influences

pain behavior makes remote observation via video recordings ideal, but this is not possible at horse shows. Interactions with handlers, spectators, and other horses and simply being in the foreign environment of an equestrian competition will all alter a horse’s behavior and potentially mask signs of pain. A visual inspection for signs of pain should include an assessment of general demeanor and posture. Signs of pain are nonspecific and may include (but are not limited to) excessive quietness or restlessness, low head carriage, weight shifting, pointing a front limb or resting a hind limb, standing hunched over or camped out, and looking at a painful area. Other signs may include bruxism (grinding of teeth), sweating, and muscle fasciculations or brief spontaneous muscle contractions (Dalla Costa et al., 2014; Gleerup et al., 2015). A horse sore in front will rarely rest a hind limb but will instead bear more weight on its hindquarters to relieve pain. Unwillingness to bear weight on a hind limb is indicative of lameness, while resting a hind limb may be attributed to other causes.

At all FEI-sanctioned events, regulatory veterinarians perform a clinical examination to assess each horse’s fitness and aptitude to compete without pain. This is determined by careful clinical observation, which may include measuring heart rate, respiratory rate, and temperature as well as the palpation of any areas considered injured or painful, based on the possible presence of swelling, redness, loss of hair/skin, or the presence of blood; palpation for hyper- and hyposensitivity of the limbs; evaluation of pain in the feet using hoof testers; passive flexion of the distal limb joints to assess the range of motion of the joint(s); and walking and trotting the horse in a straight line or a circle.

Pressure Algometry

Pressure algometry, a technique that involves administering consistent pressure to an area, is used in scientific experiments to increase the consistency and repeatability of pressure applied during palpation and has been proposed for testing horses at competitions for either hypo- or hypersensitivity. Pressure algometry has already been used to determine mechanical nociceptive thresholds (MNTs) in horses (Haussler and Erb, 2006; Haussler et al., 2008; Love et al., 2011; Schambourg and Taylor, 2020). The MNT is defined as the pressure it takes to elicit a withdrawal response by an individual. The higher it is, the more pressure the individual can tolerate at a specific site before showing a reaction. To prove repeatability, pressure is applied three consecutive times (Haussler and Erb, 2006). However, as pointed out in Chapter 2, prolonged stimulation or pressure on a painful area can produce analgesia through the secretion of local endorphins, gate control (inhibition of presynaptic nociceptive spinal neurons), and hyperstimulation analgesia (activation of descending inhibitory systems) (Melzack, 1975), which complicates pain identification. A recent study used pressure algometry to determine MNTs in pasterns of TWHs that were not sore (Haussler et al., 2008). This study found that TWHs that were not sore responded with a withdrawal reflex only to pressures greater than 10 kg/cm² (this is 10 times greater than the pressure needed to blanch the thumbnail, which is the pressure that APHIS VMOs are told to apply when palpating horses during inspections at TWH shows). This investigation also revealed that anxious TWHs did not have different mechanical nociceptive thresholds than calmer ones, which is an important factor when considering palpation at show grounds, which are foreign environments that could conceivably cause a horse to be more nervous than usual. This suggests that TWHs that were not sore tolerate a high level of pressure in their pastern region prior to responding, regardless of whether they are nervous, and that, in particular, they tolerate a much higher pressure than would be produced with palpation using a thumb. Similar work has not been done in sore TWHs but it would be expected that MNTs in sore TWHs would be well below 10 kg/cm², which could be used as an objective cutoff during inspections should pressure algometers be used. However, recently the direct digital palpation of epaxial muscles of horses by three experienced individuals was deemed superior to palpation with an algometer in terms of the repeatability of the painful response (Merrifield-Jones et al., 2019). Once again this shows the importance of familiarity and training for an adequate interpretation of the results of palpation.

Gait Analysis—Kinematics, Kinetics

Another key factor in determining a horse’s fitness to compete safely is the confirmation of the absence of lameness, or pain causing an irregular gait (Adams, 2015). In most official equestrian competitions, including racing, this is done by careful inspection of the horse at trot in a straight line, on a loose lead, and in hand and by observing for asymmetric head, limb, and pelvic movements. Veterinarians use subjective lameness grades, most commonly the five-point American Association of Equine Practitioners (AAEP) lameness scale, to grade the lameness. Any horse showing consistent lameness at the trot (grade 3 AAEP lameness) is excluded from competition. However, bilateral lameness may confound the ability to detect asymmetry, and therefore in the research and clinical setting, more sophisticated biomechanical analysis is used predominately in order to increase the sensitivity of the detection of lameness. The added challenge in assessing TWHs for lameness is that they are gaited, and usually do not trot, which requires additional expertise to visually evaluate their gait for lameness.

Kinetic analysis (related to forces acting on the body) combined with kinematic analysis (related to the movement of the body) is considered the gold standard approach to lameness diagnosis. Various commercial systems combining inertial sensors, high-speed video analysis, accelerometers, and in-ground force plates measuring ground reaction forces have been developed to aid gait analysis in sport horses at various gaits (walk, trot, canter, gallop) and movements (jumping, piaffe, passage) (Roepstorff et al., 2009; Rhodin et al., 2017; Hardeman et al., 2019). However, to the committee’s knowledge, only few kinematic (Nicodemus et al., 2002) and no kinetic studies have been conducted in TWHs and information about such studies and the characteristic gait of the TWH is lacking in the scientific literature. Additionally, TWHs are only assessed briefly for irregular gait at the flat walk and not at the running walk, which decreases the ability to detect lameness in this breed.

Finding 3-14: Pressure algometry has been used to determine pain thresholds in TWHs that are not sore. A study³ has shown that TWHs that were not sore responded with a withdrawal reflex only to pressures greater than 10 kg/cm² (10 times greater than the pressure needed to blanch the thumbnail, which is what APHIS VMOs are prescribed to apply when palpating horses during inspections at TWH shows).

Finding 3-15: There is a lack of kinetic and kinematic research studies in TWHs that are needed to establish gait characteristics of TWHs that are and are not sore.

Conclusion 3-10: The absence of studies to differentiate pain from stress in TWHs indicates a need for further research.

Conclusion 3-11: Further research is needed on using pressure algometry in TWHs with sore limbs. Kinetic and kinematic research in normal TWHs and those with sore limbs is also needed to establish gait characteristics in this breed.

RECOMMENDATIONS

Recommendation 3-1: Designating an inspection area that has as few distractions as possible will reduce the effect of the environment on the horse’s response to pain during examination. It is important that inspectors observe the horse’s response to the show environment and to restraint before starting the inspection and consider the horse’s behavior in the decision-making process.

___________________

³ Haussler, K. K., T. H. Behre, and A. E. Hill. 2008. Mechanical nociceptive thresholds within the pastern region of Tennessee walking horses. Equine Veterinary Journal 40(5):455–459.

Recommendation 3-2: To help improve accuracy of soreness detection, the horse inspector should ensure that custodians are following guidelines that prohibit stewarding while the horse is being inspected, and should closely monitor horse custodians for violations.

Recommendation 3-3: Pain assessment using facial expressions is a new area of research, and scientific investigations of these methods have not been performed in TWHs. However, evidence supports the use of facial expressions of pain as supplemental information, if video is available to review or if a second inspector is present.

Recommendation 3-4: To improve consistency across inspectors, science-based information about behavioral and facial indicators of pain in horses should be incorporated into inspectors’ training.

Recommendation 3-5: Research is needed to study validity and potential utility of using facial grimace for assessing pain in TWHs and to distinguish pain from other sources of distress. To accomplish this, researchers could, under show conditions, apply new clinical pain assessment technologies and score the horse’s behavior and facial expressions during the inspection. Facial expressions of pain are expected to correlate with findings from other currently used methods to detect soreness, such as palpation. For this purpose, it is important to capture the horse’s head in the inspection videos.

Recommendation 3-6: The decision to disqualify a horse due to soreness should be driven by an experienced veterinarian, such as a VMO, and should be based on diagnosis of local pain detected on palpation but should also include a more thorough gait or lameness assessment to identify other sources of pain. Signs of pain that should be observed include excessive quietness or restlessness, low head carriage, weight shifting, pointing a front limb or resting a hind limb, standing hunched over or camped out and looking at a painful area, bruxism, sweating, and muscle fasciculations.

REFERENCES

Dalla Costa, E., D. Stucke, F. Dai, M. Minero, M. C. Leach, and D. Lebelt. 2016. Using the Horse Grimace Scale (HGS) to assess pain associated with acute laminitis in horses (Equus caballus). Animals (Basel) 6(8):47.

Dalla Costa, E., D. Bracci, F. Dai, D/Lebelt, and M. Minero. 2017. Do different emotional states affect the Horse Grimace Scale score? A pilot study. Journal of Equine Veterinary Science 54:114–117.

Dalla Costa, E., R. Pascuzzo, M. C. Leach, F. Dai, D. Lebelt, S. Vantini, and M. Minero. 2018. Can grimace scales estimate the pain status in horses and mice? A statistical approach to identify a classifier. PLoS ONE 13(8):e0200339.

De Grauw, J. C., and J. P. A. M. van Loon. 2016. Systematic pain assessment in horses. Veterinary Journal 209:14–22.

Dyson, S., J. Berger, A. D. Ellis, and J. Mullard. 2018. Development of an ethogram for a pain scoring system in ridden horses and its application to determine the presence of musculoskeletal pain. Journal of Veterinary Behavior 23:47–57.

Ekman, P., and W. Friesen. 1978. Facial action coding system: A technique for the measurement of facial movement. Palo Alto, CA: Consulting Psychologists Press.

Everett, J. B., J. Schumacher, T. J. Doherty, R. A. Black, L. L. Amelse, P. Krawczel, J. F. Coetzee, and B. K. Whitlock. 2018. Effects of stacked wedge pads and chains applied to the forefeet of Tennessee walking horses for a five-day period on behavioral and biochemical indicators of pain, stress, and inflammation. American Journal of Veterinary Research 79(1):21–32.

FEI (International Federation for Equestrian Sports). 2020. Veterinary regulations 2020. https://inside.fei.org/fei/regulations/veterinary (accessed July 31, 2020).

Fenner, K., S. Yoon, P. White, M. Starling, and P. McGreevy. 2016. The effect of noseband tightening on horses’ behavior, eye temperature, and cardiac responses. PLoS ONE 11(5):e0154179.

Garra, G., A. J. Singer, A. Domingo, and H. C. Thode. 2013. The Wong-Baker Pain FACES Scale measures pain, not fear. Pediatric Emergency Care 29(1):17–20.

Gleerup, K. B., B. Forkman, C. Lindegaard, and P. H. Andersen. 2015. An equine pain face. Veterinary Anaesthesia and Analgesia 42(1):103–114.

Gleerup, K. B., P. H. Andersen, and J. Wathan. 2018. What information might be in the facial expressions of ridden horses? Adaptation of behavioral research methodologies in a new field. Journal of Veterinary Behavior 23:101–103.

Hall, C., D. Goodwin, C. Heleski, H. Randle, and N. Waran. 2008. Is there evidence of ‘learned helplessness’ in horses? Applied Animal Behaviour Science 11(3):249–266.

Hardeman, A. M., F. M. Serra Braganca, J. H. Swagemakers, P. R. van Weeren, and L. Roepstorff. 2019. Variation in gait parameters used for objective lameness assessment in sound horses at the trot on the straight line and the lunge. Equine Veterinary Journal 51(6):831–839.

Haussler, K. K., and H. N. Erb. 2006. Mechanical nociceptive thresholds in the axial skeleton of horses. Equine Veterinary Journal 38(1):70–75.

Haussler, K. K., T. H. Behre, and A. E. Hill. 2008. Mechanical nociceptive thresholds within the pastern region of Tennessee walking horses. Equine Veterinary Journal 40(5):455–459.

Heliczer, N., O. Lorello, D. Casoni, and C. Navas de Solis. 2016. Accuracy and precision of noninvasive blood pressure in normo-, hyper-, and hypotensive standing and anesthetized adult horses. Journal of Veterinary Internal Medicine 30(3):866–872.

Hoegh, M., D. A. Seminowicz, and T. Graven-Nielsen. 2019. Delayed effects of attention on pain sensitivity and conditioned pain modulation. European Journal of Pain 23(10):1850–1862.

Ijichi, C., L. M. Collins, and R. W. Elwood. 2014. Pain expression is linked to personality in horses. Applied Animal Behaviour Science 152:38–43.

Ijichi, C., K. Griffin, K. Squibb, and R. Favier. 2018. Stranger danger? An investigation into the influence of human-horse bond on stress and behaviour. Applied Animal Behaviour Science 206:59–63.

Jure, F. A., F. G. Arguissain, J. A. B. Manresa, and O. K. Andersen. 2019. Conditioned pain modulation affects the withdrawal reflex pattern to nociceptive stimulation in humans. Neuroscience 408:259–271.

Kappesser, J., E. Kamper-Fuhrmann, J. de Laffolie, D. Faas, H. Ehrhardt, L. S. Franck, and C. Hermann. 2019. Pain-specific reactions or indicators of a general stress response. Clinical Journal of Pain 35(2):101–110.

Keating, S. C., A. A.Thomas, P. A., Flecknell, and M. C. Leach. 2012. Evaluation of EMLA cream for preventing pain during tattooing of rabbits: Changes in physiological, behavioural and facial expression responses. PLoS ONE 7(9):e44437.

Kennedy, D. L., H. I. Kemp, D. Ridout, D. Yarnitsky, and A. S. Rice. 2016. Reliability of conditioned pain modulation: A systematic review. Pain 157(11):2410–2419.

Koolhaas, J. M., and C. G. Van Reenen. 2016. Interaction between coping style/personality, stress, and welfare: Relevance for domestic farm animals. Journal of Animal Science 94:2284–2296.

Koolhaas, J. M., S. M. Korte, S. F. De Boer, B. J. Van Der Vegt, C. G. Van Reenen, H. Hopster, I. C. De Jong, M. A. W. Ruis, and H. J. Blokhuis. 1999. Coping styles in animals: Current status in behavior and stress-physiology. Neuroscience & Biobehavioral Reviews 23(7):925–935.

Langford, D. J., A. L. Bailey, M. L. Chanda, S. E. Clarke, T. E. Drummond, S. Echols, S. Glick, J. Ingrao, T. Klassen-Ross, M. L. LaCroix-Fralish, and L. Matsumiya. 2010. Coding of facial expressions of pain in the laboratory mouse. Nature Methods 7(6):447–449.

Latremoliere, A., and C. J. Woolf. 2009. Central sensitization: A generator of pain hypersensitivity by central neural plasticity. Journal of Pain 10(9):895–926.

Levin, K. H., and P. Chauvel. 2019. Clinical neurophysiology of neuromuscular junction disease. In K. Levin and P. Chauvel (eds.), Clinical neurophysiology: Diseases and disorders. New York: Elsevier. Pp. 291–303.

Lima, L. V., T. S. Abner, and K. A. Sluka. 2017. Does exercise increase or decrease pain? Central mechanisms underlying these two phenomena. Journal of Physiology 595:4141–4150.

Lloyd, A. S., J. E. Martin, H. L. I. Bornett-Gauci, and R. G. Wilkinson. 2008. Horse personality: Variation between breeds. Applied Animal Behaviour Science 112(3–4):369–383.

Love, E. J., J. Murrell, and H. R. Whay. 2011. Thermal and mechanical nociceptive threshold testing in horses: A review. Veterinary Anaesthesia and Analgesia 38(1):3–14.

Luna, S. P., C. Lopes, A. C. Rosa, F. A. Oliveira, N. Crosignani, P. M. Taylor, and J. C. Pantoja. 2015. Validation of mechanical, electrical and thermal nociceptive stimulation methods in horses. Equine Veterinary Journal 47(5):609–614.

Maier, S. F., and M. E. Seligman. 1976. Learned helplessness: Theory and evidence. Journal of Experimental Psychology: General 105(1):3–46.

McGreevy, P. D., and A. N. McLean. 2009. Punishment in horse-training and the concept of ethical equitation. Journal of Veterinary Behavior: Clinical Applications and Research 4(5):193–197.

Meier, A., M. de Laat, C. Pollitt, D. Walsh, J. McGree, D. B. Reiche, M. von Salis-Soglio, L. Wells-Smith, U. Mengeler, D. M. Salas, and S. Droegemueller, S. 2019. A “modified Obel” method for the severity scoring of (endocrinopathic) equine laminitis. PeerJ 7:e7084.

Melzack, R. 1975. Prolonged relief of pain by brief, intense transcutaneous somatic stimulation. Pain 1:357–373.

Merrifield-Jones, M., G. Tabor, and J. Williams. 2019. Inter-and intra-rater reliability of soft tissue palpation scoring in the equine thoracic epaxial region. Journal of Equine Veterinary Science 83:102812.

Nicodemus, M. C., K. M. Holt, and K. Swartz. 2002. Relationship between velocity and temporal variables of the flat shod running walk. Equine Veterinary Journal 34(S34):340-343.

Obel, N. 1948. Studies on the histopathology of acute laminitis. [Dissertation]. Uppsala, Sweden: Almqvist & Wiksells Boltryckeri, AB.

Price, J., J. M. Marques, E. M. Welsh, and N. K. Waran. 2002. Pilot epidemiological study of attitudes towards pain in horses. Veterinary Record 151(19):570–575.

Raekallio, M., P. M. Taylor, and R. C. Bennett. 1997. Preliminary investigations of pain and analgesia assessment in horses administered phenylbutazone or placebo after arthroscopic surgery. Veterinary Surgery 26(2):150–155.

Reid, J., A. M. Nolan, and E. M. Scott. 2018. Measuring pain in dogs and cats using structured behavioural observation. Veterinary Journal 236:72–79.

Reid, K., C. W. Rogers, G. Gronqvist, E. K. Gee, and C. F. Bolwell. 2017. Anxiety and pain in horses measured by heart rate variability and behavior. Journal of Veterinary Behavior 22: 1–6.

Rhodin, M., A. Egenvall, P. Haubro Andersen, and T. Pfau. 2017. Head and pelvic movement asymmetries at trot in riding horses in training and perceived as free from lameness by the owner. PLoS ONE 12(4):e0176253.

Riddell, R. P., N. Racine, and B. J. Stevens. 2013. Acute pain management in infants. In G. F. Gebhart and R. F. Schmidt (eds.), Encyclopedia of pain. Berlin, Heidelberg: Springer.

Rietmann, T. R., M. Stauffacher, P. Bernasconi, J. A. Auer, and M. A. Weishaupt. 2004. The association between heart rate, heart rate variability, endocrine and behavioural pain measures in horses suffering from laminitis. Journal of Veterinary Medicine Series A 51(5):218–225.

Roepstorff, L., A. Egenvall, M. Rhodin, A. Byström, C. Johnston, P. R. Van Weeren, and M. Weishaupt. 2009. Kinetics and kinematics of the horse comparing left and right rising trot. Equine Veterinary Journal 41(3):292–296.

Sator-Katzenschlager, S. 2014. Pain and neuroplasticity. Revista Médica Clínica Las Condes 25(4):699–706.

Schambourg, M., and P. M. Taylor. 2020. Mechanical nociceptive thresholds in endurance horses. Veterinary Record 186(4):124.

Seksel, K., 2007. How pain affects animals. In Australian Animal Welfare Strategy Science Summit on Pain and Pain Management Proceedings, May 2007. http://www.huisdiergedrag.be/docs/Seksel,%202007.pdf (accessed September 20, 2020).

Seligman, M. E., and G. Beagley. 1975. Learned helplessness in the rat. Journal of Comparative and Physiological Psychology 88(2):534–541.

Seligman, M. E., and S. F. Maier. 1967. Failure to escape traumatic shock. Journal of Experimental Psychology 74(1): 1–9.

Seligman, M. E., R. A. Rosellini, and M. J. Kozak. 1975. Learned helplessness in the rat: Time course, immunization, and reversibility. Journal of Comparative and Physiological Psychology 88(2):542–547.

Sotocinal, S. G., R. E. Sorge, A. Zaloum, A. H. Tuttle, L. J. Martin, J. S. Wieskopf, J. C. Mapplebeck, P. Wei, S. Zhan, S. Zhang, and J. J. McDougall. 2011. The Rat Grimace Scale: A partially automated method for quantifying pain in the laboratory rat via facial expressions. Molecular Pain 7:1744–8069.

Squibb, K., K. Griffin, R. Favier, and C. Ijichi. 2018. Poker face: Discrepancies in behaviour and affective states in horses during stressful handling procedures. Applied Animal Behaviour Science 202:34–38.

Tabor, G., K. Nankervis, J. Fernandes, and J. Williams. 2020. Generation of domains for the equine musculoskeletal rehabilitation outcome score: Development by expert consensus. Animals (Basel) 10(2):203.

Taylor, P. M., P. J. Pascoe, and K. R. Mama. 2002. Diagnosing and treating pain in the horse. Where are we today? Veterinary Clinics of North America. Equine Practice 18(1):1–19.

Torcivia, C., and S. McDonnell. 2020. In-person caretaker visits disrupt ongoing discomfort behavior in hospitalized equine orthopedic surgical patients. Animals 10(2):210.

TWHBEA (Tennessee Walking Horse Breeders’ and Exhibitors’ Association). 2020. History of the Tennessee walking horse. https://www.twhbea.com/breed-general-info/ (accessed June 18, 2020).

van Loon, J. P., and M. C. Van Dierendonck. 2015. Monitoring acute equine visceral pain with the Equine Utrecht University Scale for Composite Pain Assessment (EQUUS-COMPASS) and the Equine Utrecht University Scale for Facial Assessment of Pain (EQUUS-FAP): A scale-construction study. Veterinary Journal 206(3):356–364.

van Loon, J. P. A. M., and M. C. Van Dierendonck. 2017. Monitoring equine head-related pain with the Equine Utrecht University scale for facial assessment of pain (EQUUS-FAP). Veterinary Journal 220:88–90.

van Loon, J. P. A. M., and M. C. Van Dierendonck. 2018. Objective pain assessment in horses (2014–2018). Veterinary Journal 242:1-7.

van Loon, J. P., and M. C. Van Dierendonck. 2019. Pain assessment in horses after orthopaedic surgery and with orthopaedic trauma. Veterinary Journal 246:85–91.

Van Dierendonck, M. C., and J. P. van Loon. 2016. Monitoring acute equine visceral pain with the Equine Utrecht University Scale for Composite Pain Assessment (EQUUS-COMPASS) and the equine Utrecht University Scale for Facial Assessment of Pain (EQUUS-FAP): A validation study. Veterinary Journal 216:175–177.

Waran, N., P. D. McGreevy, and R. A. Casey (eds.). 2002. Training Methods and Horse Welfare. Dordrecht: Kluwer Academic Publishers.

Wathan, J., A. M. Burrows, B. M. Waller, and K. McComb. 2015. EquiFACS: The Equine Facial Action Coding System. PLoS ONE 10(8):e0131738.

Woolf, C. J. 1989. Recent advances in the pathophysiology of acute pain. British Journal of Anaesthesia 63(2):139–146.

Yoon, J. S., J. Park, R. Song, and D. Yu. 2019. Substance P as a potential biomarker of pain assessment in dogs. Iranian Journal of Veterinary Research 20(4):289–292.