Diagnostic Criteria for Equine Thoracolumbar Myofascial Pain Syndrome: A Foundational Study

Diagnostic Criteria for Equine Thoracolumbar Myofascial Pain Syndrome: A Foundational Study | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 14 December 2025 V1 Latest version Share on Diagnostic Criteria for Equine Thoracolumbar Myofascial Pain Syndrome: A Foundational Study Authors : María Resano-Zuazu , Jorge U. Carmona , César Fernández-de-las-Peñas , and david arguelles [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.176570292.22857795/v1 377 views 229 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Background: Diagnostic criteria for myofascial pain syndrome (MPS) are well stablished in humans but under-investigated in horses. Objectives: To identify which MPS clinical signs are present in the equine thoracolumbar region, to determine their prevalence, to assess internal coherence among clinical indicators and to evaluate diagnostic relevance of behavioural, demographic, and clinical variables. Study design: In vivo experiments. Methods: A population of 120 horses was included with at least one trigger point (TrP) on the thoracolumbar region. Owner-reported horse behaviours were obtained through a questionnaire. Taut band, hypersensitive spot, local twitch response (LTR), jump sign, restricted range of motion (ROM), pain score (0-5), and ROM score (0-4) were assessed by manual palpation. Machine-learning (ML) regression models examined multivariate predictive patterns. Results: Taut band and hypersensitive spot were present in all horses, but LTR was absent. Jump sign was identified in 71.7% and restricted ROM in 69.2% of horses. Median pain score was 4 (IQR 3-4). Jump sign correlated with pain score (ρ=0.814, p<0.001) and restricted ROM with ROM score (ρ=0.749, p<0.001). Owner-reported behaviours showed a weak relationship with pain on palpation (ρ=0.218, p=0.017). Stabled horses demonstrated higher pain scores (p=0.044) and greater ROM restriction (p=0.013) than horses kept outdoors. ML identified the jump sign as the dominant pain predictor (>90% importance). Main limitations: The clinical assessment relied on a single experienced veterinarian, the specific facial and body pain-related behavioural responses were not described. Conclusions: Taut band, hypersensitive spot, and the jump sign emerged as essential diagnostic criteria for equine thoracolumbar MPS. ML models highlighted the jump sign as a robust clinical marker. This study provides the first evidence-based diagnostic framework and underscores improved owner recognition of horse pain. ††journal: The Astrophysical Journal \crefname sectionSectionSections \crefnameequationEq.Eqs. \crefnamefigureFig.Figs. \crefnametableTableTables Original Article Diagnostic Criteria for Equine Thoracolumbar Myofascial Pain Syndrome: A Foundational Study Summary Background: Diagnostic criteria for myofascial pain syndrome (MPS) are well stablished in humans but under-investigated in horses. Objectives: To identify which MPS clinical signs are present in the equine thoracolumbar region, to determine their prevalence, to assess internal coherence among clinical indicators and to evaluate diagnostic relevance of behavioural, demographic, and clinical variables. Study design: In vivo experiments. Methods: A population of 120 horses was included with at least one trigger point (TrP) on the thoracolumbar region. Owner-reported horse behaviours were obtained through a questionnaire. Taut band, hypersensitive spot, local twitch response (LTR), jump sign, restricted range of motion (ROM), pain score (0-5), and ROM score (0-4) were assessed by manual palpation. Machine-learning (ML) regression models examined multivariate predictive patterns. Results: Taut band and hypersensitive spot were present in all horses, but LTR was absent. Jump sign was identified in 71.7% and restricted ROM in 69.2% of horses. Median pain score was 4 (IQR 3-4). Jump sign correlated with pain score (ρ=0.814, p<0.001) and restricted ROM with ROM score (ρ=0.749, p<0.001). Owner-reported behaviours showed a weak relationship with pain on palpation (ρ=0.218, p=0.017). Stabled horses demonstrated higher pain scores (p=0.044) and greater ROM restriction (p=0.013) than horses kept outdoors. ML identified the jump sign as the dominant pain predictor (>90% importance). Main limitations: The clinical assessment relied on a single experienced veterinarian, the specific facial and body pain-related behavioural responses were not described. Conclusions: Taut band, hypersensitive spot, and the jump sign emerged as essential diagnostic criteria for equine thoracolumbar MPS. ML models highlighted the jump sign as a robust clinical marker. This study provides the first evidence-based diagnostic framework and underscores improved owner recognition of horse pain. Keywords back pain; horse; artificial intelligence; muscle; palpation; trigger points 1. Introduction Musculoskeletal disorders are a common cause of poor performance in horses. 1,2 Back pain is highly prevalent, affecting up to 94% of ridden horses, 3 and can lead to chronic pain, reduced performance, and impaired work capacity, representing a major clinical concern for veterinarians. 4,5 The association between back pain and lameness is well recognised, 1,6,7 with 74% of horses with back pain presenting with lameness and back problems identified in 32% of lame horses. 6 Thoracolumbar pain, particularly in the caudal thoracic region, is the most frequently observed. 8 Primary causes of thoracolumbar pain in horses include impinging dorsal spinous processes, ventral spondylosis, articular process osteoarthritis, intervertebral disc disease, nerve impingement, supraspinous/intraspinous desmopathy, vertebral fractures, sacroiliac pathology, conformational abnormalities, and epaxial muscle strain. 5,9,10 However, myofascial pain syndrome (MPS) remains unrecognised in the core veterinary literature as a distinct primary cause. MPS is characterised by myofascial trigger points (TrPs), defined as hyperirritable spots within skeletal muscle, associated with palpable nodules located in taut bands, and producing pain, referred pain, motor and autonomic disturbances upon compression. 11,12 In humans, TrPs are present in several muscles, e.g., quadratum lumborum or gluteus medium in up to 55% of individuals with low back pain. 13 MPS may be primary or secondary, 14 and reciprocal relationships exist between spinal pathology and TrPs. 15 Literature on MPS and its treatment in veterinary medicine remains limited, 16–19 although back problem syndromes involving pain, epaxial hypertonicity, intervertebral stiffness and poor performance have been described in horses. 17 A characteristic pattern of TrPs has been reported in horses with sore backs, 16 with paravertebral extensor muscles frequently affected in response to overuse or stress. 18,19 Recent findings indicate a high prevalence of TrPs in the equine longissimus thoracis muscle, reaching 71% in dressage and 86% in show jumpers. 20 MPS should be recognised as a specific clinical diagnosis rather than a general term for soft tissue pain. 21 In humans, diagnosis relies mainly in manual palpation, 22 and requires at least the presence of one active TrP. 23 Classical criteria include tenderness in a taut band, pain recognition, pain referral pattern, local twitch response (LTR), limited range of motion (ROM), jump sign and muscle weakness without atrophy. 21 Later analyses identified that a restricted group of criteria was considered as clinically relevant, 24 although the reliability varied considerably among muscles. 25–28 In 2018, a Delphi panel proposed TrP diagnosis requires at least two of the following criteria: a taut band, a hypersensitive spot and referred pain. 29 An updated systematic review highlighted the combination of tender spot, referred pain, and LTR as the most commonly diagnostic criteria used in clinical trials of MPS. 22 All these diagnostic frameworks, extensively studied in humans, have not been thoroughly evaluated in equines. Accordingly, the aims of this study were to: (1) identify which clinical signs of MPS are present in the equine thoracolumbar region; (2) determine the prevalence of MPS within a horse population; (3) assess the internal coherence among clinical features, including the relationship between pain on palpation and thoracolumbar ROM; and (4) evaluate the diagnostic relevance and predictive value of behavioural, demographic, and clinical variables using statistical, non-parametric analyses, and exploratory machine-learning (ML) regression models. 2. Materials and Methods 2.1 Horse population A convenience sample of 120 equines from several equestrian centres and private properties underwent clinical assessment of the thoracolumbar region as part of their routine veterinary consultation. Informed consent form was signed by each owner or person responsible for the animal. The study population included mares, geldings, and stallions of various breeds and disciplines. They weighed between 194 and 648 kg and ranged in age from 4 to 24 years. All animals were housed under varying conditions depending on the management routines and facilities of their respective centres. Their athletic level ranged from leisure to high-performance competition. Horses over 4 years of age, engaged in routine veterinary work, with at least one TrP in the thoracolumbar area, were included. Animals were excluded if they had received analgesic or anti-inflammatory drugs within the previous 48 hours, had a diagnosis of bone, joint, or ligament pathology, had a diagnosis of neurological disorders, were in poor general health or had an inadequate physical condition, were pregnant mares, displayed aggressive behaviour, or had undergone physiotherapeutic treatments within the last week. 2.2 Assessment of MPS 2.2.1 Questionnaire All assessments and data collection were performed by a veterinarian specialized in equine rehabilitation (M.R.Z). Owners were questioned by the veterinarian using a binary (yes/no) format to determine whether they believed their horse exhibited abnormal behaviour in the box, during saddling, girthing, and riding. 30 (Data S1). In cases where the response was affirmative, owners were asked to specify the nature of the observed behaviour and to indicate its frequency, categorized as rarely, sometimes, often, or always. The veterinarian provided a standardized list of pain-related behaviours associated with each phase to facilitate accurate reporting. The initial binary question was asked first to ensure that the response of the owner was not influenced by the suggested list. 2.2.2 Manual palpation procedure Data concerning sex, age, breed, weight, housing, and discipline were recorded for each animal. All horses were examined in a familiar environment, and their coat was dry. Palpation was performed outside the box, in a square position and on a firm and flat surface. 31 During the clinical assessment, each animal was fitted with a halter and was either held by the owner using the lead rope or tied. In both cases, the lead rope was kept loose, 31 to allow the horse to express behavioural reactions. The clinician remained in a close and safe position near the animal. Each horse was examined for approximately 15 minutes. The epaxial muscles were assessed on both sides, starting with the left and then the right side. Palpation was performed starting at the base of the withers in a caudal direction toward the lumbosacral joint. 32 A preliminary palpation was performed with the flat of the hand and relaxed fingers, 33 gently gliding once over the muscle on each side. This brief initial contact allowed the horse to become accustomed to clinician contact and promoted comfort during assessment. Palpation was performed with the intention of confirming the following myofascial TrP clinical signs: (1) a palpable taut band, identified when a taut cord-like band could be found during palpation; (2) a hypersensitive spot, identified when the animal showed pain during palpation; (3) a LTR, which is a transient contraction of the palpable taut band visualized or palpated through the skin; and (4) a jump sign, which is a characteristic behavioural response to pressure on a TrP and is identified when the animal withdraws from digital pressure. 34,35 Referred pain and reproduction of the usual pain were not evaluated due to the non-verbal nature of the animals. Findings were graded and recorded on a present/absent basis. 36 To identify a taut band, palpation was performed using a light, firm touch, 31 by drawing the fingertips of the examining hands forward and back, perpendicular to the muscle fibres. 37,38 The taut band was felt to be firmer on palpation than the surrounding muscle. 39 Once the taut band was identified within the muscle, direct pressure was applied to each band to confirm the presence of a TrP. 40 The TrP was identified as the most tender spot, indicated by the animal moving away from the applied pressure or showing other signs of discomfort. 39 (Figure 1). Pressure was applied to the muscle for at least two seconds. 41 The painful response was observed by the clinician and scored 0-5 according to the muscular tone and local and behavioural responses: soft and low tone (0), normal (1), increased muscle tone but not painful (2), increased muscle tone and/or painful (3), painful (4) and very painful (5). 42 (Table S1). Spasms associated with palpation were considered positive if they lasted at least two seconds under the area of pressure or were appreciated in some distant area away from the pressure applied. 31 The jump sign was considered positive when a pain grade of 4 or higher was observed. 2.2.3 Passive movement tests Mobility tests were performed to assess ROM, the last clinical sign evaluated for MPS. They included flexion, extension and lateral flexion with rotation. 43 Dorsal flexion was performed with digital stimulation along the ventral midline over the sternum or cranial portion of the linea alba to induce elevation of the cranial thoracic region. 44,45 Application of bilateral digital pressure in the form of tickling at the base of the withers and at T16-T18 was performed for dorsal extension. 43 Lateral flexion and rotation involved gently pulling the base of the tail towards the clinician while simultaneously applying pressure against the body of the animal at the level of the last ribs. 43 Mobility was scored 0-4: hypermobility (0), normal (1), mild (2), moderate (3), and severe (4). 46 (Table S2). 2.3 Statistical analysis All statistical analyses were performed using IBM SPSS Statistics v31 (IBM Corp., Armonk, NY, USA) and JASP (version 0.95. (Intel) (University of Amsterdam & others, Netherlands)). The analytical plan consisted of three components: (1) descriptive statistics, (2) classical inferential tests, and (3) exploratory multivariate modelling using ML regression. Categorical variables (sex, age category, breed, weight category, housing system, discipline, behavioural indicators and dichotomized clinical signs) were summarized as counts and percentages. Ordinal clinical variables (pain score and ROM score) were summarized as medians with interquartile ranges (IQR). Because two indicators - hypersensitive spots and taut bands - were present in 100% of the horses, and LTRs were absent in 100% of them, they showed no variability and were excluded from inferential analyses. Associations between demographic/management variables (e.g., sex, age, breed, weight, housing system, and discipline) and clinical indicators (jump sign, restricted ROM, ROM score, and pain score) were evaluated using chi-square tests. Fisher’s exact test was applied whenever expected cell counts fell below 5. Ordinal–ordinal relationships were examined using Spearman’s rank correlation coefficients. Behavioural domains reported by owners (box, saddling, girthing, riding) and the composite variable global behaviour (≥1 behavioural alteration) were analysed using these same methods. Cross-domain associations between owner-reported behaviours and veterinary findings (jump sign, restricted ROM, ROM score, and pain score) were tested using chi-square and Spearman correlations depending on the variable type. Given the ordinal nature of several clinical variables (pain score, ROM score) and the non-normal distribution observed upon graphical and descriptive inspection, additional non-parametric analyses were conducted to complement the categorical and correlational results. Differences in median pain and ROM scores across demographic and management categories (sex, age category, breed, weight category, housing system, and discipline) were evaluated using the Kruskal–Wallis test for factors with more than two levels and Mann–Whitney U tests for binary comparisons. Classical regression modelling was explored but proved only partially feasible. Logistic regression models using jump sign or global behaviour as binary outcomes were attempted, but the extreme imbalance of several predictors (e.g., 100% prevalence variables, highly skewed pain categories) resulted in unstable estimates and poor model convergence. Generalized linear mixed‐effects models (GLMMs) were tested using discipline as a random effect; however, models produced singular fits due to the small number of grouping levels (n=4) and limited within-group variability. Horse ID was not considered an appropriate random factor, as each horse contributed a single observation. Consequently, no mixed-effects model provided a reliable solution. The significance threshold was set at p < 0.05 for all classical statistical tests. Due to the limitations of classical modelling, supervised ML regression was performed to evaluate whether nonlinear or high-dimensional patterns could predict the global pain score. Three algorithms were used: Boosting Regression, Random Forest Regression, and Support Vector Machine (SVM) Regression. All models were trained using identical predictor sets (behavioural, clinical, and demographic variables) and the default training parameters in JASP. A 70/30 train–test split was used for model evaluation. Performance metrics included test mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and coefficient of determination (R²). ML analyses do not produce p-values and were not interpreted inferentially. 3. Results 3.1. Demographic profile of the equine population The study population consisted of 120 equines, including 118 horses and two ponies. The population included 48 mares (40%), 62 geldings (51.7%) and 10 stallions (8.3%), with a median (min-max) age of 11.6 (4-24) years and a median (min-max) body weight of 476.8 (194-648) kg. The breeds represented were 32 warmbloods (26.7%), 39 Iberians (32.5%), 9 Arabians (7.5%), 6 other pure breeds (5%), and 34 crossbreeds (28.3%). Housing conditions were 27 outdoors (22.5%) and 93 in stables (77.5%). Regarding their use, 32 were used for showjumping (26.7%), 44 for dressage (36.7%), 8 for endurance (6.7%), and 36 were leisure animals (30%). 3.2. Owner questionnaire–based behavioural findings According to the owners’ responses to the four questions, 9/120 horses (7.5%) exhibited box behaviours, 25/120 (20.8%) showed saddling behaviours, 32/120 (26.7%) displayed girthing behaviours, and 22/120 (18.3%) presented riding behaviours. Regarding the owners’ perception of saddling behaviours, among the 25 affected horses, 16/25 (64%) always displayed these behaviours, 5/25 (20%) exhibited them often, and 4/25 (16%) showed them sometimes. Of these 25 horses, 6 (24%) showed two distinct behavioural problems, while the remaining 19 (76%) exhibited only a single conduct alteration. About the owners’ perception of girthing behaviours, among the 32 affected horses, 62.5% (20/32) always displayed these behaviours, 9.4% (3/32) exhibited them often, 15.6% (5/32) showed them sometimes, 9.4% (3/32) showed them rarely, and 1/32 (3.1%) had no report for this behavioural perception. Of these 32 horses, 18.7% (6/32) showed two distinct behavioural problems, while 81.3% (26/32) exhibited only a single conduct alteration. About the owners’ perception of riding behaviours, among the 22 affected horses, 17/22 (77.3%) always displayed these behaviours, 3/22 (13.6%) exhibited them often, and 2/22 (9.1%) showed them sometimes. Of these 22 horses, 1/22 (4.5%) showed three behavioural alterations, 2/22 (9.1%) exhibited two different behavioural alterations, while the remaining 19/22 (86.4%) exhibited only a single conduct alteration. Beyond individual behavioural categories, a combined analysis of all four questions revealed that 58 out of 120 horses (48.3%) exhibited at least one behavioural alteration. Co-occurrence patterns indicated that these behaviours were not mutually exclusive (Table 1). Most horses with behavioural abnormalities showed only a single altered behaviour (35/120; 29.2%), whereas 17 horses (14.2%) presented two concurrent behavioural problems, and four horses (3.3%) showed alterations in three behavioural domains. Only two horses (1.7%) displayed behavioural abnormalities in all four categories simultaneously. Girthing, saddling, and riding behaviours accounted for the majority of isolated or combined alterations, while box-related behaviours were relatively infrequent either alone or in combination with equipment-related behaviours. Although nearly half of the horses exhibited at least one behavioural alteration, most demographic and management variables - including sex, breed, housing system, and discipline - were not significantly related to the presence of behavioural issues (Table 2). Age was the only factor significantly associated with the occurrence of behavioural alterations (p = 0.028), with middle-aged horses more frequently displaying abnormal behaviours than younger or older horses. Significant associations were found between the presence of each behavioural category and both the number and frequency of behavioural indicators reported by the owners. Horses exhibiting a given behavioural problem -particularly saddling, girthing, or riding behaviours-tended to display a higher number of behavioural signs and showed these signs more frequently. For the combined behavioural variable (global behaviour), all associations with description and frequency measures were highly significant (p < 0.001). Similar patterns were observed within each behavioural category (Table 3). 3.3. Manual assessment of myofascial pain indicators The clinical evaluation revealed a high prevalence of the key indicators traditionally associated with MPS. Hypersensitive spots and taut bands were identified in all horses (100%). Jump sign was observed in 86 of 120 horses (71.7%), while restricted thoracolumbar ROM was present in 83 horses (69.2%). LTRs were not observed in any horse. Global pain scores ranged from 2 to 5, with a median of 4 (IQR 3-4). Most horses fell within the higher pain categories, including 61 horses scoring 4 and 27 horses scoring 5. Because hypersensitive spots, taut bands, and LTRs showed no variation across the sample, no statistical comparisons were conducted for these variables. For the jump sign, no significant associations with sex (p = 0.569), age (p = 0.417), breed (p = 0.464), weight (p = 0.555), housing (p = 0.104), or discipline (p = 0.521) were found. No significant association between restricted ROM with sex (p = 0.151), age (p = 0.340), breed (p = 0.582), weight (p = 0.165), housing (p = 0.271), or discipline (p = 0.084) were neither observed. Pain score showed a significant association with weight (p = 0.031), a marginal association with sex (p = 0.049), and no significant associations with age (p = 0.212), breed (p = 0.054), housing (p = 0.113), or discipline (p = 0.580). The correlation analysis revealed that most associations between the demographic and management variables (sex, age, breed, weight, housing, and discipline) with clinical indicators of pain and mobility were weak and clinically negligible, suggesting that these classificatory factors did not consistently influence the expression of MPS findings. In contrast, two strong and significant correlations were identified among the clinical variables: a very high association between the jump sign and pain score (ρ = 0.814, p < 0.001), and between restricted ROM and the ROM score (ρ = 0.749, p < 0.001). These findings demonstrate strong internal consistency among the clinical assessment methods and support the validity of these indicators for detecting thoracolumbar myofascial involvement in horses. 3.4. Cross-domain associations between owner-reported behaviours and veterinary findings Only one significant cross-domain association was identified between owner-reported behaviours and clinical findings. Pain score showed a weak but significant positive correlation with the composite owner variable global behaviour (ρ = 0.218, p = 0.017), indicating a weak relationship between owner-reported behavioural alterations and pain detected on palpation. No significant correlations between owner-reported behaviours and jump sign, restricted ROM, or ROM score were found. 3.5. Non-parametric differences in pain and mobility outcomes Non-parametric analyses were performed to further examine differences in pain score and ROM score across clinical, demographic, and behavioural categories. Horses exhibiting a jump sign showed significantly higher pain scores than those without this sign (p < 0.001, Figure 2), whereas no significant differences were detected for ROM score between these groups. Sex and breed categories did not show statistically differences in either pain score or ROM score, and the same pattern was observed across the different disciplines. In contrast, housing conditions showed a significant effect on pain score, sine stabled horses had higher pain scores than outdoor-kept horses (p = 0.044, Figure 3). ROM score also differed between housing categories, showing greater thoracolumbar restriction among stabled horses (p = 0.013, Figure 4). Restricted ROM, when used as a grouping variable, performed as anticipated: while pain scores showed no significant differences between horses with or without restricted motion, the ROM score itself exhibited a highly significant separation between groups, confirming the internal consistency and discriminative validity of the ROM-based classification (p < 0.001, Figure 5). Finally, the presence of at least one owner-reported behavioural alteration (global behaviour) was associated with higher median pain scores (p = 0.018, Figure 6), while ROM score did not differ between behaviour-positive and behaviour-negative horses. 3.6. Predictive modelling by machine learning Three supervised ML algorithms-Boosting Regression, Random Forest Regression, and Support Vector Machine (SVM) Regression-were applied to evaluate whether multivariate combinations of veterinary, behavioural, and demographic variables could predict the global pain score. All models were trained using identical predictor sets and the default JASP training parameters. 3.6.1. Boosting Regression The Boosting Regression model showed a good level of predictive performance, with a test MSE of 0.251, RMSE of 0.501, MAE of 0.378, and an R² of 0.718. Observed and predicted pain scores demonstrated a clear positive correspondence (Figure 7). Variable importance revealed a highly asymmetric contribution pattern, with jump sign accounting for 90.5% of the total predictive influence. All remaining variables - including global behaviour, discipline, sex, age, breed, body weight, housing, ROM score, and restricted ROM - showed minimal contributions (<6% each; Figure 8). 3.6.2. Random Forest Regression Random Forest achieved the best overall performance, yielding a test MSE of 0.138, RMSE of 0.372, MAE of 0.266, MAPE of 7.97%, and an R² of 0.754. Predicted values closely mirrored the observed clinical pain scores (Figure 9). Across all variable-importance metrics (mean decrease in accuracy, node purity, and dropout loss), jump sign again emerged as the dominant predictor, substantially surpassing behavioural, demographic, and ROM-related variables, which all showed marginal, or near-zero influence. 3.6.3. Support Vector Machine (SVM) Regression The SVM model showed comparatively lower predictive accuracy (test MSE = 0.271, RMSE = 0.520, MAE = 0.313, R² = 0.678). Feature-contribution diagnostics confirmed a pattern consistent with the ensemble models: jump sign exerted the strongest upward shift on predicted pain scores, while global behaviour, ROM variables, and demographic factors contributed only small deviations from the baseline prediction. 4. Discussion This study was performed given the underdiagnosis of equine thoracolumbar myofascial pain, largely due to the lack of established clinical diagnostic criteria and limited research. It aimed to investigate which clinical signs of MPS can be identified in the equine thoracolumbar region, their prevalence and internal coherence, including the relationship between pain on palpation and ROM, and the diagnostic relevance of behavioural, demographic and clinical variables integrating innovative statistical machine learning approaches. 4.1. Clinical findings of MPS in horses The presence of taut bands and hypersensitive spots in all horses aligns with results from human studies focused on establishing a consensus on diagnostic criteria. These signs were included as essential criteria for the diagnosis of MPS in humans, supporting their relevance in the present study. 22,29,47 Spot tenderness, manifested as a hypersensitive spot, taut band, or tender spot in a taut band, was reported in 96.9% of selected clinical studies, suggesting that it is the most common diagnostic criterion for TrPs in humans. 22 In our study, LTR were not detected in any horse. In humans, some studies do not consider it essential but rather confirmatory due to the poor reliability. 26,29,36 However, a recent update has listed LTR among the three most common criteria used in MPS clinical trials, 22 in agreement with previous findings. 24 A moderate to excellent reliability combining a taut band, a hypersensitive spot, LTR, and referred pain for identifying TrPs in epicondylalgia and ankle muscles has been reported. 48,49 In humans, the pressure needed to elicit a LTR varies by muscle. 29 In the equine back, flat-hand palpation and the depth of the musculature limit the observation of LTRs, whereas in the equine brachiocephalicus muscle, they are more easily elicited and detected, 39 as this muscle is more superficial and can be palpated using a pincer technique. Interestingly, human studies have reported that LTRs in the upper-trapezius muscles can be detected through both ultrasonography and visual inspection, whereas in lower-back muscles many LTRs are only observed with ultrasonography during TrP injection. 50 Based on the clinical experience of the author, LTRs can be observed in equine back muscles during dry needling. For all these reasons, the absence of palpable LTRs in our study does not necessarily indicate their true absence in equine back muscles, suggesting that while LTRs may not be essential for palpation, they could serve as a confirmatory criterion after treatment. The jump sign and its correlation with the degree of pain are of great importance in this study. This contrasts with data reported in humans, where it is not considered an essential diagnostic criterion: in a Delphi study, only 6.5% of specialists agreed on including it, and in another review, it was reported as common in only 7.8% of trials (10 out 129 studies). 22,29 This discrepancy may be explained by the expression of pain in horses primarily through facial and behavioural signs, given their non-verbal nature, with the jump sign representing a more common behavioural reaction to pain. This aligns with findings form previous studies on musculoskeletal pain in horses, which report that the combination of facial expressions and body behaviours is a strong predictor of pain. 51–53 Moreover, in horses, the jump sign and facial and behavioural reactions to pain were considered in the palpation assessment of TrPs in the brachiocephalicus and pectoral muscles. 39,40 Restricted ROM was the least frequently observed sign. This is consistent with findings in humans, where it is not considered as an essential diagnostic criterion: only 6.5% of specialists agreed on including it, and it was reported as common in 22.5% of cases (out of 129 studies). 22,29 Human literature considers restricted ROM a non-specific diagnostic criterion for MPS, as its association with TrPs has been reported in some muscles but not in others. 22 In the present study, the absence of a significant relationship between pain and ROM scores is consistent with previous findings reported in human myofascial cervical pain. It has been suggested that ROM may decrease only at high levels of pain; however, this hypothesis has not been consistently supported. 54 Based on current literature, no studies that specifically evaluate the relationship between myofascial pain intensity and thoracolumbar ROM have been published. Interestingly, an increase in thoracolumbar mobility was observed in some painful horses in the present study, which aligns with previous descriptions of hypermobility in acutely sensitive and painful horses. 55 A potential explanation for this finding is that ROM depends on the anatomical position of the joint, as recent studies in the equine thoracolumbar spine have demonstrated, 56 and may arise not only from TrPs but also from pathological collagen cross-linking and fibrous fascial adhesions, as observed in humans. 57 Additionally, it is suggested that posture and individual expressions of pain may also influence the correlation between ROM and pain scores. Prolonged inactivity, associated with muscle stiffness and reduced flexibility in humans, increases the risk for MPS, 58 consistent with the present study, in which stabling increased pain and reduced ROM. 4.2. Owner-reported behaviours and veterinary assessment In the current research, a modest but significant association between owner-reported behaviours and clinical pain findings was observed, suggesting that owners were generally unable to detect the specific clinical manifestations of myofascial pain identified during the veterinary examination. This is consistent with previous findings indicating that most horse owners are unable to detect abnormal behaviours signs, particularly during tacking-up or mounting, 30 and behavioural expression is frequently subtle and potentially missed or misinterpreted. 59 These results emphasize the importance of professional training to enhance owners’ ability to recognize pain-related behaviours. Although demographic and management factors did not appear to contribute meaningfully to the expression of the behaviours evaluated in this study, behavioural alterations may be influenced by age-related factors. According to the owner questionnaire results, middle-aged horses were more frequently affected, consistent with human reports of higher MPS prevalence at 27-50 years. 60 Muscle overuse is a recognised risk factor for MPS in humans, 61 suggesting that improper training over time may increase susceptibility to developing TrPs in middle-aged horses. 4.3. Predictors of MPS diagnosis in horses Machine learning comprises a set of computational methods capable of identifying complex, non-linear patterns within datasets that may elude traditional statistical approaches. Its use has expanded rapidly in human and veterinary medicine, particularly in diagnostic fields where multifactorial conditions require integrative analysis of behavioural, demographic, and clinical variables. In musculoskeletal medicine, ML has been increasingly applied to improve the detection of subtle pain-related features, predict injury risk, classify movement abnormalities, and refine diagnostic decision-making by weighting the relative importance of clinical signs. 62–68 We incorporated ML into the current study to complement classical analyses and to explore whether multidimensional modelling could reveal hidden relationships among palpation findings, behavioural indicators, and horse-level factors. This approach is especially valuable for emerging conditions such as equine MPS, where diagnostic criteria are not yet standardised and clinical signs may present with overlapping or heterogeneous patterns. ML therefore offers an innovative framework to assess the predictive contribution of each variable and to validate the robustness of proposed diagnostic indicators beyond conventional statistical constraints. The integration of ML in this study aimed to determine whether complex multivariate patterns could improve the prediction of clinical pain beyond the capacity of classical statistics. Across all algorithms-Random Forest, Boosting Regression, and SVM Regression-the jump sign emerged overwhelmingly as the main determinant of pain score, contributing more than 90% of the total variable importance. This result is clinically meaningful: it indicates that, even when considering multiple behavioural, demographic or mobility-related variables, the horse’s immediate behavioural reaction to palpation remains the dominant marker of thoracolumbar MPS. Interestingly, ROM-related variables, owner-reported behaviors, and demographic factors contributed minimally to model performance (<6% in all), reinforcing the limited diagnostic value of these indicators and supporting the weak correlations observed in classical analyses. ML approaches demonstrated good predictive performance (R² up to 0.75 in Random Forest), confirming that pain expression in equine MPS follows a relatively stable and identifiable clinical pattern. Importantly, the inability of more complex features to improve prediction beyond palpation findings suggests that equine MPS may be less influenced by external factors than previously assumed, with clinical signs offering the highest diagnostic yield. Taken together, ML results provide convergent validity for the diagnostic criteria proposed here and highlight the jump sign as a core indicator whose predictive strength persists even in high-dimensional modelling environments. 4.4. Strengths and limitations The main strengths of this study include its conduction under real clinical conditions with an experienced examiner, which enhances reliability as reported in humans, 69 and the use of a substantial population of horses, increasing the applicability and robustness of the findings. However, the main limitation was the absence of additional assessors, which would have allowed for a potential inter-rater reliability assessment of MPS diagnosis. Although some studies have reported good intra-rater reliability in equine spinal palpation, 70 the inclusion of additional assessors which, would have strengthened the findings, as previously reported. 42 Additional limitations include that this study did not analyse which types of behavioural signs were most frequently observed, as this was considered outside the scope of the present study. Therefore, future research should conduct an international Delphi study among experts to work toward standardised diagnostic criteria, providing a more detailed description of the behavioural signs most frequently associated with MPS and examining the relationship between MPS diagnostic criteria and ridden work. 5. Conclusion This study establishes, for the first time, clinical diagnostic criteria for MPS in the equine thoracolumbar region. The presence of a taut band, a hypersensitive spot, and the jump sign were identified as essential diagnostic indicators. Additionally, owners exhibited limited capacity to recognize pain-related behaviours during routine management, and stabling conditions were identified as a risk factor for MPS. These findings reinforce the fundamental role of veterinarians not only in the accurate diagnosis of this condition but also in educating owners to improve early detection and management. Communicating this knowledge within the veterinary community is crucial to ensure that MPS is consistently incorporated into differential diagnosis processes. ML-based modelling can complement traditional clinical assessments and may support future development of objective, standardised diagnostic criteria for equine MPS. Funding information This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Acknowledgments The authors thank the owners of the horses and the equestrian centres involved in this study. This work is part of a doctoral thesis in the Biosciencies and Agroalimentary Sciences Program at the {masked for review}, Spain. Conflict of interest statement The authors have declared no conflicting interests. Author contributions {masked for review}: Conceptualization; methodology; formal analysis; investigation; resources; writing-original draft; writing-review and editing; visualization; project administration. {masked for review} : Formal analysis; writing-original draft; writing-review and editing; visualization; supervision. {masked for review} : Writing-review and editing; visualization; supervision. {masked for review} : Formal analysis; writing-original draft, writing-review and editing; visualization; supervision; p roject administration. All authors have read and agreed to the published version of the manuscript. ††journal: The Astrophysical Journal \crefname sectionSectionSections \crefnameequationEq.Eqs. \crefnamefigureFig.Figs. \crefnametableTableTables Data integrity statement {masked for review} and {masked for review} had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of data analysis. Ethical animal research No specific ethical approval was required, as per Spanish animal welfare policies, since all horses were presented for veterinary examination as part of routine professional procedures (Real Decreto 53/2013, of February 1 st , which establishes the basic standards for the protection of animals used in experimentation and other scientific purposes, including teaching . Official publication: Boletín Oficial del Estado (B.O.E); number: 34/2013; publication date: 8 February 2013; page number: 11370-11421). All interventions were performed in accordance with the Code of Good Veterinary Practice 71 . Informed consent All owners provided informed consent for their horses’ data to be used in the study. Table 1. Co-occurrence of behavioural alterations across the four owner-assessed categories ( n = 120 horses). No behaviours 0 0 0 0 62 51.7% Only riding 0 0 0 1 12 10.0% Only girthing 0 0 1 0 13 10.8% Only saddling 0 1 0 0 8 6.7% Only box 1 0 0 0 2 1.7% Girthing + riding 0 0 1 1 3 2.5% Saddling + riding 0 1 0 1 3 2.5% Saddling + girthing 0 1 1 0 8 6.7% Box + girthing 1 0 1 0 3 2.5% Box + saddling 1 1 0 0 1 0.8% Saddling + girthing + riding 0 1 1 1 2 1.7% Box + saddling + girthing 1 1 1 0 1 0.8% All four behaviours 1 1 1 1 2 1.7% Table 2. Association between individual behavioural variables, the combined behaviour variable, and demographic/management factors. ††journal: The Astrophysical Journal \crefname sectionSectionSections \crefnameequationEq.Eqs. \crefnamefigureFig.Figs. \crefnametableTableTables Box behaviour 0.473 0.844 0.702 0.395 0.078 Saddling behaviour 0.274 0.422 0.137 0.158 0.101 Girthing behaviour 0.093 0.121 0.564 0.692 0.350 Riding behaviour 0.512 0.318 0.949 0.591 0.176 Global behaviour 0.359 0.028 0.713 0.646 0.254 Table 3. Chi-square associations between behavioural categories and the number and frequency of behavioural indicators. Box behaviour ns ns 0.019 0.014 ns ns Saddling behaviour < 0.001 < 0.001 < 0.001 0.004 ns ns Girthing behaviour < 0.001 0.014 < 0.001 < 0.001 ns ns Riding behaviour ns 0.025 ns ns < 0.001 < 0.001 Global behaviour < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 Figure legends Figure 1. Pain-related behaviours observed during clinical assessment of the thoracolumbar region (attempt to bite, ears flattened, tense muscles above the eyes and tail swishing). Figure 2. Comparison of pain scores between horses with and without a jump sign. Raincloud plot illustrating the distribution of global pain scores according to the presence or absence of a jump sign. Horses displaying a jump sign showed markedly higher pain scores. 0=absent; 1=present. a−b = Groups with different lowercase letters denote significant differences (p < 0.05) by the Mann–Whitney U test. Figure 3. Pain score distribution according to housing conditions. Raincloud plot comparing pain scores between stabled horses and horses kept outdoors. Stabled animals demonstrated visibly higher pain values. 1=box; 2=outdoor. a−b = Groups with different lowercase letters denote significant differences (p < 0.05) by the Mann–Whitney U test. Figure 4. Range of motion (ROM) score differences between housing categories. Raincloud plot showing cervical range-of-motion restriction in stabled versus outdoor-kept horses. Stabled horses exhibited greater ROM limitation. 1=box; 2=outdoor. a−b = Groups with different lowercase letters denote significant differences (p < 0.05) by the Mann–Whitney U test. Figure 5. ROM score distribution in horses with and without restricted ROM. Raincloud lot demonstrating the expected divergence in ROM scores when grouped by the clinician-assigned restricted ROM classification. 0=absent; 1=present. a−b = Groups with different lowercase letters denote significant differences (p < 0.05) by the Mann–Whitney U test. Figure 6. Pain score comparison based on the presence of owner-reported behavioural alterations. Raincloud plot illustrating pain score distributions for horses with versus without at least one reported behavioural abnormality. Horses with behavioural issues showed higher pain levels. 0=absent; 1=present. a−b = Groups with different lowercase letters denote significant differences (p < 0.05) by the Mann–Whitney U test. Figure 7. Boosting Regression model: Predicted vs. observed pain scores. Scatterplot showing the relationship between observed veterinary pain scores and those predicted by the Boosting Regression model. The diagonal line represents the line of perfect agreement. The model demonstrates moderate predictive accuracy, with most points clustering near the line (test RMSE = 0.501; R² = 0.718). Figure 8. Variable importance in the Boosting Regression model. Relative influence of each predictor variable included in the Boosting model. Jump sign overwhelmingly dominated model importance (90.5%), whereas all other variables (global behaviour, discipline, sex, age, breed, body weight, housing, ROM score, and restricted ROM) contributed minimally (<6% each). Figure 9. Random Forest model: Predicted vs. observed pain scores. Scatterplot showing the Random Forest model’s predictive performance. Observed and predicted values show a strong linear relationship, confirming high model accuracy (test RMSE = 0.372; R² = 0.754). Random Forest was the best-performing model across all machine-learning approaches evaluated. List of legends for Supplementary items Data S1. Owner Questionnaire Table S1. Palpation scoring scale used to evaluate the thoracolumbar region. Table S2. Mobility scoring scale used to evaluate the thoracolumbar range of motion (ROM). References 1. Boado A, Pollard D, Dyson S. A Retrospective Study of the Evolution of Orthopaedic Injuries in 70 Dressage Horses. Animals . 2025;15(12). doi:10.3390/ani15121740 2. Putnam JRC, Holmes LM, Green MJ, Freeman SL. Incidence, causes and outcomes of lameness cases in a working military horse population: A field study. Equine Vet J . 2014;46(2):194-197. doi:10.1111/evj.12084 3. Haussler KK. Chiropractic evaluation and management of musculoskeletal disorders. In: Ross MW, Dyson SJ, eds. Diagnosis and Management of Lameness in the Horse . First edition. Elsevier; 2003:803-811. 4. Riccio B, Fraschetto C, Villanueva J, Cantatore F, Bertuglia A. Two multicenter surveys on equine back-pain 10 years a part. Front Vet Sci . 2018;5(AUG). doi:10.3389/fvets.2018.00195 5. Marshall-Gibson ME, Durham MG, Seabaugh KA, Moorman VJ, Ferris DJ. Survey of equine veterinarians regarding primary equine back pain in the United States. Front Vet Sci . 2023;10. doi:10.3389/fvets.2023.1224605 6. Landman MAAM, de Blaauw JA, van Weeren PR, Hofland LJ. Field study of the prevalence of lameness in horses with back problems. Vet Rec . 2004;155(6):165-168. doi:10.1136/vr.155.6.165 7. Dyson S. Evaluation of poor performance in competition horses: A musculoskeletal perspective. Part 1: Clinical assessment. Equine Vet Educ . 2016;28(5):284-293. doi:10.1111/eve.12426 8. Harrison LM, Sole-Guitart A, Ahern B, Goff LM. Functional anatomy of the equine thoracolumbar spine related to equine back rehabilitation. Journal of Equine Rehabilitation . 2025;3:100027. doi:10.1016/j.eqre.2025.100027 9. Mayaki AM, Intan-Shameha AR, Noraniza MA, Mazlina M, Adamu L, Abdullah R. Clinical investigation of back disorders in horses: A retrospective study (2002-2017). Vet World . 2019;12(3):377-381. doi:10.14202/vetworld.2019.377-381 10. Domańska-Kruppa N, Wierzbicka M, Stefanik E. Advances in the Clinical Diagnostics to Equine Back Pain: A Review of Imaging and Functional Modalities. Animals (Basel) . 2024;14(5). doi:10.3390/ani14050698 11. Donnelly JM., Simons DG. Travell, Simons & Simons’ Myofascial Pain and Dysfunction : The Trigger Point Manual . Wolters Kluwer Health; 2019. 12. Rivers WE, Garrigues D, Graciosa J, Harden RN. Signs and Symptoms of Myofascial Pain: An International Survey of Pain Management Providers and Proposed Preliminary Set of Diagnostic Criteria. Pain Med . 2015;16(9):1794-1805. doi:10.1111/pme.12780 13. Monclús-Díez G, Díaz-Arribas MJ, Fernández-de-las-Peñas C, et al. Prevalence of Myofascial Trigger Points in Patients with Radiating and Non-Radiating Low Back Pain: A Systematic Review. Biomedicines . 2025;13(6):1453. doi:10.3390/biomedicines13061453 14. Gerwin RD. Classification, epidemiology, and natural history of myofascial pain syndrome. Curr Pain Headache Rep . 2001;5(5):412-420. doi:10.1007/s11916-001-0052-8 15. Lam C, Francio VT, Gustafson K, Carroll M, York A, Chadwick AL. Myofascial pain - A major player in musculoskeletal pain. Best Pract Res Clin Rheumatol . 2024;38(1):101944. doi:10.1016/j.berh.2024.101944 16. Ridgway K. Acupuncture as a treatment modality for back problems. Vet Clin North Am Equine Pract . 1999;15(1):211-221. doi:10.1016/S0749-0739(17)30173-6 17. Haussler KevinK. Efficacy of Manual Therapies in the Treatment of Thoracolumbar Dysfunction. In: AAEP Annual Convention . 2011:141-152. Accessed October 19, 2025. Downloaded from https://cabidigitallibrary.org by 88.10.51.151 18. Ridgway KJ. Diagnosis and Treatment of Equine MusculoSkeletal Pain. The Role of the Complementary Modalities: Acupuncture and Chiropractic. In: AAEP Annual Convention . 2005:403-408. Accessed October 19, 2025. Downloaded from https://cabidigitallibrary.org by 88.10.51.151 19. Rogers P. Acupuncture for Equine Paraspinal Myofascial Pain. American Journal of Traditional Chinese Veterinary Medicine . 2012;7(1):69-75. 20. Portier K, Schiesari C, Gauthier L, Yeng LT, Tabacchi Fantoni D, Formenton MR. Comparison of the Prevalence and Location of Trigger Points in Dressage and Show-Jumping Horses. Animals . 2025;15(11). doi:10.3390/ani15111558 21. Simons DG., Travell JG., Simons LS. Travell & Simons’ Myofascial Pain and Dysfunction : The Trigger Point Manual . Williams & Wilkins; 1999. 22. Li L, Stoop R, Clijsen R, et al. Criteria Used for the Diagnosis of Myofascial Trigger Points in Clinical Trials on Physical Therapy: Updated Systematic Review. Clin J Pain . 2020;36(12):955-967. doi:10.1097/AJP.0000000000000875 23. Mayoral del Moral O, Torres Lacomba M, Russell IJ, Sánchez Méndez Ó, Sánchez Sánchez B. Validity and reliability of clinical examination in the diagnosis of myofascial pain syndrome and myofascial trigger points in upper quarter muscles. Pain Medicine (United States) . 2018;19(10):2039-2050. doi:10.1093/pm/pnx315 24. Tough EA, White AR, Richards S, Campbell J. Variability of criteria used to diagnose myofascial trigger point pain syndrome–evidence from a review of the literature. Clin J Pain . 2007;23(3):278-286. doi:10.1097/AJP.0b013e31802fda7c 25. Myburgh C, Larsen AH, Hartvigsen J. A systematic, critical review of manual palpation for identifying myofascial trigger points: evidence and clinical significance. Arch Phys Med Rehabil . 2008;89(6):1169-1176. doi:10.1016/j.apmr.2007.12.033 26. Lucas N, Macaskill P, Irwig L, Moran R, Bogduk N. Reliability of physical examination for diagnosis of myofascial trigger points: a systematic review of the literature. Clin J Pain . 2009;25(1):80-89. doi:10.1097/AJP.0b013e31817e13b6 27. Rathbone ATL, Grosman-Rimon L, Kumbhare DA. Interrater Agreement of Manual Palpation for Identification of Myofascial Trigger Points: A Systematic Review and Meta-Analysis. Clin J Pain . 2017;33(8):715-729. doi:10.1097/AJP.0000000000000459 28. McEvoy J, Huijbregts PA. Reliability of myofascial trigger point palpation: A systematic review. In: Dommerholt J, Huijbregts PA, eds. Myofascial Trigger Points: Pathophysiology and Evidence-Informed Diagnosis and Management . Jones and Bartlett Publishers; 2011:65-88. 29. Fernández-de-las-Peñas C, Dommerholt J. International consensus on diagnostic criteria and clinical considerations of myofascial trigger points: A delphi study. Pain Medicine (United States) . 2018;19(1):142-150. doi:10.1093/pm/pnx207 30. Dyson S, Bondi A, Routh J, et al. Do owners recognise abnormal equine behaviour when tacking-up and mounting? A comparison between responses to a questionnaire and real-time observations. Equine Vet Educ . 2022;34(9):e375-e384. doi:10.1111/eve.13471 31. Martin BB, Klide AM. Physical examination of horses with back pain. Vet Clin North Am Equine Pract . 1999;15(1). doi:10.1016/S0749-0739(17)30163-3 32. Munroe GA. Equine Neck and Back Pathology: Diagnosis and Tre Atment . Second. (M.D. Henson F, ed.). John Wiley & Sons, Ltd; 2018. doi:10.1002/9781118974520 33. Goff L. Physiotherapy Assessment for the Equine Athlete. Veterinary Clinics of North America: Equine Practice . 2016;32(1):31-47. doi:10.1016/j.cveq.2015.12.002 34. Al-Shenqiti AM, Oldham JA. Test-retest reliability of myofascial trigger point detection in patients with rotator cuff tendonitis. Clin Rehabil . 2005;19(5):482-487. doi:10.1191/0269215505cr791oa 35. Bron C, Franssen J, Wensing M, Oostendorp RAB. Interrater reliability of palpation of myofascial trigger points in three shoulder muscles. J Man Manip Ther . 2007;15(4):203-215. doi:10.1179/106698107790819477 36. Gerwin RD, Shannon S, Hong CZ, Hubbard D, Gevirtz R. Interrater reliability in myofascial trigger point examination. Pain . 1997;69(1-2):65-73. doi:10.1016/s0304-3959(96)03248-4 37. Gerwin RD. Diagnosis of myofascial pain syndrome. Phys Med Rehabil Clin N Am . 2014;25(2):341-355. doi:10.1016/j.pmr.2014.01.011 38. Barbero M, Schneebeli A, Koetsier E, Maino P. Myofascial pain syndrome and trigger points: evaluation and treatment in patients with musculoskeletal pain. Curr Opin Support Palliat Care . 2019;13(3):270-276. doi:10.1097/SPC.0000000000000445 39. Macgregor J, Graf von Schweinitz D. Needle electromyographic activity of myofascial trigger points and control sites in equine cleidobrachialis muscle–an observational study. Acupuncture in medicine : journal of the British Medical Acupuncture Society . 2006;24(2):61-70. doi:10.1136/aim.24.2.61 40. Bowen AG, Goff LM, McGowan CM. Investigation of Myofascial Trigger Points in Equine Pectoral Muscles and Girth-Aversion Behavior. J Equine Vet Sci . 2017;48:154-160.e1. doi:10.1016/j.jevs.2016.04.095 41. Kothari SF, Kothari M, Zambra RF, Baad-Hansen L, Svensson P. Standardization of muscle palpation- methodological considerations. Clin J Pain . 2014;30(2):174-182. doi:10.1097/AJP.0b013e31828c893d 42. Merrifield-Jones M, Tabor G, Williams J. Inter- and Intra-Rater Reliability of Soft Tissue Palpation Scoring in the Equine Thoracic Epaxial Region. J Equine Vet Sci . 2019;83. doi:10.1016/j.jevs.2019.102812 43. Guedes JRB, Vendruscolo CP, Tokawa PKA, Carvalho AM, Johnson PJ, Faleiros RR. The Detection of Thoracolumbar Spine Injuries in Horses with Chronic Laminitis Using a Novel Clinical-Assessment Protocol and Ultrasonographic Examination. Animals . 2024;14(9). doi:10.3390/ani14091364 44. Haussler KK, Manchon PT, Donnell JR, Frisbie DD. Effects of Low-Level Laser Therapy and Chiropractic Care on Back Pain in Quarter Horses. J Equine Vet Sci . 2020;86. doi:10.1016/j.jevs.2019.102891 45. Maldonado MD, Parkinson SD, Story MR, Haussler KK. The Effect of Chiropractic Treatment on Limb Lameness and Concurrent Axial Skeleton Pain and Dysfunction in Horses. Animals . 2022;12(20). doi:10.3390/ani12202845 46. De Heus P, Van Oossanen G, Van Dierendonck MC, Back W. A Pressure Algometer Is a Useful Tool to Objectively Monitor the Effect of Diagnostic Palpation by a Physiotherapist in Warmblood Horses. J Equine Vet Sci . 2010;30(6):310-321. doi:10.1016/j.jevs.2010.04.010 47. Rivers WE, Garrigues D, Graciosa J, Harden RN. Signs and Symptoms of Myofascial Pain: An International Survey of Pain Management Providers and Proposed Preliminary Set of Diagnostic Criteria. Pain Medicine . 2015;16(9):1794-1805. doi:10.1111/pme.12780 48. Sanz DR, Lobo CC, López DL, Morales CR, Marín CS, Corbalán IS. Interrater Reliability in the Clinical Evaluation of Myofascial Trigger Points in Three Ankle Muscles. J Manipulative Physiol Ther . 2016;39(9):623-634. doi:10.1016/j.jmpt.2016.09.002 49. Mora-Relucio R, Núñez-Nagy S, Gallego-Izquierdo T, et al. Experienced versus Inexperienced Interexaminer Reliability on Location and Classification of Myofascial Trigger Point Palpation to Diagnose Lateral Epicondylalgia: An Observational Cross‐Sectional Study. Evidence-Based Complementary and Alternative Medicine . 2016;2016(1). doi:10.1155/2016/6059719 50. Rha D wook, Shin JC, Kim YK, Jung JH, Kim YU, Lee SC. Detecting Local Twitch Responses of Myofascial Trigger Points in the Lower-Back Muscles Using Ultrasonography. Arch Phys Med Rehabil . 2011;92(10):1576-1580.e1. doi:10.1016/j.apmr.2011.05.005 51. Bussières G, Jacques C, Lainay O, et al. Development of a composite orthopaedic pain scale in horses. Res Vet Sci . 2008;85(2):294-306. doi:10.1016/j.rvsc.2007.10.011 52. Ask K, Rhodin M, Tamminen LM, Hernlund E, Haubro Andersen P. Identification of Body Behaviors and Facial Expressions Associated with Induced Orthopedic Pain in Four Equine Pain Scales. Animals . 2020;10(11):2155. doi:10.3390/ani10112155 53. Auer U, Kelemen Z, Vogl C, et al. Development, refinement, and validation of an equine musculoskeletal pain scale. Frontiers in Pain Research . 2024;4. doi:10.3389/fpain.2023.1292299 54. Wilke J, Niederer D, Fleckenstein J, Vogt L, Banzer W. Range of motion and cervical myofascial pain. J Bodyw Mov Ther . 2016;20(1):52-55. doi:10.1016/j.jbmt.2015.04.003 55. Tabor G. Outcome measures and their importance for effective equine back pain rehabilitation. Journal of Equine Rehabilitation . 2024;2:100016. doi:10.1016/j.eqre.2024.100016 56. Baudisch N, Schneidewind L, Becke S, et al. Computed tomographic study analysing functional biomechanics in the thoracolumbar spine of horses with and without spinal pathology. Anat Histol Embryol . 2024;53(1). doi:10.1111/ahe.13016 57. Simmonds N, Miller P, Gemmell H. A theoretical framework for the role of fascia in manual therapy. J Bodyw Mov Ther . 2012;16(1):83-93. doi:10.1016/j.jbmt.2010.08.001 58. Lalchhuanawma A. Finding the possible link between physical activities, dietary nutrients and health on myofascial pain syndrome. Curr Med Res Pract . 2021;11(4):176-180. doi:10.4103/cmrp.cmrp_41_20 59. Hall C, Randle H, Pearson G, Preshaw L, Waran N. Assessing equine emotional state. Appl Anim Behav Sci . 2018;205:183-193. doi:10.1016/j.applanim.2018.03.006 60. Lam C, Francio VT, Gustafson K, Carroll M, York A, Chadwick AL. Myofascial pain – A major player in musculoskeletal pain. Best Pract Res Clin Rheumatol . 2024;38(1):101944. doi:10.1016/j.berh.2024.101944 61. Steen JP, Jaiswal KS, Kumbhare D. Myofascial Pain Syndrome: An Update on Clinical Characteristics, Etiopathogenesis, Diagnosis, and Treatment. Muscle Nerve . 2025;71(5):889-910. doi:10.1002/mus.28377 62. CİHAN P, GÖKÇE E, KALIPSIZ O. Veterinerlik Alanında Makine Öğrenmesi Uygulamaları Üzerine Bir Derleme. Kafkas Univ Vet Fak Derg . Published online 2017. doi:10.9775/kvfd.2016.17281 63. Chada G. Machine Learning Models for Abnormality Detection in Musculoskeletal Radiographs. Reports . 2019;2(4):26. doi:10.3390/reports2040026 64. Burti S, Banzato T, Coghlan S, Wodzinski M, Bendazzoli M, Zotti A. Artificial intelligence in veterinary diagnostic imaging: Perspectives and limitations. Res Vet Sci . 2024;175:105317. doi:10.1016/j.rvsc.2024.105317 65. Xiao S, Dhand NK, Wang Z, et al. Review of applications of deep learning in veterinary diagnostics and animal health. Front Vet Sci . 2025;12. doi:10.3389/fvets.2025.1511522 66. Shariatnia MM, Bagherieh S, Semnani F, et al. Artificial Intelligence Applications in Musculoskeletal Imaging. Curr Rev Musculoskelet Med . 2025;19(1):4. doi:10.1007/s12178-025-09997-0 67. Dhumale A, Shinde S, Ambali MP, Patil P. Integration of Artificial Intelligence for Diagnostic Methods in Musculoskeletal Conditions: A Systematic Review. Cureus . Published online February 20, 2025. doi:10.7759/cureus.79391 68. Feierabend M, Wolfgart JM, Praster M, Danalache M, Migliorini F, Hofmann UK. Applications of machine learning and deep learning in musculoskeletal medicine: a narrative review. Eur J Med Res . 2025;30(1):386. doi:10.1186/s40001-025-02511-9 69. Hsieh CY, Hong CZ, Adams AH, et al. Interexaminer reliability of the palpation of trigger points in the trunk and lower limb muscles. Arch Phys Med Rehabil . 2000;81(3):258-264. doi:10.1016/s0003-9993(00)90068-6 70. Seffinger MA, Najm WI, Mishra SI, et al. Reliability of Spinal Palpation for Diagnosis of Back and Neck Pain. Spine (Phila Pa 1976) . 2004;29(19):E413-E425. doi:10.1097/01.brs.0000141178.98157.8e 71. European Veterinary Code of Conduct and Veterinary Act. Accessed November 18, 2025. https://www.fve.org/cms/wp-content/uploads/ap12_a_i_Code-of-Conduct_forGAVote.pdf Information & Authors Information Version history V1 Version 1 14 December 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Authors Affiliations María Resano-Zuazu no affiliation View all articles by this author Jorge U. Carmona Universidad de Caldas Facultad de Ciencias para la Salud View all articles by this author César Fernández-de-las-Peñas Universidad Rey Juan Carlos Departamento de Fisioterapia Terapia Ocupacional Rehabilitacion y Medicina Fisica View all articles by this author david arguelles [email protected] Universidad de Cordoba - Campus Agroalimentario Cientifico y Tecnico de Rabanales View all articles by this author Metrics & Citations Metrics Article Usage 377 views 229 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation María Resano-Zuazu, Jorge U. Carmona, César Fernández-de-las-Peñas, et al. Diagnostic Criteria for Equine Thoracolumbar Myofascial Pain Syndrome: A Foundational Study. Authorea . 14 December 2025. DOI: https://doi.org/10.22541/au.176570292.22857795/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.176570292.22857795/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a028511d3bdc09d6',t:'MTc3OTkxOTQ4Mg=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();