ADVANCING VETERINARY TRAUMA PATIENT CARE THROUGH A TRANSLATIONAL LENS

Abstract

Trauma is a leading cause of morbidity and mortality in humans and dogs. Trauma care requires rapid assessment of multi-organ dysfunction and management of sequelae that can develop after the injury. Although trauma represents an economic and years-life-lost burden, it remains underfunded compared to other diseases. In addition, the heterogeneity of traumatic injury limits the efficacy of pre-clinical models to capture the complexity of trauma. The objective of this dissertation is to leverage translational science and translational research to identify pathways (interventions and systems) for improving trauma outcomes in pets and people. This dissertation presents research that evaluates the dog as a natural animal model for human trauma and develops systems that lead to better research methodology. Each chapter, anchored by 1-2 aims, addresses critical topics in human trauma research and knowledge gaps in veterinary clinical medicine. Aim 1: Describe the predictors of massive transfusion in dogs. Hemorrhagic shock is a leading cause of death in dogs and people. Massive transfusion is uncommon in human and veterinary medicine but requires recognition of critical bleeding patterns and a plan for effective hemostatic product administration. Little is known about how to predict massive transfusion in dogs or how these dogs are resuscitated. This retrospective study from 3 university veterinary hospital studied critically bleeding dogs, defined as dogs who received ≥ 25 mL/kg of blood products for treatment of hemorrhagic shock caused by blood loss. Trauma was the most common cause of critical bleeding. Median plasma to red blood cell ratio was 0.8 (0-4) and median non-blood product resuscitation fluid to blood product ratio was 0.5 (0-3.6). Massive Transfusion (MT) was given to 47.8% of dogs. Survival rate was 40.6%. No predictors of MT were identified. Critical bleeding in dogs was associated with a wide range of resuscitation patterns and carries a guarded to poor prognosis. This research documents the importance of a dynamic definition of massive transfusion that considers size of the patient and acuity of the bleed. Aim 2: Describe the predictors of blood production administration at veterinary trauma centers. A blood transfusion is a major intervention in the trauma patient. The clinical decision to give a unit of blood requires the integration of a wide range of examination and clinicopathologic data and finding clear indicators for transfusion need is important for improving outcomes. In a retrospective study using Veterinary Committee on Trauma registry data, 30, 617 cases were examined to evaluate predictors of transfusion in dogs. Animal Trauma Triage score – a calculation on severity of injury – was the best univariate predictor (AUROC 0.82). This highlights the importance of performing regular injury severity scoring at the time of presentation. The study also assessed the performance of a machine learning model called XGBoost in predicting transfusion needs. XGBoost demonstrated the highest predictive accuracy (AUROC 0.96), outperforming multivariate logistic regression, heuristic models, and univariate predictors. Integrating advanced algorithms into clinical practice may enhance transfusion decision-making and resource management, ultimately improving outcomes for canine trauma patients. Aim 3: Describe the clinical and prognostic characteristics of dogs with traumatic brain injury and with head injury. Traumatic brain injury has the highest associated morbidity for human trauma patients, but short-term and long-term effects are largely uncharacterized in veterinary patients. In human medicine, a TBI is defined as a derangement in the Glasgow Coma Scale score. This study assessed the utility of a corollary definition of TBI in veterinary patients, using any decrease in the modified Glasgow Coma Scale score. This registry-based retrospective study described a large multicenter cohort of 9, 607 dogs with head injury and dogs with a change in MGCS as an exploratory definition of TBI. Dogs with a Modified Glasgow Coma Scale (MGCS) score of less than 18 had a significantly lower survival rate (67.7%) compared to those with an MGCS score of 18 (97.8%). Age and size were significantly correlated with MGCS <18, with younger and smaller dogs being more likely to present with lower scores. Additionally, the incidence of head injuries was markedly higher in dogs with MGCS <18 (32.8%) compared to those with MGCS = 18 (8.4%). This epidemiological study provides valuable insights into the clinical and prognostic characteristics of dogs with altered MGCS scores and head injuries. While a change in MGCS is often associated with traumatic brain injury (TBI), it is not always a reliable surrogate for TBI diagnosis and may not align consistently with the Animal Trauma Triage (ATT) neurologic score or the presence of head injury. These findings highlight the challenges of accurately recognizing TBI in canine patients and emphasize the need for improved diagnostic approaches. Aim 4: Determine the feasibility of measuring four candidate plasma markers of EGC injury (i.e., syndecan-1, hyaluronan, heparan sulfate, VE-cadherin) in canine trauma patients and describe initial data on the effects of naturally occurring trauma and treatment on those plasma markers in dogs over time. Biomarkers for endothelial injury are valuable tools to diagnose shock and track the utility of interventions. This observation study evaluated the feasibility of measuring plasma concentrations of four endothelial glycocalyx components and their correlation to patient and treatment characteristics in canine trauma patients. Dogs with traumatic injury and healthy-age matched controls had plasma samples taken over 24 hours. The biomarkers showed different temporal patterns. Hyaluronan was significantly lower at T0 versus T24 and demonstrated a significant increase over all time points. There was a significant negative correlation between heparan sulfate and lactate. VE-cadherin was significantly higher in penetrating trauma than blunt trauma at T0 and T6. Measurement of the four markers of endothelial glycocalyx damage in canine trauma patients was generally feasible, and further study may determine clinical application such as use in acute traumatic coagulopathy or to track fluid resuscitative Aim 5: Compare the plasma and bone marrow oxylipin composition before and after surgical repair in dogs with traumatic fracture and dogs with CCL rupture. Metabolomics research allows for the evaluation of a large number of compounds that contribute to the metabolic response in traumatic injury. Long bone fractures represent a major component of polytrauma in both human and veterinary patients, contributing to systemic inflammation and poor outcomes. The objective of this observational study characterized the systemic and local lipidomic responses to traumatic fracture and cranial cruciate ligament in dogs. Canine patients presenting with traumatic long bone fractures or undergoing elective tibial plateau leveling osteotomy (TPLO) were prospectively enrolled. Plasma and bone marrow samples were collected at standardized perioperative timepoints. Targeted lipidomic analysis was performed using ultra-high-performance liquid chromatography tandem mass spectrometry. Systemic plasma profiles revealed minimal oxylipin differences associated with fracture and surgical timing. In contrast, localized plasma samples from the surgical site in TPLO demonstrated substantial metabolic remodeling, with strong separation between pre- and post-surgical states. Bone marrow lipidomics showed distinct profiles between fractured and TPLO limbs, characterized by upregulation of several lipid species implicated in inflammation and tissue remodeling. These findings suggest that while systemic lipidomic changes after long bone injury are subtle, localized environments such as the marrow and periarticular plasma undergo significant remodeling following surgical intervention. The identified lipid species may represent candidate biomarkers or therapeutic targets for modulating inflammation and healing in both canine and human orthopedic trauma. Aim 6: Examine the efficacy of a modified RECOVER CPR hands on simulation training to help learners feel confident in emergent settings and work competently with their team. Non-technical skills have garnered significant research interest in human medicine. This study evaluates confidence before and after a modified RECOVER (Reassessment Campaign on Veterinary Resuscitation) advanced life support (ALS) training at a university veterinary teaching hospital A second objective was to measure competence in communication, interpersonal relationships, leadership, and debrief skills in the same setting. Simulation-based CPR training enhances both confidence and competence, with confidence levels remaining high over a 6-month period. Some aspects of competence showed decay over time, emphasizing the need for ongoing training and practice. Aim 7: Describe the differences between One Health and IPE at veterinary and medical schools within the same institution to better understand the different IPE offerings in human medical and veterinary medical programs. Interprofessional Education (IPE) and One Health are two complementary frameworks for fostering collaborative practice across disciplines. While IPE is widely implemented in human medical institutions, One Health is more prevalent in veterinary and graduate programs. However, the intersection of these frameworks is still underexplored in both academic scholarship and professional practice. This prospective, qualitative study investigates the integration of IPE and One Health within institutions belonging to the Clinical and Translational Science Award (CTSA) One Health Alliance (COHA). COHA comprises veterinary schools partnered with human medical institutions under the National Institutes of Health CTSA funding mechanism, designed to advance the understanding of diseases shared by humans and animals. Twenty-four interviews were conducted with professionals from eight different disciplines. Participants identified key barriers to implementing IPE, including limited awareness, accessibility, efficacy, and challenges in extending its application beyond the classroom. The study also revealed gaps in institutional competency and the absence of consistent, validated evaluation tools. A lack of a shared mental model for IPE and One Health was identified across medical professions, along with significant obstacles to integrating these frameworks into professional curricula. Operationalizing One Health through IPE principles offers a promising pathway for addressing systemic health challenges and fostering transdisciplinary leaders in human and animal medicine. Conclusions: The overarching goal of this dissertation was to investigate the use of dogs as a natural animal model for human trauma research, with the broader aim of contributing to the One Health/One Medicine framework and improving trauma care for both pets and people. Through retrospective and prospective studies, this research addressed key aspects of canine trauma care across seven distinct aims, including predictors of massive transfusion, blood product administration, traumatic brain injury prognosis, endothelial glycocalyx injury biomarkers, oxylipin lipidomics, CPR simulation training, and interprofessional education in veterinary and medical schools. These findings provide critical insights into the pathophysiology, clinical management, and translational potential of canine trauma research, highlighting opportunities to advance both veterinary and human trauma care. This work has laid the foundation for the next phase of developing the canine model for trauma research By fostering collaboration between human and veterinary clinician-scientists, this work strives to advance trauma research and education, laying the groundwork for innovative solutions to shared challenges in trauma care.